/Passle/MediaLibrary/Images/2026-01-09-17-39-34-817-69613d561e42b4e613742f58.jpg)
Second bite of the Apple: CAT partially approves certification of failed Riefa claim
8 minute read
The Competition Appeal Tribunal has approved part of a collective proceedings originally brought by Professor Riefa against Apple and Amazon, which has now been revived by Mr Le Patourel.
Background
As we reported in our earlier article, when this claim first came before the Tribunal (the Riefa Claim), the Tribunal refused to grant a collective proceedings order (CPO) on the grounds that the authorisation condition was not met (in particular that the PCR had not “demonstrated sufficient independence or robustness so as to act fairly and adequately in the interests of the class”). At that time, Professor Riefa was the sole director of the SPV which was acting as the proposed class representative (PCR).
Following the Tribunal’s refusal to certify the Riefa Claim, the PCR changed its name, and Mr Le Patourel was appointed as its sole director instead. In this form, the PCR made a fresh application for a CPO to bring the proceedings against Apple and Amazon (the JLP Claim).
The claim
The claim alleges that Apple and Amazon entered into a series of agreements which had the object or effect of preventing, restricting or distorting competition by:
excluding from Amazon all resellers of Apple products, other than a limited number of Apple-authorised resellers based on criteria which are neither objective nor non-discriminatory; and
limiting the ability of Apple’s competitors to purchase advertising space on the Amazon marketplace to advertise their own products, as well as imposing other advertising restrictions.
Those agreements are said to have caused prices to rise on Apple products purchased through Amazon’s marketplace (On-Amazon) and through other retail channels (Off-Amazon) causing loss to the proposed class members.
Strike out/summary judgment, the Microsoft test and Opt-Out
A key battleground was whether the Off-Amazon claim should be:
struck out/summarily dismissed; or
refused certification on the basis that it did not pass the “Microsoft test” (or, alternatively, not certified on an opt-out basis).
The Tribunal found that it would be premature to strike out/summarily dismiss the Off-Amazon claim. To fully determine the issue, there would need to be an assessment of complex factual and economic evidence which was not available at certification. Instead, the Tribunal only had before it a preliminary consideration of the evidence that might be available at trial.
However, the Tribunal went on to refuse to certify the Off-Amazon claim on the basis that it did not pass the Microsoft test. Rather than providing a clear blueprint to trial, the Off-Amazon claim was found to rest on a “complex and speculative” theory of harm and the PCR’s expert had not established a “clear and coherent methodology” grounded in the facts of the case, with at least some evidence of the availability of data that would be required.
What is the “Microsoft test”?
The “Microsoft test” originated from Canadian case-law, and was adopted in the UK in Mastercard v Merricks [2020] UKSC 51. In short, the Tribunal will only approve certification if:
- the expert methodology offers a realistic prospect of establishing loss on a class-wide basis;
- the expert methodology is more than purely theoretical and grounded in the facts of the particular case; and
- there is some evidence of the availability of the data that the methodology requires. The test has been considered and refined in several cases since the Merricks decision, as set out in paragraph 51 of this judgment.
The Tribunal also had significant concerns as to the independence of the PCR’s expert. By way of background, the expert had been “closely involved in instigating the case” and had approached the PCR’s lawyers to propose a UK claim. Whilst this did not “inevitably” mean that the expert could not fulfil their duties to the Tribunal in these collective proceedings, it did mean that the expert should have been “particularly mindful” of such duties.
Apple and Amazon had disclosed sales data, which they said showed the share of UK sales of new Apple products made via Amazon was so low that it could not “conceivably have had an impact on the Off-Amazon pricing of Apple products”. Rather than engaging objectively with the disclosure of this new data and Apple and Amazon’s comments in respect of it, the Tribunal found that the expert instead relied “heavily on assumptions and assertions” without undertaking any empirical analysis to test the hypothesis for the Off-Amazon claim.
The Tribunal elected to use its powers to order the disclosure of the instructions to the expert, and ultimately placed limited weight on his third expert report which dealt with this issue. In contrast, the Tribunal found that the theory of harm for the On-Amazon claim was credible and the expert had provided “a preliminary empirical analysis which is consistent with that head of loss”.
Having refused to certify the claim, the Tribunal did not need to consider Apple and Amazon’s alternative case that the Off-Amazon claim was not strong enough to be certified on an opt-out basis.
Distribution
In light of the Tribunal’s prior experience in the Gutmann v First MTR and Waterside case (summarised in our respective articles: Gutmann and Waterside), the Tribunal emphasised the importance of a “simple and effective” distribution method being canvassed at the certification stage. For the On-Amazon claim, the Tribunal was satisfied that Apple and Amazon would likely have the requisite data to facilitate effective distribution by way of direct credits to class members. In contrast, the Tribunal was not convinced that there would be significant take-up in the Off-Amazon claim, given that this seemed likely to require more substantive class member engagement. This further reinforced the Tribunal’s decision not to certify the Off-Amazon claim.
Cost-benefit analysis
The Tribunal paid close attention to the overall cost/benefit analysis in the context of its decision whether to grant a CPO. The consideration of the cost/benefit analysis as a significant factor at certification has been a notable recent trend - see, for example, our article on the refusal to certify the Waterside claim for this reason.
In the Riefa Claim, the Tribunal had found that the PCR should bear the costs of its unsuccessful CPO application. Nonetheless, the JLP Claim funding arrangements enabled the funder, insurers and legal advisers to be paid Riefa Claim costs (the inclusion of which, taking into account the funders’ return on these costs, totalled c.£36m). The Tribunal found that this was “manifestly unreasonable and inappropriate” and “sufficient in itself” for the Tribunal to refuse certification unless addressed.
On the basis that the parties would remove Riefa Claim costs from the funding agreement package (by way of formal contractual amendments), the Tribunal concluded that the benefits outweighed the costs of the litigation. Whilst not a decisive factor, the Tribunal noted in this context that the PCR was also now seeking injunctive relief, which would bring the ongoing alleged anti-competitive agreements to an end. This was a relevant matter to take into account in the cost-benefit analysis because, even though injunctive relief would not involve a payment by Apple and Amazon to the class members, it would put to an end any ongoing breach of competition law if the claim were successful at trial.
Authorisation of the PCR
Apple and Amazon submitted that Mr Le Patourel’s decision to include the Riefa Claim costs in the funding arrangements called into question his judgment, with the potential implication that the PCR should not be authorised as a class representative.
The Tribunal did not consider Mr Le Patourel’s evidence of his attempts to negotiate the removal of these terms from the funding package was “entirely compelling”. This was particularly given that the stakeholders agreed to remove the Riefa Claim costs within two weeks of the CPO hearing once the Tribunal had criticised them.
However, the Tribunal found that it was not unreasonable for Mr Le Patourel to take the view that the Tribunal could refuse to permit the Riefa Costs at a suitable stage in proceedings, even though he had not personally managed to negotiate this.
The Tribunal also rejected Apple and Amazon’s submissions that Mr Le Patourel had not given sufficient consideration to instructing a fresh expert and legal team.
The Tribunal therefore authorised the PCR to act as class representative.
PCR remuneration
Building on an analysis of the recent case-law (in Waterside Class v Mowi [2026] CAT 32 and Or Brook v Alphabet [2026] CAT 65), other comparators submitted by the parties, and a survey of class representatives, the Tribunal concluded that it would not be reasonable for the PCR to charge more than £170 per hour, and imposed a daily cap of £750 (with it being reasonable for these amounts to increase annually in line with inflation). Going forward, the Tribunal made clear that it would expect the PCR’s hourly rate and any daily cap to be published on the relevant claims website.
Key takeaways
This judgment serves as a practical reminder of the differences between the strike out/summary judgment test, and the Microsoft test. A claim may survive the strike out/summary dismissal threshold but nevertheless not be certified due to failing the Microsoft test. The distinction is relevant because if the Off-Amazon claim had been struck out/summarily dismissed that would have put an end to it. However, because certification of that part of the claim was refused on the basis the Microsoft test was not met, the PCR (or another proposed class representative) could potentially still revive this aspect of the claim if they can formulate a revised and improved expert methodology. Any re-formulated claim would also need to respond to the Tribunal’s criticisms of the proposed distribution for the Off-Amazon claim.
At the certification stage, the Tribunal continues to be interested in cost-benefit and ensuring that collective proceedings are primarily initiated for the benefit of class members, not funders and lawyers. This tension has also been identified by the Tribunal in recent collective settlement judgments (see, for example, Consumers’ Association v Qualcomm Incorporated [2026] CAT 50).
The Tribunal’s focus on the PCR’s remuneration is part of a wider trend of the Tribunal paying close attention to costs, with the Tribunal considering a wide range of comparators in its judgment. It remains to be seen whether any broader consultation exercise is carried out on this topic, and whether the limits placed by the Tribunal in this case have any impact on the willingness of potential class representatives to act more generally.
The judgment also shows that, following Practice Direction 3/2025 (Expert Evidence), the Tribunal continues to focus on the need for experts to be independent, and is prepared to place little weight on expert reports where there is evidence that the expert has shown a “lack of objectivity”.
Authors
Related topics
Like what you are reading?
Stay up to date with our latest insights, events and updates – direct to your inbox.
How can we help you?
Browse our people by name, team or area of focus to find the expert that you need.