Potentially the biggest and most complex claim in British legal history, according to some commentators, the Competition Appeal Tribunal (“CAT”) declined to certify consumer claims against Mastercard as eligible for inclusion in opt-out collective proceedings. Permission to appeal the decision has recently been refused. The claim sought damages of approximately GBP 14 billion arising indirectly in relation to overcharging by Mastercard of multilateral interchange fees, which it was alleged were passed on by merchants to consumers, by charging higher prices for goods and services, from 1992 – 2008.

In this article we analyse the implications of this judgment.


Mastercard operates a four party card payment scheme. A cardholder presents a card to a merchant, and details of the transaction are passed by the merchant to its acquiring bank and then to the cardholder’s issuing bank. The issuing bank transmits payment to the acquiring bank, less a transaction fee known as the interchange fee (“IF”). The acquiring bank deducts the IF, along with its own fee from the payment that it then makes to the merchant. In the absence of alternative agreement, the IF, between the acquiring and issuing bank, will default to the multilateral interchange fee (“MIF”) set by Mastercard.

The EU Commission decided in December 2007 (which was upheld by the EU Court of Justice) that the setting of the EEA MIF (an MIF applying where a card is issued in one EEA Member State and used in another Member State) by Mastercard was in breach of competition law, and in effect set a minimum price that merchants had to pay their acquiring bank for accepting Mastercard branded cards.

In this case the applicant Walter Merricks (a former FOS chief ombudsman), who was seeking to be the class representative, claimed that the higher EEA MIF caused a higher UK MIF (where a card issued in the UK is used to pay a merchant in the UK), that the higher fee was fully passed to merchants (which was not in issue), and that the merchants passed through this increase in retail prices charged to customers. The class of claimants was defined as all those individuals who purchased goods or services from merchants that accepted Mastercard (not just those who used a Mastercard as it was said the pass-through of the charge would affect all of the merchants’ products) but excluded those under 16 (on the basis they were unlikely to be spending their own money), those not resident in the UK for at least 3 months, and those no longer alive.

UK Competition Law Regime

The Consumer Rights Act 2015, in force from 1 October 2015, amended the Competition Act 1998, substituting a new s47A and s47B to give rights to individuals to bring private damages actions and to allow authorised class representatives to bring opt-in or opt-out collective actions on their behalf in the CAT. The question before Mr Justice Roth was whether a collective proceedings order should be made in the Mastercard case. An order can be made by the CAT under s47B Competition Act if:

1. the CAT considers that the person who brought the proceedings is a person who the Tribunal could authorise to act as a representative; and

2. the claims are eligible for inclusion in proceedings (i.e. that they raise “the same, similar or related issues of fact or law and are suitable to be brought in collective proceedings”). The Competition Appeal Tribunal Rules 2015 (the “CAT” rules) provide that the CAT may certify claims as eligible for collective proceedings where the claims: a. are brought on behalf of an identifiable class of persons;

The Competition Appeal Tribunal Rules 2015 (the “CAT” rules) provide that the CAT may certify claims as eligible for collective proceedings where the claims:

a. are brought on behalf of an identifiable class of persons;

b. raise common issues; and

c. are suitable to be brought in collective proceedings. 

In determining whether the claims are suitable the CAT can take into account all matters it thinks fit including: whether collective action is an appropriate means for the fair and efficient resolution of the common issues; the costs and benefits of continuing the collective proceedings; whether separate proceedings have been brought by members of the class; the size and nature of the class; whether it is possible to determine whether any person is a member of the class; and whether the claims are suitable for an aggregate award of damages.

In July 2017, the CAT held that the correct approach to apply when considering the expert evidence adduced in support of the stated common issues was that of the Canadian courts. In particular, the Canadian case Pro-Sys Consultants Ltd v Microsoft Corp (2013) was referred to. In that case, it was held that the expert methodology must offer a realistic prospect of establishing loss on a class-wide basis, so that if loss is established at a trial of the common issues there is a means by which to demonstrate that it is common to the class. 

Suitability for collective action

The CAT found that, although there were a number of issues in the claim that were common to all claimants, two issues were very different: (i) how much each merchant passed through the overcharge to its customers; and (ii) the amount that each claimant spent at these merchants. 

A key finding of the CAT was that there was no requirement that all significant issues in the claim should be common or that the common issues should predominate (in contrast to the position in the US). However, in view of the difficulty in determining the two issues in question, the CAT was not satisfied that the claims were suitable for an aggregate award of damages. 

The claimant’s submissions, that the expert evidence demonstrated that difficulties regarding the pass-through issue could be dealt with by applying a methodology to calculate the overcharge and a weighted average passthrough percentage, were rejected. The court also found that it was impossible to see how the payment of damages to an individual claimant from the aggregate damages could be determined on any reasonable basis. The individual’s level of expenditure, merchants from whom they purchased and the mix of products purchased over the relevant period would all be relevant and there was no plausible way of reaching even a rough and ready approximation of loss suffered by an individual.

Authorisation of the class representative

Although it was not strictly necessary to decide this issue, the CAT went on to consider Mastercard’s submissions that the applicant should not be authorised as the class representative. Mastercard argued that the litigation funding agreement that the applicant (Walter Merricks) had entered into meant that he would not be able to fund the litigation or pay Mastercard’s costs if ordered to do so as the agreement could be terminated by the funder; the limit of GBP 10 million for funding Mastercard’s costs was inadequate; and that the terms of the funding agreement gave rise to a conflict of interest.

The CAT rejected these arguments, following an amendment made to the funding agreement. The CAT found that a payment that has to be made to a third party funder in consequence of its funding commitment did constitute a cost or expense incurred in connection with proceedings. It found that commercial returns for the funder can be recovered from residual funds left unclaimed from the damages, and such funding returns were recoverable where the applicant’s obligation to pay funding costs is contingent, provided the obligation was a direct liability on the applicant. 

Also rejected were Mastercard’s submissions that the applicant was subject to a conflict of interest, as there was an obligation to ensure there was a sufficient amount of unclaimed damages for the funder to receive its “Total Investment Return” in conflict with the interests of the class to maximise the amount of damages recovered and distributed. The CAT found that, although a term in the funding agreement requiring the applicant to use his best endeavours to distribute the proceeds to the class would have been desirable, given the powers of the CAT and the applicant’s evidence that he would act in the best interests of the class there was no realistic prospect that the applicant would be constrained from acting.


Permission to appeal the decision was refused by the CAT on 28 September 2017. The CAT found that the legislation provides no route of appeal from the refusal to certify the class. The CAT was of the view that this reflected a deliberate policy, on the basis that experience from other jurisdictions, in particular the United States and Canada, is that decisions refusing or allowing a class to be certified typically generate appeals. Therefore, the CAT was of the view that in order to craft an effective system of collective redress for the UK, the legislature had restricted the procedure to a specialist tribunal and confined the right of appeal to decisions on the substantive claims thus precluding prolonged litigation. Although not strictly necessary, the CAT also went on to comment that it would have refused permission in any case. This was for a number of reasons. In particular the CAT stated that its finding had not been that it was necessary for the applicant to assess the loss suffered by individual members of the class. The introduction of a collective proceedings regime had not changed the fundamental nature of damages for breach of competition law as being compensatory. The applicant had been unable to propose a method of distribution of the aggregate award of damages that, even on a rough and ready basis, would lead to payments on a compensatory basis.


The case was seen as a test of the class action regime, and is the second heard by the CAT since the rules were amended (the first, relating to mobility scooters, also failed to be certified as a collective action). The case also fits a trend we have seen in the UK for increasing involvement of US funders and lawyers in UK litigation, and the US plaintiff bar is of course much more familiar with a class action regime. 

The outcome in the case is due to the particular difficulties that arose in assessing individual damages, and it is of interest that the CAT made reference to Canadian case law.

In Canadian class actions (subject to some nuances between provinces), plaintiffs must put forward a “credible or plausible” methodology for addressing class wide issues of loss, which is typically an expert methodology in consumer claims. This must establish “some basis in fact” to prove the existence of a common issue regarding loss on a classwide basis, a standard which is lower than the balance of probabilities. Although this cannot be purely theoretical or hypothetical if there is conflicting evidence, the matter is one that is suitable for trial, and certification is not generally held to be an obstacle to the claim proceeding. Courts in Canada are flexible about certification of class actions where damages may require individual assessment, as specifically contemplated in the legislation. Canadian class proceedings are generally interpreted purposefully to promote the efficient management of legal proceedings and access to justice. Where the courts are dealing with a large number of claims of modest value, which are unlikely to be suitable for individual litigation, access to justice tends to be strongly promoted. 

In the Mastercard case, the CAT noted that it had been argued that as it was totally impractical for members of the class to bring individual claims without a collective proceedings order, a vast number of individuals who suffered loss would get no compensation. However, the CAT stated that this was the case in most cases of widespread consumer loss following competition infringements, and did not mean an application to bring collective proceedings would always be granted, as each case had to be considered on its own terms.

It will be interesting to see the effect of the decision, and whether it has a serious dampening effect on other claims following this route. Indeed the CAT collective action regime has been mooted as a potential option for dealing with foreign exchange rate benchmark manipulation cases. There is no doubt that funders and claimant lawyers will be scrutinizing potential claims very carefully, in light of the outcome in this case.  

Funders will note the acceptance of the funding arrangements in this case. The CAT considered debate in the House of Lords on whether funding should be permitted in collective actions, noting that the Government clearly intended many collective actions would be dependent on funding, and this could not be achieved unless the class representative incurred a conditional liability for the funder’s costs that could be discharged from unclaimed damages. There are some lessons as to relevant terms that funders might wish to include in agreements, including the criticism of the convoluted terms of a funding agreement concerning consumers.

Assens Havn v Navigators: Court of Justice of the European Union holds that third party bringing a direct action against a marine insurer is not bound by a jurisdiction/choice of law agreement between the insurer and insured

Under Regulation 44/2001 (which has since been replaced by Regulation 1215/2012, but the relevant provisions in this case have remained the same), special jurisdiction rules apply to insurance (but not reinsurance) contracts. Broadly, an insured can only be sued in the place of his domicile. However, the insured can sue its insurers in its own domicile, or that of the insurers, or in the place of the loss (usually that position applies even if there is a valid jurisdiction clause, but there are some exceptions, one of which is referred to further below). 

The Regulation further provides that an injured third party which is allowed under local law to bring a direct action against an insurer is also permitted to sue the insurers, in his own domicile, or that of the insurers or in the place of the loss. Of issue in this case was whether a valid jurisdiction clause in the insurance contract could override that position in respect of the injured third party. The Court of Justice of the European Union has now held that it cannot.

The facts of the case are that a Swedish charterer took out liability insurance with a UK insurer. The vessel caused damage to the Port of Assens in Denmark, which sought to bring a subrogated claim against the insurer under Danish law when the Swedish charterer/insured went into liquidation. The insurance policy contained an English choice of law and jurisdiction clause and the issue was whether the Danish courts nevertheless could hear the claim brought by the Port of Assens.

The fall-back provisions regarding jurisdiction and insurance referred to above can be departed from by (amongst other things) a jurisdiction agreement which relates to a policy which covers risks set out in Article 14 of Regulation 44/2001 (which includes “any liability…. arising out of the use or operation of ships…”). However, it was held that that article does not apply to direct action claims brought by an injured third party – it only applies to actions between the insurer and the insured. Hence the third party was not bound by the English jurisdiction clause: “The view must therefore be taken that an agreement on jurisdiction made between an insurer and an insured party cannot be invoked against a victim of insured damage who wishes to bring an action directly against the insurer before the courts for the place where the harmful event occurred… or before the courts for the place where the victim is domiciled”.

The Third Parties (Rights against Insurers) Act 2010 is the UK equivalent of the Danish Act which gave rise to the third party direct action against insurers in this case. It applies where the insolvency procedure takes place under the law of one of the parts of the UK. However, jurisdictional issues regarding where that claim should be brought are governed by Regulation 44/2001 or, now, Regulation 1215/2012. Although marine and aviation and “large risks” insurers can in certain circumstances contract out of the jurisdictional rules laid down by the Regulations in their policies, this case confirms that that contracting out will not affect the ability of an injured third party, which can bring a direct action against insurers to rely on those jurisdictional rules. Whilst (as the Court pointed out) the third party never directly contracted out of jurisdiction in the first place, the decision is noteworthy because the third party’s claim is still a subrogated/assigned claim and usually third parties can have no better rights than the insured into whose shoes they step. Nor will insurers be able to protect themselves against this risk, in the absence of the third party’s agreement.