[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> MOL (Europe Africa) Ltd & Ors v Mark McLaren Class Representative Ltd [2022] EWCA Civ 1701 (21 December 2022) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2022/1701.html Cite as: [2022] EWCA Civ 1701, [2023] Bus LR 318, [2023] WLR(D) 3

[New search] [Printable PDF version] [Buy ICLR report: [2023] Bus LR 318] [View ICLR summary: [2023] WLR(D) 3] [Help]

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM The Competition Appeal Tribunal
Mrs Justice Falk DBE, Dr William Bishop & Eamonn Doran
[2022] CAT 18

		Royal Courts of Justice Strand, London, WC2A 2LL
		21/12/2022

Lord Justice Lewison, Lord Justice Green & Lord Justice Snowden :

A. Introduction

The appeal

1. This is the unanimous judgment of the Court. It concerns an appeal from the judgment of the Competition Appeal Tribunal ("the CAT") dated 18^th February 2022 ([2022] CAT 10) ("the Judgment") in which it made a Collective Proceedings Order ("the CPO") and rejected applications for strike out/reverse summary judgment made by the appellants. The CPO combines claims under section 47A Competition Act 1998 ("CA 1998") for damages caused by the appellants' breaches of statutory duty for infringing Article 101 of the Treaty on the Functioning of the European Union ("TFEU") and Article 53 of the Agreement on the European Economic Area ("AEEA"). These are "follow-on" claims where the existence of the breach of duty has been determined by the European Commission ("the Commission") in a prior infringement decision adopted on 21^st February 2018 in Case AT.40009 – Maritime Car Carriers ("the Decision"), following a settlement between the Commission and the appellants which involved an admission of breach. The Decision concerned the provision by the appellants of deep sea carriage of new motor vehicles (cars, trucks and high and heavy vehicles) on various routes to and from the European Economic Area ("EEA") which at the relevant time included the UK. Under section 47A CA 1998 the Commission findings are dispositive of breach.
2. The Class Representative is the respondent and cross-appellant ("the Class Representative"). It is a company incorporated under the laws of England and Wales specifically for the purposes of bringing the proposed collective proceedings on behalf of consumers who bought new motor vehicles which had been carried by the appellants during the period which the cartel operated. Its sole director and sole member is Mr Mark McLaren, who has experience of working in consumer-related roles. The appellants were defendants before the CAT and divide into two groups: (i) the 1st – 3rd and the 5th – 11th appellants ("MNW" or "the MNW appellants"); and (ii), the 4th appellant ("KK" or "the KK appellant"). The estimate of aggregate loss (before interest) is between £57m and £115m with the range due to uncertainty as to the applicable overcharge figure. The claim increases to between £71m and £143m if simple interest is included.

The evidence relied upon

3. The CPO Application was filed on 20th February 2020 by way of a collective proceedings claim form. It was accompanied by documents and evidence setting out the "methodology" which the Class Representative argued sufficed to establish the impact of the cartel upon motor manufacturers and whether and to what extent any impact upon price was passed on to the members of the proposed class of claimants, so as to justify the grant of a CPO. It was supported by a witness statement by Mr McLaren, joint industry expert evidence by Mr Andrew Goss and Mr Anthony Whitehorn ("Goss & Whitehorn"), and expert economic evidence from Mr Tom Robinson ("Robinson"). Mr Goss and Mr Whitehorn are experienced in the automotive industry. Mr Goss is currently Chairman of Vertu Motors PLC (a substantial dealership group) and has previously held senior roles at Jaguar Land Rover, Porsche and Toyota. Mr Whitehorn has held senior roles at Hyundai and Toyota. Mr Robinson is a Director in the forensic practice of BDO LLP. KK served two witness statements: (i) from Mr James Dent ("Dent"), the Retail Sales Leader at Stratstone BMW, Harrogate, which was a franchisee retailer of BMW in the UK; and (ii), from Mr Neil Cunningham ("Cunningham"), a self-employed consultant in the vehicle rental and credit hire sector offering consultancy services to private equity investors and financial companies. No factual evidence was served by the MNW appellants in response to the economic and industry expert evidence of the Class Representative. KK served a short, 14-page, expert report from Dr Majumbar of RBB Economics which set out alleged deficiencies in the proposed methodology. MNW served a short report from Dr Tosini of NERA, which was not relied on for this appeal.

The Decision

4. The Decision found that the appellants had breached Article 101 TFEU and Article 53 AEEA by participating in a single and continuous infringement consisting of the co-ordination of prices and the allocation of customers with regard to the provision of deep-sea carriage of new motor vehicles to and from the EEA. The infringement lasted from 18^th October 2006 to 6^th September 2012 ("the Cartel Period").
5. The Decision describes the activities of the cartel members as being involved to varying degrees in conduct that sought to: (i) coordinate the prices of tenders; (ii) allocate the business of customers; and (iii) reduce capacity by coordinating the scrapping of vessels. The conduct followed a "rule of respect" whereby the cartelists would not compete with the business of an incumbent carrier. In paragraphs [30] and [32] the Commission explained:

"(30) … According to that principle, shipments of new motor vehicles related to already existing businesses on certain routes for certain customers would continue to be carried by the undertaking traditionally carrying it (the incumbent).

…

(32) The parties applied the rule of respect as a guiding principle for their practices. Some carriers were considered to be incumbents concerning specific routes and/or specific customers. [confidentiality claim pending], the carriers would respect the business of the incumbent carrier, by either providing a quote above the incumbent's rates, or refraining from quoting. The conduct also covered single and general Requests for Quotations ("RFQs") (or tenders) issued by certain vehicle manufacturers. … In some cases, the carriers followed the rule of respect only in order to avoid possible conflict among themselves."

6. The modus operandi of the cartel was summarised in paragraphs [34] – [38]:

"(34) The parties engaged in various types of contacts, during which they, to varying degrees:

(a) coordinated rates for certain routes and for certain customers, except for CSAV that was engaged in this type of conduct only as of June 2011 onwards. In addition, other participants than CSAV were engaged in coordination concerning the BAF (Bunker Adjustment Factor) and CAF (Currency Adjustment Factor) for certain routes and for certain customers.

(b) allocated various RFQs, and the business of certain customers (including agreements on which party should win the RFQ or business or a certain share thereof and the details of the offers) as well as replies submitted in the framework of contract renewals and annual price negotiations;

(c) discussed and coordinated capacity reductions through scrapping of vessels, except for CSAV; and

(d) exchanged commercially sensitive information as a means to support the conduct described in points (a), (b) and (c) above.

(35) The various types of contacts consisted of the following:

– Four Carriers Meetings ("FCMs");

– "3J" meetings;

– bilateral contacts.

(36) A significant part of the coordination took place at the FCMs. The FCMs were usually held on a monthly basis in Japan and were attended by the representatives of MOL, NYK, "K" Line and WWL. In addition to the conduct related to routes from Japan (and certain other Asian countries) to the EEA, the FCMs also touched upon operational issues, which fall outside the scope of this Decision.

(37) Trilateral meetings took place between the "3Js", i.e. the three Japanese carriers: MOL, "K" Line and NYK. Without forming a separate set of arrangements, those discussions concerned certain issues/contracts relevant to the three carriers.

(38) In addition to these, multiple bilateral contacts took place between parties to varying degrees. As not all carriers were present on all trades and did not serve all customers, the carriers participating in these contacts depended on the route and customer involved."

7. In paragraphs [51], [56] and [57] the Commission found that one aim of the cartel was to maintain or increase prices on a durable basis:

"(51) Through a combination of multi-lateral and bi-lateral contacts, structured around the "rule of respect", MOL, "K" Line, NYK, CSAV and WWL and EUKOR engaged with varying intensity, in market sharing, price fixing, customer allocation and capacity reduction, concerning deep sea car carrier services. The parties engaged in such practices with the aim of restricting competition on the market and maintaining the status quo, that is to say, ensuring that the car carriers would keep their respective businesses for certain customers and/or certain routes. They also aimed to preserve their position in the market and to maintain or increase prices, including by resisting requests for price reduction from certain customers.

…

(56) The evidence available shows that the conduct between MOL, "K" Line, NYK, CSAV and WWL and EUKOR was in pursuit of an identical object, namely to avoid price decline and to maintain the existing balance of business between carriers. To do this, the parties held various multi-lateral and bi-lateral contacts and used the "rule of respect". The parties shared the common understanding not to undercut each other's prices in the responses to certain RFQs and other requests from vehicle manufacturers, either by refraining from quoting or by quoting higher than the incumbent carrier, and to respect each other's positions concerning specific vehicle manufacturers and/or routes.

(57) The conduct was an on-going process and did not consist of isolated or sporadic occurrences. The contacts between MOL, "K" Line, NYK, CSAV and WWL and EUKOR were of a continuous nature, with numerous and regular contacts. The different elements of the infringement were in pursuit of a single anticompetitive object as described above, which remained the same throughout the entire period of infringement, although the contacts were route or customer specific and not all parties were involved in every exchange. The existence of a single and continuous infringement is further supported by the fact that the cartel followed the same pattern and there was a continuity in and similarity of the arrangements between the parties."

8. The Decision is a "short form" decision adopted as part of settlement proceedings between the cartelists and the Commission. Such decisions are short form in that they predicate liability upon the "object" of an agreement, not its "effect", which is the alternative trigger for liability. This is because object violations are easier to establish than effects-based violations but still engage the jurisdiction of the Commission to impose substantial financial penalties for violations during the Cartel Period. The Decision arrives at no conclusions about effects. There are no findings, for instance, that the overcharge was "x" Euros or that there was pass on of the overcharge which did cause downstream consumer detriment, etc. These are matters remaining to be proven at trial.

Appeal on points of law

9. We address one matter up front. The Court of Appeal only has jurisdiction to hear appeals on a "point of law": Section 47 CA 1998. The judgments in Merricks v Mastercard Inc [2020] UKSC 51 ("Merricks") and Le Patourel v BT Group PLC and another [2022] EWCA Civ 593 ("Le Patourel") at paragraphs [50] – [57] have provided guidance as to the identification of points of law over which the Court has jurisdiction. All the issues arising on this appeal amount to points of law. The MNW appeal raises questions about the correct approach to compensatory damages under the aggregate damages regime. A similar issue was treated as a point of law in Merricks (ibid) paragraph [64(e)] and see also Le Patourel (ibid) at paragraph [52]. The MNW and the KK appeals raise questions about the level of detail required in a methodology to meet the statutory certification requirements. Issues relating to the overall, high level, approach that the CAT should take were classified as points of law in Le Patourel (ibid) paragraph [55]. This included the level of detail required to be set out in a methodology to satisfy the statutory requirements: see e.g. LSER and others v Gutmann [2022] EWCA Civ 1077 ("Gutmann") at paragraphs [44] – [79]. The Respondent's Notice concerns the scope and effect of two judgments of the Supreme Court in relation to the application of principles of causation: Globalia Business Travel S.A.U. of Spain v Fulton Shipping Inc of Panama [2017] UKSC 43, [2017] 1 WLR 2581 ("Fulton") and Sainsbury's Supermarkets Ltd v Mastercard Inc [2020] UKSC 24, [2020] Bus LR 1196 ("Sainsbury's"). The issues arising concern the scope and effect of those judgments and in particular how they apply in the context of collective proceedings. We treat these as points of law. The cross appeal of the Class Representative concerns a novel issue of limitation as applied to the position of deceased persons and the correctness of a prior judgment of the High Court. We are clear that this cross appeal raises a point of law.

B. The CAT's judgment on the pricing issues

The nub of the dispute: How to determine loss in a consumer pass on case

10. The core of the appeal concerns how, in collective proceedings, loss is to be determined in the case of a good or service which has been charged at an unlawful, supra-competitive, rate ("the overcharge"). That good or service forms one component out of many making up a finished product which is ultimately sold to consumers at the end of a chain of distribution where the overcharge might (or might not) be passed on down to be borne by the consumer. In this case the overcharge to the manufacturers per vehicle would on any view amount to a tiny fraction of the final price paid by consumers. The CAT observed (Judgment paragraph [126]): "For a purchaser, the amounts are likely to be very small as compared to the overall price: for example a £20 increase in the delivery charge of a £20,000 vehicle would represent a price change of just 0.1%.".

The three issues arising on the appeal concerning pricing: silo pricing v overall pricing

11. The appeals of MNW and KK and the Respondent's Notice of the Class Representative all raise issues relating to the proof of loss in a pass on case. At the risk of oversimplification, they concern arguments about two theories of pricing: "silo pricing" and "overall pricing". The Class Representative argues that consumers are charged separately for delivery which includes any unlawful overcharge. Charges for delivery occur in a "silo" and are unaffected by the pricing of the vehicle itself. MNW argues that there is no such thing as silo pricing; cars are purchased by the negotiation of a single "overall" price including all component costs and charges, which includes delivery.
12. There are three matters that we address in relation to pricing:

(i) The MNW appeal: MNW complains that the Class Representative failed to address the appellant's alternative theory of "overall pricing" in its methodology and the CAT erred in certifying the claim on such an inadequate basis. MNW couches its argument both as an issue of law and principle but also as one of methodology.

(ii) The Class Representative Respondent's Notice: It is argued by the Class Representative that, in view of findings made by the CAT that its silo theory of pricing was plausible, the CAT's rejection of its argument - that any benefit consumers might gain from negotiating overall car prices downwards was irrelevant when calculating the quite different loss based upon delivery overcharge, - was clearly inconsistent and wrong.

(iii) The KK appeal: KK argues that in relation to the "silo pricing" theory the methodology constructed by the Class Representative was inadequate, premised upon "extreme" and untenable assumptions, and the CAT erred in certifying the claim and making the CPO.

Both MNW and KK argue that in light of the CAT's errors the claim should be struck out or set aside. A consequence of this would be that, since the original claim ran hot on the heels of the expiry of the relevant limitation period, any new claim would be time barred.

The relevant distribution chain and pass on

13. As a term, "pass on" describes how various persons in a chain of distribution seek to avoid loss by passing on the burden of the overcharge to their own customers. The issue before the CAT in collective proceedings is whether the members of the class have suffered any loss. Pass on is relevant in this regard as a question of fact in working out how much of the overcharge found its way down the distribution chain to be borne by end consumers. The CAT encapsulated how the relevant distribution systems operated and how pass on was alleged to have occurred:

"12. The proposed class members ("PCMs") obviously did not contract directly with any of the Respondents. The claims therefore depend on the class members having suffered loss as a result of the "passing-on" of any overcharge down a supply chain. It is uncontroversial that the relevant supply chain in this case is, at least in most cases, as follows. Original equipment manufacturers ("OEMs") enter into agreements with vehicle carrier operators such as the Respondents to transport vehicles to a central distribution location for the relevant national market. Once transported, vehicles are generally passed down the supply chain to national sales companies ("NSCs"). NSCs are typically, but not always, members of the same corporate group as the relevant OEM. NSCs in turn supply retailers (dealerships). It is the dealerships that sell the vehicles to private or business purchasers, such as the PCMs. In essence, the Applicant claims that the full effects of the cartel were passed down the supply chain to PCMs via delivery charges which were either paid directly or were borne via lease or other financing payments on the vehicles they acquired. As discussed below, although delivery charges are levied by dealers, recommended delivery charges are set by the NSCs."

The methodology advanced at the certification hearing

14. There is a requirement upon an applicant for certification to place before the CAT a methodology setting out how the claim will be advanced at trial. This forms part of the material which is evaluated by the CAT to decide whether the statutory conditions for certification are met. The test to be applied to determine its adequacy is the "Microsoft" test which derives from a judgment of the Supreme Court of Canada (Pro-Sys Consultants Ltd v Microsoft Corp [2013] SCC 57) as endorsed by the Supreme Court and Court of Appeal in this jurisdiction: See Gutmann at paragraphs [45], [46] and [81]-[83] for a summary of the law and practice.
15. The methodology in this case was summarised in the Judgment in paragraphs [77] – [94]. It fairly reflects the case of the Class Representative. The methodology was tentative in that it was prepared prior to disclosure and in some respects confined itself to explaining what sorts of data would be needed in the future and how this, when received, would then be modelled. In paragraph [77] the CAT explained how the methodology would seek to show as a first stage that actual prices were higher than in the counterfactual (where there was no cartel):

"77. In summary, the first stage of the Applicant's proposed methodology would use a comparator-based approach to estimate the size of any overcharge arising from the operation of the cartel during the Relevant Period, the aim being to demonstrate the extent to which shipping costs would have been lower in the counterfactual situation of no cartel having existed. Mr Robinson proposes to identify a control period after the Relevant Period and apply a regression technique to compare cartelised and non-cartelised pricing by controlling for movements in price attributable to extraneous factors. The analysis would use information that should be available from the Respondents on disclosure. Mr Robinson anticipates that this will enable him to calculate an aggregate overcharge per brand. The analysis could in due course be broken down between different periods of time if the effect of the cartel changed during it, and could if appropriate take account of the level of overcharge being different between different OEMs. As already indicated, the Relevant Period includes a run-off period after the Cartel Period because the cartel is expected to have continued to have some effect after the end of the Cartel Period, bearing in mind that shipping contracts entered into before it ended would have remained in place for a time. This additional run-off period is currently assumed to be three years. In making his preliminary estimate of loss, Mr Robinson has made adjustments to the proportion of vehicles assumed to be the subject of an overcharge during the run-off period, reflecting the fact that shipping contracts signed during the Cartel Period would gradually have been replaced."

16. In paragraphs [80] to [83] the CAT summarised the evidence relied upon by the Class Representative to show that the overcharge was passed on to class members:

"80.  Starting at the top end of the supply chain, the shipping charges to OEMs will reflect the nature of the individual vehicles transported (and so, for example, are likely to be affected by the weight and size of a particular model), as well as other factors such as the length of the route. The OEMs will pass on these charges, together with other costs of transport to the relevant local market, to NSCs by charging them a price for each vehicle that includes those costs. This would therefore include any overcharge. Where vehicles of a particular model are manufactured in more than one location a blended cost will be used.

81.  NSCs set not only basic list prices for vehicles but also a recommended delivery charge payable by the consumer to the retailer. The delivery charge might be identified as a separate item or as part of an overall "on the road price" which includes the basic price of the vehicle, the delivery charge and other charges such as vehicle excise duty, registration fees, number plates and fuel. (In the case of some NSCs, delivery charges are instead included in the list price.) It is the Applicant's position that the full cost of transporting the vehicle, including shipping costs, is passed to the end customer as part of the delivery charge. This is achieved by setting delivery charges at a level that covers: (a) the OEM's logistics charges to the NSC (including the OEM's margin); (b) the NSC's own costs at the point of import and of onward distribution; (c) the NSC's margin on the cost of delivery; (d) what the NSC considers to be a reasonable margin for retailers on the delivery element; and (e) an allowance for pre-delivery inspection by the retailer. The result is then benchmarked against delivery charges for equivalent brands, VAT is added and there is rounding up to the nearest £5 or £10.

82.  Generally, recommended delivery charges are the same across all models of a particular brand, and the NSC adopts the same approach in determining its charge to the retailer. So whilst the charge by the OEM to the NSC will be model specific, the NSC's charge to the retailer, as well as the recommended delivery charge to the consumer, will be determined by a calculation that involves dividing total projected logistics costs by total projected unit sales across the brand in question, irrespective of the size and origin of the particular vehicle or model. This is the reason why the Applicant maintains that although only 13% of vehicles registered in the UK in the Relevant Period were manufactured outside the UK and Europe (and so were likely to be affected by excessive deep sea shipping charges), it had an impact on the delivery charge of 81.4% of all vehicles registered. It also explains the concept of Excluded Brands in the proposed class definition: that concept only comprises brands which did not ship any vehicles to the EEA during the Relevant Period.

83.  The experience of the industry experts is that increases in vehicle carrier or other distribution costs are typically reflected in an increase in the delivery charge at the earliest opportunity. However, this will usually not occur if any cost increases are offset by decreases in other costs. Further, delivery charges will generally remain static if overall costs fall. In other words, a minimum margin, which the evidence indicates would be a fixed amount rather than a percentage, would be maintained but there would be no price reduction to remove any increased margin caused by falling costs."

17. In paragraphs [84] to [91] the CAT summarised how the Class Representative's proposed methodology would apply this evidence:

"84.  Mr Robinson proposes to apply this evidence in the following way. Having measured the aggregate overcharge by brand for a given year (see above), he would divide it by the number of vehicles of that brand registered with the DVLA to arrive at an overcharge per vehicle. … The illustrative calculations indicate that the per vehicle figure could range from a few pence to approaching £60, depending on the brand and the level of overcharge.

85.  Mr Robinson would then observe the next increase in the delivery charge for that brand, and would calculate the overcharge to the end customer as the lower of the overcharge per vehicle and the increase in the delivery charge per vehicle. This would provide figures for the overcharge per vehicle for each brand and for each year. Aggregate damages would be calculated by multiplying the figures by the number of vehicles affected and totalling the results.

86.  The choice of the lower of these two numbers is designed to limit the amount claimed only to overcharges passed on, rather than overcharges that are absorbed higher in the chain through lower margins as compared to the counterfactual. This is best understood by examples.

87.  The simplest scenario is where shipping costs increased as a result of the cartel and other costs remained the same. If the NSC raised delivery charges by an amount equal to the increase to restore the margin, then at that stage there would be full pass-on.

88.  The position is more complex if the effect of the cartel was to maintain shipping costs at an artificially high level, when in the counterfactual they would have decreased. In that case, in the counterfactual the delivery charge would not have been reduced (because delivery charges are generally not reduced when costs fall: see above) so there would have been no pass-on at that point. Rather, a higher margin would be earned by the NSC in the counterfactual as compared to the actual position with the cartel in place. However, if other costs increased then in both the actual and counterfactual scenarios the NSC would increase the delivery charge by whatever amount was required to restore its margin. To the extent that the result was a higher delivery charge in the actual rather than counterfactual, Mr Robinson's approach would attribute that element to the overcharge."

The CAT's observations on the methodology

18. The CAT made the significant (and in our view correct) observation that proportionality and practicability govern the construction of a methodology:

"74. A key point to bear in mind is that there can be no bright line distinction between methodology and data. The two are closely linked. In particular, the methodology chosen will be informed by the likely availability of data to which it can be applied. If it appears that data that would be required to apply a particular methodology will not be available, or will not be available without disproportionate cost, then that would indicate that that methodology is inappropriate. It would not meet the Microsoft test. A lack of data may therefore mean that a theoretically preferable methodology cannot be selected in practice.

75. In those circumstances the use of an alternative methodology which will be capable of being applied in practice should not be prevented simply because a better one might be available in economic theory. Any such alternative methodology will need to be assessed on its own merits, having regard to the availability of data to enable it to be applied. Further, any chosen methodology may need to be adapted as data becomes available, or perhaps proves not to be available in exactly the way that was previously anticipated. The possibility of this occurring does not preclude certification. … some gaps in data may ultimately turn out to be unbridgeable, so that nothing might be recovered for part of a claim. But the Tribunal's task is to do the best it can with the evidence."

The CAT added (paragraph ([76]) – again in our view correctly) that there was no rule confining a methodology to a particular econometric technique or to the expert evidence of economists. It might be derived for example from a combination of industry expert evidence and expertise in economics.

The CAT's reasoning

19. The reasoning of the CAT in the Judgment which has been the subject of close and critical scrutiny is found in paragraphs [114] - [127] (in relation to the MNW appeal and the Respondent's Notice) and paragraphs [104] – [110] (in relation to the KK appeal).
20. We start by summarising the reasoning in relation to the MNW appeal and the Respondent's Notice. In paragraph [114] the CAT recorded MNW's "overall pricing" argument. This was said to be "… a hard edged legal point that was fatal to the Applicant's case". The CAT summarised the argument under the heading "Overall price vs. delivery charge":

"114. Class members bought vehicles, and nothing else. They did not pay for a vehicle and also for a delivery service. The vehicles they bought were already in the country when they acquired them. The Applicant's proposed methodology was defective because it failed to address the fact that there was a single transaction for a single price. It did not therefore determine the question whether class members paid more for their vehicles than they would have done in the absence of the infringement. The fact that there might be a separate line item on an invoice for a delivery charge made no difference. Money is fungible, and the question whether loss was suffered could not turn on how an invoice might or might not be itemised."

The methodology was said to be defective at the level of principle and, it followed, the CAT erred in approving it.

21. The appellants advanced a second line of attack. The CAT erred in approving the methodology because it failed to grapple with the MNW "overall pricing" model. The CAT summarised the argument:

"118 … given the single transaction of acquisition of a vehicle, it could not be right for the methodology to consider only the delivery charge, even where that exists as a separate line item (which it does not in all cases). The proposed methodology was defective because it failed to permit any examination of whether the overall price was set in a way that resulted in overcharges not being passed on, or not being passed on in full. In particular, list prices might be set in a way that offset a higher delivery charge, or the class member could receive a discount on the price of the vehicle with the same result.

119… List prices [for particular models] would need to be set, or discounts negotiated, to ensure [the model] was competitive. The methodology had to be capable of investigating whether, notwithstanding the competition, higher overall prices were charged [for a particular model] than would have been the case absent the cartel. The Applicant's methodology did not allow causation to be established by reference to the only transaction entered into, namely the purchase of the vehicle.

22. The CAT then turned to what it termed the "principal" response of the Class Representative which was that the principles of causation laid down by the Supreme Court in Fulton and in Sainsbury's applied. Insofar as there were negotiations for an overall vehicle price which affected (downwards) the overall price, they were irrelevant when calculating the loss flowing from the overcharge:

"120. … the fact that a class member might have achieved a good price for the vehicle overall was not relevant, because it was a benefit conferred by a third party, and such benefits should not be taken into account unless in some sense caused by the breach of duty. [The Class Representative] … relied in particular on Globalia Business Travel S.A.U. of Spain v Fulton Shipping Inc of Panama [2017] UKSC 43 ("Fulton") where, following termination of a charterparty after a repudiatory breach, the defendant charterers sought to limit the claim against them by bringing into account the benefit the owners achieved by selling the vessel in 2007 at a higher price than would have been achieved had it been sold in 2009 after the charterparty had run its course. The Supreme Court, agreeing with Popplewell J at first instance, held that there was no requirement to give credit for the benefit."

23. The CAT did not agree. Neither Fulton nor Sainsbury's provided an answer to the overall pricing argument:

"121. We are not persuaded that this addresses the Respondents' point. Not only is there no separate transaction or event of the kind considered in Fulton (because class members entered into single transactions to acquire the vehicles), but it rather assumes the answer to the question posed. Fulton emphasises that the issue is one of causation. To be brought into account, a benefit "must have been caused either by the breach … or by a successful act of mitigation" (Lord Clarke's judgment at [30]). In Fulton the benefit of avoiding the fall in value of the vessel was not legally caused by the repudiation, nor was it an act of mitigation. This was because the vessel could have been sold at any time, including during the charterparty, and the decision to do so was a commercial one at the owners' risk, independent of the charterparty and its termination.

122. Ms Ford also relied on a discussion in Sainsbury's Supermarkets Ltd v Mastercard Incorporated [2020] UKSC 24; [2020] 4 All ER 807 ("Sainsbury's") from [192] onwards which also referred to Fulton (see [202], [213] and [219]). But in that case there was no dispute that the overcharge was passed to the merchants. The discussion was about the relevance of the merchants' response to the overcharge that was undoubtedly imposed on them, whether by reducing their margins, raising prices or cutting other costs. It was held that the merchants could plead the overcharge as the prima facie measure of their loss without proving a consequential loss of profit (at [199]) because profitability was not the relevant measure of damage (at [203]). However, in some scenarios steps taken would be taken into account in determining the question of mitigation, in respect of which the merchants did not have the burden of proof (at [206]-[216]). That is different to the point being made by the Respondents here, which relates to whether and how a prima facie measure of loss can be established, in the form of an overcharge passed on to class members.

123. If it was the case, for example, that any discount that a class member was able to negotiate would have been the same amount irrespective of any overcharge, then in principle the discount would not affect the class member's claim. However, if it was the case that pass-on did not occur because discounts were negotiated, or list prices were set, in a way that would have differed in the counterfactual as compared to the actual, such that any overcharge was not passed on or was passed on a lower amount, then it is likely that the claim would be affected.

124. The Applicant submitted that this could only be demonstrated in a case where the customer specifically negotiated away the delivery charge, which the industry expert evidence indicates would be a rare occurrence, rather than achieving a discount on the overall price. Again, however, this rather assumes the answer to the question, namely whether list prices or any discounts on the overall price were affected by the cartel."

24. Nonetheless, the CAT did not decline certification. The alternative argument of the Class Representative was that, on the facts, delivery charges were treated as a discrete cost which had to be recovered so that, in effect, there was simply no connection between the pricing of all other vehicle costs, and the pricing of delivery. The CAT found on the evidence that this silo model of pricing was plausible (paragraphs [125]-[127]), a conclusion based upon four matters. First, the evidence of the Class Representative which established the existence of silo pricing was preferred by the CAT over that of the appellants which disputed it. Secondly, its acceptance of the broad economic case advanced by the Class Representative focusing upon the absence of elasticity of demand in the distribution chain from which it was reasonable to infer that a carriage overcharge could easily be passed on down the chain to consumers. Thirdly, the failure of the appellants' adduced evidence to contradict the Class Representative's silo pricing case in certain material respects. Fourthly, the CAT concluded that were the appellants to prevail at trial then the CAT would be able to perform adjustments to quantum. We elaborate upon these four reasons.
25. The evidence: The CAT set out provisional conclusions of fact based upon the evidence (paragraph [125]): (i) it was "highly unlikely" that OEMs and NSCs would not recover their delivery costs in full; (ii) it was "rare" for a retailer to discount the delivery charges; (iii) there was "no customer expectation" that retailers would not recover delivery charges; (iv) delivery charges were considered to be separate cost items; (v) delivery charges "must be recovered"; (vi) increases in delivery costs were not "absorbed"; (vii) delivery charges were not "simply wrapped up in, or considered as part of, a single undifferentiated price"; and (viii) delivery prices were "considered separately".
26. Economic principle: The CAT (paragraph [126]) accepted that the case of the Class Representative was consistent with broad economic theory. Because the overcharge was a very small component of the overall vehicle price there would be insufficient upstream elasticity of demand to make customers (of the carriage service) sensitive to the existence of increases in prices resulted from the overcharge and these could be passed on down the line to ultimate consumers who would also be indifferent to the overcharge because it was such a small component of the final vehicle price. Put another way the existence of the overcharge in a final selling price would not persuade consumers to switch to sellers of other vehicles where there was no overcharge:

"126. Further … the negotiating characteristics of the parties are the same in the counterfactual and actual scenarios. The seller's motivation to recover costs (and maintain margins) would be the same. As already indicated, the industry expert evidence indicates that delivery charges are considered to be a separate cost item which must be recovered. For a purchaser, the amounts are likely to be very small as compared to the overall price: for example a £20 increase in the delivery charge of a £20,000 vehicle would represent a price change of just 0.1%. In order to show an impact on the purchaser's position there would need to be high elasticity of demand associated with small changes in price, meaning that customers would have to be very sensitive to price changes such that sellers would lose customers if they tried to recover the increased cost. Mr Robinson suggests, based on a previous study, that it is unlikely that buyers of vehicles would be as sensitive as would be required to make a difference. The obvious differences that exist between models of different brands, even where those models are in direct competition with each other, reinforces this."

27. The relative strengths of the evidence: The CAT was unimpressed with the state of the appellant's evidence (paragraph [127]). No evidence had been provided indicating that discounts would have differed in amount in the counterfactual (where there was no cartel). There was no evidence about how list prices were set in practice. KK had served an expert report from Dr Majumdar of RBB Economics which set out various points in the distribution chain at which it was suggested negotiations would take place to reduce or eliminate the overcharge, but the analysis was largely abstract and unsupported by actual evidence. The report did not have the "the same grounding in evidence of practice in the industry" as that of the Class Representative's expert, Mr Robinson. The report did indicate that discounts might have been negotiated especially by large business purchasers of vehicles, but it did not "obviously undermine key elements of the [Class Representative's] evidence". Indeed, as was pointed out before us, the trade witness evidence of KK positively averred that negotiations did not generally occur in relation to delivery charges (see paragraph [38] below).
28. Ability of the CAT to adjust quantum at trial: Finally, in relation to the ability of the CAT to make adjustments at trial to take account of victories on the part of the appellants, the CAT concluded:

"127 … if that were so, and it appeared that the discount would have been lower in the counterfactual (as might be the case if, for example, discounts were calculated in percentage terms by reference to the overall price, or increases in delivery charges were negotiated away) then an adjustment could be made to any award to reflect the lower level of pass-on that that would imply."

The CAT made additional observations about its ability to perform an exercise in adjustment at trial in which it referred to the possibility of no loss claimants:

"111. … At this stage we also have no basis to conclude that it would not be feasible for any successful evidential challenge to the quantum of loss claimed to be addressed, whether by adjusting the methodology or by making some other adjustment to the quantum of any award of damages to arrive at a reasonable estimate of loss. To take one example that Mr Singla relied on, if it were the case that it was established at trial that certain categories of class member (such as car rental companies) tended to achieve discounts that reduced or eliminated the effect of any overcharge, or that they negotiated away the effect of any increase in delivery charges, then that could be taken into account in determining the quantum of any award."

29. In relation to the KK appeal paragraphs [104] – [110] of the Judgment set out the CAT's analysis. KK had argued that the silo pricing theory was theoretical, based upon untenably extreme assumptions and, in any event, the methodology failed to address in any acceptable detail how the case was to be proven. The CAT disagreed. It started by acknowledging that the intensity of analysis by the CAT when considering certification of a methodology was more than "symbolic scrutiny". However, KK's "criticisms of the proposed methodology go too far" (paragraph [104]). There was no rule confining the concept of methodology to a particular econometric technique or to the expert evidence of economists. With limited exceptions it was not the role of the CAT "to determine the merits of the case" at the certification stage and this included the merits and robustness of the methodology. Ultimately, if the Class Representative's expert evidence was successfully challenged at trial, the claim might simply fail. The Microsoft test was not so onerous that the CAT was bound to reject any methodology that might later on "break down in the face of a challenge". That was not the "low threshold" that the test was intended to set (paragraph [106]). The task of the CAT was to determine whether the methodology offered a "realistic prospect of assessing loss on a class-wide basis":

"Realistic prospect" means just that. It does not mean that the Tribunal must satisfy itself that the methodology is bound to work, or will work on a balance of probabilities, whatever the evidential challenges. The Tribunal is not conducting a mini-trial."

30. Importantly, the CAT acknowledged that the object of the methodology was to establish loss on an aggregate basis under section 47C CA 1998 and there was no need to assess loss individually. It was not therefore fatal that some class members might be proved to have suffered no loss (paragraph [108]). The CAT summarised its reasons for rejecting the KK argument:

"110. We have scrutinised the proposed methodology, including the Applicant's industry expert evidence, in detail and have addressed the specific challenges raised by MNW. Both of the Applicant's industry experts have a great deal of experience in the motor industry. Their evidence is clear that any overcharge would have been passed on, and that NSCs would seek to maintain a minimum margin. We would not characterise those as extreme factual assumptions. In principle, they are plausible. Further, whilst some evidence about practice in the industry has been adduced on behalf of KK we have seen nothing that obviously undermines key elements of the Applicant's evidence, such that the methodology would not meet the "realistic prospect" threshold in the Microsoft test. We specifically reject Mr Singla's submission that points of difference between the witnesses means that the Applicant's methodology cannot be "grounded in the facts" as the Microsoft test requires. Apart from that comment being directed at the availability of data, it is not our role at this stage to find the facts, beyond determining whether the threshold just referred to is met."

C. The MNW appeal – the overall pricing argument

The two limbs of the MNW appeal

31. We turn now to the MNW appeal. This was put in two different ways. First, it is said that the CAT made an error of law and principle in failing to take sufficient account of the basic principles of tortious compensation. Ms Demetriou KC argued that once the CAT had rejected the Class Representative's causation argument (see paragraph [54] – [62] below) it was illogical and wrong "as a matter of law, economics or common sense" to permit the claim to proceed. Secondly, it is said that, in any event, the CAT erred because the methodology was defective in that it did not address the appellant's overall pricing case but confined itself to explaining the silo pricing theory. The CAT should not have approved the methodology on such a partial and inadequate basis.

Issue of law or fact?

32. We start by addressing the objection that the CAT made an error of law.
33. It is clear that the CAT understood the appellant's overall pricing argument. It is fairly encapsulated in paragraph [114] of the Judgment (see paragraph [20] above). The CAT did not however reject it holding only that its merits were to be determined at trial. If it succeeded to any relevant degree damages would then be adjusted. The issue before us is therefore relatively straightforward. Was the CAT correct in treating the MNW argument as an issue of fact for trial rather than a hard edged issue of law and principle? There are three reasons why we consider that the CAT correctly treated this as an issue of fact for trial, not law.
34. First, MNW's starting point is that at the level of first principle damages are assessed upon the basis of a but for counterfactual where there is no unlawful cartel and no overcharge. In this counterfactual, because money is "fungible", vehicles are sold as composite products for a single overall price and the CAT erred in paragraph [125] in failing to accept this as a reason to reject the CPO. We disagree. We do not see that the fungibility of money is of direct relevance. In a typical case the use of a counterfactual involves a comparison of the pricing of the same product or service in the actual and but for counterfactual markets. The MNW counterfactual however does not involve a like for like comparison. MNW declines to disentangle the delivery charge in the counterfactual so that it can be compared with actual charges because, MNW argues, in the real world there is no such thing as a delivery price, only a single overall price. However, a counterfactual where there is no unlawful cartel, and therefore no overcharge in relation to carriage of vehicles by sea, will be neutral as to whether vehicles are sold downstream at a single composite price including delivery, or whether there is silo pricing for delivery. Both are possible in a no cartel/no breach scenario. If there is a tough negotiation in the counterfactual over, say, number plates or tyres, which results in the final vehicle price being the same as the end vehicle price in the real world, that might be pure happenstance. The final price could have been higher or lower. It might be difficult to infer anything from this as to the level of the (hypothetical) competitive delivery charge absent the cartel in the counterfactual so as to be able to compare it with the delivery charge in the real world to decide whether there is an alleged overcharge concealed somewhere in the final price. Ultimately, the premise behind the overall pricing theory does not prima facie flow from the but for counterfactual where there is no breach. It is a different factual proposition that can only be determined at trial.
35. Secondly, this is a claim for aggregate damages under section 47C CA 1998 which as the Supreme Court explained in Merricks entails a "radical" departure from normal principles of compensation (ibid paragraphs [58] and [76]). Aggregate loss is not computed by reference to the traditional bottom-up position of individual consumers. Instead, it is determined by reference to the top-down position of the class as a whole. On such an approach there might well be individuals or groups of individuals who suffer no loss when their positions are compared with the counterfactual, but that is not fatal to the claim, as the CAT pointed out in its Judgment. The Supreme Court in Merricks (ibid paragraph [77]) and in Lloyd v Google LLC [2021] UKSC 50 ("Lloyd") (at paragraph [32]) and the Court of Appeal in Le Patourel (ibid paragraph [32]), made clear that the CAT was not bound by traditional principles of compensation. When both quantifying and distributing aggregate damages the CAT might, wielding its broad axe, work with new techniques and principles to achieve practical justice. In Lloyd (ibid at paragraph [32]) the Supreme Court accepted for example that the CAT could apply a "broad brush" approach to distribution of damages leading to an equal division amongst all class members and a person or category of persons might therefore acquire an entitlement to compensation even if there was no proof of loss. To take a simple example if the CAT at trial were to find using its broad powers of estimation that (say) 25% of all class members suffered no loss the CAT might reflect this by reducing the overall aggregate damages but then dividing the discounted aggregate sum (according to some formula to be determined) across the entire class thereby resulting in payment of damages to no-loss members of the class. In such a case, defendants would not over-pay, in an aggregate sense, but there would be a flattening out of the sums distributed by way of damages. In Flynn Pharma and others v Competition and Markets Authority [2020] EWCA Civ 339 at paragraphs [118] – [125]) the Court had to rule upon the test to be applied to decide whether a drug price was excessive and an abuse of dominance. It rejected the submission that a hypothetical, but for, counterfactual was mandatory and held that a wide range of evidence might be relevant and that there was no single, compulsory, test for identifying and quantifying an overcharge. The choice of test depended critically upon the availability of data and other evidence. The case concerned the identification of an overcharge for the purpose of establishing abuse; it was not a damages case. Nonetheless, the approach adumbrated is closely analogous to that taken by the courts in relation to aggregate damages and sheds light upon how a suitable counterfactual might be chosen. In Merricks (ibid paragraph [77]) the Supreme Court suggested that in many cases the "fairest method" might only be determined at the distribution stage when the size of the class and the aggregate award were known facts highlighting that the approach to be applied is highly contextual and dictated by pragmatism and flexibility. On the facts of this case that is an apposite observation. At this stage it is not possible to conclude that there is only one way of computing damages, as MNW in effect submits.
36. Finally, MNW's argument ignores the facts found by the CAT in paragraph [125] of the Judgment which were that the Class Representative had established that there were in effect two pricing silos and that these did not affect each other. The CAT recorded that there was a plausible case that delivery charges were not "simply wrapped up in, or considered as part of, a single undifferentiated price" and that they were "considered separately". These findings contradict the MNW overall pricing counterfactual which assumes, as a fact, the existence of only one silo into which all costs are combined which then forms the basis of price negotiations. However, Ms Demetriou KC, for MNW, did not argue that the findings in paragraph [125] of the Judgment were capable of challenge. They were accepted as provisional findings based upon evidence adduced by the Class Representative which the CAT concluded was superior to that tendered by the appellants. Indeed, Ms Demetriou fairly acknowledged (at least at this stage) that there would be classes of consumer who had suffered loss. This being so even on MNW's case there may still need to be a trial of the evidence to disentangle delivery from the overall price to enable a direct like for like comparison to be made.

The challenge to the methodology

37. We turn now to the second way in which the argument was put. The complaint is methodological and is that the CAT erred in permitting the claim to go to trial because the methodology failed to grapple with the overall pricing argument: see Judgment paragraphs [118] and [119] set out at paragraph [21] above. In oral argument Ms Demetriou KC emphasised that the silo pricing theory of the Class Representative was framed in all or nothing terms. If it failed at trial, there was no fall back and it was accordingly an error for the CAT to hold that if at trial the battle between the silo and the overall pricing theories went in favour of the appellants this would result only in the CAT adjusting damages (downwards). That was wrong because, since the class had no fall back, if it lost on its silo pricing theory at trial it necessarily lost the whole case. Ms Demetriou KC argued further that the failure of the methodology to address overall pricing was also fatal in that it went directly to whether the class had suffered any loss and in law the burden of proving loss lay with the Class Representative, which had not discharged that burden. The CAT was bound to address this fatal omission at the CPO stage and had it done so it would or should have held that the CPO application failed.
38. Ms Ford KC for the Class Representative rejected this description of the claim. It was accepted that at the certification point the claim, based squarely on the evidence, was that there was a silo embracing delivery charges. This was supported by the evidence tendered by the Class Representative but also by that tendered by KK in the evidence of Mr Dent who in his witness statement expressly recognised that, generally, there was a lack of negotiability over delivery charges:

"27. I do not believe that customers were prejudiced by the fact that the delivery charge may not have reflected the actual transportation cost, because ultimately the customer could negotiate discounts far in excess of the delivery charge. I tell customers that the delivery charge is included in the advertised price and cannot be discounted. However on a number of occasions (four to six times a year on average), I have had customers who have specifically focused on at that cost and said that they would not buy the car if they had to pay the delivery charge: although the system does not allow me to change the delivery charge from GBP 825 to zero, I could achieve the same effect by putting an additional amount of GBP 825 in the special allowance section of the invoice. It makes no difference to a dealer where the discount is applied; it is, however, physically impossible to change certain line items."

To the extent, in an aggregate damages case, that there were negotiations which reduced overall prices or which in some way neutralised the overcharge these were "outliers" which under Section 47C CA 1998 would be of no relevance when the CAT computed aggregate damages.

39. Ms Ford also rejected MNW's argument that the burden of proof lay on the Class Representative to address any alternative theory of pricing. The CAT's findings in Judgment paragraph [125] that a plausible prima face case of loss based upon silo pricing had been established put that argument to rest. MNW's overall pricing model was, properly analysed, a defence because it sought to meet the prima facie case of loss by demonstrating that the overcharge had not in fact been incurred. In law the burden of proving a defence to a claim of unlawful overcharge lay with the defendants to the action, not the Class Representative. There was no duty on the Class to advance a pre-emptive case on this, at least at the certification stage. In his responsive second economic report Mr Robinson recognised that following disclosure the Class might need to reconsider the position. This was a case management position the CAT had properly accepted in its discretion.

Analysis: No setting aside of the CPO

40. We turn to our conclusions. The motivation behind the appeal is that if it is right then, MNW argues, we should set aside the claim and the CPO and, given the expiry of limitation, that would bring to an end litigation risk for the appellants. We reject the submission that the claim should be struck out or set aside.
41. The Supreme Court in Merricks made emphatically clear that once the CAT had concluded that a claim was arguable and was not to be dismissed on the merits there was an entitlement or right on the part of the Class Representative to have the claim tried. Once the strike out hurdle has been passed the "…claimant is entitled to have the court quantify their loss almost ex debito justitiae" (ibid paragraph [47]). The right to quantification in order to do justice is "unavoidable" and the CAT cannot deny a class the right to a trial merely because of "forensic difficulties" in quantification "however severe" (ibid paragraph [50]) or "formidable" (ibid paragraph [70]).
42. The judgment of the CAT that there was an issue to be tried was plainly correct. Liability had been established by the Decision. The Cartel Period lasted 6 years and prima facie it is reasonable to infer that the parties considered that to some degree the cartel achieved its object of keeping prices artificially high, otherwise they would not have taken the considerable financial, legal and reputational risk of persisting in this covert, unlawful, conduct for so long. The economic case that there would be pass on of the overcharge is also entirely plausible (as the CAT held – see paragraph [26] above) and this being so there is a real possibility that at least some of the overcharge was transferred down the distribution chain and borne by consumers. It follows that there is a prima facie claim to be tried.
43. As observed, MNW argues that the overall pricing argument goes directly to the very existence of loss and hence engages the legal burden of proving loss which is on the Class Representative. In Sainsbury's (which was not a collective action) the Supreme Court held (ibid paragraph [216]) that the prima facie measure of loss was the unlawful overcharge, and the legal burden was upon the defendant to plead and prove that the claimant had mitigated the loss, including by passing it on. The Court rejected an argument by the defendant (similar to the argument advanced by MNW in this appeal) that because proof of pass on or mitigation went to the very existence of loss the burden of proof lay with the claimant (ibid paragraph [207]). Instead, the Court (ibid paragraph [211]) endorsed the conclusion of the Court of Appeal ([2018] EWCA Civ 1536 at paragraph [324]) that: "Whether or not the unlawful charge has been passed on is a question of fact, the burden of proving which lies on the defendant … who asserts it." The Supreme Court also focused upon practicalities. Once the legal burden on the defendant had been met by the raising of an issue of mitigation there was a "heavy evidential burden" on the retailer (the claimant) to provide evidence of how they had dealt with the recovery of costs within their business. This arose because the evidence relevant to the issue would lie "exclusively in the hands of the merchant". If the merchant did not produce this evidence adverse inferences might be drawn by the Court. In NTN and others v Stellantis [2022] EWCA Civ 16 ("Stellantis") at paragraph [53] the Court of Appeal explained that, in the light of Sainsbury's, the whereabouts of relevant disclosure would affect how the burden of proof applied in practice. How all of this applies in the context of the present case is not something we express a definitive view upon and will be for the CAT to determine at trial. The CAT here held that the Class Representative has established a plausible case of loss. The overall pricing theory is a response to that prima facie case designed to establish that any overcharge was in fact neutralised and/or offset elsewhere. It is at least arguable that this is a defence for which the appellants have the burden of proof and if it turns out that the Class Representative does not have any or any significant relevant direct disclosure to give that might affect how the CAT evaluates the appellants' evidence.

Analysis: Case management issues

44. We do however share some of the concerns expressed by Ms Demetriou KC about the lack of detail in the Judgment as to how the silo and overall pricing theories are to be addressed in the future as the case proceeds. We are of the view that whilst none of the criticisms made go to certification, they do amount to an error of law in the way in which the CAT understood and approached the principles governing its gatekeeper and case management responsibilities.
45. The duty on the CAT as gatekeeper in collective proceedings is proactive as well as reactive. Once the CAT has decided to make a CPO that is not the end of the gatekeeper role. A CPO "… is neither the beginning or the end of measures whereby the CAT may case manage collective proceedings" (Merricks (ibid) paragraph [28]). A class representative might not have to overcome a very high hurdle to obtain a CPO but the CAT should nonetheless ensure that from the certification stage the case proceeds efficiently to trial. This role might well entail the CAT imposing substantial case management burdens on the parties at an early stage.
46. In Gutmann the Court of Appeal, when seeking to pull the threads together from the case law, endorsed the proactive gatekeeper role of the CAT (e.g. ibid paragraphs [60] and [61]). In Merricks, endorsing Canadian authority, the Supreme Court emphasised the strong public interest element in collective actions (ibid paragraph [37]); see also the summary in Le Patourel (ibid paragraph [29]). There are clearly established strong public interest benefits in the CAT performing an active elucidatory role which include: ensuring that large scale litigation is run efficiently; ensuring that defendants are not confronted with baseless claims; and ensuring that potentially sprawling cases do not absorb an unfair amount of judicial resource.
47. In such cases, the methodology advanced by the class representative at the certification stage will be an important feature of the process. The level of detail of a methodology required by the CAT will always be fact and context sensitive and will turn upon such matters as the availability of evidence. However, underlying the Microsoft test is the proposition that if a claim is certified then the methodology offered by the class representative will provide an initial blueprint for the parties and the CAT of the way ahead to trial. That is of course not to say that the class representative's methodology is cast in stone. It can, as in the instant case, be challenged by the defendants, and Rule 85 of the CAT Rules contains "wide powers" for the CAT to stay, vary or revoke a CPO (Merricks (ibid)). In short, the CAT has power at any point to revisit the methodology.
48. In the instant case, clear battle lines were drawn in relation to the methodology at the CPO stage. The Class Representative advanced a relatively inflexible case based upon its theory of silo pricing, and it seems almost inevitable that it will in due course have to modify or adapt its methodology to address the appellants' overall pricing case. The CAT said as much when it recorded that the methodology was provisional pending disclosure and evidence. The MNW appellants, equally, advanced a relatively rigid theory about overall pricing. They have not set out what evidence they will adduce to prove the counterfactual or why and how it will establish that there would be no difference in outcome. The submission that there will be no difference between actual and counterfactual pricing might rest upon some hefty factual assumptions given what is presently known about the evidence.
49. Neither the class, who are consumers, nor the appellants, who are carriers, will have much, if any, direct disclosure to give on the issue of how car prices are actually set by those in between. Attention will lie with alternative or proxy forms of evidence. None of the parties set out in any real detail how they proposed to address this evidential lacuna, or what the proxy forms of evidence would be. Nor did they address how they proposed that the CAT make appropriate findings of fact, or, once facts were found, what methodologies might, in an aggregate damages case, enable the CAT to arrive at conclusions on quantum. Nor have they considered what sorts of adjustments might need to be made should the appellants prevail on some issues for example relating to the extent to which there is pass through of the overcharge, or as to the existence of possible classes of no loss claimant, or as to the possibility of partial off-setting of overcharges by reductions elsewhere.
50. In its Judgment, the CAT identified the battle lines, but said that the battle along those lines was for trial. In our judgment this was an error in approach. Once it had decided to grant certification, the CAT should have gone on to address the ramifications of the challenges to the Class Representative's methodology. At the CPO stage it was clear that this represented the pivotal dispute in the case.
51. In this regard, if the CAT was of the view that it lacked sufficient information to perform this elucidatory role it could, exercising its broad case management powers, have directed the Class Representative to set out more fully its response to the overall pricing case, as presented by the appellants. If, however, it considered that the appellants had not sufficiently particularised or evidenced their overall pricing case, it could have directed them to provide further detail and then directed the Class Representative to respond. Either approach would have enabled the CAT fully to exercise its gatekeeper role and at the outset lay down a more developed judicially approved trial preparation pathway. Instead, we consider that the CAT did err in simply stopping in its tracks when confronted with two starkly opposing pricing theories and holding that they were for trial.
52. In our judgment it is appropriate, on the facts of this case, to remit the issue now to the CAT before additional significant steps are taken by way of preparation for trial. There are a number of reasons for this. First, further consideration by the CAT at this juncture will provide substantial clarity to the parties going forward. It will sharpen the focus on disclosure and evidence preparation and in due course should improve the management of the trial and assist in making the proceedings more efficient and less costly. Secondly, the CAT lost an opportunity to lay down early guidance for other cases as to how consumer pass on disputes should be prepared and case managed. The issues arising in this case are likely to reflect issues arising in other similar cases. A more detailed assessment by the CAT now could be valuable in guiding how putative Class Representatives construct their methodologies and as to the way in which defendants seek to counter them. In coming to this conclusion, we are not seeking to prejudice in any way how the CAT goes about addressing the remittal or as to the conclusions it might arrive at, and nothing we have said affects the threshold to be met for the grant of a CPO.
53. Our conclusion therefore is that the CAT needed, as part of its gatekeeper role, to set out more clearly how it expected the trial to proceed.

D. The Class Representative's Respondent's Notice - The causation issue

The issue

54. We turn now to the Respondent's Notice of the Class Representative. This concerns causation between the breach of duty and benefits accruing to class members as a result of the breach which might be set off against damages otherwise to be paid. The CAT dismissed, seemingly outright, the argument of the Class Representative, based upon the judgments of the Supreme Court in Fulton and Sainsbury's, that if there were negotiations which resulted in a reduction in the overall price of a vehicle this was irrelevant to the calculation of loss in respect of the overcharge: See Judgment paragraphs [121] – [124] set out at paragraph [23] above. MNW argued that given this conclusion of the CAT it erred in holding that the claim should not be struck out. Ms Ford KC for the Class Representative pointed out in response that the findings of fact in Judgment paragraph [125] were not challenged and it logically followed from these that there was a clearly arguable case which, if established at trial, would result in the conclusion that any benefits accruing from the negotiation of the overall vehicle price were immaterial to the calculation of loss for the delivery overcharge. In other words, there would be no sufficient connection between the breach and the benefit.

The law

55. We start by setting out the law. In Fulton the claimant chartered a vessel to the defendant the term of which was extended by agreement. The defendant then refused to accept the agreed extension and sought to redeliver the vessel to the owner who treated the return as a repudiatory breach and sued for damages. The owner then sold the vessel for $23.7m. At the end of the extension period (about 2 years later) the sale price of the vessel would only have been c$7m. The defendant claimed that the difference was a benefit serving to diminish the damages arising from the breach of contract. The Court of Appeal reversed the judgment of the High Court which had found that the benefit was irrelevant and did not serve to reduce damages otherwise payable, holding instead that the incremental sale revenue was a benefit to be credited against the damages. The Supreme Court allowed an appeal reinstating the judgment of the High Court. Lord Clarke, for a unanimous court, observed:

"30. … The essential question is whether there is a sufficiently close link between the two and not whether they are similar in nature. The relevant link is causation. The benefit to be brought into account must have been caused either by the breach of the charterparty or by a successful act of mitigation."

56. Lord Clarke endorsed the analysis of the judge at first instance. There had to be a "sufficiently close link" between the benefit and the breach for the benefit to count against damages. The test was not whether they were similar in nature. On the facts: (i) there was nothing about the premature termination of the charterparty which made it necessary to sell the vessel which could have been sold during the term of the charterparty; (ii) the sale was a commercial decision at the owner's risk; (iii) the interest in the vessel was not part of the subject matter of the charterparty and had nothing to do with the charterers; (iv) the premature termination of the charter party was the occasion for the sale but not the cause of it; and (v) the sale was not an act of mitigation.
57. In Sainsbury's the claim was brought by retailers who had been unlawfully overcharged for credit card services. The defendant argued that the claim was one of pure economic loss and therefore to be calculated by reference to the loss of profit retailers would have enjoyed in the counterfactual, i.e. but for the alleged wrongful act. The Court rejected this argument. The prima facie measure was the overcharge. Whether there was pass on in whole or part was a question of fact. The court had to make findings about this to avoid the risk of multiple recovery. The normal rule of compensatory damages applied (ibid paragraphs [195] and [196]). The Court then stated:

"197. There are sound reasons for taking account of pass-on in the calculation of damages for breach of competition law. Not only is it required by the compensatory principle but also there are cases where there is a need to avoid double recovery through claims in respect of the same overcharge by a direct purchaser and by subsequent purchasers in a chain, to whom an overcharge has been passed on in whole or in part."

58. The prima facie measure of overcharge as the recoverable loss might however be subject to adjustment. On the facts that had been found at first instance ([2016] CAT 11 at paragraphs [434] and [435]), and which were not in dispute on appeal, that there were well established means by which retailers sought to off-set the overcharge:

"205…(i) a merchant can do nothing in response to the increased cost and thereby suffer a corresponding reduction of profits or an enhanced loss; or (ii) the merchant can respond by reducing discretionary expenditure on its business such as by reducing its marketing and advertising budget or restricting its capital expenditure; or (iii) the merchant can seek to reduce its costs by negotiation with its many suppliers; or (iv) the merchant can pass on the costs by increasing the prices which it charges its customers. Which option or combination of options a merchant will adopt will depend on the markets in which it operates and its response may be influenced by whether the cost was one to which it alone was subjected or was one which was shared by its competitors…."

59. If options (iii) or (iv) were adopted these were capable of reducing the loss suffered. Paragraph [205] continued:

"… If the merchant were to adopt only option (i) or (ii) or a combination of them, its loss would be measured by the funds which it paid out on the overcharge because it would have been deprived of those funds for use in its business. Option (iii) might reduce the merchant's loss. Option (iv) also would reduce the merchant's loss except to the extent that it had a "volume effect", if higher prices were to reduce the volume of its sales and thereby have an effect on the merchant's profits."

In either case the court was required to take account of the effect "… and there [would] be a question of mitigation of loss" (paragraph [206]).

60. If, by engaging in acts of mitigation, a claimant acquired an "additional" benefit the possibility arose that the benefit should be set off against the loss (ibid paragraph [212]). There were broadly two types of case where the benefit might be set against the loss: (i) where the benefit was sufficiently causally connected to the wrong (which it would not be if the benefit arose from an independent commercial decision by the claimant) (ibid paragraph [212]); and (ii) where there was a failure on the part of the claimant to act reasonably in response to its loss i.e. a failure of the duty to take reasonable steps to mitigate (ibid paragraph [214]). In paragraph [215] the Court highlighted that the issue was one of fact. The question for the court was to determine "factual loss" i.e. the sum actually lost. The test was one of "proximate or legal cause". In determining actual loss, the court would determine whether the claimant had "avoided all or part of the loss". In paragraph [219] the Court referred to the legal or proximate causal link having to be a "close" one. The judgment in Fulton was endorsed as an example of a case where the owners' claim for loss of profits from the charter party was unaffected by the overall profitability of the claimant arising from the sale. The Court endorsed the observation of Lord Clarke that: "The relevant link is causation. The benefit to be brought into account must have been caused either by the breach of the charterparty or by a successful act of mitigation", and (at paragraph [219]) also endorsed the conclusion that the causative link between the wrong and the benefit had to be "close".

Analysis

61. How does this apply in the present case? Can it be confidently said at the CPO stage that on the facts of this case there is a sufficiently close causal link between the breach and any benefit that might arise in an overall price negotiation which would warrant that benefit being set off against the overcharge? In Judgment paragraph [121] the CAT stated that there was "no separate transaction or event of the kind considered in Fulton (because class members entered into single transactions to acquire the vehicles)" but then in paragraph [125] it held that increases in delivery costs were not "absorbed" and delivery charges were not "simply wrapped up in, or considered as part of, a single undifferentiated price" and they were "considered separately". Ms Ford KC argued that it followed from the finding in this latter paragraph that it was inconsistent and illogical for the CAT to say that there was nonetheless a sufficiently causal link between the breach of duty by the cartel in respect of delivery charges and any benefit arising from negotiation in the non-delivery silo such that a non-delivery benefit was relevant to offset the overcharge. How, she argued rhetorically, can there be a sufficiently causal link when the CAT itself held that there was no linkage between the overcharge and pricing elsewhere? We see some force in this argument.
62. In our judgment the CAT erred. Whether there is a sufficiently close causal connection is, on the facts of this case, an analysis that can only realistically be formed after the facts have been found. In this case, which involves a complex evidential dispute, the CAT formed a conclusion on the factual issue of causality at the CPO stage and became prematurely sucked into the merits of the point. This was inappropriate in the light of authority which makes plain that (outside of a strike out/ summary dismissal application or a case under CAT Rule 79(3)(a) on opt-in/out – Merricks (ibid) paragraphs [59], [60]) it is not the role of the CAT at the CPO stage to make merits findings. The definitive preclusion of a possible argument about causation is a merits conclusion about the strength of the claim. On the facts as presently accepted in paragraph [125] there is a real argument that, if proven at trial, the CAT would conclude that there were two unconnected pricing processes involved in the sale of a vehicle to a consumer and that what happened in the overall pricing silo was unrelated to pricing in the delivery charge silo. In coming to a negative conclusion on causation the CAT thus ignored the implications of its own findings. We would add only that any analysis of causality between the breach and the benefit must take into account that when the purchases in question occurred the purchaser was oblivious to the overcharge which was, by its nature, covert. There was therefore no incentive to argue that the overcharge was not payable because it was unlawful or, in the face of seller resistance, to seek to offset it elsewhere, for instance in a negotiation over tyres or number plates. A similar observation was made by the Court of Appeal in Stellantis (ibid) at paragraphs [19], [67] and [75]. In short, it was premature to seek to decide this issue. For these reasons we conclude that the CAT erred in its findings at paragraphs [121] – [124] of the Judgment.

E The KK appeal – The methodology in relation to silo pricing relies upon extreme and unsustainable assumptions

The issue

63. We turn now to the third issue arising in relation to methodology. This raises a quite different point to the MNW appeal which is about the alleged failure of the Class Representative and the CAT to address the appellant's overall pricing case. Here, the challenge is as to the level of detail set out in the methodology relating to the silo pricing theory of the Class Representative and whether it meets the threshold for certification. KK argues that the:

"…Tribunal ought to have taken into account the lack of sensitivity in the proposed methodology when assessing whether it was a credible and plausible means of estimating aggregate damages. This did not require any "mini-trial" or detailed assessment of the facts but rather it required an assessment of the proposed methodology's ability to take into account plausible variations in the facts which will be argued at trial. In failing to carry out such an assessment, the Tribunal failed to properly apply the [Microsoft] test and failed properly to discharge its gatekeeping role at the certification stage"

64. Mr Singla KC, for KK, contended that the CAT erred in failing to rule that the methodology was deficient. He argued that in Gutmann (ibid) this Court had made clear that the methodology had to set out all of the "issues" (paragraph [56]) and articulate a method which would enable them all to be tried in a workable fashion (paragraph [60]) and that as part of its gatekeeper function the CAT had to "closely scrutinise" the methodology (paragraph [61]).
65. KK complains that the methodology for the silo pricing theory failed to estimate pass-on using empirical techniques (e.g. a regression analysis) nor did it establish a causal link between the alleged overcharge and the increase in delivery charges. Instead, the methodology computed the level of pass-on assuming, but without proving, that the factual premises underpinning the assumptions were correct either throughout the market or for the duration of the claim. KK listed these unproven "extreme" assumptions: (i) each OEM always passed-on 100% of the overcharge on vehicles sold in the UK to the NSCs it worked with throughout the relevant period, irrespective of whether the NSCs were subsidiaries of the OEM or not; (ii) all NSCs that identified delivery charges as a line item cost had the same pricing strategy for setting the level of the delivery charge which was invariably implemented regardless of market conditions; (iii) NSCs that did not have itemised delivery charges (e.g. Ford, Nissan and Mitsubishi) nevertheless set overall prices for their vehicles in a manner which incorporated an implicit charge set in exactly the same way; and (iv) all retailers always passed on 100% of the charge to customers and they were never negotiated. KK adds that even the claimant's expert accepted in the report that there might be variations on these themes, but he did not then proceed to explain how the methodology would be adapted or adjusted to take such variations into account. KK contended before the CAT, by reference to witness statement evidence, that at all stages there was rigorous negotiation of prices so that the essential premise behind the claimant's analysis was false.
66. KK served a short expert report from Dr Majumdar of RBB Economics. He expressed the opinion that the methodology was defective because it made the assumption that there was 100% pass on as between OEMs and NSCs and no attempt had been made to estimate pass on by NSCs to dealers. He was of the view that an assumed 100% at this stage was "unrealistic". He also expressed the opinion that the methodology did not estimate pass on to the end consumer instead, once again, assuming a 100% pass through which was also unrealistic. He also decried as unrealistic the assumption that there was zero pass-on by the class. Finally, he said that all of these assumptions needed to be addressed at a more "granular" level.

Analysis

67. We reject the KK appeal essentially for the reasons given by the CAT (see paragraphs [29] and [30] above). Our conclusions can be briefly stated.
68. First, the argument collides with the conclusion of the CAT which was that the claim was not based upon extreme assumptions and was plausible based upon the actual evidence before the CAT which indeed included evidence from the KK, which corroborated aspects of the version of events described by the Class Representative (see the evidence set out at paragraph [38] above). In contrast the expert report of Dr Majumdar was not grounded in facts as was the evidence of the Class Representative and the CAT preferred this latter evidence.
69. Secondly, the approach to certification proposed by KK takes out of context observations in Gutmann. It ignores observations elsewhere in the judgment concerning the relevance of disclosure, the emphasis on issues rather than answers, the need for practical justiciability, and the relatively low standard required by threshold test (see e.g. paragraphs [55], [56], [60] and [61]). KK posits a test for certification which is too demanding.
70. So far as the gatekeeper role of the CAT is concerned, the intensity with which the CAT might scrutinise a methodology is different to the conclusion it might subsequently arrive at in relation to the height of the hurdle to be overcome. The criticism here is really no more than that the Class Representative should have been required by the CAT to surmount a higher hurdle than is required by law. In our judgment in relation to the level of detail set out in the methodology concerning silo pricing the CAT, having considered the evidence carefully and thoroughly, was well within its legitimate discretion to conclude that it was adequate.

F. The limitation issue

The issue

71. We turn now to the cross appeal. The Class Representative cross-appeals against the CAT's decision to exclude from the definition of the class of persons whose claims are included in the collective proceedings (the "Class Definition") natural persons who died before the Claim Form was issued on 20 February 2020.
72. The issue arises because the limitation period for bringing claims expired just after the Claim Form was issued. The relevant CAT Rule dealing with such a situation is Rule 38, which provides in relevant part:

"38.—(1) The Tribunal may grant permission to remove, add or substitute a party in the proceedings.

…

(6) After the expiry of a relevant period of limitation, the Tribunal may add or substitute a party only if—

(a) that limitation period was current when the proceedings were started;

and

(b) the addition or substitution is necessary.

(7) The addition or substitution of a new party, as the case may be, is necessary for the purpose of paragraph (6)(b) only if the Tribunal is satisfied that—

(a) the new party is to be substituted for a party who was named in the claim form by mistake;

(b) the claim cannot properly be carried on by or against the original party unless the new party is added or substituted as claimant or defendant; or

(c) the original party has died or had a bankruptcy order made against it and its interest or liability has passed to the new party."

73. The proposed Class Definition in the Claim Form was:

"All Persons (other than Excluded Persons) who during the period 18 October 2006 to 6 September 2015 either Purchased or Financed, in the United Kingdom, a New Vehicle or a New Lease Vehicle, other than a New Vehicle or New Lease Vehicle produced by an Excluded Brand.

For these purposes: Persons means all persons, whether legal or natural."

74. There was no dispute before the CAT that the claims of persons who purchased or financed a relevant vehicle between 18^th October 2006 and 6^th September 2015 ("the Relevant Period") and who were alive on 20 February 2020, but who died or die after that date whilst the proceedings are on foot, should fall within the scope of the collective proceedings. Accordingly, in the Order made by the CAT, the proposed Class Definition was amended pursuant to Rule 38(7)(c) so as to read:

"All Persons (other than Excluded Persons) who during the period 18 October 2006 to 6 September 2015 either Purchased or Financed, in the United Kingdom, a New Vehicle or a New Lease Vehicle, other than a New Vehicle or New Lease Vehicle produced by an Excluded Brand or, in the event such a Person has died on or after 20 February 2020, their Personal Representative."

"Personal Representative" was defined to mean the executor or administrator of an estate.

75. The dispute which is the subject of the cross-appeal relates to persons who acquired a vehicle during the Relevant Period but who died before the Claim Form was issued on 20^th February 2020.

The CAT Judgment

76. At paragraph [173] of the Judgment, the CAT held, referring to the analysis of a similar issue in the remittal judgment in Merricks v MasterCard Inc [2021] CAT 28, [2021] 5 CMLR 17, that although section 1 of the Law Reform (Miscellaneous Provisions) Act 1934 provides that the claims of a person survive their death for the benefit of their estate, claims cannot be commenced in the name of a deceased person after their death, and that any claim brought in such a way would be a nullity. That reasoning is well-established by authority: see e.g. Clay v Oxford (1866) LR 2 Exch 54 at 55.
77. Following Merricks, the CAT then held that this meant that claims in the name of persons who died before the commencement of the instant proceedings could not form part of the collective proceedings under section 47B CA 1998, because collective proceedings comprise a bundle of the claims which could have been brought individually under section 47A.
78. These conclusions were not challenged before the CAT or on appeal before us.
79. There was also no challenge on appeal to the CAT's conclusion (at paragraph [175] of the Judgment) that the proposed Class Definition could not be construed as it stood to include the personal representative of persons who had already died before the proceedings were commenced. On its plain reading, the proposed Class Definition only included persons who acquired a vehicle during the Relevant Period: a personal representative would not have done so in that capacity and there was no express reference to such a person.
80. The CAT then went on to consider whether there was power under the Rules of the CAT to add or substitute the personal representatives of parties who died before 20^th February 2020 as new parties to the collective proceedings under Rule 38. The relevant provisions of Rule 38 are set out above.
81. The CAT dealt with the argument concisely at paragraph [181] of the Judgment, focussing on the requirement that such addition or substitution after the expiry of a limitation period is only permissible under Rule 38(6)(b) if it is "necessary" as defined by Rule 38(7):

"181. Since the limitation period has expired, rule 38(6) must be satisfied. What is "necessary" is exhaustively defined by rule 38(7). In our view the [Class Representative] has not demonstrated that any of the provisions of that rule apply. It was not suggested that there was a mistake in the collective proceedings claim form within sub-paragraph (a). Both of sub-paragraphs (b) and (c) require there to be an "original party". But where a claim is brought on behalf of a person who is already dead, there is no such original party because claims brought on behalf of deceased persons are a nullity."

82. In its Respondent's Notice, the Class Representative contends, by way of cross-appeal:

"That was an error of law. Rule 38(7) is identical terms to CPR r.19.5(3)(b) and falls to be interpreted in the same way. There is clear authority (albeit only at first instance) that CPR r.19.5(3)(b) can cover a situation where the proceedings would otherwise be a nullity: see AIG Europe Limited v McCormick Roofing Limited [2020] EWHC 943 (TCC) at paragraph 41."

The case law

83. The case now relied upon by the Class Representative, AIG Europe Limited v McCormick Roofing Limited ("AIG"), was not cited to the CAT and did not feature in its decision.
84. The AIG case related to a fire on 31^st August 2013 which was allegedly caused by the negligence of roofing contractors at premises of which the leasehold owner was an English company, AIG Europe Limited ("Limited"). Before proceedings were issued against the contractors, and as part of the restructuring of the AIG Group's business in advance of Brexit, Limited entered into an agreement for a cross-border merger with AIG Europe SA ("AESA") an EU-based subsidiary of its parent company, pursuant to the Companies (Cross-Border Mergers) Regulations 2007 (SI 2007/2974) (the "CBM Regulations"). Under the terms of the merger agreement and in accordance with Regulation 17 of the CBM Regulations, Limited's cause of action in negligence was transferred to AESA as the surviving transferee company. Limited, as transferor company, was then dissolved and ceased to exist on 5^th December 2018.
85. In error, however, when proceedings were issued against the contractors on 27^th August 2019, shortly before the expiry of the limitation period, they were commenced in the name of Limited. The proceedings were not served until after the limitation period had expired, by which time the error had been noticed, and AESA applied to be substituted as claimant in place of Limited pursuant to CPR rule 19.5(3).
86. The application was heard by Roger ter Haar QC sitting as a Deputy High Court Judge, who permitted the substitution sought. He did so primarily on the basis of CPR rule 19.5(3)(b) (which is the equivalent of Rule 38(7)(b) of the CAT Rules). Although the Deputy Judge also thought that CPR Rule 19.5(3)(a) (which is the equivalent of Rule 38(7)(a) of the CAT Rules) dealing with the naming of a party by mistake might be applicable, as indicated above, the Class Representative does not suggest that Rule 38(7)(a) is applicable in the instant case.
87. At paragraphs 40 and 41 of his judgment, the Deputy Judge rejected the contractors' argument that CPR 19.5(3)(b) was inapplicable because Limited had ceased to exist at the time that the proceedings were commenced and hence the proceedings were a nullity. The Deputy Judge did not, however, explain his conclusion other than to state that:

"A situation where the proceedings would otherwise be a nullity is well within the situation in respect of which CPR Rule 19.5(3)(b) is intended to provide a remedy."

88. The basis for the Deputy Judge's view that such was the intention behind CPR Rule 19.5(3)(b) was not stated. The Deputy Judge had, however, referred earlier in his judgment to a number of authorities which it must be surmised were the basis for his decision. Those authorities were Parkinson Engineering Services PLC v Swan [2010] Bus LR 857 ("Parkinson"), Irwin v Lynch [2011] 1 WLR 1364 ("Irwin") and Insight Group v Kingston Smith [2014] 1 WLR 1448 ("Insight").
89. In Parkinson, the liquidator of a company caused a negligence claim to be issued in the name of the company against insolvency practitioners who had previously been appointed as its administrators. The defendants contended that the claim by the company was barred by a release that they had obtained when the administration order had been discharged under what was then Section 20 of the Insolvency Act 1986. In order to overcome that defence, the liquidator applied for himself to be substituted for the company as claimant, and for permission to amend the claim form to include a claim by him under section 212 of the Insolvency Act 1986. Section 212 is a procedural section that enables a liquidator to apply to the court using a summary procedure for an order to be made against (among others) a former administrator of the company who has been guilty of misfeasance or breach of duty in relation to the company. The liquidator's application was, however, made after the expiry of the limitation period in respect of some of the actions complained of.
90. In determining that this situation fell within CPR Rule 19.5(3)(b), Lloyd LJ stressed that the claim that the liquidator wished to assert under section 212 was precisely the same as that which the company already asserted in the proceedings. He took the view that the claim could not be carried on by the company because of the defence based upon the release, but that subject to permission being given, it could be carried on by the liquidator who was not bound by the release. He held that as a matter of jurisdiction, this fell within the scope of CPR Rule 19.5(3)(b). Lloyd LJ's reasoning appears from the following passage in paragraph [28]:

"28.  However, on the question of jurisdiction, whether it is open to the court to permit the substitution, it seems to me that this is a case in which the substitution is necessary in terms of section 35(5)(b) [of the Limitation Act 1980] as well as of CPR rule 19.5(3)(b). The original action, asserting the company's claim against the former administrators, cannot be determined without the substitution of the liquidator whereas if brought by the liquidator under section 212 it can. Without that substitution it could only, and would be bound to be, determined in favour of the defendants because of the section 20 defence. The claim would be struck out, because of that defence, and it could not be decided on its merits, either way, as the proceedings stand. In terms of the rule, it cannot properly be carried on by the original party, the company, whereas it can be maintained and carried on if the liquidator is substituted. No more than minimal change is necessary to the statement of case: substitution of references to the liquidator as claimant and references to the company in the third rather than the first person, so to speak, together with consequential changes as regards the relief sought. It is the same claim, in every respect, despite the fact that it is asserted by the liquidator on behalf of the company, rather than in the name of the company itself."

91. It is therefore readily apparent that Parkinson concerned a cause of action vested in a company which was still in existence and which had itself brought the proceedings within the limitation period. Nothing in the reasoning of Lloyd LJ bears upon a case in which a claimant had died or ceased to exist before the proceedings were commenced, or where the proceedings might otherwise be thought to be a nullity.
92. In Irwin, the administrator of a company issued an application, purportedly under section 212 of the Insolvency Act 1986, contending that a transaction under which, prior to administration, the company had done work on a property owned by the directors, had been a misfeasance and breach of duty by the directors for which they ought to be ordered to compensate the company. When it was pointed out by the defendants that administrators are not among the persons able to invoke section 212 of the 1986 Act, the administrator applied to amend to substitute the company itself as claimant. The application was met by the argument that the limitation period for a claim by the company had expired, and that CPR 19.5(3)(b) was inapplicable because the original claim was vitiated by the fact that the administrator lacked the necessary standing to assert it on behalf of the company. The first instance judge refused the application to substitute the company for the administrator and struck out the claim.
93. The Court of Appeal allowed an appeal. Lloyd LJ held, at [24]-[26]:

"24. It seems to me that the present case is one in which the substitution is necessary for the determination of the original claim because the particular claim cannot be maintained unless the company is substituted as claimant. The original claim is a claim that the directors were in breach of duty in causing the company to enter into the contract, thereby causing the company loss. The claim, as amended with the substituted claimant, is identical. The original claim cannot be maintained successfully; the new claim can be maintained successfully, subject obviously to proof of the facts. If it is so asserted, it is the identical claim but with a substituted and correct claimant.

25. It seems to me that some of what I said in [Parkinson], para 28 applies here with all the necessary consequential changes:

[Lloyd LJ quoted Parkinson (above)]

26.  Here the original claim was liable to be struck out, as it has indeed been, because of lack of standing, but I see no good reason to regard the reason for the striking out as being a critical distinction between that case and this. I would also reject the contention that the cause of action is not the same because of the identity of the claimant. Sometimes the identity of the party might be, indeed often it might be, a vital distinction, but here Mr. Irwin plainly asserted the company's cause of action and asserted it on behalf of the company, just as the substituted liquidator did in the Parkinson case. So the cause of action is identical; it is already pursued for the benefit of the company, but it is doomed to failure because of the lack on Mr. Irwin's part of the necessary locus standi. It seems to me that it is possible and appropriate for the court to exercise its discretion under rule 19.5 to allow the joinder of the company so as to assert the relevant claims."

94. Again, it is apparent that Irwin was not a case in which a claim was being asserted in the name of a non-existent person. The cause of action was that of an existing company. The problem was that the claim could not be continued because the person asserting the company's cause of action lacked standing to do so under the particular statutory procedure that he was attempting to use. Lloyd LJ considered that CPR rule 19.5(3)(b) could be used to overcome that problem so that the company's claim could be continued, but he did not treat the original proceedings as a nullity.
95. The third case was Insight. In that case, a claim in professional negligence had been made by a company against a limited liability partnership ("the LLP"). It was then realised that almost all of the allegedly negligent acts had been committed before the LLP had come into existence by members of a partnership ("the firm") who had advised the claimant company and whose business had been acquired by the LLP. The claimant applied under CPR 19.5(3)(a) to substitute the members of the firm as defendants in place of the LLP. By the time that the application came to be heard, the limitation period had expired in respect of some of the claims. The Master refused the application but an appeal was allowed by Leggatt J (as he then was).
96. The primary basis for Leggatt J's decision was that CPR Rule 19.5(3)(a) permitted substitution in such circumstances on the basis of mistake. However, Leggatt J also held (obiter) that on the basis of Parkinson and Irwin, substitution would in theory have been possible under CPR Rule 19.5(3)(b) if the cause of action to be asserted against the members of the firm had been the same as that asserted against the LLP, but that this was not so on the facts. Leggatt J explained his reasoning in paragraphs [96] – [97] as follows:

"96.  The principle which I derive from these two decisions of the Court of Appeal is that the court has power to order substitution under section 35(6)(b) [of the Limitation Act 1980] and CPR r 19.5(3)(b) if: (1) a claim made in the original action is not sustainable by or against the existing party; and (2) it is the same claim which will be carried on by or against the new party.

97.  Applying this test to the facts of the present case, it is common ground that the claims made in this action were unsustainable against the LLP. The first requirement was therefore satisfied. However, the second requirement was not satisfied, as the claims which the claimants sought to carry on against the firm were not the same claims as were made against the LLP. I have concluded earlier that the claims originally made against the LLP alleged that the LLP had been negligent in auditing the accounts of the second claimant and providing administrative and fiduciary services during the relevant period. In contrast, the claims asserted against the firm after the claimants had realised their mistake alleged that the firm (and not the LLP) acted as auditor and provided the relevant services. The new claims, therefore, allege different facts and are not identical to the original claims."

97. Like Parkinson and Irwin, Insight was not a case of a claim brought in the name of (or even against) a non-existent person. The claim was by an existing company against an existing LLP. The reason that Leggatt J would have refused permission for the members of the partnership to be substituted as defendants under CPR rule 19.5(3)(b) was because the claim against them was not the same cause of action as that which had been asserted against the LLP, and hence the substitution was not necessary in order to enable the original claim to be pursued. As with Parkinson and Irwin, Insight therefore provides no guidance as to the applicability of CPR rule 19.5(3)(b) in the case where the original claim is a nullity because it is brought in the name of a non-existent person.
98. For completeness, it should be noted that in AIG the Deputy Judge also referred briefly to the decision of the Supreme Court in Roberts v Gill [2010] UKSC 22, [2011] 1 AC 240 ("Roberts").
99. His first reference was to the dictum of Lord Walker at paragraph [104] that:

"In the ordinary case of a simple assignment or transmission of a cause of action after proceedings have been commenced, no question of limitation arises."

That statement is, however, of no assistance where the issue arises because the assignment or transfer of a cause of action in question has taken place before the commencement of proceedings.

100. The Deputy Judge's second reference to Roberts was in rejecting an argument by the contractors' insurer (which had been joined to the application as an interested party) to the effect that CPR rule 19.5(3)(b) could not apply because Limited had no original claim, having assigned it to AESA. The Deputy Judge rejected the argument on the basis that the Supreme Court had recognised in Roberts:

"… that an assignment is very much the situation where rule 19.5(3)(b) will apply."

101. It is certainly true that in setting out the background to the enactment of Section 35 of the Limitation Act 1980 and the rules of court that implement it, Lord Collins remarked, at paragraph [2] of Roberts, that one of the objectives of the legislation was:

"… to enable parties to be added out of time, in cases where joinder of the new party was necessary if the plaintiff's claim was to succeed, for example where the plaintiff was an equitable assignee and had omitted to join the assignor prior to the expiry of the limitation period."

102. However, that is a situation very far removed from the case before the Deputy Judge. The example identified by Lord Collins was not one in which the assignor had ceased to exist, and it was very different from the facts of AIG in which Limited was the assignor, and its right of action had been transferred to AESA in accordance with the CBM Regulations. In such a case, the only appropriate and necessary party to the bringing of a claim was AESA as legal assignee.

Analysis

103. Accordingly, in our judgment, the decision in AIG cannot be supported on the basis of the authorities to which the Deputy Judge referred. In our view, it was wrongly decided and should not be followed. The decision in AIG therefore casts no doubt upon the correctness of the Judgment that Rule 38 of the CAT rules does not permit the inclusion in the collective proceedings of claims by the personal representatives of persons who died before the Claim Form was issued on 20^th February 2020.
104. The cross-appeal is therefore dismissed.

D. Conclusions/disposition

105. It follows from the above that:

(i) The MNW appeal is refused to the extent that it seeks to have the claim struck out or otherwise dismissed.

(ii) The claim is remitted to the CAT in order that it might reconsider case management of the dispute between the parties as to the approach to pricing.

(iii) In relation to the Respondent's Notice paragraphs [121] – [124] of the Judgment are set aside.

(iv) The KK appeal is dismissed.

(v) The cross-appeal of the Class Representative is dismissed.