Le Patourel v BT Group PLC (No 2):
The Court of Appeal clarifies the evidential burden in excessive-pricing cases and the limits of appellate intervention

1. Introduction

This commentary examines the Court of Appeal’s decision in Le Patourel v BT Group PLC & Anor ([2025] EWCA Civ 1061) – “Le Patourel II”. The judgment refuses permission to appeal against the Competition Appeal Tribunal’s (“CAT”) 2024 dismissal of a £600 million collective action alleging that BT over-charged “stand-alone fixed voice” (SFV) customers. Central to the appeal were:

• Whether the CAT misapplied the Cost-Plus test (Limb 1) and mis-evaluated justifications for the price differential (Limb 2).
• The extent to which appellate courts may draw or compel adverse inferences from a dominant firm’s failure to adduce cost evidence.
• The weight that may be given to expert evidence built on imperfect datasets, and
• Whether the CAT’s approach to “economic value”, switching, “distinctive value”, brand, and so-called “Gives” complied with United Brands/Phenytoin principles.

Although the judgment is procedurally an application for permission, the Court of Appeal (Newey, Phillips & Arnold LJJ) delivered a fulsome, 96-paragraph reasoned refusal which will undoubtedly be cited as substantive authority on a range of competition-law evidential issues.

2. Summary of the Judgment

• The Court of Appeal held that each ground advanced by the Class Representative (“CR”) was, in substance, a challenge to evaluative fact-finding, not a point of law. Section 49(1) Competition Act 1998 confines appeals from the CAT to points of law; mere disagreement over weighting of evidence cannot found an appeal.
• Even where a dominant undertaking withholds potentially relevant cost data, the CAT enjoys a broad discretion to decide whether to draw Blatch v Archer adverse inferences. On the facts the CAT was entitled to decline to do so.
• The tribunal’s blended use of two irreconcilable expert methodologies (RFS vs SAC-Combi) was legitimate “broad-axe” estimation, not irrationality.
• In assessing Limb 2, the CAT was entitled to treat (a) observed switching, (b) “Gives”, and (c) incremental brand value as cumulatively sufficient to render an average 38 % excess prima facie “reasonable”, even though those items were largely unquantified.
• Proposed complaints on compound interest were academic once permission was refused on liability.

3. Analytical Discussion

3.1 Precedents Cited or Clarified

The Court of Appeal knitted together a rich tapestry of older authorities and very recent competition-damages rulings:

• Sainsbury’s v Visa [2020] UKSC 24 – heavy evidential burden rests on the party holding the information.
• Royal Mail v DAF Trucks [2024] EWCA Civ 181 – application of adverse-inference doctrine in cartel disclosure.
• Phones 4U v EE & Ors [2025] EWCA Civ 869 – appellate deference to trial judge on adverse inferences.
• Efobi v Royal Mail [2021] UKSC 33 – drawing inferences is “common sense, not technical law”.
• Cinven v CMA [2025] EWCA Civ 578 – modern restatement of United Brands two-limb excessive-pricing test.
• Hydrocortisone, Phenytoin, and Generics v CMA – on “economic value” and the “willingness-to-pay” fallacy.
• PPC [2011] CAT 5 – caution against over-reliance on combinatorial stand-alone cost tests.

Le Patourel II synthesises these strands: it applies Sainsbury’s but emphasises that the inference is discretionary; it reiterates Cinven that cost-plus is only a start; and it cautions against importing the United Brands concept of “unfairness” into mere differences of expert modelling.

3.2 Core Legal Reasoning

a) Adverse Inference and Evidential Burden

“In drawing or refusing to draw adverse inferences an appellate court must ask: did the fact-finder act irrationally? The tribunal may be critical of a party’s disclosure yet still consider itself able to ‘cope’.” (para 25)

The Court accepts that BT could – at modest cost – have produced granular SFV-specific indirect-cost data, but upholds the CAT’s decision that:
(i) expert reconstruction (Dr Jenkins) was an acceptable case-management substitute; and
(ii) the CR had chosen not to seek disclosure orders pre-trial, making it harder to complain on appeal.

b) Cost Allocation – “40 % common costs”

Despite excoriating the SAC-Combi model as “seriously deficient”, the CAT cited it only as a marker when settling on 40 %. The Court of Appeal held that merely “bearing in mind” faulty data is not the same as “relying on” it; blended, judgmental allocation falls within the CAT’s economic expertise.

c) Limb 2: Economic Value, Switching and “Distinctive Value”

• Switching evidence. The CAT reasoned that observed capacity to switch negated the “captivity” assumption and allowed a modest inference of brand/value appreciation. The Court endorsed this nuanced, fact-sensitive approach.
• “Distinctive value”. Borrowing language from Hydrocortisone, the CAT asked whether features of BT’s SFV product offered value “in contradistinction” from rivals. The Court saw no misdirection: “distinctive” is a yardstick, not a legal test.
• “Gives” and Brand. While the CAT lacked granular quantification, it adopted a holistic view: on-shoring of call centres, Call Protect, and a fault-fix guarantee collectively provided “real positive value”. The Court accepted that the law does not require exact arithmetic where the differential is relatively modest (c. £6 per month).

d) Appellate Restraint

Running throughout the judgment is a theme of institutional competence. The CAT – a specialist panel blending legal and economic expertise – is better placed than the Court of Appeal to weigh the “broad-axe” estimates typical of pricing-abuse litigation. The judgment therefore fortifies the high threshold a would-be appellant must surmount.

3.3 Broader Impact

The case is likely to influence UK competition litigation in four principal ways:

1. Disclosure strategy. Dominant firms remain incentivised to provide cost data, but Le Patourel II confirms they will not be automatically penalised if they do not – unless the gap prevents the CAT from deciding the case.
2. Expert methodology. Tribunals may mix-and-match competing models, provided they explain the blend. Parties should prepare sensitivities and “cross-runs” of their opponent’s model.
3. Collective actions. The ruling signals that once a CAT trial is lost on fine-grained economic facts, an appeal is uphill. Class Representatives must assemble robust evidence before trial, not rely on appellate second-guessing.
4. Limb 2 articulation. The Court implicitly accepts that “impressionistic” reasoning is inevitable when the alleged excess is not vast; future claimants may need consumer-survey or conjoint-analysis evidence to pin a monetary value on non-price factors.

4. Complex Concepts Simplified

4.1 Cost-Plus (Limb 1)

• Cost: All incremental and common costs of producing the service.
• Plus: A “reasonable” return on capital (profit margin).
• The tribunal compares Cost-Plus with the actual price to see if there is a significant and persistent excess.

4.2 Limb 2 (“Unfairness”)

Even if a price exceeds Cost-Plus, it is not necessarily abusive. The supplier may justify the excess by showing consumers receive extra value – e.g. superior quality, brand reassurance, product features, or that analogous competitive markets produce similar margins.

4.3 Adverse Inference

A court may treat missing evidence as harmful to the party who could (and reasonably should) have produced it. Whether to do so is discretionary; it depends on:
(a) the seriousness of the omission;
(b) whether the tribunal can still reach a fair decision; and
(c) overall proportionality.

4.4 “Willingness-to-Pay” Fallacy

It is erroneous to assume a price is fair merely because customers pay it; captive customers may have no realistic alternative. Evidence of switching (or its absence) helps determine whether payment reflects true valuation or practical compulsion.

5. Conclusion

Le Patourel II does not create dramatic new doctrine; its significance lies in consolidating and clarifying existing principles:

• The CAT’s economic judgment will be accorded wide latitude on appeal.
• Adverse inferences are a flexible tool, not an automatic sanction.
• Where the alleged excess is moderate, tribunals may permissibly rely on qualitative evidence of value.
• Collective claimants must anticipate and fill evidential gaps at trial; the Court of Appeal will rarely interfere post-hoc.

The decision will be cited alongside Cinven and Phenytoin as guidance on excessive pricing, and alongside Phones 4U on appellate deference. For practitioners, it emphasises that the road to success in an excessive-pricing claim is paved with rigorous, early-stage data gathering and nuanced economic analysis – not with hope of an appellate re-run.