Royal Mail Group Limited v DAF Trucks Limited & Ors

This appeal concerns the quantum of follow-on damages awarded by the Competition Appeal Tribunal (“the CAT”) in its judgment dated 7 February 2023 to Royal Mail Group Limited (“Royal Mail”) and companies in the BT Group (“BT”) (to which I will refer together as the Claimants). The claims arise following the determination of the European Commission in 2016 that five truck manufacturers, including the defendants, DAF, had colluded on pricing arrangements and had carried out a single and continuous infringement of Article 101 TFEU and Article 53 of the EEA Agreement between 1997 and 2011.

The CAT found that this collusion caused a 5% increase in prices charged to the Claimants (for truck and truck “body” purchases and lease payments) during this period, and dismissed DAF’s arguments that the Overcharge was mitigated or offset by the Claimants passing on the Overcharge. DAF appeals, either with the permission of the CAT or with permission granted by me, on the basis that the Tribunal erred in assessing the quantum of the loss; in its inclusion of truck “bodies” (in addition to trucks) in the Overcharge; in finding there was no mitigation by way of supply pass on of the Overcharge; and in its findings as to the cost of financing the Overcharge.

On 19 July 2016, the European Commission adopted the Settlement Decision in Case 39824 – Trucks, (“the EC Decision”) finding that five European truck manufacturing groups (DAF, MAN, Daimler, Iveco, and Volvo/Renault) had infringed Article 101 of the TFEU and Article 53 of the EEA Agreement between 17 January 1997 and 18 January 2011. The manufacturers, including DAF, signed up to the EC Decision, thereby admitting the infringement. The infringement was “by object” rather than “by effect”.

On 1 December 2016 and 13 October 2017, Royal Mail and BT respectively issued follow-on claims for damages in the Chancery Division. In June and July 2018 respectively, those claims were transferred by consent to the CAT. The trial of those claims in the CAT took place over 25 days in May and June 2022. Because the infringement constituted a breach of statutory duty giving rise to a cause of action in tort, if the Claimants could prove actionable harm or damage caused by the infringement, they would be entitled to damages. Accordingly, the trial was concerned only with issues of causation and quantum of damage, including mitigation.

As already noted, judgment was handed down on 7 February 2023. In broad terms the CAT found that the infringement caused loss to both Claimants in the form of an Overcharge which it assessed at 5% for both Royal Mail and BT on their value of commerce over the whole of the relevant period. DAF’s various “mitigation” defences including supply pass-on failed. However, there was a dissenting opinion from one of the members of the CAT, Mr Ridyard, on the supply pass-on issue. The Orders of the Chair, Michael Green J, concerning damages and interest for both Royal Mail and BT were made on 3 March 2023. In a Permission to Appeal Ruling dated 16 May 2023, the CAT granted DAF permission to appeal on one ground (supply pass-on) on the basis there was another “compelling reason” for the appeal (its impact on other competition cases where supply pass-on is an issue) but refused permission on its other grounds.

DAF then applied for permission to appeal to the Court of Appeal on three of its remaining grounds. On 11 July 2023, I granted permission to appeal on all those grounds, on the basis that they had a real prospect of success or alternatively there was a compelling reason for them to be heard so that this Court could give guidance on quantum-related issues and issues arising out of the EC Decision.

The CAT produced a careful and detailed judgment running to some three hundred pages. In the introductory sections, it set out an outline of the EC Decision and of the infringement. The facts relating to the claimants were set out at Section F and the facts relating to DAF at Section G. In that Section at [102] to [107] the CAT noted that DAF called four witnesses of fact but that none of them had given evidence explaining how the cartel operated and all had denied any knowledge of the cartel or the unlawful exchange of confidential information between the cartelists. The CAT recorded that Mr Beard KC, counsel for DAF, had speculated as to the motives of the culpable individuals but said that whatever their motives, that did not alter the facts or the data and it was those that determined whether there was an Overcharge, not the subjective intentions and beliefs of those involved in the infringement. The CAT did not accept that analysis, saying at [108]:

“We do not wholly accept that. DAF’s expert evidence on the theory of harm is based on speculation as to how the Infringement would have worked within DAF and then draws conclusions on such speculation as to how the Infringement would not have had an effect on prices. We think that any such theory would be more soundly based on what actually happened factually within DAF in terms of how the information was used and how the Infringement managed to continue over such a long period, presumably for the mutual benefit of all the Cartelists.”

The CAT went on to refer to speculation by Mr Beard KC in his closing submissions as to how the confidential information might have been used to beneficial effect within DAF, for example in better understanding the relative positioning in the market of DAF’s products as against those of the other cartelists. The CAT rejected this speculation, saying at [116]-[117]:

“We take no account of this speculation and it is an inappropriate way of approaching this issue by DAF. The burden remains on the Claimants to prove causation but where DAF has elected to call no evidence as to how the Cartel was operated by DAF and how it used the information to its advantage it is not open to its Counsel to speculate as to what actually happened. This was highly commercially sensitive information that was disclosed among the Cartelists over a long period of time. The Commission found that this information enabled the Cartelists to be better able to calculate their competitors’ approximate net prices. Further, the basis of a finding of an infringement by object is that it is very likely to have had negative effects on transaction prices. Therefore, in our view, this means that, if DAF wished to argue that, because of the way it used the confidential information obtained through the Cartel, there was no effect on prices, it would have had to adduce factual evidence to such effect. In other words, DAF’s admissions and the Settlement Decision establish a prima facie case that the Cartel had an adverse effect on transaction prices.

117.

That is not to say that DAF is unable to rely on its expert evidence to argue that the data shows that there was no Overcharge paid by these Claimants. But even their expert was unable to explain or come up with a rational economic basis for DAF’s participation in the Cartel over such a long period. While Prest does not entitle the Claimants to say that they have therefore proved that DAF’s participation in the Cartel led to higher prices it does mean that it is not open to DAF to argue that, as a matter of fact, the information was not used by it to achieve prices that were higher than they would otherwise have been without that information exchange.”

The CAT considered the general principles of law at Section H. It dealt with causation and quantum at [167] to [175]. As to causation, it explained that the Claimants’ cause of action is in tort and damages are compensatory, referring to Sainsbury’s Supermarkets Ltd v Mastercard Incorporated [2020] UKSC 24 (“Sainsbury’s SC”) at [194]. The Claimants must establish both (a) a breach of competition law and (b) actionable harm or damage caused by that breach. In this case, the former “is established by the findings of the Infringement in the Settlement Decision”, but the latter must also be proved by the Claimants, which “will not accrue until there has been actionable damage”. They “must satisfy the test for causation before there can be consideration of the quantification of their actual loss” ([168]). The CAT cited the explanation of Marcus Smith J at [424]-[427] of BritNed Development Ltd v ABB AB and ors [2018] EWHC 2616 (“BritNed”) as to whata claimant has to prove in terms of actionable damage ([169]). It concluded at [172] that the Claimants “are required to establish that they suffered monetary harm as a result of the Infringement and they must do so on the balance of probabilities”.

Turning to quantum, the CAT considered the application of the “broad axe” principle to the quantification of damages:

“173.

Once the cause of action has been established in this way, the quantification of damages has to be considered. There is no dispute that damages are to be assessed on a “broad axe” basis rather than on the balance of probabilities. As Marcus Smith J said in BritNed at [12(6)]:

“During this quantification exercise, English law moves away from the balance of probabilities. An assessment or quantification of damages involves the taking into account of all manner of risks and possibilities…”

And in Michael O’Higgins FX Class Representative Ltd v Barclays Bank Plc [2022] CAT 16 at [172], Marcus Smith J (sitting in the CAT with Mr Paul Lomas and Professor Anthony Neuberger) said:

“Actionable loss has nothing to do with the quantification of damages. If the necessary elements of the tort are made out, the claimant or claimants have a right to damages, no matter how difficult or recondite the assessment process”.

174.

The “broad axe” principle originated from Lord Shaw’s statement in Watson Laidlaw & Co Ltd v Pott, Cassels & Williamson [1914] SC (HL) (18) that quantification of damage is to be “accomplished to a large extent by the exercise of sound imagination and the practice of a broad axe”. This was specifically endorsed by the Supreme Court in Sainsbury’s [218] for competition claims. And in the trucks case collectively, in Dawsongroup plc v DAF Trucks NV [2020] CAT 3 at [40(3)], the CAT referred to the necessary use of averages, extrapolations and aggregates. This “broad axe” approach, largely based on expert econometric evidence, is necessary to accommodate the difficulties of proof inherent in the quantification of competition law damages. It is also required by the principle of effectiveness and the overriding objective that cases should be dealt with proportionately (see Sainsbury’s at [217]).”

The CAT next considered the mitigation defences in detail from [176] to [230], noting that these issues were engaged principally in respect of the issue of Supply Pass-On (“SPO”), “in which the claimants are said to have passed on all of the Overcharge to their customers through higher downstream prices” ([176]). The legal principles in relation to pass-on have been reviewed and explained by the Supreme Court in Sainsbury’s SC. The CAT considered at [178] that the ability to reduce or avoid loss because of commercial decisions in the pricing of the Claimants’ own products “is curious from a legal perspective” and the connection between the Overcharge and a decision to increase prices “might appear remote”, but “the Supreme Court seem to have accepted that this could be sufficient as a matter of both legal and factual causation”.

As the parties disagreed as to whether, given Sainsbury’s SC, an issue of legal causation arose in this context, it was “necessary to look at some of the basic legal principles in relation to mitigation of loss and the requisite connection between the upstream and downstream prices for the purposes of proving causation at law” ([179]). The CAT continued:

“180.

SPO only becomes relevant if the Claimants have proved the Overcharge… Damages for breach of statutory duty or tort are compensatory. The Claimants are entitled to damages that will put them in the position they would have been had the tort not been committed – see Sainsbury’s at [194]. The prima facie measure of the Claimants’ loss is the Overcharge – see Sainsbury’s at [198] – [199] – and it is not necessary for them to prove any consequential loss of profit.”

The CAT considered that the Supreme Court had clarified some general principles in relation to pass-on, which it set out at [181] with cross-references to paragraph numbers in the Supreme Court judgment:

“(1)

It is an aspect of the assessment of damages, rather than a defence strictly so-called; it is a form of mitigation of loss: “pass-on is an element in the calculation of damages and the normal rule of compensatory damages applies to claims for breach of statutory duty” – [196];

(2)

In relation to national claims for damages for breach of competition law, Member States may lay down procedural rules governing actions which safeguard rights derived from EU law, but those national rules must comply with the principles of equivalence and effectiveness. The principle of effectiveness requires that the rules of domestic law “do not make it practically impossible or excessively difficult to exercise rights guaranteed by EU law” - [188];

(3)

Claimants suffering from an overcharge are not required to prove that they have suffered an overall loss of profits as a result. If it were otherwise, claimants might face an insurmountable burden in establishing their claims and “such a domestic rule [… ] would very probably offend the principle of effectiveness. It is the duty of the court to give full effect to the provisions of Article 101 by enabling the claimant to obtain damages for the loss which has been caused by anti-competitive conduct”: [209];

(4)

The legal burden is on the defendant to plead and prove that the claimants have mitigated their losses by passing-on the overcharge, although the Supreme Court went on to say that this should not be overstated and there is a “heavy evidential burden” on claimants once the defendant has raised the issue of mitigation: [211] and [216].”

The CAT examined the decisions of the Courts in Sainsbury’s in further detail from [182] to [200]. It highlighted the importance of the context of Sainsbury’s, which concerned losses sustained by merchants from the payment of multilateral interchange fees (“MIFs”) which were passed on by the acquiring bank to the merchant as part of the merchant service charge. At first instance the CAT “addressed head-on the central difficulty with pass-on which is the risk of under-compensation if the defendant proves pass-on or over-compensation, if it does not”. At [184] the CAT cited [484(4) & (5)] of the judgment of the CAT in Sainsbury’s, including:

“(4)… We consider that the legal definition of a passed on cost differs from that of the economist in two ways:

(i)

First, whereas an economist might well define pass-on more widely (i.e. to include cost savings and reduced expenditure), the pass-on defence is only concerned with identifiable increases in prices by a firm to its customers.

(ii)

Secondly, the increase in price must be causally connected with the overcharge, and demonstrably so.

…

[The] risk of under-compensation, we consider, to be as great as the risk of overcompensation, and it informs the legal (as opposed to the economic) approach. It would also run counter to the EU principle of effectiveness in cases with an EU law element…

(5)

Given these factors, we consider that the pass-on “defence” ought only to succeed where, on the balance of probabilities, the defendant has shown that there exists another class of claimant, downstream of the claimant(s) in the action, to whom the overcharge has been passed on. Unless the defendant (and we stress that the burden is on the defendant) demonstrates the existence of such a class, we consider that a claimant’s recovery of the overcharge incurred by it should not be reduced or defeated on this ground.” (emphasis added by the CAT)

The CAT noted that it appears that the Court of Appeal in Sainsbury’s disagreed with the CAT’s conclusion at (5) that the burden was on the defendant to show the existence of another class of claimant to whom the overcharge had been passed on and that “this was not an essential condition for establishing pass on”, albeit the CAT’s finding that there was no pass-on of MIFs was not challenged on appeal. The CAT in the present case went on:

“186.

Importantly, the Court of Appeal did not interfere with (indeed it seemed to endorse it – see [340]) the test for causation set out in [484(4)] of CAT Sainsbury’s that there must be an identifiable increase in prices charged by the merchant and that such increases are “causally connected with the overcharge, and demonstrably so.” At [332] of CA Sainsbury’s, there are two references to there needing to be a “sufficiently close causal connection” between the overcharge and any increase in the prices charged to customers. This indicates that the connection must be close enough that the particular price and/or the increase in that price and the persons paying that price (and thereby suffering loss) must be sufficiently identifiable.

187.

We think that the Court of Appeal was only deciding that there was no self-standing requirement (i.e. distinct from the requirements in [484(4)]) for a defendant to identify a purchaser or class of purchasers to whom the overcharge had been passed. It left open whether it may be necessary for a defendant to identify such persons so as to demonstrate the requisite causal connection between the overcharge and the price increase…

188.

… it seems to us that the Court of Appeal has not ruled out the possibility that [484(5)] of CAT Sainsbury’s was not a separate requirement but was instead part of the overarching legal test for causation in this area. Therefore, in order to establish the requisite causal connection, it may be necessary for a defendant to identify who has suffered the loss from the overcharge in order to prove that the claimant has not actually lost the full extent of the overcharge.”

The CAT considered at [189] that “some of the problems in this area have been caused by an element of confusion between the economic concept of ‘pass-on’ of a business’s costs and the legal test for causation in relation to mitigation of loss”. The CAT therefore turned to the analysis in Sainsbury’s by the CAT and the Supreme Court as to the four principal ways in which a business might react to an increase in a specific cost from [190] to [194]. The Supreme Court at [205] of its judgment had listed these as:

“(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.”

The CAT noted at [192] that the Supreme Court considered that “relevant pass-on will only arise in categories (iii) and (iv) (which are not distinguished).” Accordingly it continued:

“193.

The focus therefore of the Supreme Court’s approach is the effect on profit margins of the overcharge. However, the loss itself is measured by reference to the overcharge, not the claimant’s loss of profits. There is a slight mismatch in that the loss is the overcharge whereas the mitigation is assessed by reference to whether decisions taken by the claimant in response to the overcharge have served to lessen the initial loss of profit. As noted above, the Supreme Court held that a claimant was not required to prove the effect of the overcharge on its profits, as this might offend the principle of effectiveness, and yet it seems that for pass-on it has to meet a case that it has not taken consequential steps that would have resulted in it suffering lower profits as a result of the overcharge.

…

195.

We are concerned in this case with category (iv), as they were in Sainsbury’s. There is no dispute that this is a species of pass-on but as the Supreme Court recognised, and is a live issue in this case, such a form of pass-on may result in reduced downstream sales volume and consequent further losses to the Claimants. That is why the decision to increase prices in response to a specific increased cost is not a straightforward one as it will necessarily involve the balancing of a number of considerations, in particular whether profit maximisation is best achieved by passing on the full increase in costs or whether it is better to pass on less or none of the increase to ensure no loss, or a reduced loss, of volume. Businesses have to make these multi-factorial decisions the whole time and the law needs to delineate how close the connection should be for causation to be established.”

The CAT pointed out at [196] that, by contrast with MIFs which were not a secret overcharge, in this case “the Overcharge was secret and unknown to the Claimants. Hence there was no conscious response to the Overcharge and, although at an abstract level one can envisage that truck costs do form a (small) part of the operating cost of providing a mail delivery or telephone service, there is only an indirect connection between the trucks bought from DAF and the items bought by the Claimants’ customers, such as postage stamps or line rentals.”

The CAT next considered the Supreme Court’s approach to the necessary connection for the purposes of legal causation, citing at [199] the Supreme Court’s judgment at [215]-[216], including:

“215.

… The issue of mitigation which arises is whether in fact the merchants have avoided all or part of their losses. In the classic case of British Westinghouse Electric and Manufacturing Co Ltd v Underground Electric Railways Co of London Ltd [1912] AC 673, at 689 Viscount Haldane described the principle that the claimant cannot recover for avoided loss in these terms:

“[W]hen in the course of his business [the claimant] has taken action arising out of the transaction, which action has diminished his loss, the effect in actual diminution of the loss he has suffered may be taken into account …” (emphasis added by the Supreme Court).

Here also a question of legal or proximate causation arises as the underlined words show. But the question of legal causation is straightforward in the context of a retail business in which the merchant seeks to recover its costs in its annual or other regular budgeting. The relevant question is a factual question: has the claimant in the course of its business recovered from others the costs of the MSC, including the overcharge contained therein?

216.

The legal burden lies on the operators of the schemes to establish that the merchants have recovered the costs incurred in the MSC. But once the defendants have raised the issue of mitigation, in the form of pass-on, there is a heavy evidential burden on the merchants to provide evidence as to how they have dealt with the recovery of their costs in their business…”
(second emphasis added by the CAT).

The CAT noted at [200] that, the Supreme Court having opened the door to claims of pass-on within its category (iii), the defendants had sought to plead this in general terms in their defences, and DAF sought to amend its pleadings in this case to allege that the claimants had mitigated their losses by negotiating lower input costs from their other suppliers. The CAT stated at [201] that it had dismissed that application in its May 2021 judgment, noting that, in doing so, the CAT had made “important observations” at [35]-[36] of the May 2021 judgment about the test for causation in relation to pass-on in the light of Sainsbury’s, and concluded at [36]that given the Supreme Court’s analysis:

“…for a defendant to be permitted to raise a plea of mitigation in this way in general terms, there must be something more than broad economic or business theory to support a reasonable inference that the claimant would in the particular case have sought to mitigate its loss and that the steps taken by it were triggered by, or at least causally connected to, the overcharge in the direct manner required by the British Westinghouse principle.”

The CAT also noted that, at [43] of the May 2021 judgment, it had considered that the sort of evidence required to satisfy the test for pleading this sort of mitigation was:

“…some plausible factual foundation for the application of the broad economic theory in the way required to satisfy the British Westinghouse test that is relied upon, and for there being a causative connection between overcharge and cost cutting.” (emphasis added)

At [204] the CAT referred to the issues of knowledge and size of the Overcharge addressed in [44] of the May 2021 judgment, contrasting the situation in this case with that in Sainsbury’s:

“…where the overcharge was not only covert but a tiny fraction of Royal Mail’s and BT’s expenditure, it is inherently unlikely that it would have been specifically addressed, but rather fed into the overall expenditure of the regulated or unregulated parts of the business. As DAF accepts, that general principle that all costs of all inputs are fed into business planning is insufficient to establish the necessary causative connection for a plea of mitigation of loss”. (Emphases added by the CAT).

At [205]-[206] the CAT noted that the Court of Appeal had to consider a similar argument from the defendants in NTN v Stellantis [2022] EWCA Civ 16, [2022] 2 All ER (Comm) 706, where this Court approved the CAT’s judgment in that case and the May 2021 judgment in the present case, disallowing a proposed category (iii) amendment. Green LJ stated:

“33.

Pulling the strands together, the burden of proof when pleading causation is on the defendant to demonstrate: (a) that there is a legal and proximate, causal, connection between the overcharge and the act of mitigation; and (b), that this connection is "realistic" or "plausible" (the two phrases being interchangeable) and carries some "degree of conviction"; and (c) that the evidence is more than merely "arguable". The assessment will be fact and context specific and, to foreshadow a point I refer to later, may depend upon the characteristics of the industry or sector in question. It may be easier to show a pleadable case of mitigation in some circumstances than in others.” (emphasis added by the CAT)

As to the relevance of knowledge of the overcharge, Green LJ said at [48] of Stellantis that it was “plainly relevant” that the issue of MIF pass-on had been live in the industry for many years and there was nothing secret about the imposition of a MIF: this was “clearly relevant to the burden facing a defendant in this sector seeking to raise a realistic case of mitigation”, and the facts “therefore contrast with those of a typical, secret price fixing cartel”.

The CAT next considered the test for legal causation. It noted at [207] to [209] that, prior to Sainsbury’s, both Rose J and Roth J at CMCs in the present proceedings “seem to have been of the view that the legal test for causation in this area requires there to be a link between the product bought from the cartelist and the product sold by the purchaser. We agree with that point but the question is, in the light of Sainsbury’s, what sort of link there needs to be” ([209]). As to this, the CAT said at [210]:

“The Supreme Court in [215] relied on British Westinghouse, a contract case, and emphasised the words in that judgment “arising out of the transaction” to indicate that, for this form of mitigation, there had to be a causal link between the “transaction” in that case, or the breach of statutory duty in competition cases, and the action taken to diminish the loss. It is the next sentence that is a little surprising: “But the question of legal causation is straightforward in the context of a retail business in which the merchant seeks to recover its costs in its annual or other regular budgeting.” What the Supreme Court seems to be saying is that there is no live issue of legal causation in relation to any business where the pass-on is in either category (iii) or (iv). The only issue is a factual one and that is whether the merchant did actually recover its costs by reducing other costs or increasing its prices.”

The CAT noted at [211] that in the MIF Umbrella Judgment [2022] CAT 31,the CAT had concluded that the Supreme Court had decided that there was no issue of legal causation where it was shown as a matter of fact that the mitigating conduct had reduced or eliminated the claimant’s loss. However, the CAT said at [212] that it was “important to see what the CAT considered the issue of legal causation to be” which, where factual causation has been proved and the overcharge has been passed on to the claimant’s customers, was whether there was any “policy reason” why the claimant should be able to continue to claim the overcharge from the defendant. The CAT considered the MIF Umbrella Judgment was “heavily influenced” by the fact that the ultimate customers, credit card holders, were bringing their own claims in respect of the overcharge allegedly passed on to them, and the CAT was principally concerned with effective case management of all the claims before it.

The CAT pointed out at [213] that both Sainsbury’s and the MIF Umbrella Judgment concerned proceedings about MIFs which it was known retailers were seeking to pass on to customers. It stated:

“213… The Supreme Court limited its comment as to legal causation being straightforward to a retailer seeking “to recover its costs”. Green LJ in Stellantis at [77] considered that “what the Supreme Court described as “straightforward” was causation in relation to the MIF”…

…

215.

In our view, the sentence in the Supreme Court’s judgment has been taken out of context and needs to be read with what came before and after. The Supreme Court, after the quote from British Westinghouse, said that there is a “question of legal or proximate causation” that arises in relation to pass-on. The rest of the paragraph seeks to explain how that question is resolved. It is not entirely clear what is meant by “legal causation” and the Supreme Court does not use the phrase “factual causation” by way of contrast. Rather it said that “the relevant question is a factual question” and then proceeded to define that factual question and upon which party the legal and evidential burdens lie. What the Supreme Court does not do is set out the legal test for causation. But by delineating the factual question to whether the claimant has in that case “recovered from others” the MIF through categories (iii) and (iv) thereby transferring “all or part of its loss to others”, the Supreme Court has effectively held that that is the legal test of causation.

216.

As was said in Stellantis, the defendant must demonstrate “a legal and proximate, causal connection between the overcharge and the act of mitigation”; or as the CAT said in its May 2021 Judgment, there must be a “direct causative link” between the Overcharge and, in that category (iii) case, the reduction in the costs of other supplies. It is insufficient, as was admitted by Mr Beard KC at that hearing, merely to allege the Claimants were seeking to recover their costs. We do not therefore think that the Supreme Court was suggesting otherwise in saying that “legal causation is straightforward” in that case. It was still necessary for the defendant to prove a sufficient causal connection on the facts to satisfy the legal test for causation.”

The CAT went on to say at [217] that the difficulty “is finding a test that enables the court or tribunal to work out who has actually suffered loss as a result of the Overcharge”. Unlike in the MIF Umbrella Judgment, there were no claims from the Claimants’ customers. The CAT recognised that “the task we face is to decide if the Claimants have truly suffered loss from the imposition of the Overcharge or whether they have avoided some or all of that loss by passing it on to someone else”.

The CAT agreed with the May 2021 judgment that the particular factors relevant to category (iii) and (iv) cases (and endorsed by the Court of Appeal in Stellantis) were (i) the claimants’ knowledge of the nature and amount of the Overcharge such that they would seek to address it and (ii) the size of the Overcharge as a proportion of the claimants’ relevant expenditure and/or price/cost margin ([220]). The CAT did not consider these “necessary requirements”, but “if they are present then a defendant would be far more likely to be able to prove that the claimant was seeking to address the overcharge by taking the mitigating action that it did” ([222]). It continued:

“223.

Accordingly, we consider that DAF must prove that there was a direct and proximate causative link between the Overcharge and any increase in prices by the Claimants. That means that there must be something more than reliance on the usual planning and budgetary process, into which the Overcharge was input and at some point prices increased. We think that there is substance to the point made in CAT Sainsbury’s as to the identification of persons to whom the Overcharge has been passed as being a relevant factor in relation to the strength of the causal connection. The process is more properly one of identifying the persons who have suffered loss by paying the Overcharge and therefore who should be compensated by the defendant.”

The CAT summarised its conclusion on the legal test for causation at [228] to [230]:

“228.

By way of summary on the legal test for causation in relation to a pass-on form of mitigation defence, we respectfully conclude that DAF must prove a direct and proximate causative link between the Overcharge and any increase in prices by the Claimants. It is not enough for DAF to say that all costs, including increases in costs, are fed into the Claimants’ or their regulators’ business planning and budgetary processes. There must be something more specific than that and there are a number of potentially relevant factors that it can rely on including:

(1)

Knowledge of the Overcharge or the specific increase in the cost in question;

(2)

The relative size of the Overcharge against the Claimants’ overall costs and revenue;

(3)

The relationship or association between what the Overcharge is incurred on and the product whose prices have been increased; and/or

(4)

Whether there are identifiable claims by identifiable purchasers from the Claimants in respect of losses caused by the Overcharge.

229.

This is not an exhaustive list of factors but they do seem to us to be the most relevant ones to this case. In relation to the last point (4), we think that, even though there are no such claims before us, we need to be mindful of the effect of our decision in relation to pass-on defences on other claims. The danger that is inherent throughout this process if we decide against any such defences is that the Claimants are overcompensated and the potential other claimants are deprived of their claims. The converse is equally fraught, in that if we allowed some or all of the pass-on defences, DAF may escape paying compensation to all those who suffered loss as a result of the Overcharge.

230.

We consider the above factors when dealing with this issue below and particularly in our analysis of the expert evidence… What the experts were seeking to show was whether the downstream prices charged by the Claimants were higher in the actual world with the Overcharge than they would have been in the counterfactual where there was no Overcharge. But even if that can be shown, it will have to be demonstrated whether there is the necessary proximate and direct causative link required by the legal test for causation, based on the above factors.”

In the next Section of the judgment at I, the CAT set out its assessment of the parties’ expert evidence. It expressed some concerns as to the “marked” way in which particularly the economic experts came to conclusions that favoured their clients, and addressed the Claimants’ attacks on the independence of the Defendants’ economic expert Professor Neven, who had been acting as an expert for DAF for some years since 2013, but had failed to disclose in a transparent way details of his long-standing professional relationship with DAF or of the information he had received from DAF about the operation of the cartel and its effects. The CAT noted at [238] that none of this was referred to in his reports and only came to light during cross-examination. Even after protracted correspondence between the parties’ solicitors, it remained unclear to the CAT what were the terms of Professor Neven’s original engagement and what he had been advising on in the past.

The CAT set out, in detail, the history and background to its concerns from [237] to [257]. This provided support for the conclusion that Professor Neven showed a “lack of candour”. He had been involved in advising DAF for nearly a decade and from an early stage was expressing opinions and advice upon potential theories of harm that would assist DAF. The CAT observed that it was “… fairly obvious that he was their favoured economic expert and this seems likely to have been influenced by his opinions on the plausibility of there having been any effect of the infringement.” [237]). The CAT also concluded at [249] that Professor Neven’s theories of harm were based upon what DAF had told him and their “narrative” (advanced before the European Commission) that there was no anticompetitive effect flowing from the cartel. The CAT recorded at [250] the evidence of Professor Neven in cross-examination that he had asked DAF why it had participated in the cartel. He was, apparently, told by DAF that it enabled the company to know its competitors’ list prices so that they could test their relative competitiveness. However, Professor Neven did not go further because he rejected this explanation as “ex post rationalisation” which was irrelevant to his economic analyses and theories of harm ([251]). The CAT concluded at [254] that an understanding of DAF’s behaviour in the cartel was “surely potentially relevant to any theory of harm and his lack of curiosity in this respect is troubling”. The CAT stated:

“256.

… This situation provided Professor Neven with insights and access that, as an independent expert, we could reasonably have expected him to use in order to assist us. We examine in detail the theory of harm that he puts forward in his evidence in this case and it is safe to say that his conclusion that it is implausible that there were any effects in the UK and on the Claimants from the Infringement is a surprising one. His theory provides a justification for the conclusion that he draws from the data that there was no Overcharge throughout the period of the Infringement. But we are left with the lingering suspicion that, as was disclosed very late on in these proceedings, he had come up with his theory of harm back in 2013 or 2014 (and certainly well before he had access to detailed empirical data), and that has shaped his approach to the expert evidence he has provided on the central issues in relation to the Overcharge.”

Nonetheless, the CAT adopted a balanced approach. It did not reject Professor Neven’s opinion outright, but said that his failure to disclose precisely what he was doing for DAF in the early period and his surprising lack of curiosity as to DAF's motives and conduct “to a certain extent” undermined his credibility. The CAT was cautious about accepting what he said at face value ([257]).

In Section J, the CAT addressed the experts’ assessment of the plausible theory of harm, which both experts had used to underpin their econometric analyses of the Overcharge. Since this is not relevant to the issues on the appeal, it is not necessary to consider it further.

At Section K, the CAT dealt with Overcharge. This analysis formed a substantial part of the judgment. Both economic experts, Mr Harvey (for the claimants) and Professor Neven (for DAF), relied on statistical models that employed multiple regression techniques which analysed large amounts of pricing and market data to determine what, if any, was the effect of the infringement on prices and any pass-on through the relevant period. The CAT heard evidence from the experts concurrently by way of hot-tubbing. At [327] it identified a number of issues with which it dealt in turn, starting with how regression models work from [328] to [335], and the approach of using market-wide data to estimate price effects on the claimants at [336] to [345].

It then assessed the different models used by each expert at [346] to [372]. The CAT summarised that both experts sought to assess the impact of the cartel by comparing the outcomes that arise in the cartel period with those that occur outside of it (in this case, the periods before and after the cartel), though they differed in their approach: Mr Harvey used two models: (a) a Before-During (“BD”) model of truck prices between 1995 and 2003, the Infringement period having begun on 17 January 1997; and (b) a During-After (“D-A”) model of truck prices between 2004 and 2017. Professor Neven also carried out a D-A model for the same period, but because of problems with the available data for the early period, and although he agreed in principle that a B-D model would be appropriate, he instead used a Before-During-After (“B-D-A”) model for the period 1995 to 2017. The CAT addressed the parties’ arguments in favour of their respective approaches, and noted in respect of the Claimants’ arguments at [353] that “there are significant problems with the available data for a robust B-D analysis. We therefore think it sensible to have a D-A analysis as well while bearing in mind that there may have been an overhang effect that could lead to an underestimate of the Overcharge”.

The CAT then turned to assessing the two experts’ approaches from [357] to [372]. In respect of Mr Harvey’s approach it said:

“365.

In our view, while it is clear that there are problems around the AS/400 dataset and the consequent lack of granularity on truck level costs, it is necessary to attempt a B-D analysis because of the advantages of comparing prices unaffected by the Infringement. Some caution must therefore be exercised in relying on the results obtained in such an analysis but we are satisfied that with the sensitivities carried out by Mr Harvey, we can draw inferences from such evidence, despite its imperfections.

366.

Mr Harvey’s B-D model does have the advantage of not being tainted by problems associated with the GFC (as to which see further below) and the use of separate B-D and D-A models might also capture the possibility that the Infringement effect was not symmetrical.”

It then set out Professor Neven’s approach before concluding at [372] that:

“…there are advantages and disadvantages of the B-D and B-D-A models and it cannot be said that it was wrong for each expert to have used the model that he did… in view of the complexity described above, we do not think it is possible or necessary to reach a definitive view on which was the better model to use. The more significant debate is over the three main issues that we deal with below – exchange rate effects, the [global financial crisis] and emissions premia – all of which had the biggest impact on the experts’ estimation of the effects of the Infringement and the size of the Overcharge”.

The first of those three main issues with which the CAT then dealt was exchange rates. As transactions occurred in multiple currencies, both sides were agreed that the modelling had to be conducted in a single currency. Mr Harvey decided to conduct his analysis in Euros; Professor Neven in Sterling, and as the CAT said at [373]: their choice “has a very significant effect on the outcome of their modelling”. It explained at [376] that: “the treatment of currency factors in the regression models was a critical issue because of the ‘identification’ problem that arises when trying to disentangle exchange rate effects from cartel effects. This was notably the case at the start of the Cartel period, when coincidentally the Pound strengthened against the Euro over the period from 1996 to 1998.”

The CAT considered that the approach of both experts to the problem was imperfect:

“381.

How long it actually takes for Pound prices to adjust to exchange rate changes through the competitive process depends on a variety of institutional and competitive factors, but Mr Harvey’s effective assumption that the adjustment is instantaneous can be seen as an extreme one. Hence, his approach creates a risk that he has found a cartel effect in the early part of the Cartel period when such did not exist.

382.

By contrast, Professor Neven’s approach of using DAF’s budget exchange rates in his regression model, means that the influence of the exchange rate change on incentives is suppressed for a year…

…

385.

DAF and Professor Neven criticised Mr Harvey’s modelling approach for imposing an extreme solution... But Professor Neven was not able to offer a definitive account of how quickly any such profit windfall should be dissipated under normal competitive conditions. This is indeed a complex question that does not generate an obvious or simple solution.

386.

Viewed this way, the choice between Mr Harvey’s and Professor Neven’s approaches is one between two imperfect alternatives. Both are capable of reaching a misleading conclusion about cartel effects, but in opposite directions, and we note that the bias in each case happens to assist the experts’ respective clients’ positions. Importantly, neither approach fully solves the underlying identification problem that arises from the coincidence of the start of the Infringement and an appreciable shift in the exchange rate.”

The CAT analysed each approach in more detail from [387] to [406]. In respect of Mr Harvey’s analysis, it noted:

“396.

Mr Harvey did not, and could not, introduce a standalone control variable for the market exchange rate in order to address the identification problem. He said that his model cannot control for the exchange rate because of the correlation between the exchange rate and the Infringement. He therefore accepted that there are difficulties in distinguishing between the exchange rate and the Infringement. Professor Neven agreed that in the context of Mr Harvey's model it would be difficult to disentangle the exchange rate and the Infringement, because of the correlation between these variables. However, he said, correctly, that this is a problem of Mr Harvey's own making.”

It concluded:

“407.

There was voluminous evidence adduced on this subject... But it really comes down to the best method of capturing what was actually going on, consistent with the way that DAF operated its business within a context where all of the significant suppliers to the UK market relied substantially or entirely on production costs that were incurred in currencies other than the Pound. The objective is to identify any effect on prices caused by the Infringement and in this context that means how best to remove the effect of exchange rate changes on prices. Whether the conversion is to Pounds or Euros has a highly significant effect on the outcome, which in itself perhaps indicates that either route is an extreme one and that the correct conclusion is somewhere between those extremes. As we said above there may not be a right way of doing this but we recognise that the experts, in order to perform their regression analyses, had to use one currency for all the variables in the model. In that sense they are both driven to an extreme position.

408.

We consider that Mr Harvey’s approach has a superficial attraction... But in terms of the identification problem, Mr Harvey’s approach to exchange rate changes is probably more problematic than Professor Neven’s in this regard. Mr Harvey acknowledged that it is implausible that actual Pound price adjustment in the truck prices negotiated between UK customers and DAF would be instantaneous. Mr Harvey’s approach does “hardwire” the adjustment from Pounds to Euros prices...

409.

Professor Neven’s modelling approach has greater flexibility... However, as we note above, there is no definitive solution to the challenge of how to solve the identification problem caused by simultaneous exchange rate and Infringement events, and it is arguable that Professor Neven’s reliance on DAF’s budget rates is also a restriction that could mask the way that price competition might work between truck suppliers in a competitive market.

410.

Accordingly, we do not say that one approach is right and the other wrong. Instead, we are left with the feeling that the answer is more nuanced than that and that the Infringement effect lies somewhere between the two positions on the basis that neither is truly capable of addressing the problems and difficulties inherent in this situation of having to convert into one currency or another.”

The next main issue which the CAT addressed was the Global Financial Crisis (“GFC”). The CAT explained that “[w]hen demand for a product increases, for example due to an upturn in the economy, one would expect that prices and margins would also increase” and vice versa: accordingly, “both experts’ models included a demand effect to control for this non-cartel influence on prices” ([411]). It was “common ground that the GFC had a significant impact on demand” ([415]), and it created “another identification problem” because the GFC coincided with the end of the Infringement period ([418]).

The CAT summarised the experts’ approaches as follows:

“416.

In relation to the GFC, Professor Neven relied simply on his existing standard demand controls. He considered they were sufficient to capture the effect of the GFC as it was essentially a demand shock. Mr Harvey adopted a radically different approach because of the “unprecedented” event of the GFC which he considered could not adequately be dealt with by way of standard demand controls. Instead, Mr Harvey used dummy variables for each of the years 2008, 2009 and 2010 which had the effect of taking those years out of account for the purpose of measuring the Overcharge.”

As to Mr Harvey’s approach, the Tribunal said:

“420.

Mr Harvey’s initial intuition was that the standard demand controls would be sufficient for the GFC effect. It is concerning that this only emerged at the hearing while he was giving evidence on this area. He disclosed for the first time that he originally ran his model with the standard demand controls in place and arrived at an Overcharge estimate of between 1 and 2%. This result was not referred to in his Reports.

421.

He then decided that the demand shift during the GFC was so profound that his demand controls were not adequate in these years, so he made an ad hoc adjustment to his model to include additional dummy variables in the three GFC years: 2008 to 2010. In effect, this meant that his model gave up on any attempt to measure the effect of the Cartel in this period, since the dummy variables for these years would be equally effective in capturing the GFC and Infringement effects that arose in those years…

422.

The adoption of GFC dummy variables in Mr Harvey’s model also crucially means that the options to assess a GFC effect separately from the end of the Infringement is eliminated, creating a clear identification problem. Similarly, Mr Harvey’s GFC dummy variables also compromise the ability of the model to measure the impact of the sharp fall in the value of the Pound against the Euro in 2008…

423.

Having implemented this change to his model, Mr Harvey then found a higherInfringement effect of 6-14%, depending on the truck family...

424.

Professor Neven criticised this ad hoc approach because it effectively absorbed all the variability in prices in trucks for these years...

425.

There is obvious appeal to this criticism, and it is inescapable that Mr Harvey’s approach does appear to have had the effect of shifting the goalposts ex post after his original model using the standard demand controls reached an inconvenient result.

….

427.

However, the fact that there are concerns with Mr Harvey’s GFC dummy variables does not rule out the possibility that the standard demand controls might be unreliable in capturing the abnormal effects of the GFC… This is an issue that can best be informed by reference to the factual evidence rather than the technical dispute between the experts’ views on the specification of their regression models.”

The CAT then considered whether the factual evidence supported Mr Harvey’s assessment from [429]-[435]. It considered on review that “there is some justification for Mr Harvey’s view that the GFC had a somewhat unique effect on pricing that might not be captured by the standard demand controls” ([435]).

Turning to Professor Neven’s approach, the CAT considered:

“436.

Professor Neven did seek to address this by proposing various ways to define “abnormal” demand... This is in principle a reasonable and constructive approach to the GFC problem.

437.

However, the specific alternative approaches employed by Professor Neven never really identified a satisfactory alternative measure…

438.

The wider concern of [Professor Neven’s] approach is as to whether demand levels can really capture the dynamics that might drive changes in pricing policies, and the possible interactions of demand levels and order cancellations…”

The CAT’s conclusion was:

“439.

In summary, we have concerns about Mr Harvey’s approach to the GFC problem but understand why he has done that and do not wholly reject it, as DAF invites us to do. Whilst we are unhappy with the way in which Mr Harvey changed his modelling approach only after discovering that the standard approach yielded results that were unhelpful to his client, and with the lack of transparency in the way this was done, and whilst DAF makes valid criticisms of the rather blunt methodology adopted by Mr Harvey of using the dummy variables for the full three years, 2008 to 2010, of the GFC, we consider that the GFC plausibly did have effects on pricing dynamics that would not be well captured by demand controls that work across normal demand fluctuations.

440.

Like with the exchange rate debate, there are legitimate arguments on both sides and we do not accept Mr Beard KC’s characterisation of Mr Harvey’s approach as “plainly misconceived”. Again, the actual answer may be found somewhere between the opposing positions which is more likely to reflect the true impact of the GFC on DAF’s pricing.”

The third main issue with which the CAT dealt was Emissions Premia at [441] to [462] where it concluded:

“462.

Overall, we find that the increase in price-cost margins that both experts agreed arose when new emissions standards were introduced, coupled with the admitted and plausible evidence that truck manufacturers did seek to coordinate on the truck price increases that should be associated with these standards, provide a compelling case for the emissions premia to be treated as part of the Overcharge.”

Since the issue of Emissions Premia did not really feature in the argument on this appeal it is not necessary to consider it further.

The CAT then dealt with the Value of Commerce (“VoC”) i.e. the total expenditure on trucks bought from DAF during the infringement, for which the disputed issues only concerned Royal Mail and, in particular, whether the price included in the VoC should be the price paid for the complete truck or should exclude the price of truck bodies, as DAF contended. The CAT considered first whether truck bodies were subject to the Overcharge and concluded at [473]:

“473.

However, we consider that DAF’s failure to provide any evidence as to how the Cartel operated, and particularly in this respect as to how it affected the pricing for the whole truck, including bodies and other options, leaves us in the dark on this question. The most natural approach to take here, given the absence of that evidence is to consider truck bodies equivalently to other extras that were purchased from DAF by Royal Mail. If the Infringement took effect through a mechanism linked to list price changes, there is no convincing evidence for us to reject the possibility of such an effect also applying to truck bodies. Had DAF chosen to share more information on the coordination infrastructure surrounding the Infringement, one might have been able to take an alternative view, but this is not the case. We do think that DAF should be held to what it signed up to in the Settlement Decision and this did not clearly exclude bodies from the scope of the Infringement.”

At [475] to [486] the CAT set out its overall conclusion on Overcharge. It noted at [475] that with the imperfections in the evidence, and insoluble practical problems, it was not possible to arrive at a definitive figure. However, the process of the expert evidence had yielded useful insights as to reasons for the differences between the experts and enabled the CAT to reach a better-informed view on the critical question of the Overcharge. It concluded at [477] that on a balance of probabilities the evidence pointed to the existence of a cartel Overcharge, saying:

“There are sound a priori reasons for expecting that a concerted attempt by all the major European truck suppliers to restrict price competition that persisted over a 14-year period would to some extent have succeeded in materially affecting transaction prices. Further, whilst there are legitimate criticisms to be levelled at Mr Harvey’s estimates of the effect, particularly with regard to the way his analysis approached exchange rate issues, we also consider it is clear that these criticisms do not justify the extreme approach of dismissing all positive Overcharge results.”

Accordingly, the CAT found that the claimants had established the requisite causation to complete their cause of action. The CAT then applied the broad axe approach to placing a value on the Overcharge, stating at [479]:

“…As we made clear in the sections above on exchange rates and the GFC, the true value of the Overcharge we believe lies somewhere between the two experts’ diametrically opposed positions. In the circumstances, we have no choice but to make a judgment based both on the evidence that was presented in the experts’ models, and on a wider appreciation of the factual context and witness evidence.”

Adopting the broad axe approach, the CAT concluded at [484]-[485]:

“484.

In relation to the exchange rates issue we concluded that, whilst neither expert’s approach was right, Professor Neven’s position has more merit than that adopted by Mr Harvey; GFC was more evenly split. We therefore consider that a fair and reasonable broad axe view on Overcharge comes out at 5% for both Claimants (ie approximately half of what they are claiming).

485.

We have no reason to adjust the profile of this Overcharge between the different years of the Infringement, and indeed to do so would imply a greater precision to the broad axe approach than we consider is justified, given the substantial imperfections in the data available and the complexity of the task.”

The CAT went on to deal with the Complements issue at Section L of the judgment and the Resale Pass-On issue at Section M. These issues are not relevant to the appeal. The CAT then dealt in detail with SPO, on which as already noted, it was divided, at Section N. As the majority judgment of Michael Green J and Sir Iain McMillan recorded at [550], the CAT had been unanimous as to the law on SPO set out at Section H and summarised at [11] to [30] above. It recorded that, for the purposes of legal causation, a number of factors needed to be taken into account which included the four relevant factors already set out at [228] (cited at [30] above).

The CAT dealt with those factors in turn. It noted at [551] that there was no dispute about knowledge: “neither of the Claimants knew anything about the Infringement or the Overcharge at the time”, nor of any “particular increase in their truck costs”. Therefore, “[t]hey could not be said to be specifically seeking to address the costs increase”. At [552] the majority judgment recorded that it was also accepted that “the size of the Overcharge was, for both Claimants, tiny relative to their overall costs and revenue”. DAF argued “size is irrelevant if, as a matter of fact, it was passed on to the downstream customers”, whereas the Claimants said their prices were not “fine-tuned” enough to be able to conclude they were actually higher as a result of the Overcharge.

As to the relationship between the trucks bought and the products sold, it could “fairly be said” that customers who (for example) purchased stamps from Royal Mail are essentially purchasing transportation services for their mail; that could not really be said for BT’s products although it was treated as an input cost internally allocated to the vast array of its products. At [554] the majority said that:

“The question of identifiable claims by identifiable purchasers is an important albeit not necessary factor. The factual question with which we are concerned is, as defined by the Supreme Court in Sainsbury’s, whether by passing-on the cost to their customers the Claimants “transferred all or part of [their] loss to others”. It might be thought that the “others” should be identified, whether as a class or not, together with at least an approximation as to the amount of the loss that was so transferred.”

At [555] the majority noted that the experts on this issue, Mr Bezant for DAF and Mr Harvey for the Claimants:

“…were agreed that they were engaged in an analysis of what would have happened in the counterfactual if there was no Overcharge. As there were regulatory price controls in place for much of the Infringement period, the question is largely whether there would have been a different outcome to the respective regulator’s price control in the counterfactual, as the Claimants tended to price up to the price cap.”

Contrary to the arguments of DAF, the majority considered at [556] that the small size of the Overcharge in this case “is a highly significant factor”. The majority summarised the experts’ arguments and criticisms at [568] to [570], and considered that before turning to analyse the expert evidence, it was relevant to consider “where the expert evidence actually takes us in this case in relation to SPO, given our conclusions on the law”, concluding:

“572.

As we have already said, in relation to the four factors identified in [550] above, none of them are present in this case. The absence of knowledge, together with the tiny size of the Overcharge, means that there was obviously no specific decision by the Claimants to increase prices in response to the increase in costs. Nor is there any direct association between truck costs and the products sold by the Claimants, even though an element is properly attributable to each product. And even if it can be shown that there was an increase in prices because of an increase in costs, it will be impossible to identify which prices in relation to which specific products actually increased because of the Overcharge. Therefore, we find it difficult to see how there can be sufficiently identifiable purchasers from the Claimants who could make a claim in respect of the Overcharge or to whom it could be said that the loss suffered by the Claimants had been transferred.

573.

In the circumstances, we do not think that DAF can satisfy the legal test for causation which requires the Overcharge to be a direct and proximate cause of the increase in specific prices. Even if, as a matter of forensic accountancy, DAF is able to show that the miniscule Overcharge can be traced through the series of internal steps, judgments and regulatory intervention resulting in a higher price setting, the absence of the four factors means that the Overcharge is too remote from the downstream prices. While the four factors are not themselves decisive or necessary, we think that in a situation where none are present, the evidence of factual causation needs to be that much stronger so that the requisite proximity can be established.

574.

We will still examine the evidence and the experts’ opinions to see if it is strong enough to overcome the absence of the relevant factors we have identified. Despite Mr Bezant’s careful, meticulous and professional approach to the material that he had, we are clear that his evidence does not sufficiently bridge the gap between the Overcharge and downstream prices so as to establish on the facts the requisite proximity to satisfy the legal test for causation.”

The majority then turned to analyse the expert evidence in detail. In respect of Royal Mail, in general the majority considered that there was a lack of evidence that demonstrated a tiny increment in one costs input would lead to an increase in downstream prices, and that Mr Bezant, DAF’s expert, had not “demonstrated that it is highly likely, or at least more likely than not” ([605]) that the price set would have been different without the tiny Overcharge. The conclusion was reached by the CAT at [606] in relation to PC2, one of the price controls imposed on Royal Mail by the regulator:

“There is a further problem for DAF’s case. The price control is an overall cap on Royal Mail’s revenue but it does not dictate what prices Royal Mail must set. It was accepted by Royal Mail that it generally sought to recover the maximum allowed revenue and therefore would price up to the cap. But this could be quite challenging as Royal Mail would have to predict sales volumes and the mix of products and assess the impact on demand of price changes. This requires a substantial degree of commercial judgment and imprecision is inherent in the process. Far from it being a mechanical exercise that an increase in a tiny amount of costs will inevitably feed through to the price cap, the price setting process for each product is much more complicated than that and involves judgment, both commercial and regulatory, as well as inherent uncertainty and imprecision. It will also be impossible to identify into which of Royal Mail’s products the Overcharge was passed on and therefore to whom the loss was transferred.”

A similar conclusion was reached in relation to PC3 at [616]-[617] that DAF had not proved factual causation.

In respect of BT, the majority rejected DAF’s case on SPO in relation to the Openreach products. They dealt with the use by Mr Bezant of probability analysis at [658]-[659]:

“658.

However, DAF bears the burden of proof of showing that the very small Overcharge actually did make a difference. The charge controls are not set down to the fullest possible level of cost granularity. They may be set in whole pounds or pence. An RPI–X may be set to a whole number or one decimal place, meaning that it cannot capture a tiny cost increase. That is why the probability analysis came to the fore of Mr Bezant’s argument.

659.

The probability analysis is concerned with the rounding of the value of X in the glidepath control. Mr Bezant’s argument is that at some point the rounding would tip over into the next level, at which stage the Overcharge, however small, would likely be recovered. On the face of it, the fact that Mr Bezant is driven to having to make this sort of argument seems to demonstrate DAF’s inability to trace the Overcharge through into downstream prices and therefore to be a long way from the proximity required to satisfy the legal (or factual) test for causation. Nevertheless, we will explore it a little further.”

The CAT then considered Mr Bezant’s evidence in some detail, concluding at [667]-[668]:

“667.

In our view, this is wholly inadequate evidence upon which to prove that there has been SPO, let alone 100% SPO, in relation to the Overcharge attributed to Openreach. DAF cannot show this actually happened. Even if it is more likely than not that one (or more) charge control “tipped”, DAF cannot show which one did and when it happened. It is impossible to identify which downstream customers may have ended up paying the Overcharge or who may have a claim against DAF. It cannot be said that BT has recovered the Overcharge from others and so factual causation has not been established. Even if BT might have hit the jackpot at some point, that cannot represent the recovery of the Overcharge and we do not think that that could have been the intention of Ofcom as to the way it would work.

668.

We also think that this cannot amount to sufficient proximity between the Overcharge and the prices charged to Openreach’s customers to satisfy the legal test for causation. Accordingly, we reject DAF’s case on SPO in relation to Openreach.”

They then went on to conclude that it was impossible to say that there had been SPO in relation to other products.

The majority’s overall conclusion on SPO was at [688]-[691]:

“688.

We have rejected DAF’s case on SPO in relation to both Royal Mail and BT for all periods and lines of business. We have found on the balance of probabilities that DAF has not established on the facts that the prices charged to the Claimants’ customers would have been lower in the counterfactual absent the Overcharge. We have also been clear that, as a matter of law, we do not consider there to be the necessary proximate and direct causative link between the Overcharge and the downstream prices so as to satisfy the legal test for causation.

689.

In coming to those conclusions, we paused to consider the impact both on potential downstream customer claims and whether the Claimants might therefore be overcompensated for the losses they actually suffered...

690.

But our conclusion on the evidence before us is that there was no SPO. We cannot rule out the possibility that the Claimants’ customers might try to claim in the future, whether by class or individually, and our findings will not be binding on them. We agree with Mr Ridyard that the monetary size of the Overcharge together with the number of downstream customers makes it virtually impossible for them to mount a viable claim even if they were able to prove SPO... But we cannot shrink from such conclusions because of their potential impact on unknown other claims. It necessarily follows from our findings that we do not think that, in the words of the Supreme Court in Sainsbury’s, the Claimants have “recovered from others” the Overcharge or “transferred all or part of [their] loss to others”. In the circumstances, the Claimants are not being overcompensated.

691.

As we have said above, it is important to distinguish between the economic concept of pass-on and the legal test for causation in relation to mitigation of loss. The former is likely to be much broader than the latter which requires there to be demonstrated a proximate causal connection between the Overcharge and an increase in downstream prices. Mere recovery of costs is insufficient proof of such a connection. Something more is required and we are satisfied that DAF has not in the end provided us with anything more than that the increase in truck costs represented by the Overcharge was taken into account in the price setting process, whether by the respective regulators or the Claimants themselves. A number of other factors were also taken into account as well as costs and these were overlain with regulatory, public interest and commercial judgments being made. It is not possible to say that an increase in truck costs, however small, was likely to have led to an increase in prices. And if that is the case, there can be no SPO defence of mitigation.”

In his dissenting opinion, Mr Ridyard set out at [692] that, although he agreed with the majority on the overall conclusion on DAF’s SPO defence, he disagreed as to the reasoning:

“…I believe, contrary to the majority view, that it is likely that both Claimants did pass on a substantial amount of the Overcharge to their downstream consumers, and that there is a sufficiently close causal connection between the Overcharge and a likely SPO. However, I am not persuaded that the SPO argument should be used to impose a reduction in the damages awarded to the Claimants because, given the specific facts associated with this case, to do so would jeopardise the principle of effectiveness.”

He evaluated the SPO issues against the four factors set out at [30] above. In broad overview, his conclusions were as follows:

“I. Knowledge

…

704.

… in the current case the Claimants had no knowledge of the trucks Cartel or of DAF’s Overcharge... Hence, whilst I agree that the visibility of an overcharge would make it more straightforward to establish a causal link with any consequent change in claimant behaviour, it cannot be regarded as a necessary condition. I do not place significant weight on this factor in my assessment of the current case.

II. Relative size

705.

The second factor, the size of an overcharge relative to the value of the claimant’s downstream business, has an obvious influence on the ability to measure and identify a pass-on effect. In the current case there is no dispute that the Overcharge we have found, whilst substantial in its own right at somewhere in the region of £15 million in historic values, is extremely small relative to the value of the Claimants’ downstream businesses. This factor renders any attempt to measure pass-on empirically hopeless, and it plainly presents the biggest obstacle to proving the existence of pass-on.

706.

The key question is whether this practical impossibility of measuring the specific downstream impact of a pass-on effect is sufficient to prove (on a balance of probabilities test) that such an effect does not exist. I believe this is the factor that has most influenced my colleagues to reach their conclusion on pass-on, and I fully understand and respect their rationale. However, I do not agree that the fact that an effect is too small to be measured or separately identified within the price of the downstream product means that it must be unlikely to exist. To make that assessment, it is necessary to look to other contextual evidence that might reveal the existence of a likely pass-on mechanism at work.

III. Relationship between upstream costs and downstream prices

….

710.

.... Trucks are purchased by the Claimants in order to enable them to provide their downstream postal and telecommunications services, so they are in both cases “components or costs” that are directly used in the downstream activities in which the SPO is alleged to occur. The proposition that trucks were an input used by both Claimants in providing their respective downstream services was accepted by the Claimants’ expert Mr Harvey. The fact that in both cases the impact of these components is dramatically diluted by the costs associated with all the other inputs that also go in to the provision of the Claimants’ downstream operations does not negate the fact that they are related. I consider that this applies equally to both Claimants…

711.

… Truck costs formed a part of vehicle costs for both Claimants... it seems clear to me that the Overcharge must also have been included in them during the relevant time period, whereas in the counterfactual it would have to be deducted.

712.

The next question is to address the causal connection between the Overcharge and downstream prices. As Mr Bezant’s evidence makes clear, that must be addressed primarily by examining the way the Claimants’ businesses were regulated… There are many facets to this assessment, but a central premise is that the regulated firms should be entitled to recover reasonably incurred (efficient) levels of cost from their monopoly activities…

…

720.

In my assessment… none of these complicating factors fundamentally undermines the conclusion that the revenues earned by the Claimants in their respective downstream markets were substantially dependent on a regulatory process that was designed to remunerate reasonably incurred (efficient) costs. Since the trucks Cartel was unlawful and covertly implemented, I do not see any basis on which the Overcharge paid by the Claimants could have been regarded as anything other than a reasonably incurred cost of providing their downstream services.

721.

This is not to say that the Claimants would automatically have achieved 100% SPO, but in terms of the balance of probabilities I regard it as overwhelming likely – and certainly more likely than not - that a substantial part of any Overcharge would have found its way into the regulatory system and have been reimbursed through the price caps and constraints.

…

724.

By choosing the actual outcomes as the relevant benchmark, Mr Harvey’s approach places the burden on the counterfactual assessment to show how the absence of the Overcharge would be “fine-tuned” to deliver a different outcome. As Mr Bezant observed, one likely consequence of this approach is that, to the extent that there is any inertia in the setting of the Claimants’ downstream prices, Mr Harvey’s “what changes?” question requires the Overcharge itself to overcome that inertia in order to establish the likelihood of an SPO effect. Given the very small scale of the Overcharge relative to the downstream market value, it is unsurprising that the Overcharge is unable to overcome this inertia. However, I consider that Mr Harvey’s “what changes?” question is the wrong one to ask when addressing the legal question, and that this biases the assessment in a way that is likely to understate the degree of pass-on.

…

727.

Throughout all the above, I acknowledge that the small size of the Overcharge relative to downstream market values is problematic for DAF’s pass-on argument. My point is that small pass-on effects can exist even if they are not easily identifiable, and that pass-on arguments should be able to succeed if there is a sufficiently clear factual basis for establishing that such pass-on occurs. In my assessment, Mr Bezant’s evidence of a causal connection between the Claimants’ input costs and downstream prices is sufficient to meet that test.

IV. Identifiability of downstream claims and implications for the principle of effectiveness

728.

To the extent that the Claimants did pass on some or all of the Overcharge in their downstream markets, the passed-on cost (and hence damage) was in most cases likely to have been felt by customers of the Claimants’ businesses...

729.

For reasons I describe further below, I do not think it is necessary to arrive at a specific value of the damage that is passed on to these downstream customers, but… it is evident that the passed on damage to any individual customer will be very small, and a matter of a few pence in the case of individual consumers or households.

731.

However, simply identifying the downstream claimants for any pass-on in this case does not in itself establish that they would be able to make a viable claim against DAF. In this respect, it is important to note that the guidance issued by the Supreme Court in Sainsbury’s included the need to ensure that any approach on pass-on did not offend the principle of effectiveness. Specifically, it is necessary to consider whether the prospects of a successful claim from downstream customers against DAF would be “excessively difficult or impossible”.

732.

I think it is obvious that there is a very high risk that downstream claims for any passed on damage in this case would indeed fail this test. Individual claims would be far too small in value to be viable, and even a collective action on behalf of Royal Mail and/or BT consumers would be likely to face extreme difficulty...”

Mr Ridyard considered at [733] that this conclusion created “an obvious dilemma”: if a substantial degree of pass-on is more likely than not, then a damages award that pays the full Overcharge to the Claimants would involve over-compensation. On the other hand, if a successful claim from downstream customers for their share of the passed on damage would also be excessively difficult or impossible, then an award that covers only a part of the damage caused by DAF’s unlawful act “would seem to fall foul of the principle of effectiveness” . He suggested at [734] possible avenues to resolve this dilemma, by making full payment to the Claimants but with some sort of set-aside in respect of potential claims by downstream customers.

Mr Ridyard’s conclusions are summarised at [738]. He considered that: legal proof of an SPO effect does not require precision; proof cannot be assumed but requires evidence of a causal link; the fact that trucks are an input employed in downstream services is critically important to assessing whether there is a causal link and that category (iv) [from the categories in Sainsbury’s SC as set out at [16] above] pass-on has a stronger basis in economic theory than others; the fact that the Overcharge is relatively very small makes it more challenging to identify a causal link and virtually impossible to measure the effect. His disagreement with the majority arose because (1) he did not believe that an effect that is too small to measure cannot exist; and (2) DAF’s expert had shown it was likely that regulatory processes would have allowed the Claimants to pass on their costs in downstream prices. The principle of effectiveness created a dilemma, albeit this did not need to be resolved given the majority’s findings.

The CAT addressed at [739] to [752] the issue of whether an adjustment to any SPO impact should be made to cover loss of volume. It did not arise given the findings on SPO and is not relevant to this appeal.

The CAT then considered the additional claims for damages in respect of the cost of financing the overcharge. As it noted at [755], the Claimants parted company. BT claimed simple interest pursuant to section 35A of the Senior Courts Act 1981 whereas Royal Mail claimed it should be compensated for its historic losses by way of compound interest based on its weighted average cost of capital (“WACC”). The CAT had no difficulty in favouring a compound interest calculation over simple interest, as this accorded with economic reality. However, it rejected Royal Mail’s argument that WACC was the appropriate measure of its financing losses. It went on to consider an alternative measure proposed by Royal Mail based on a combination of its cost of debt finance and its returns on short term investments over the relevant finance period. The CAT’s findings on this measure are at issue on the appeal.

The Tribunal noted at [798] that this alternative measure “is broadly what happened in CAT Sainsbury’s and involves assumptions as to how Royal Mail would have used the additional funds that it would have had in the absence of the Overcharge”. The experts were agreed that the relevant question was whether the funds would have been used to increase actual investments in short-term investments or to reduce the amounts of debt Royal Mail had held in the past. The only issues between the experts were Royal Mail’s cost of debt from 2013/14 to 2021/22 and the appropriate weighting to be applied as between short-term investments and debt.

The cost of debt issue is addressed from [801] to [818] and is not appealed. The issue of the relevant weighting is addressed from [819] to [824]. In [819] it was noted that:

“This area of disagreement between the experts concerned how Royal Mail’s cost of debt and return on short-term investments should be combined into a single figure for the appropriate interest rate to apply to past financing losses”.

The CAT first addressed the approach of Royal Mail’s expert, Mr Earwaker, noting that he was arguing for a “very binary approach” to Royal Mail’s finance costs that split the relevant period into two with 100% weighting to short-term investment returns and 0% weighting for debt or vice versa: for the period from 1997 to 2007/08 the rate was wholly based on Royal Mail’s actual returns achieved on various short-term investments; and for the period 2008/09 to the present, it was wholly based on its cost of debt.

At [820] the CAT said that the rationale for this approach was that, in the earlier period, Royal Mail was not reliant on borrowings and had spare cash which it deployed in short-term investments; whereas in 2007 its short-term investments shrank rapidly and in this latter period it became dependent on external loan finance. Mr Earwaker argued that in the counter-factual earlier period Royal Mail would have committed more funds to these short-term investments (and hence lost out on those returns), whereas in the latter period the most practical assessment is to assume it would have borrowed less absent the Overcharge (and hence would have paid less interest). He explained that its short-term investments “shrank considerably during 2007 as postal operations faced increasing financial challenges” and from 2008 onwards it “took on additional borrowing to finance its UK operations as its previously profitable business started to record losses”. His understanding from the witness statement of Mr Jeavons, the Chief Financial Officer of Royal Mail, was that its remaining short-term investments in that period represented a “basic level of working capital” for short term access, “rather than the kinds of surpluses of excess cash” in the earlier period. It therefore held cash reserves in short-term investments at the same time as it was a net borrower of money because it needed to have some short-term funds available for liquidity purposes.

The CAT considered at [821] that: “Mr Earwaker took a similar approach to that taken in CAT Sainsbury’s, using a ‘broad axe’ to assess how Royal Mail would have been likely to have used the extra funds in the counterfactual”.

At [822] the CAT noted that by contrast Mr Delamer, DAF’s expert, adopted a “blended approach in which debt and investment income are both considered relevant across the entire period”, applying weights based on the relative values of Royal Mail’s debts and short-term investments at any point in time. The CAT noted the parties’ arguments in respect of this approach: where Mr Beard KC submitted that this had the merit of not speculating on what Royal Mail would have done with the extra funds in the counterfactual (particularly where its evidence had not addressed the point), Mr Lask for Royal Mail submitted that Mr Delamer’s approach was “blunt and oversimplistic”, because it assumed Royal Mail would have used the additional funds to make short-term investments and reduce debt in precisely the relative proportions that they bore to each other.

The CAT concluded on this issue:

“824.

We prefer Mr Earwaker’s approach which is based on how a rational business such as Royal Mail would have used extra funds that it had at the relevant time. His two-period characterisation of Royal Mail’s financial position, as a net investor in the first period and a net borrower in the second, is credible on the evidence and it would therefore be more likely that Royal Mail would use the funds in one direction rather than two. That therefore is a reasonable way to assess Royal Mail’s actual cost of financing the Overcharge.”

The CAT’s judgment as to BT’s simple interest claim, addressed at [825] to [830], is not appealed.

DAF pursues four grounds of appeal. The CAT granted permission on Ground 2 and I granted permission on Grounds 1, 3 and 4:

Mr Daniel Beard KC on behalf of DAF began his submissions on the quantum issue raised by Ground 1 with a citation of the applicable principle set out in [52-001] of McGregor on Damages (21^st edition):

“The claimant has the burden of proving both the fact and the amount of damage before they can recover substantial damages. This follows from the general rule that the burden of proving a fact is upon the person who alleges it and not upon the person who denies it, so that where a given allegation forms an essential part of a person’s case the proof of such allegation falls on them. Even if the defendant fails to deny the allegations of damage or suffers default, the claimant must still prove their loss.”

He posed the question as to how the Court deals with issues of uncertainty in relation to proof of the quantum of damage and referred to three principles which arise in this context: the principle of Armory v Delamirie (1721) 1 Str 505, the principle of the broad axe and the principle, derived from European law, of effectiveness. On the first two, he cited [37] and [38] from the judgment of Lord Reed in One Step v Morris-Garner [2018] UKSC 20; [2019] AC 649:

“37.

The quantification of economic loss is often relatively straightforward. There are, however, cases in which its precise measurement is inherently impossible. As Toulson LJ observed in Parabola Investments Ltd v Browallia Cal Ltd (formerly Union Cal Ltd) [2010] EWCA Civ 486; [2011] QB 477, para 22:

“Some claims for consequential loss are capable of being established with precision (for example, expenses incurred prior to the date of trial). Other forms of consequential loss are not capable of similarly precise calculation because they involve the attempted measurement of things which would or might have happened (or might not have happened) but for the defendant’s wrongful conduct, as distinct from things which have happened. In such a situation the law does not require a claimant to perform the impossible, nor does it apply the balance of probability test to the measurement of the loss.”

An example relevant to the present case is the situation where a breach of contract affects the operation of a business. The court will have to select the method of measuring the loss which is the most apt in the circumstances to secure that the claimant is compensated for the loss which it has sustained. It may, for example, estimate the effect of the breach on the value of the business, or the effect on its profits, or the resultant management costs, or the loss of goodwill: see Chitty on Contracts, 32nd ed (2015), paras 26-172 - 26-174. The assessment of damages in such circumstances often involves what Lord Shaw described in Watson, Laidlaw at pp 29-30 as “the exercise of a sound imagination and the practice of the broad axe”.

38.

Evidential difficulties in establishing the measure of loss are reflected in the degree of certainty with which the law requires damages to be proved. As is stated in Chitty, para 26-015, “[w]here it is clear that the claimant has suffered substantial loss, but the evidence does not enable it to be precisely quantified, the court will assess damages as best it can on the available evidence”. In so far as the defendant may have destroyed or wrongfully prevented or impeded the claimant from adducing relevant evidence, the court can make presumptions in favour of the claimant. The point is illustrated by the case of Armory v Delamirie (1721) 1 Str 505, where a chimney sweep’s boy found a jewel and took it to the defendant’s shop to find out what it was. The defendant returned only the empty socket, and was held liable to pay damages to the boy. Experts gave evidence about the value of the jewel which the socket could have accommodated, and Pratt CJ directed the jury “that, unless the defendant did produce the jewel, and shew it not to be of the finest water, they should presume the strongest against him, and make the value of the best jewels the measure of their damages: which they accordingly did”.

In relation to the Armory v Delamirie principle, Mr Beard KC referred to the judgment of Hamblen J (as he then was) in Porton Capital Technology Funds v 3M UK Holdings Ltd [2011] EWHC 2895 (Comm) at [240] to [245] as demonstrating the limited scope of the principle which should not be extended so as to undermine the test of the balance of probabilities.

In relation to the broad axe principle, he asked the Court to note that its origin in the speech of Lord Shaw of Dunfermline in Watson v Laidlaw was in the context of loss of reputation or personal injury, but as the Court pointed out, that was a patent infringement case, so that it was a case of financial loss in which the principle applied. Mr Beard KC referred to a number of cases where the broad axe principle has been applied, including Sainsbury’s SC, but submitted that its application did not mean that a claimant was relieved from proving the substance of its loss.

At one point in his submissions about the CAT’s judgment Mr Beard KC appeared to be submitting that the CAT had not applied the balance of probabilities test, but the broad axe, in determining whether the Claimants had suffered a loss at all. However, on being pressed by the Court, he confirmed that he accepted that, on the evidence, the CAT was entitled to find that there was a loss, but he submitted that it was not entitled to reach the conclusion that the loss was sizeable. Mr Beard KC submitted that the broad inference that there was such a loss from the fact of the infringement for a substantial period of time referred to in [477] of the judgment was not justified, because the EC Decision was that this was an object infringement not an effects one. Where the CAT went wrong was that it concluded there was a loss, but did not like either Mr Harvey’s approach or Professor Neven’s approach, so it split the difference. Mr Beard KC submitted that this was not justified because the CAT had not made a finding that one could make out a cartel effect from Mr Harvey’s evidence. Although the Court pointed out that that was what the CAT had found in [476] to [478], Mr Beard KC maintained his position that the CAT had impermissibly applied a broad axe approach, rather than concluding that the Claimants had to prove that there was a substantial loss. He submitted that in the light of the fundamental identification issues with Mr Harvey’s approach (which was the basis for the assertion that there was a substantial loss) the CAT should have concluded that it had not been established that there was an Overcharge, therefore the claim failed. He submitted that [476] and [477] were dealing with causation and the CAT then jumped at [479] to the broad brush approach to quantification of loss, missing out a middle section dealing with the burden and standard of proof.

In relation to Ground 2, Mr Beard KC referred to the various options in relation to an Overcharge set out in Sainsbury’s SC at [205] of its judgment, although as Green LJ pointed out, the Supreme Court was really reciting the findings of fact made by the CAT in that case. The present case is only concerned with option or category (iv). Mr Beard KC also referred to the distinction between factual causation and legal causation identified at [215] and [216] of the Supreme Court judgment as clarified by the CAT in the MIF Umbrella Proceedings judgment, where at [50(2)] the CAT said:

“…causation (which, as we have said, the second aspect of mitigation turns on) itself has two aspects, “legal” causation and “factual” causation:

(i)

Factual causation is the more obvious of the two: it involves consideration of whether the effect of the alleged mitigating conduct was, as a matter of fact, to reduce or eliminate B’s loss.

(ii)

Legal causation concerns the question of whether – even if the effect of the alleged mitigating conduct was, as a matter of fact, to reduce or eliminate B’s loss – as a matter of legal policy it should serve to reduce or eliminate the amount of damages that A should pay B. The question arises quite frequently and is an elusive one. Thus, the fact that a claimant receives an indemnity by virtue of a contract of insurance is regarded as “collateral” to the defendant’s liability and thus will not affect it. In personal injury cases, the fact that the claimant receives some benefits as a result of his or her injury is also generally regarded as “collateral”.

Having identified the two aspects of causation, the Supreme Court then proceeded to say – in a single sentence in paragraph [215] – that no issue of legal causation arose: “…But the question of legal causation is straightforward in the context of a retail business in which the merchant seeks to recover its costs in its annual or other regular budgeting.” It seems to us very difficult to identify any policy reason why B should nevertheless continue to be able to claim the Overcharge from A, despite having passed it on to C. Indeed, one can see very strong reasons for not permitting B to persist in such a claim, because (as we have described) on these facts C will have a claim against A, and A should not be obliged to pay twice over. Frankly, we can see exactly why the Supreme Court regarded this as a “no brainer”. As the Supreme Court noted, the difficult question is that of factual causation.”

Mr Beard KC submitted that Mr Ridyard’s dissenting opinion had correctly identified that both Claimants had used the regulatory process to which they were subject at the relevant time to pass on higher truck prices to their downstream customers through elevated prices for their services. Unlike the majority of the CAT, he considered that the fact that the specific downstream impact of pass-on is too small to be measured did not mean that SPO did not exist.

He submitted that in contrast, the majority conclusion was confused and incorrect. They had been wrong to conclude at [573] that there was no legal causation in this case. They had also been wrong to conclude that, because the four factors they identified at [550] were absent here, there needed to be stronger evidence of factual causation.

In relation to Ground 3, Mr Beard KC submitted that the CAT erred in including truck bodies in the VoC because the truck bodies are not within the infringement found in the EC Decision. This Ground only concerns rigid trucks supplied by DAF to Royal Mail. It does not affect BT. There were, as the CAT found, a number of separate body manufacturers who supplied the bodies for the Royal Mail rigid trucks, none of whom was party to the infringement. However, Mr Beard KC submitted, Royal Mail had convinced the CAT to treat these third party bodies as part of the truck that was subject to the infringement, notwithstanding that DAF only had two customers, Royal Mail and Morrisons, to whom it sold the third party bodies at all. Apart from those two customers, bodies were purchased directly from the third party manufacturers.

He submitted that at [473], the CAT had been wrong to conclude that DAF had not produced any evidence that bodies were not included within the infringement. The evidence of Mr Ashworth, the Managing Director of DAF UK, in his witness statement was that DAF either charged Royal Mail at cost for the bodies which were fitted onto the trucks or took a small margin to reflect the work DAF had to do in liaising with the body manufacturers. That unchallenged evidence was referred to in the last sentence of [472]. Mr Beard KC submitted that this illustrates that bodies were not within the cartel arrangements.

Furthermore, he submitted that the CAT had been wrong to conclude in [473] that the EC Decision did not clearly exclude bodies from the scope of the infringement. Nothing in the EC Decision referred to truck bodies at all. He referred to a number of provisions. First, Recital (5) which provided:

“The products concerned by the infringement are trucks weighing between 6 and 16 tonnes ("medium trucks") and trucks weighing more than 16 tonnes ("heavy trucks") both as rigid trucks as well as tractor trucks (hereinafter, medium and heavy trucks are referred to collectively as "Trucks"). The case does not concern aftersales, other services and warranties for trucks, the sale of used trucks or any other goods or services sold by the addressees of this Decision.”

He submitted that truck bodies sold by third parties were within “other goods or services”.

Recital [26] provided:

“…Trucks are not commodity products but are specified according to individual customer requirements and are inherently complex. All of the Addressees offer a range of trucks and hundreds of different options and variants.”

The options and variants being referred to were in relation to the truck itself, leaving aside issues to do with bodies and trailers.

Recital (27) referred to the pricing mechanism in the trucks sector, which he submitted was referring to the manufacturers’ own pricing, not to third party product pricing. In relation to recital (28) Mr Beard KC noted that the Claimants relied on the last sentence, which provided: “The EEA price lists contained the prices of all medium and heavy truck models as well as all factory-fitted options that the respective manufacturer offered.” He argued that the factory-fitted options being referred to were not the bodies but the options and variants like engine capacity, cab size and length.

In relation to Ground 4, Mr Beard KC referred the Court to the decision of the Privy Council in Sagicor Bank Jamaica Ltd v Seaton [2022] UKPC 48; [2023] 1 WLR 1759, which was not cited before the CAT and which was a claim for damages for the loss of use of money. In the judgment at [23] to [28] there was extensive citation from the speeches in the House of Lords in Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34; [2008] 1 AC 561 (“Sempra”). In relation to claims for compound interest as damages for breach of contract, the Board said at [37]:

“In summary, interest, including compound interest, may be awarded as damages for breach of contract. A plaintiff seeking interest as damages where the defendant has withheld money in breach of contract must plead and prove its loss. If a plaintiff pleads that it has incurred loss by having to borrow replacement funds, what it must prove are facts and circumstances from which a court may properly infer on the balance of probability that it has borrowed funds to replace that which has been withheld from it.”

Mr Beard KC submitted that the approach of the CAT to this issue in its judgment (as cited at [74] and [75] above), which preferred Mr Earwaker’s evidence to the effect that in the period up to 2007/2008, Royal Mail would have preferred to make more short-term investments and in the period thereafter to pay off debt, was not supported by any evidence. Whilst there was evidence about how the money was used in the relevant periods, there was no evidence about what Royal Mail would have done differently if it had not suffered the putative Overcharge. The CAT had simply unduly wielded the broad axe. In his reply submissions, Mr Beard KC returned to this point and said that there was no factual evidence from Mr Jeavons, (who gave evidence) that supported the binary split put forward by Mr Earwaker.

Mr Tim Ward KC dealt with the first three grounds of appeal on behalf of the Claimants. He submitted that the appeal suffers from three fundamental flaws. First, it disregards binding high authority on matters such as the meaning of the broad axe. Second, it cannot overcome the doctrine of appellate restraint. There was no error of law and DAF has to establish that the CAT’s approach to the evidence was irrational. It was not. On the contrary its findings are carefully reasoned and compelling. Third, it proceeds by mischaracterising the CAT’s reasoning.

Mr Ward KC began with the issue of Overcharge. He observed that the CAT’s finding of Overcharge was based on more than 600 pages of expert evidence, expert hot-tubbing and cross-examination. He pointed out that DAF does not challenge the CAT’s finding that at least material harm was established on a balance of probabilities, but its case was that the Claimants should still walk away empty handed. He submitted that this was based on a fundamental misapprehension of the law and, although Mr Beard KC had cited a number of authorities, he had not taken the Court to the judgment of the majority of the Supreme Court in Mastercard v Merricks [2020] UKSC 51 which contains an authoritative statement as to how the broad axe should be applied. Difficulties in quantification should not defeat attempts to quantify the loss and victims of the cartel have the right to have the loss quantified. There is no gap between causation and quantum as Mr Beard KC had suggested.

Mr Ward KC submitted that it is wrong in principle and unfair to pick out isolated features of the approach and reasoning of the CAT without placing them within the broader factual context. This is a point which was well made in a case of Overcharge by this Court in BritNed at [123]:

“As will now be apparent, we have found it necessary to review the parts of the main judgment dealing with the assessment of the overcharge at considerable length, but we make no apology for doing so. In order to ascertain the extent of the loss caused to BritNed by the cartel, as reflected in the price which it paid for the cable element of the Interconnector project, it was necessary for the judge to conduct a wide-ranging and multi-factorial evaluation of all the evidence deployed before him during a four week trial. It would be wrong in principle, and unfair to the judge, to pick out isolated features of his approach and reasoning, without placing them within the broader context of the full picture which he so painstakingly constructed. It is also essential for us to keep firmly in mind the well-known principles of appellate restraint in relation to questions of fact, including the evaluation of primary facts and the inferences to be drawn from them, which have been emphatically restated in a plethora of recent cases of the highest authority. Those principles apply just as much to cases in the field of competition law as they do in other areas of civil litigation. They also apply to the assessment of expert opinion evidence no less than they do to findings based on the evidence of witnesses of fact.”

The CAT was well aware of this broader context, as it said in the last two sentences of [479] of its judgment quoted at [52] above.

Mr Ward KC highlighted five factors from the factual section of the judgment. First, that the cartel was conducted in secret for a number of years, so that the CAT was looking at modelling which went back more than twenty five years. It is scarcely surprising that there were data challenges and gaps in the documentary record. Second, although this was a follow-on claim, the CAT only had a Settlement Decision to go on, which is in short form, a point made by Rose LJ (as she then was) in AB Volvo v Ryder [2020] EWCA Civ 1475 at [83] quoted by the CAT at [15] of its judgment. Third, DAF had taken a strategic decision, as recorded in [18] of the CAT judgment, not to produce evidence about how it used the information obtained from its competitors or to call any witnesses who knew about or had participated in the cartel.

Mr Ward KC noted that DAF’s case was that it was not plausible the cartel had any effect, but he referred to what the CAT had said about this at [38] of its judgment:

“All the unlawful exchanges and agreements between the Cartelists were carried out for the same purpose, namely restricting price competition in the whole of the EEA. For this to have been sustained in such a concerted manner by all the Cartelists for 14 years without any of them leaving, and taking very considerable risks in the process, it would be most unlikely to think that they were not each receiving substantial benefits for continuing with it for so long.”

As I pointed out in argument, this was a point to which the CAT returned several times in its judgment, including at [477] quoted at [51] above.

Mr Ward KC asked the Court to note the two presumptions which the CAT said should be made against DAF. The first, referred to in [109] is the so-called Anic presumption(derived from the decision of the European Court of Justice in Commission v Anic Partecipazioni [1999] ECR I-4125)that undertakings which take part in collusive arrangements “take account of the information exchanged with their competitors in determining their conduct on that market”. At [110] the CAT recorded what Mr Ward KC had described as a significant concession by Mr Beard KC that: “…in relation to gross list price exchanges, it has never been part of our case that we just ignored them…we accept that in relation to gross list pricing information that these matters were taken into account. We are not saying we ignored the information we received in that regard.” As Mr Ward KC commented to the Court, that was a case that the gross price list exchanges were taken into account but this just had no effect on prices. That evidently did not impress the CAT which said at [112]:

“…The fact that DAF admits that it took account of the information when determining its conduct on the market must be part of the relevant matrix of fact in considering whether that conduct influenced transaction prices.”

The second presumption is the one deriving from Prest v Petrodel Resources Ltd [2013] UKSC 34 that adverse inferences can be drawn from DAF’s failure to adduce evidence as to how the cartel operated and how DAF used the information received for its benefit. At [115] the CAT recorded the explanation Mr Beard KC had given as to why the relevant directors were not called to give evidence, but the CAT rejected that as speculation concluding at [116]-[117]:

“…in our view, this means that, if DAF wished to argue that, because of the way it used the confidential information obtained through the Cartel, there was no effect on prices, it would have had to adduce factual evidence to such effect. In other words, DAF’s admissions and the Settlement Decision establish a prima facie case that the Cartel had an adverse effect on transaction prices.

117.

That is not to say that DAF is unable to rely on its expert evidence to argue that the data shows that there was no Overcharge paid by these Claimants. But even their expert was unable to explain or come up with a rational economic basis for DAF’s participation in the Cartel over such a long period. While Prest does not entitle the Claimants to say that they have therefore proved that DAF’s participation in the Cartel led to higher prices it does mean that it is not open to DAF to argue that, as a matter of fact, the information was not used by it to achieve prices that were higher than they would otherwise have been without that information exchange.”

As Mr Ward KC said, none of this was referred to by Mr Beard KC before this Court and there is no appeal against any of these conclusions. Instead there was extended reference to Armory v Delamirie, which was not relied upon at all by the CAT. Mr Beard KC had sought to argue that competition damages claims do not engage the public interest but the refutation of that argument was in the decision of the European Court of Justice in Courage v Crehan (2001) [2002] QB 507 at [26]-[27]:

“26.

The full effectiveness of Article 85 of the Treaty and, in particular, the practical effect of the prohibition laid down in Article 85(1) would be put at risk if it were not open to any individual to claim damages for loss caused to him by a contract or by conduct liable to restrict or distort competition.

27.

Indeed, the existence of such a right strengthens the working of the Community competition rules and discourages agreements or practices, which are frequently covert, which are liable to restrict or distort competition. From that point of view, actions for damages before the national courts can make a significant contribution to the maintenance of effective competition in the Community.”

The fourth factor to which Mr Ward KC referred was that, before the CAT, DAF even sought to contest that it was bound by the EC Decision at least in respect of recitals which did have an obvious bearing on pricing, recitals [27] and [47]. The CAT held that DAF was bound by the Decision including those recitals, in effect holding at [45] to [52] that DAF’s approach was an abuse of process. There is no appeal against those conclusions.

The fifth and final factor was the nature and duration of the cartel itself, so long lasting that it spanned major currency movements and the GFC. The currency movements caused considerable difficulties for both experts, not just Mr Harvey and that was in part why the CAT concluded that neither expert’s regression analysis could yield a definitive solution.

Mr Ward KC next made submissions about the approach of the CAT to the assessment of the Overcharge. The starting point is that the CAT found at [477] to [479] that the Claimants had proved, on the balance of probabilities, the existence of material harm. Accordingly DAF’s challenge in relation to the Overcharge is entirely about the exercise of quantification. He pointed out that quantification proceeded by way of regression models on both sides, which is a methodology which is inherently uncertain. As I said in this Court in UK Trucks Claim Limited v Stellantis NV [2023] EWCA Civ 875 at [96]:

“This approach ignores the fact that any regression analysis and determination will be highly sensitive to the assumptions made and data input. There is an inevitable element of subjectivity both in the selection of the data and these assumptions. Without in any way being critical of or doubting the integrity of Dr Davis, complete objectivity in expert economic evidence cannot really be achieved. This was a point made by the CAT in Royal Mail in relation to the expert evidence there on overcharge at [475] to [480]…there is no single, objectively ascertainable, "right" answer to the overcharge pass-on issue…”

He submitted that the law is clear that, having established harm on the balance of probabilities, there is a right to compensation and the courts will do their best on the available evidence. This was clear from the judgment of Lord Briggs JSC for the majority of the Supreme Court in Merricks at [47] to [52]. Mr Ward KC placed emphasis on several passages in that part of the judgment to demonstrate that it was a complete answer to DAF’s case on Overcharge. At [47] the Supreme Court stated that: “Once that hurdle [of showing a triable issue that more than nominal loss has been suffered] is passed, the claimant is entitled to have the court quantify their loss, almost ex debito justitiae. There are cases where the court has to do the best it can upon the basis of exiguous evidence.”

The Court continued at [48]:  “A resort to informed guesswork rather than (or in aid of) scientific calculation is of particular importance when (as here) the court has to proceed by reference to a hypothetical or counterfactual state of affairs” continuing at [49]: “This principle of entitlement to quantification notwithstanding forensic difficulty has stood the test of time and outlasted the involvement of civil juries in the assessment of damages.” At [50] the Court made the point in relation to the authorities: “In none of these cases does the court throw up its hands and bring the proceedings to an end before trial because the necessary evidence is exiguous, difficult to interpret or of questionable reliability.”

At [51] the Supreme Court cited a passage from the judgment of Popplewell J (as he then was) in ASDA Stores Ltd v Mastercard Inc [2017] EWHC 93 (Comm):

“The ‘broad axe’ metaphor appears to originate in Scotland in the 19th century. The more creative painting metaphor of a ‘broad brush’ is sometimes used. In either event the sense is clear. The court will not allow an unreasonable insistence on precision to defeat the justice of compensating a claimant for infringement of his rights.”

Mr Ward KC also relied upon the judgment of Green LJ in London & South Eastern Railway Ltd v Gutmann [2022] EWCA Civ 1077 in relation to the broad axe principle at [59]:

 “…It is not so much a substantive principle of law as a description of a well-established judicial practice whereby judges eschew artificial demands for precision and the production of comprehensive evidence on all issues and instead use their forensic skills to do the best they can with limited material to achieve practical justice.”

The final authority to which Mr Ward KC referred was One Step and the passage from Lord Reed JSC’s judgment quoted at [79] above. He placed particular emphasis on the passage cited from the judgment of Toulson LJ in Parabola: “the law does not require a claimant to perform the impossible, nor does it apply the balance of probability test to the measurement of the loss.”He submitted that the approach advocated in these appellate authorities was exactly the approach of the CAT in the present case. At [172]-[173] the CAT concluded that the Claimants “are required to establish that they suffered monetary harm as a result of the Infringement and they must do so on the balance of probabilities” but once the cause of action has been established in that way, the quantification of damages is on the basis of the broad axe and the Claimants have a right to damages however difficult the assessment process. The suggestion in [32] of DAF’s Skeleton Argument that the broad axe could not replace the requirement for the Claimants to prove the level of their loss is completely contrary to all these authorities.

Mr Ward KC submitted that, in its Conclusions on Overcharge at [475] to [486] (the relevant passages from which are quoted at [51] to [53] above) the CAT had made a positive reasoned finding on quantum. It had not simply split the difference between the two experts as Mr Beard KC had suggested. Also, contrary to Mr Beard KC’s submissions, the CAT had not rejected Mr Harvey’s evidence. In a number of respects to which Mr Ward KC drew attention, it had found his evidence helpful and informative. He referred to [365]-[366] (quoted at [37] above) where, in relation to the question of how to model the first part of the cartel, the CAT recorded that some caution must be exercised in relying on the results from a Before/During analysis (in which Mr Harvey had engaged) but said: “we are satisfied that with the sensitivities carried out by Mr Harvey, we can draw inferences from such evidence, despite its imperfections”.

In relation to the currency/exchange rate issue, Mr Ward KC went through the CAT’s analysis, culminating in the conclusions it drew at [407] to [410] quoted at [42] above. He submitted that this was a carefully reasoned and balanced assessment of the expert evidence leading the CAT to make its own positive evaluation of where the answer might lie given that it was not possible to reach an authoritative position on the issue. He emphasised that, contrary to Mr Beard KC’s submission, the CAT had not rejected Mr Harvey’s evidence on exchange rates but concluded at [484] that Professor Neven’s position had more merit, whilst they were “more evenly split” on GFC. The CAT had then concluded: “We therefore consider that a fair and reasonable broad axe view on Overcharge comes out at 5% for both Claimants (ie approximately half of what they are claiming).” As Mr Ward KC said, this was not simply splitting the difference but was, from what the CAT had said as a whole in this section of its judgment, a reasoned, informed judgment.

In relation to the GFC, Mr Ward KC noted that the question was whether it was so unusual that it had an effect on prices that would not be captured by the ordinary demand control of the level of DAF sales. Mr Harvey tested for that with dummies that turned out to be statistically significant. Professor Neven’s view was that there was nothing here beyond a normal demand effect. The CAT rejected that extreme view and concluded at [439]-440] (quoted at [48] above) that although it had concerns about Mr Harvey’s approach, it did not wholly reject his evidence as DAF had invited it to do. Again the CAT concluded that the actual answer may be found somewhere between the opposing positions and, as with the evidence on exchange rates, it helped the CAT to determine its own figure using the broad axe. Mr Ward KC submitted that what the evidence showed was that, contrary to Mr Beard KC’s submissions, there was no gap between the finding of harm and quantification of the loss. This was not a case where there was nominal harm and then the CAT plucked a figure out of the air. Rather the CAT had formed its own view of the true value of the Overcharge based on all the available evidence.

In relation to Ground 2, Mr Ward KC submitted that DAF’s case on SPO was, at best, strikingly ambitious. Given that DAF’s case was that there was no Overcharge at all, its argument was a hypothetical one, namely that, if there was an Overcharge, the Claimants would have passed it on. There was no allegation that the Overcharge was noticed by the management of Royal Mail or BT and prices were adjusted accordingly. Undeterred, DAF’s case was that between 75% and 139% of any Overcharge was passed on by Royal Mail and 100% by BT. However, on the basis of the CAT’s conclusion that the Overcharge was 5%, this was never more than about 0.025% of Royal Mail’s revenues and 0.0015% of Openreach’s revenues. And yet DAF’s case was that it could not only trace that tiny amount to the price of the Claimants’ individual products, but that it could show that this tiny cost increment actually caused a price increase.

The majority of the CAT had rejected that argument. Mr Ward KC pointed out that DAF argued that in doing so they had erred in law but an important starting point is that the CAT was unanimous in its approach to the law. He submitted that this approach disclosed no error of law at all. On the facts, there had been 2,000 pages of expert evidence on this topic, regulatory documents, pricing papers and so forth, none of which was in the bundles before this Court. Mr Ward KC submitted that the appeal was largely a dispute about the evaluative judgments of the CAT and DAF’s case fell a long way short of the required irrationality threshold. The fact that Mr Ridyard reached a different conclusion on that material does not establish that the conclusion of the majority was irrational.

He submitted that Mr Beard KC had recognised that problem by seeking to argue that there was an error of law which coloured the majority’s factual analysis, but that was not correct and was based on a highly selective reading of the majority judgment. Mr Ward KC also asked the Court to note how opportunistic this SPO argument was. Six years on from the EC Decision, DAF was not facing any claims from disgruntled customers of Royal Mail or BT to whom the Overcharge had allegedly been passed on and even Mr Ridyard thought such claims would be impossible to quantify.

He submitted in relation to causation that it appeared to be common ground that DAF needed to show that Royal Mail’s and BT’s prices were actually higher because of the Overcharge than would have been the case otherwise and even if that was not accepted, it was clear from [205] of Sainsbury’s SC where the Supreme Court set out the four options. Option (iv) was that the merchant could pass on the overcharge by increasing its price to its customers, in other words, there has to be causation. The burden of proof in relation to SPO is on DAF.

Mr Ward KC referred to the passages from earlier authorities, including the decision of this Court in Stellantis set out at [204] to [206] of the CAT judgment and quoted at [23] to [24] above. He submitted that DAF’s argument was that the Supreme Court in Sainsbury’s SC effectively stripped the requirement of legal causation of any content and reduced it to factual causation. However, that criticism was wrong. The CAT had carefully considered the law and followed this Court in Stellantis. He submitted that the CAT had applied the right test which was the British Westinghouse principle, as applied by this Court in Stellantis, whether there was a close and a legal and proximate causal connection between the steps alleged to have been taken by the claimant, here SPO, and the Overcharge. The CAT in the present case had correctly cited and applied that test as restated by the CAT in [36] of its May 2021 judgment (quoted at [20] above).

Mr Ward KC took the Court through the CAT’s analysis of the law and submitted that, in [223] set out at [29] above, the CAT synthesised its unanimous view of the law: “we consider that DAF must prove that there was a direct and proximate causative link between the Overcharge and any increase in prices by the Claimants. That means that there must be something more than reliance on the usual planning and budgetary process, into which the Overcharge was input and at some point prices increased.” He submitted that it was in this that Mr Beard KC had to find an error, which he could not do. Mr Ward KC submitted that the difference between the factual and the counter-factual here was a tiny amount of overcharge which, unlike the charge in the case of MIFs, was completely unknown and so reduced the likelihood of the business having done something different.

Mr Ward KC referred to the summary of the legal test for causation at [228] to [230] of the CAT’s judgment (quoted at [30] above) and the four potentially relevant factors which it identified. Again, he submitted that there was no conceivable error of law in any of that. In relation to the first factor, knowledge of the overcharge, he submitted that that was a reference back to what was said by Roth J in the CAT in the May 2021 judgment at [42]: “the Claimants’ knowledge of the nature and amount of the Overcharge, such that they would seek to address it.”

He took the Court to the passage in the judgment at [572] to [573] (quoted at [58] above) where the CAT went through the four non-exhaustive factors and concluded none of them was present, going on to say that where the factors were absent, the evidence of factual causation needed to be that much stronger. This passage had been subjected to criticism by Mr Beard KC, but Mr Ward KC submitted that this was an example of “sedulously picking and criticising individual words and phrases” in a judgment rather than reading the reasoning fairly in context, citing Stuart Smith LJ at [61] of R (Milburn) v Local Government Ombudsman [2023] EWCA Civ 207; [2023] PTSR 1250.

Although DAF had sought to rely upon the dissent of Mr Ridyard, Mr Ward KC pointed out that the CAT had been unanimous as to the legal test for SPO and had unanimously stated in relation to DAF’s application for permission to appeal on this Ground at [11] of its Ruling on Permission to Appeal of 16 May 2023:

“DAF then goes on to suggest that the majority asked the wrong question for the purpose of assessing causation. But this is a challenge to the majority’s evaluative judgment as to the expert and factual evidence, upon which an appeal court would exercise substantial “appellate restraint” in deciding whether the Judgment was wrong. Furthermore the majority was testing, as was the minority, whether the Claimants’ prices would have been lower in the counterfactual without the Overcharge.”

Mr Ward KC also referred to the conclusion of the majority on SPO at [688] of the judgment, quoted at [62] above, which he submitted was an unimpeachable formulation of the role of the counter-factual.

Mr Ward KC took the Court, as had Mr Beard KC, to two of the price controls imposed by the regulator on Royal Mail, PC2 and PC3. Although some time was taken on the detail of this, as Green LJ said in the course of argument, the essence of the point is the conclusion reached by the CAT at [606] in relation to PC2, quoted at [59] above. Mr Ward KC submitted that, contrary to Mr Beard KC’s submission, there was no point of law here and nothing in the reasoning discloses any confusion about the correct counter-factual. As already noted, a similar conclusion was reached in relation to PC3 at [616] and [617]. As Green LJ said, these conclusions are findings of fact, evaluations by the CAT of the evidence.

In relation to BT, Mr Ward KC noted what was said at [658]-[659] of the judgment, quoted at [60] above and the conclusion of the majority rejecting the probability analysis at [667]-[668]. He submitted that this was an evaluative judgment open to the CAT which discloses no error of law whatsoever.

Mr Ward KC then discussed briefly Mr Ridyard’s dissenting opinion. As he said, he did not need to show that it was erroneous, only that the majority’s judgment was lawful. Nonetheless, he submitted that the problem with Mr Ridyard’s analysis is that it proceeds at a high level of generality, at which it looks unobjectionable, but it does not grapple, as did the majority, with whether this tiny unknown overcharge caused prices to be higher. He also noted that the CAT was unanimous that there were no identifiable downstream purchasers, from which it followed that Mr Ridyard’s dissent posits an impossible claim that no-one would be able to bring. He submitted that at this point, the principle of effectiveness does come into play because DAF is trying to rely on this impossible to quantify, non-existent downstream claim as a way to defeat the claim by the actual direct purchaser which is before the Court.

Mr Ward KC then turned to Ground 3, which he submitted proceeded by mischaracterising the CAT’s reasoning and engaging in a selective reading of the decision and of DAF’s own evidence. He asked the Court to note that Royal Mail had bought trucks, including bodies, from DAF and that in fact, for four years of the cartel from 2007, DAF had itself manufactured bodies at its UK plant. However, whether the bodies were manufactured by DAF or a third party, they were supplied to Royal Mail by DAF, as he put it, on the DAF invoice, in the same way as would be the tyres, supplied by DAF, albeit manufactured by, say, Michelin.

He referred to the EC Decision noting in relation to Recital (5) set out at [88] above that the submission on this appeal that the bodies were “other goods” within that provision was not one which he had heard advanced before. Furthermore, if DAF had been concerned that the settlement included bodies, they could have pulled out of the settlement negotiations or made submissions that it should not include bodies. However, it had not done that, but as recorded in Recital (43) the settlement submission of each addressee including DAF included: “an acknowledgement in clear and unequivocal terms of the Addressee's liability for the infringement summarily described as regards its object, the main facts, their legal qualification, including its role and the duration of its participation in the infringement in accordance with the results of the settlement discussions”.

He also submitted that it was clear from the EC Decision that the collusion extended to truck options. He referred to the reference to “factory fitted options” in Recital (28) and to “available options” to which price increases related in Recital (56). He showed the Court documents exchanged with other cartelists which included information about different body options in price lists. He referred to Mr Ashworth’s witness statement which refers to the fact that DAF had to deliver fully built trucks including the body and tail lift together with other items and that these were fitted onto the truck in DAF’s factory. He also said the price negotiation with Royal Mail would come down to what the customer was willing to pay for the whole truck.

Mr Ward KC drew our attention to [120] of the judgment where the CAT said:

“…the exchange of gross list price information and in particular proposed list price increases formed a major part of the Infringement. Mr Ashworth frankly admitted that, in the case of direct customers, such as the Claimants, a “very good rule of thumb” was that DAF UK normally expected to achieve “about half” of a list price increase in the form of increased transaction prices.”

As he pointed out, the CAT then recorded that Mr Beard KC had sought to contend that Mr Ashworth had made an honest mistake about this, but [121] said that his evidence was significant and that DAF should not be undermining its own witness, who knew nothing about the cartel. He submitted that what this evidence showed was a negotiation over the price of the whole truck under pressure from the cartelist DAF N.V. to increase prices using gross list prices, which was one of the core elements of the cartel.

Mr Ward KC submitted that the CAT’s reasoning in dealing with this issue at [473] (quoted at [50] above) was unimpeachable. Particularly in the absence of evidence from DAF as to how the cartel operated and how that affected the pricing for the whole truck, the natural approach was to consider bodies in the same way as other extras. If the infringement took effect through a mechanism linked to list price changes, there was no convincing evidence on which the CAT could reject the possibility of the effect also applying to bodies.

Mr Ben Lask KC dealt with Ground 4 on behalf of Royal Mail. He submitted that the CAT had evaluated the claim for compound interest carefully, considering the authorities and weighing up the expert and factual evidence. This ground of appeal relates only to the CAT’s decision as to the appropriate weighting between the cost of debt and the return on short term investments. Mr Lask KC submitted that this required the CAT to draw inferences from the evidence as to what Royal Mail would have done in the counter-factual with the money used to pay the unlawful Overcharge, which was quintessentially a matter for the CAT as an expert tribunal.

He submitted that, recognising the formidable obstacle imposed by the principle of appellate restraint, DAF is taking the extreme position of contending that there was no evidence to support the CAT’s approach to weighting. This position is both inherently unlikely and demonstrably incorrect. As Mr Lask KC said, the CAT in finding that there was ample evidence to support the claim for compound interest said at [767]: “As was said in CAT Sainsbury’s, the court or tribunal draws ‘broad axe’ inferences as to what the claimant would have done in the counterfactual with the money it had to use to pay the Overcharge.” He submitted that the CAT’s approach was entirely consistent with the authorities. He drew attention to the passage at the end of [95] in the judgment of Lord Nicholls in Sempra quoted at [765] of the judgment:

“Whatever form the loss takes the court will, here as elsewhere, draw from the proved or admitted facts such inferences as are appropriate. That is a matter for the trial judge. There are no special rules for the proof of facts in this area of the law.”

The Privy Council in Sagicor made the same point at [33]:

“The Board agrees with Males J that, in assessing a claim for financial loss caused by the failure to pay money that is contractually due, the law does not require a detailed examination of a plaintiff’s financial affairs and that an extensive process of disclosure by the plaintiff to make or verify that assessment is likely to be unhelpful and is in any event disproportionate. The question of what evidence is required from which a court can infer that a plaintiff has suffered financial loss in the form of the incurring of borrowing costs will depend upon the circumstances of the particular case, as Lord Nicholls recognised in para 95 of his speech in Sempra Metals”.

Mr Lask KC took the Court through the section of the judgment where the CAT dealt with the appropriate weighting, which I have summarised at [70] to [75] above. He submitted that, contrary to DAF’s contention, there was factual evidence to support the approach of Royal Mail’s expert Mr Earwaker. He showed the Court the passage in Mr Earwaker’s second report which dealt with this and his Table 6 setting out Royal Mail’s short term investments and debt over the period from 1997 to 2020. He submitted that this showed that Mr Earwaker’s approach was firmly rooted in factual evidence as to the scale and evolution of Royal Mail’s debt and short term investment over the relevant period. The data set out in Table 6 showed a contrast between two periods: the earlier period up to 2007 when Royal Mail held significant cash and had limited debt and the later period after 2007 when that general trend was reversed. This was why Mr Earwaker took the approach he did. His approach was based on extensive factual material including accounts and annual reports and individual debt instruments.

In the Joint Expert Statement Mr Earwaker said that: “Prior to 2008, Royal Mail held significant amounts of surplus cash and borrowed only to finance international acquisitions.” As Mr Lask KC pointed out, this was not challenged by DAF at trial. What it demonstrated, as I pointed out in argument, is that if Royal Mail had had a whole load of additional cash, it would not have used it to fund overseas investments because they were already covered. There was evidence summarised in Table 4 in Mr Earwaker’s second report which showed Royal Mail’s borrowings over time and which supported what he had said in the Joint Expert Statement.

As for Mr Beard KC’s submission referred to at [92] above that there was no evidence about what Royal Mail would have done differently had it not been for the overcharge, this was not correct. Mr Lask KC showed the Court what Mr Earwaker had said in cross-examination. He submitted that the suggestion that there was no factual evidence to support Mr Earwaker’s approach was clearly wrong. He pointed out that the data which Mr Earwaker summarised in his Table 6 reflected a clear shift from short term investments to debt after 2008.

DAF complained that Mr Jeavons had not given evidence about a change of approach after 2008. Mr Lask KC submitted that, irrespective of whether Mr Jeavons had addressed the point, the change in circumstances or approach was reflected in the data summarised by Mr Earwaker. The answer to the point also made by DAF that Mr Jeavons had not given evidence about what Royal Mail would have done with the additional funds in the counter-factual was that such evidence would have been hypothetical and of limited use, a point made by the CAT at the beginning of [767]. In any event, Mr Lask KC submitted, that was not what the law requires. He noted that in [37] of Sagicor in the passage after that quoted at [91] above, the Privy Council had given some examples of the sort of evidence which may suffice. He submitted that was precisely what Royal Mail adduced.

Finally, Mr Lask KC made the point that on weighting, the CAT was invited to choose between two expert approaches. He drew the Court’s attention to the passage in the cross-examination of Mr Delamer, DAF’s expert, where he accepted that he had not considered for himself how Royal Mail may have deployed the additional funds it would have had available but for the overcharge. He submitted that this was the problem with Mr Delamer’s approach which justified his description of it, referred to by the CAT at [823] as blunt and over-simplistic. The CAT had been entitled to prefer Mr Earwaker’s approach and DAF had not identified any error by the CAT in doing so.

At the heart of DAF’s case on Ground 1 was the proposition that, since it was for the Claimants to prove that there was a substantial loss, the CAT had erred in applying the broad axe to quantification of the loss and had missed out any assessment of whether the Claimants had satisfied the burden and standard of proof. The logical consequence of that case was the counter-intuitive proposition that the Claimants had failed to establish that there had been an Overcharge at all. Although, as I said at [82] above, on being pressed by the Court, Mr Beard KC did somewhat resile from that extreme position and accept that, on the evidence, the CAT was entitled to find that there was a loss, he still submitted that the CAT had not been entitled to conclude that the loss was sizeable or substantial, a position which still involves, as Mr Ward KC pointed out, the Claimants going away empty-handed.

In my judgment, DAF’s case that the CAT failed to address the burden and standard of proof is unsustainable. In its consideration of whether there was an Overcharge, the CAT was clearly addressing the question whether the Claimants had established such an Overcharge on a balance of probabilities. That is precisely the analysis in which the CAT engaged in the last sentence of [172] quoted at [9] above and again in the first two sentences of [477]:

“The first question we need to address was whether, based on a balance of probabilities test, the evidence points to the existence of a cartel Overcharge. We conclude that it does.”

On its own admission, DAF participated in a secret cartel with other truck manufacturers to maintain prices at a supra-competitive level for some fourteen years. It is inconceivable that it would have participated in the cartel for such an extended period of time with all the financial, regulatory and reputational risk that that entailed, unless it was gaining significant financial benefit from that participation. This was a point which the CAT correctly had well in mind, referring to it several times in the judgment, for example in the passage at [38] set out at [97] above and the passage at [477] set out at [51] above.

In the circumstances, the CAT was entitled to proceed on the basis that there was an Overcharge and that it was substantial, not minimal, as DAF’s case would suggest. Mr Beard KC did not really have an answer to this. He suggested in reply, when the Court put this point to him, that there might have been other reasons for information being shared between cartelists than financial benefit, but that is frankly unsustainable speculation unsupported by any of the evidence which DAF could and should have called if it wanted to make good that contention. In the absence of such evidence, there was a strong inference, which the CAT was entitled to draw, that DAF participated in this cartel over such an extended period of time because by doing so it gained a significant financial benefit and, in turn, the victims of the cartel such as the Claimants suffered a significant financial loss. Against that background, the expert thesis advanced by Professor Neven that there was a zero Overcharge lacked credibility and the CAT’s criticisms of him were entirely justified.

Furthermore, the CAT’s conclusion that there was an overcharge was justified by the fact that this was a follow-on claim following the EC Decision that the cartel was unlawful by object. As this Court explained in O’Higgins FX Class Representative Ltd v Barclays Bank Plc and Evans v Barclays Bank Plc [2023] EWCA Civ 876 at [25] to [32], a finding by the Commission that an agreement is illegal by object involves a finding that the cartel is likely to have exerted adverse effects on competition (even though in an object case there is no duty on the Commission to go on and make findings about actual effects). As the CAT itself said at [116]: “the basis of a finding of an infringement by object is that it is very likely to have had negative effects on transaction prices.”

There was also a complete failure on the part of DAF to adduce any evidence as to how the cartel operated and what benefits it derived from its participation. In those circumstances, the CAT was entitled to draw adverse inferences from that failure to adduce evidence, applying the Prest presumption referred to at [99] above. The CAT was entitled to conclude at [116]-[117] that DAF’s failure to adduce such evidence meant that it was not open to DAF to contend that it had not used information obtained through the cartel to achieve higher prices than it would otherwise have done. As Mr Ward KC pointed out, none of that was challenged on appeal.

The CAT was entitled to conclude, on the basis of the evidence before it and on the basis of adverse inferences which it was entitled to draw, that the Claimants had established, on the balance of probabilities, that there was an Overcharge. The next question for the CAT was the quantification of the loss caused by the Overcharge and it was to that question which the CAT turned at [479]. Contrary to what Mr Beard KC submitted, there is no missing section in the CAT’s analysis.

As Mr Ward KC correctly submitted, once a Court has established loss on a balance of probabilities, the claimant is entitled to be compensated and the Court will do its best to quantify the compensation on the available evidence. This principle is well-established and by way of authority, one need look no further than the passages in the majority judgment of the Supreme Court in Merricks cited at [104] to [106] above. As is clear from that judgment, the judgment of Green LJ in Gutmann cited at [107] above and the judgment of Lord Reed JSC in One Step cited at [79] above, it is in the context of difficulties in the quantification of loss that the principle of the broad axe is deployed by the Courts, as the CAT itself recognised at [173]-[174] of its judgment and again at [479]. In my judgment, the CAT’s analysis of the applicable principles is unimpeachable and does not begin to disclose an error of law.

At [479] the CAT recognised the imprecision of the experts’ regression models and declined to score the experts issue by issue in order to arrive at a figure for the overcharge. It considered that that would be an exercise in “spurious accuracy”. Instead, the CAT sought in broad terms to indicate which party had the better of the argument in relation to the two key variables it identified of exchange rates and the GFC. In my judgment, in the light of the evidential problems it faced, this approach by the CAT was entirely warranted. The CAT went on to carefully evaluate the relevant evidence on those variables (as fairly summarised in Mr Ward KC’s submissions as recorded at [110] and [111] above) and set out the probative value which it attached to that evidence. Although it was highly critical of Professor Neven as set out at [31] to [33] above, it did not reject his evidence outright but adopted a balanced approach, giving his evidence credence and weight when it thought it proper to do so. This Court will not interfere with that evaluation of the evidence by the CAT.

Taking into account its assessment of the expert and other evidence, the CAT reached the conclusion at [484], wielding the broad axe, that a figure of 5% for both Claimants (i.e. approximately half what they were claiming) was the appropriate figure for the overcharge. In my judgment, Mr Beard KC’s criticism of this approach was unjustified. I agree with Mr Ward KC that the CAT did not simply split the difference as Mr Beard KC suggested, but in the section of its judgment setting out its Conclusions on Overcharge at [475] to [486] it made positive and reasoned findings as to the appropriate quantification of the overcharge.

This is precisely the sort of situation where wielding the broad axe is appropriate. It cannot be right for DAF to fail to disclose potentially inculpatory material which might have shed light upon the reality of its participation in the cartel and its impact on prices and for its expert then to advance a case that there was no overcharge when, as a rational business, DAF must have concluded that lengthy continuation of risky illegality was financially worthwhile. Its approach involves conducting litigation at the level of theory, not reality. This Court is entitled to ask rhetorically how an appellant can complain when the CAT plugs gaps in the evidence caused by that appellant by wielding the broad axe.

In my judgment, Mr Ward KC’s description of DAF’s case on SPO in Ground 2 as “strikingly ambitious” is entirely apt. Its case is, in effect, that there was no overcharge, but if there was one, which on its case is entirely hypothetical, it would have been passed on. As Mr Ward KC pointed out, given the level of overcharge determined by the CAT of 5%, this was no more than about 0.025% of Royal Mail’s revenues and 0.0015% of Openreach’s revenues, on any view a tiny amount. The idea that this tiny amount could not only be traced to the price of the Claimants’ individual products, but that it is then possible to establish that it caused a price increase, seems to me completely unreal. Even Mr Ridyard’s dissenting opinion concluded that the specific downstream impact of SPO was too small to be measured.

In its argument on this Ground, DAF raises the somewhat elusive distinction between legal causation and factual causation in relation to mitigating conduct. In my judgment the distinction is clearly and usefully explained by the CAT in the MIF Umbrella Proceedings judgment quoted at [83] above. Factual causation involves consideration of whether the effect of the mitigating conduct was in fact to reduce or eliminate the claimant’s loss, whereas legal causation concerns whether, even if the effect of the mitigating conduct was in fact to reduce or eliminate the claimant’s loss, as a matter of legal policy, it should serve to reduce or eliminate the damages payable by the defendant to the claimant. The classic example referred to by the CAT is an indemnity obtained by the claimant under a contract of insurance which is regarded as res inter alia acta and thus does not reduce the defendant’s liability. As Green LJ pointed out in the course of argument, other examples of legal causation are when the relevant matter is too remote or amounts to a novus actus. In my judgment, the CAT in the present case correctly identified at [212] of its judgment, on the basis of the judgment in the MIF Umbrella Proceedings, the distinction between factual and legal causation.

In terms of factual causation, DAF could only succeed in its argument on SPO if it could establish that the prices charged by Royal Mail and BT to their customers were higher because of the overcharge, in other words if it could establish (and the burden of proof is on DAF) that the overcharge had been passed on to those customers. The CAT was unanimous as to this requirement at [223] of its judgment where it said: “we consider that DAF must prove that there was a direct and proximate causative link between the Overcharge and any increase in prices by the Claimants. That means that there must be something more than reliance on the usual planning and budgetary process, into which the Overcharge was input and at some point prices increased.” I agree with Mr Ward KC that the CAT was applying the correct legal test, as recently restated by this Court in Stellantis (as cited at [23] above).

In seeking to identify how DAF might prove that direct and causative link between the Overcharge and an increase in prices charged, the CAT set out a number of non-exhaustive potentially relevant factors at [228] (cited at [30] above). The first two factors, knowledge of the Overcharge and the size of the Overcharge were the ones identified as particular factors relevant to category (iii) and (iv) cases by the CAT in its May 2021 judgment (see [29] above). In the cases involving MIFs, the overcharge in the form of the MIF is transparent and known to the merchant, so that it can seek to address that overcharge, for example by passing it on to customers. In contrast, the Overcharge as a consequence of the secret cartel was, by definition, not known to the Claimants, so they were not in a position to address it. This was not disputed before the CAT (see [551] of the judgment).

The fourth factor, the existence of identifiable claims by identifiable purchasers from the Claimants is the one which the CAT in Sainsbury’s considered a relevant factor in relation to the strength of the causal connection, as noted by the CAT in this case at [223]. However, as Mr Ward KC pointed out, it is now over seven years since the EC Decision and yet there are no identifiable claims by customers of Royal Mail or BT in the present case that the Overcharge has been passed on, as was accepted by Mr Ridyard in his dissenting opinion, which also accepted that such claims would be impossible to quantify.

The CAT concluded that none of the four factors was present in this case, a conclusion which was not challenged on appeal. However, as set out at [85] above, Mr Beard KC did submit that the majority had been wrong to conclude at [573] that: “we think that in a situation where none [of the four factors] are present, the evidence of factual causation needs to be that much stronger so that the requisite proximity can be established.” In my judgment, that submission is misconceived. In circumstances where none of the four factors which might establish the requisite degree of proximity to establish a direct causative link between the Overcharge and the prices charged by the Claimants is present, it is both logical and common sense to conclude that there would need to be some other evidence of factual causation to establish that requisite degree of proximity. Far from that majority conclusion being confused and incorrect as Mr Beard KC submitted, its analysis is clear and correct, disclosing no identifiable error of law.

Although DAF sought to rely upon Mr Ridyard’s dissenting opinion, that does not establish that there was any error of law by the majority. In that context, it is significant that the CAT was unanimous in concluding at [11] of its Ruling on Permission to Appeal (quoted at [120] above) that the majority had not asked the wrong question in assessing causation and had engaged in an evaluative judgment of the factual and expert evidence which DAF had no real prospect of successfully overturning on appeal (see [13] of the Ruling). The CAT only gave permission to appeal on this Ground on the basis of some other “compelling reason” because it thought this Court should consider the correct approach to SPO (see [14] of the Ruling).

In my judgment, DAF’s argument on this Ground is really an attack on what was an evaluative judgment by the majority of the CAT that, on the factual and expert evidence before it, DAF had not established that the prices charged to the Claimants’ customers would have been lower in the counter-factual absent the Overcharge. This Court would not interfere with that evaluative judgment of an expert tribunal in the absence of an error of law unless the conclusion was irrational, which it clearly is not, since it is founded on a careful evaluation of the evidence. Furthermore, DAF cannot derive any assistance from Mr Ridyard’s dissent, since it seems to me that Mr Ward KC is correct in his criticism of it set out at [123] above.

In relation to Ground 3, I consider that the CAT was correct in its conclusion that truck bodies were within the scope of the infringement as found by the EC Decision. As set out at [126] above, the evidence of Mr Ashworth was that DAF had to deliver a fully built truck to Royal Mail including the body and tail lift which although they may have been supplied by third party manufacturers were fitted onto the truck in DAF’s factory. In those circumstances they were clearly within the scope of “factory fitted options” in Recital (28) and “available options” in Recital (56). Mr Beard KC sought to challenge that conclusion in relation to Recital (28) in reply by saying that the arrangements with Royal Mail and Morrisons to fit the bodies were exceptional. That may well be so, but it does not make the bodies fitted on their trucks any less “factory fitted options”. It is no answer to say that the bodies were manufactured by third parties. Nothing in the EC Decision suggests factory fitted options are limited to those manufactured by DAF. As Mr Ward KC also pointed out there was evidence before the CAT of information exchanged between the cartelists which included information about different body options in price lists, demonstrating that the infringement encompassed the whole truck, including the body.

I also agree with Mr Ward KC that the CAT’s reasoning on this issue at [473] quoted at [50] above is unimpeachable. In the absence of any evidence from DAF as to how the cartel operated and how that affected the pricing of the whole truck or any evidence as to the information provided by DAF to the European Commission or its settlement discussions with the Commission, the CAT was entitled to conclude that the natural approach was to consider bodies in the same way as other extras purchased by Royal Mail from DAF. If DAF had chosen to produce evidence as to how the cartel operated which demonstrated that bodies were not within the scope of the infringement, the position might have been different. However it did not and seeking to assert that bodies were “other goods” within Recital (5) is not supported by any evidence and is an artificial construct by DAF.

In relation to Ground 4, I agree with Mr Lask KC that the CAT evaluated Royal Mail’s claim for compound interest carefully, considering the relevant law and weighing up the expert and factual evidence. There was no error of law in its analysis of the authorities, specifically Sempra. This Ground only concerns the CAT’s decision as to the appropriate weighting between the cost of debt and the return on short term investments. This involved the CAT drawing inferences from the evidence as to what Royal Mail would have done in the counter-factual, which as Mr Lask KC said, was quintessentially a matter for the CAT as an expert tribunal and this Court should exercise considerable restraint before interfering with its evaluation.

DAF sought to overcome this obstacle by taking a somewhat extreme position that there was no evidence to support the CAT’s approach to this issue. As Mr Lask KC rightly said, that is demonstrably incorrect. The expert evidence of Mr Earwaker was founded on the data set out in the tables in his expert report which showed a shift from short-term investments to debt after 2008. There was nothing in the complaint by DAF that Mr Jeavons had not given evidence about what Royal Mail would have done with the funds in the counter-factual since, as the CAT recognised at the beginning of [767] any such evidence would have been hypothetical.

As the CAT continued in that paragraph:

“There is ample evidence before the court and considered by the experts as to what Royal Mail actually did at the time in terms of investment and debt finance. For example, Mr Jeavons referred to Royal Mail having borrowed from the Government at a rate of 12% compound interest. As was said in CAT Sainsbury’s, the court or tribunal draws “broad axe” inferences as to what the claimant would have done in the counterfactual with the money it had to use to pay the Overcharge.”

There is no error of law in the adoption of that approach.

The CAT went on to consider and evaluate the rival positions of the experts on weighting and preferred the evidence of Mr Earwaker to that of Mr Delamer. At [824] it concluded:

“We prefer Mr Earwaker’s approach which is based on how a rational business such as Royal Mail would have used extra funds that it had at the relevant time. His two-period characterisation of Royal Mail’s financial position, as a net investor in the first period and a net borrower in the second, is credible on the evidence and it would therefore be more likely that Royal Mail would use the funds in one direction rather than two. That therefore is a reasonable way to assess Royal Mail’s actual cost of financing the Overcharge.”

In my judgment, that is an entirely rational approach which draws appropriate inferences as to what would have happened in the counter-factual from the expert evidence which the CAT was entitled to prefer which in turn was founded on the data set out in his report. There was no error of law on the part of the CAT in adopting this approach.

For all the reasons set out above, I consider that the appeal should be dismissed on all grounds. I should add that I have also had the opportunity to read the concurring judgment of Green LJ, with which I entirely agree.

I agree with both judgments.

I agree with the judgment of the Chancellor for the reasons he has given. In particular I associate with his observations about the approach taken to the evidence by the parties and the impact this had upon the use of the broad axe. I wish to add the following because this type of litigation presents issues the CAT will no doubt grapple with again.

The CAT criticised the approach of the parties. Five main criticisms were made focusing upon the use of expert evidence. All applied to DAF and two applied to the claimants. They were:

The decision of DAF not to give disclosure relating to the operation of the cartel (Judgment paragraph [473]) deprived the CAT of evidence which it might, otherwise, have used to plug gaps created by the limitations of the regression analyses submitted by the parties and hindered cross-examination by reference to potentially inculpatory material. This was compounded by DAF tendering witnesses of fact limited to issues relating to quantum who had no direct knowledge of the cartel. This also deprived the CAT of evidence about the normal operation of the market which it could have used to plug deficiencies in the regression analysis (Judgment paragraphs [102]-[108]), and it also curtailed cross-examination. Counsel for the claimant criticised these omissions. DAF’s response was that it had produced huge amounts of documentation both from the Commission File and in relation to DAF’s price-setting structures. It had called four senior witnesses of fact involved in decision making including at board level and in relation to list prices, list price changes and the introduction of new emissions standards. It was the claimants’ own fault if they failed to cross-examine those witnesses properly. Counsel for DAF also speculated as to the motives of the culpable individuals and postulated that evidence from participants in the cartel might not have added to the sum of knowledge about pricing, but in any event whatever those motives were it did not alter the facts nor the data and it was those that determined whether there was an overcharge and “not the subjective intentions and beliefs of those involved in the infringement” (Judgment paragraphs [106]-[107] and [115]). The CAT did not accept this analysis:

“108.

… DAF’s expert evidence on the theory of harm is based on speculation as to how the Infringement would have worked within DAF and then draws conclusions on such speculation as to how the Infringement would not have had an effect on prices. We think that any such theory would be more soundly based on what actually happened factually within DAF in terms of how the information was used and how the Infringement managed to continue over such a long period, presumably for the mutual benefit of all the Cartelists.”

Next, the CAT was concerned at a lack of independence and objectivity displayed by the experts. They persistently came to conclusions favouring their clients. The regression analyses involved “big and difficult” issues upon which experts could reasonably disagree. But as the trial progressed the experts did not modify their positions and, at least on some issues, move towards each other:

“235 ... We consider that there should have been more recognition, on certain issues, of the scope for a range of possible results and of the reasonableness of the other expert’s opinion. As they are aware, the experts’ primary duty is to assist us in understanding the factors behind their differing conclusions, rather than defending the conclusions which favoured their respective clients’ positions. When there are fine and difficult issues for us to decide, it is important that we are able to trust the independence of the experts”.

“476 … However, as we have commented above, there were a number of instances where both experts might have been more transparent and realistic in identifying and accepting the existence of some of the limitations of their regression model results, and to have done so in their exchanges prior to the hearing. The tendency of both experts to defend their positions without acknowledging the inherent difficulties in their own approach was disappointing and inconsistent with their primary duty to assist the Tribunal”.

There was a lack of disclosure by DAF’s expert of the extent of his engagement with DAF from 2013 onwards, and as to the information that he had received from DAF concerning the operation of the cartel and its effects, none of which was referred to in his reports. This came to light during cross-examination. The history and background were set out fully in the judgment (paragraphs [237]-[257]). This supported the conclusion that the expert showed a “lack of candour”. He had been involved in advising DAF for nearly a decade and from an early stage was expressing opinions and advice upon potential theories of harm that would minimise damages. The Tribunal observed that it was “… fairly obvious that he was their favoured economic expert and this seems likely to have been influenced by his opinions on the plausibility of there having been any effect of the infringement.” (Judgment paragraph [237]). The expert’s theories of harm were based upon what DAF had told him and as to their “narrative” (advanced before the European Commission) that there was no anticompetitive effect flowing from the cartel (Judgment paragraph [249]). In cross-examination he explained that he had asked DAF why it had participated in the cartel and was informed that it enabled the company to know its competitors’ list prices so that they could test their relative competitiveness (Judgment paragraph [250]). The expert rejected this as “ex post rationalisation” and irrelevant to his economic analysis (Judgment paragraph [251]). He did not pursue the matter further. The CAT concluded that an understanding of DAF’s behaviour in the cartel was “surely potentially relevant to any theory of harm and his lack of curiosity in this respect is troubling”:

“256.

… This situation provided Professor Neven with insights and access that, as an independent expert, we could reasonably have expected him to use in order to assist us. We examine in detail the theory of harm that he puts forward in his evidence in this case and it is safe to say that his conclusion that it is implausible that there were any effects in the UK and on the Claimants from the Infringement is a surprising one. His theory provides a justification for the conclusion that he draws from the data that there was no Overcharge throughout the period of the Infringement. But we are left with the lingering suspicion that, as was disclosed very late on in these proceedings, he had come up with his theory of harm back in 2013 or 2014 (and certainly well before he had access to detailed empirical data), and that has shaped his approach to the expert evidence he has provided on the central issues in relation to the Overcharge.”

The CAT’s concerns came to the fore in relation to the refusal of DAF’s expert to reflect inferences to be drawn from the long-term participation of DAF in the cartel, coupled with a refusal to take into account the conclusions of the European Commission that DAF had engaged in an infringement by object which assumed an inherent likelihood of an effect upon prices. The European Commission had found, formally, that the cartel was unlawful by object (see paragraph [142] above). The CAT addressed arguments of DAF, which the CAT considered to be speculative, as to why and how the Commission decision might have been immaterial to the analysis:

“116.

We take no account of this speculation and it is an inappropriate way of approaching this issue by DAF. The burden remains on the Claimants to prove causation but where DAF has elected to call no evidence as to how the Cartel was operated by DAF and how it used the information to its advantage it is not open to its Counsel to speculate as to what actually happened. This was highly commercially sensitive information that was disclosed among the Cartelists over a long period of time. The Commission found that this information enabled the Cartelists to be better able to calculate their competitors’ approximate net prices. Further, the basis of a finding of an infringement by object is that it is very likely to have had negative effects on transaction prices. Therefore, in our view, this means that, if DAF wished to argue that, because of the way it used the confidential information obtained through the Cartel, there was no effect on prices, it would have had to adduce factual evidence to such effect. In other words, DAF’s admissions and the Settlement Decision establish a prima facie case that the Cartel had an adverse effect on transaction prices.

117.

That is not to say that DAF is unable to rely on its expert evidence to argue that the data shows that there was no Overcharge paid by these Claimants. But even their expert was unable to explain or come up with a rational economic basis for DAF’s participation in the Cartel over such a long period. While Prest does not entitle the Claimants to say that they have therefore proved that DAF’s participation in the Cartel led to higher prices it does mean that it is not open to DAF to argue that, as a matter of fact, the information was not used by it to achieve prices that were higher than they would otherwise have been without that information exchange.”

Next, the CAT expressed concern as to the extent to which expert evidence dominated the issues at trial. The volume of such evidence was “huge” and “excessive”. The CAT accepted that this was the first of a series of trucks claims in which many of the relevant issues might be determined and that the parties therefore had to cover all their bases. Nonetheless, a “more measured approach” would have been “beneficial” (Judgment paragraph [231]). There were two consequences. First, DAF conducted the trial at the level of economic theory, not reality. The trial might have proceeded differently had DAF tendered evidence of the modus operandi of the cartel which would have made it more difficult for DAF’s expert to persist in the argument that, notwithstanding the cartel, there was zero overcharge. Secondly, reliance upon regression analysis risked generating results which were broad brush and left gaps the CAT, perforce, would have to plug. Regression analysis does not lack value but its limitations are well established: see the commentary in O’Higgins FX Class Representative Ltd v Barclays Bank plc and Evans v Barclays Bank plc [ 2023] EWCA Civ 876 at paragraph [114] and cases cited thereat. Those limitations were exacerbated by four factors undermining the reliability of assumptions underpinning the modelling: the duration of the cartel; the number of significant regulatory changes to vehicle specifications made over time; the global financial crisis of 2008; and, fluctuations in exchange rates between Sterling and the Euro. The Tribunal’s conclusion is salutary:

“475.

Despite the enormous amount of work that went into the expert process on this case, and the vast quantities of data analysed, there are numerous serious gaps and unresolved issues in the analyses which taken together makes it difficult to distil the experts’ work on Overcharge into a simple definitive figure. Nor is it feasible to specify an “ideal” regression equation, based on the various work of the experts, that could be relied upon to yield the correct answer to the Overcharge question which would navigate successfully between the rival claims and conflicting conclusions reached by the experts. There are too many imperfections in the evidence, and insoluble practical problems, to allow any such approach.”

This approach to evidence led to a position at the trial end where there was a significant spread between the overcharge calculated by the competing experts from (0%) – (6.7% -14.7%), depending on time frame and truck types, conclusions which were “clearly influenced in favour of the commercial interests of [the] respective clients”: Judgment paragraph [480]. The CAT refused to resolve the dispute by scoring the experts, a “spurious” exercise (Judgment paragraph [479]). As the Chancellor points out, resolving this sort of difference is the very sort of exercise the broad axe was designed for.

The CAT will no doubt learn from the difficulties arising. It might consider how to prevent the scenario arising from becoming an embedded feature of this sort of litigation. It might consider requiring the defendant to identify early on whether disclosure of the operation of the cartel is to be given and, if not, why not. It might consider whether to compel production and/or make clear that a failure might lead to adverse inferences being drawn. The Supreme Court in Sainsbury's Supermarkets Ltd v Mastercard Inc [2020] UKSC 24, at paragraph [216], in respect of evidence needed to determine a pass-on defence, held that if a claimant refused disclosure the Court might draw adverse inferences, a principle equally applicable to a defendant failing to provide disclosure relevant to a claim of overcharge. In NTN and others v Stellantis [2022] EWCA Civ 16 ("Stellantis") at paragraph [53] this Court explained that, in the light of Sainsbury's, the whereabouts of relevant disclosure could affect how the burden of proof applies in practice. In MOL and others v McLaren [2022] EWCA Civ 1701 (“McLaren”) at paragraph [43] the point was repeated.

The CAT might also inquire at an early stage whether a defendant intends to call, as witnesses of fact, individuals who were participants in the cartel or, otherwise, had knowledge of its operations, so that they can be cross-examined, and if not, why not.

Experts can also be required to explain whether they have requested their client to furnish internal documentation concerning the modus operandi of the cartel, which might include a contemporaneous assessment of its effectiveness. If they have, the CAT might wish to know: what evidence was provided and its completeness; the extent to which it has been taken into account; and, the extent to which it has been provided by way of disclosure in the litigation, etc. If the expert has not sought this material the CAT might wish to know why, and might also wish to ensure that the instructions given to the expert are transparent to the CAT and other parties. It was relevant in the present case that the defendant’s expert was influenced by the “narrative” of his client, tendered (by way of exculpation) during proceedings before the EC Commission, that there was no anti-competitive effect flowing from the cartel. These are matters germane to cross-examination and bear upon the weight the CAT might attach to an opinion. The CAT is entitled to infer that employees participating in a cartel might be amongst the most knowledgeable and expert of all a defendant’s employees in assessing, in real time, whether any gain was made from the cartel (cf Stellantis (ibid) paragraphs [73]). A contemporaneous view might well have higher probative value than that of an expert, instructed years after the event, who seeks to reconstruct market realities as they stood many years earlier.

The CAT might also wish to confirm that all experts understand, and respect, what is entailed in owing a duty to assist the Tribunal (cf CAT Rule 4(7) – duty of cooperation between the parties and the Tribunal). This is a duty with bite. The CAT is entitled to be assured that experts will give full and frank disclosure of anything, including prior relations with a client, impacting upon the objectivity and independence of their opinion. The CAT is also entitled to expect experts to adjust their opinions, even to the detriment of their clients, in the light of evidence as it emerges. An expert whose heels remain firmly dug in, might find such obduracy taken into account, adversely, by the CAT in the final account.