This article, available here, provides critical assessment of the UK’s emerging collective certification process. It argues that the Competition Appeal Tribunal has applied the test for certification too strictly and not in accordance with the case law surrounding the ‘Canadian model’ on which the UK certification procedure is based; and incorrectly treated the award of aggregate damages as the summation of individual damages. It also argues that the way the CAT has handled these two factors threatens to undermine the purpose and effectiveness of the UK’s new collective action regime.

Show me the money

The piece is structured as follows:

Section 2 provides an overview of the collective certification requirements.

Prior to the Consumer Rights Act 2015, which amended the Competition Act, potential litigants had limited opportunities to bring a group action against a common defendant. One might try to rely on group litigation orders – a case management device that allows a court to manage separate claims which share ‘common or related issues of fact or law’ –, representative actions – which can be brought by a representative claimant when more than one person has the ‘same interest’ in a claim – or on opt-in collective actions on behalf of consumers. In 2010, the Civil Justice Council produced a report which recommended that the UK Government introduce a better means for multiparty litigation. In 2013, the UK Government announced that it would introduce legislation to allow collective actions on the opt-out class action model. In 2015, section 47B of the UK Competition Act was amended to create such a procedural mechanism.

Under this new regime, a class representative must, in order to bring an opt out action, obtain certification from the CAT via a collective proceedings order (‘CPO’). This order ensures that the members included in the group have sufficient ‘common’ interest to constitute a suitable class of claimants. To obtain a CPO, the class representative must satisfy the Tribunal that the claims: (a) are brought on behalf of an identifiable class of persons (‘eligibility’); (b) raise ‘common issues’ (‘commonality’): and (c) are ‘suitable’ to be brought in collective proceedings (‘suitability’). The CAT Rules elaborate on the elements of ‘commonality’.

Section 3 reviews two failed certification applications before the CAT.

Up until the date when this article was published, there had been two unsuccessful CPO applications: Dorothy Gibson v. Pride Mobility Products Limited and Walter Merricks CBE v. MasterCard International. In both, the CAT said it would follow the ‘Canadian approach’ according to which: (1) ‘the expert methodology must be sufficiently credible or plausible to establish some basis in fact for the commonality requirement’; and (2) ‘There must be some evidence of the availability of the data to which the methodology is to be applied.’ The Canadian approach contrasts with class certification in the US under which, according to Federal Rules of Civil Procedure (Rule 23), the courts are required to conduct a rigorous analysis of the evidence and resolve all legal and factual disputes before they certify a claim, even if this involves a preliminary merits review.

Nonetheless, the two CPO applications were very different in scale, and the reasons why they failed were also distinct. Dorothy Gibson v. Pride Mobility Products was a follow on action from an OFT infringement decision. The OFT found that Pride Mobility Products had illegally restricted its retailers from advertising their products online. However, while the Gibson claim covered all Pride’s retailers, the OFT’s decision only related to a sample of eight of Pride’s estimated 250 to 300 retailers which had been selected by the OFT ‘for convenience’ as test cases or examples of the alleged infringement. The Tribunal rejected Gibson’s inclusion of all Pride retailers in the claim. While The Tribunal gave the applicant leave to amend its claim, Gibson abandoned the claim as the damages (circa. GBP 300k) for the smaller group was insufficient to justify any further legal expense.

As we saw last week, Merricks v MasterCard was a follow on action based on the European Commission’s MasterCard decision. In the UK, this led to a number of retailers bringing damage actions against MasterCard and Visa. Against this backdrop, Mr Merricks brought a claim for GBP 14 billion on behalf of 46.2 million UK consumers for excessive UK domestic default MIFs. The CAT dismissed the application because a) the test for commonality had not been satisfied; and b) the proposed method of distributing damages to members of the class was not compensatory. This decision was appealed to the Court of Appeal and, subsequently, to the Supreme Court.

At both CPO hearings, the only witnesses called to give oral evidence were the economic and accounting experts, who were subject to intense questioning. Only after both hearings were the CPO decisions adopted, which revealed that the CAT had adopted the Canadian approach – but with a twist. The general perception is that Canadian class certification is permissive, imposing a low evidential threshold and not requiring common issues to outweigh individual issues. The CAT focused instead on the absence of sufficient evidence and on the failure of the proposed methodologies to identify individual harm.

A related section is concerned with the appeal of the Merricks v MasterCard decision.

In deciding whether to grant a CPO in this case, the CAT accepted that a number of issues were common to all the claimants. This included:

– the level of the EEA MIFs, which had a common effect on the level of the UK MIFs (for both Mastercard credit and debit cards);

– the amount by which those MIFs were higher than the counterfactual interchange fees that would have applied in the absence of an infringement, which could be treated as a common blended average; and

– the existence of full pass on of the MIFs in the Merchant Service Charge charged by Acquiring Banks to the merchants where the claimants bought goods and services.

On the other hand, the CAT did not accept that the degree of passing on of overcharges and their impact on prices for each merchant at which the claimant purchased goods and services was a common issue.

Further, the CAT found that the experts’ report did not adequately address the availability of data and doubted that there existed sufficient data to undertake an economy-wide assessment of pass on. However, while the adequacy of data is a valid concern, it is a concern present in all competition damage actions, where it is extensively investigated – but only at trial, when disclosure is complete. It is unclear how claimants are supposed to meet this test prior to disclosure and evidence production.

Another ground for the CAT’s rejection of Mr Merricks’ application was that the proposed method of distributing damages was not compensatory. Mr Merricks proposed a uniform per capita award to each member of the class. It would have also been prohibitively expensive and contrary to the purpose of a collective action to have to estimate each individual claim at the certification stage. The CAT considered that this method was not acceptable because it would infringe the principle of compensation for harm suffered.

According to the author, the CAT’s position contradicts the legal framework. Firstly, the legislation envisages that the award of aggregate damages will not be based on the aggregation of the individual losses of each member of the class. Secondly, the certification requirements do not take into account how the award of aggregate damages was to be disbursed to members of the class, let alone make it a determining factor. In this way, this decision revealed the inherent tension between individual and aggregate damages which affects these proceedings. The purpose of a collective action is to aggregate damages which would be too expensive to pursue individually. The loss to individual customers depends on their purchasing patterns, the merchants they bought from, and whether or not they were cardholders. Thus, the CAT’s requirement that the individual disbursement of damages be compensatory raises huge evidential problems in a mass consumer action.

Section 4 addresses the costs of certification. This is remarkably technical and of little interest from a competition law perspective, so I will not review it here. Do look it up if you are curious about the costs that pursuing these type of (preliminary) actions involve.

Section 5 deals with funding.

Collective actions are unlikely to be funded by the class representative or the class. Instead, they will be funded by others – by lawyers through contingency fee arrangements, third-party litigation funders and after-the-event insurance. These funding arrangements are critical to the success of the consumer collective redress regime.

The two CPO applications were funded in very different ways. In Gibson, the solicitors and counsel commuted their fees under contingency fee arrangements, and adverse costs and experts’ fees were covered by after-the-event insurance. In Merricks, the claimant was well-funded by a third-party litigation funder by up to nearly GBP 43.5m. The potential ‘investment’ return to the litigation funder from any undistributed proceeds and cost awards was to be: ‘equal to the sum of: (a) the greater of (i) £135,000,000; or (ii) 30% of the undistributed proceeds (i.e. the damages award amounts not claimed by victims) up to £1 billion, plus 20% of the Undistributed Proceeds more than £1 billion; plus (b) late payment interest, if any’.

This last element of the funding arrangement raised some controversy, since the default position is that undistributed damages will be paid to charity, currently the Access to Justice Foundation, whose stated aim is to facilitate access to pro-bono legal assistance for those most in need.’ The CAT nonetheless clarified that the funder was entitled to a return from the undistributed damages, and expressed no opinion as to what was a ‘reasonable’ return.


The author was the expert for the claimant in the Merricks case. As such, he is undoubtedly waiting eagerly for the Supreme Court’s decision. At the same time, that position undoubtedly influenced this paper, which reads as an argument for the Court of Appeal to overturn the CAT’s decision.

In my comment on the original CAT decision, which can be found here, I noted that the creation of a class reflects a decision to emphasise some commonalities of the members of that class and to ignore some of their individual characteristics. To some extent, the advantage of collective actions is that they permit a number of different individual claims to be brought under the same umbrella, without requiring evidence to be provided regarding each and every claim. However, this may come at the cost of losses being calculated by reference to the group and, hence, of damages being awarded that do not exactly reflect individual losses. In that comment, I pointed out that I thought it likely that courts would have to balance the benefits brought by collective actions against the individual interests which such actions will necessarily ignore.

The concept (and legal test) of commonality is thus often a crucial element in the decision to allow a collective action to proceed. For example, while there may be a presumption (based on economic theories) that direct purchasers of a cartelised product will have paid an overcharge, the question of whether a collective action can be brought on behalf of all of them will depend on the legal standard for commonality. Maybe commonality of interests arises merely because claimants were all direct purchasers; or it may be further required that there is an underlying structure underpinning differentiated prices, so that the harm suffered by class members can be identified via a common method or by reference to common evidence; or it may be required that the common characteristics of a class are such as to allow an approximately correct allocation of the damages according to individual loss. Furthermore, courts may pursue inquiries with different degrees of intensity regarding whether the adduced evidence, and economic methods and models, will allow harm to be proved by reference to class-wide evidence.

In other words, the issue of commonality is ultimately not merely about the identification of common elements between the victims. Instead, the selection of what elements are to be used to determine whether individual claims share enough commonalities is ultimately about balancing the compensatory (or deterrent) goals pursued by the tort system with the limitations that may be introduced to this system in order to ensure the effectiveness of damages claims. The identification of a threshold at which the disparity of individual interests is such as to override the commonalty of claims – and hence, to prevent a class from being certified – will ultimately depend on this balancing exercise.

This article focuses a great deal more on the challenges that the CAT’s approach creates for would-be claimants – including regarding funding their actions – than on these questions. These challenges are real, and are important to bear in mind; so much so that the Court of Appeal broadly agreed with the position outlined in this paper. However, there are other interests as well, including one not mentioned by the author – that of preventing frivolous or unmeritorious litigation, and the development of a ‘litigation culture’ – which are also important.

The balance between the various interests at stake was, to a large extent, set by the legal reform creating opt-out collective claims. I am personally very interested in seeing how the Supreme Court decides that such a balancing should be pursued at a more detailed level.


Author Socials A weekly email with competition/antitrust updates. All opinions are mine

What do you think?

Note: Your email address will not be published