Where a supplier has, in breach of duty (tortious, contractual or otherwise), charged a purchaser too much for supplies (“the overcharge”), can the supplier seek to defeat a claim for compensation brought by the purchaser by pleading that the purchaser has mitigated the overcharge by neutralising the sum in question by securing commensurately increased discounts on supplies to it from other suppliers (“off-setting”)? In particular is it permissible to plead such a defence without any actual evidence that the claimant did in fact mitigate its loss in this manner but only upon the hypothetical basis that it is a “reasonable” inference that can be drawn that the purchaser would have mitigated in this manner. This is, in a nutshell, the issue arising on this appeal.

On 19^th March 2014 the EC Commission issued a decision (“the Decision”) under Articles 101 TFEU and 53 EEA which prohibit agreements which have as their object or effect the restriction of competition. The Decision found that the defendant and 5 others had engaged in a collusive tendering cartel which spanned a 7-year period between 8^th April 2004 and 25^th July 2011. The agreement was implemented beyond the territory of the EU and EEA but, for the purpose of the Decision, it sufficed that it affected trade between the Member States of the EU and EEA. It is common ground that, as of the date when these proceedings were commenced, Articles 101 TFEU and 52 EEA conferred private law rights which could be enforced in the domestic courts in claims for damages. The claimant pleads that the infringement amounts to a breach of statutory duty.

The agreement involved the exchange between suppliers of commercially secret information about customers’ procurement processes and agreement as to how to collude to defeat attempts by customers to impose competitive tendering upon suppliers by means of Requests For Quotations (“RFQs”). The agreement related to bearings for automotive applications which were supplied to original equipment manufacturers (“OEMs”) who produced components for vehicles. Each cartelist admitted participation and fines were imposed. Because of the admissions, the central issue before the Commission was mitigation and fines. As is now relatively common, in such cases the Commission issues a short-form decision which records at a high level the infringements but does not descend into either the evidence or the effects. These decisions predicate liability upon the object of the cartel which, according to well established case law which I do not need to address, is sufficient to trigger liability without proof of effect which is the alternative condition for liability.

In light of the Decision the defendant has admitted liability, which is therefore not in issue. However the probative value of the Decision is still limited. The fact that the Decision is in short form inevitably complicates the follow-on proceedings, even if they are (as here) limited to quantum, because there are no findings about the actual effects of the cartel on the market in question. For instance, there is no finding that the cartel succeeded in raising prices by “x”% above the competitive level, a finding which, had it been made, would have short-circuited a great deal of fact finding by the court hearing the quantum claim.

On 18^th March 2019 the claimant (“FCA”) commenced proceedings in the Commercial Court (which were subsequently transferred to the Competition Appeal Tribunal (“CAT”)) claiming damages for breach of statutory duty arising from breach of Articles 101 TFEU and 52 EEA in the approximate sum of €100million (including interest). The primary defence of the defendants (“NTN”) is that no loss at all was caused by the cartel. This is because, so it is argued, FCA was successful in using the RFQ system to prevent price increases for inputs. This might appear counterintuitive given that the defendant, and fellow cartelists, colluded for about 7 years, at exceptionally high risk of severe regulatory sanction if they were discovered, with the express object of seeking to defeat their customer’s competitive RFQ tendering strategies. The aim of NTN and others was to limit price competition and raise prices above the competitive level. According to the Decision, the cartelists exchanged information about RFQs and agreed who was going to bid in response and at what level so as to avoid undercutting each other. As such, the cartelists tailored and structured the cartel so as to counter attempts by the customers to use their negotiating powers to wrest better prices from suppliers. On one view, bearing in mind these decided facts, NTN would not have accepted the ever present risk of detection or whistleblowing, over such a period, absent some significant degree of confidence that they were in fact benefiting materially from the operation of the cartel.

Nonetheless, it has become the customary starting point for many defendants in damages follow-on claims to aver that there was no loss. This is the position taken by NTN. A secondary aspect of the no-loss defence is that, if there was an overcharge, then it is alleged that the claimant off-set any increase in prices by reducing prices elsewhere i.e. from suppliers other than the defendant. It is important to be clear as to what is and is not averred. The defence is not that FCA failed to off-set and, as such, acted unreasonably and in breach of its ordinary common law duty to take reasonable steps by way of mitigation (see paragraphs [18ff] below). It is not even that there must have been off-setting by FCA because the primary defence is that there was no overcharge at all, and therefore nothing to counter through mitigation. NTN’s case is based upon the conditional hypothesis that if there was an overcharge the claimants would have mitigated the overcharge by off-setting. However, this averment is not advanced upon the basis that NTN has any actual knowledge or evidence that FCA, actually, mitigated by offsetting. Instead, NTN pleads only that it can infer (it says reasonably) that FCA would have engaged in off-setting if there was an overcharge.

FCA applied to the CAT to have the off-setting defence struck out upon the basis that the pleaded defence was theoretical, lacked realism and was implausible and that to permit such a speculative defence would add disproportionately to the burden of the trial. NTN sought to rely upon voluntary further particulars of the defence. The defence was struck out and permission was refused to NTN to amend the pleading by reference to the voluntary particulars.

Appeals from rulings of the CAT lie to the Court of Appeal on points of law only. The issue for the Court is whether the CAT, in refusing to permit the defence to be advanced, made errors of law.

This matter comes before this Court as an application for permission to appeal and, if permission is granted, for the appeal to be heard thereafter. We received full and careful oral and written argument on the issue. For my part, I would grant permission to appeal given the arguments raised, the significance of the issue, and the fact that this is the first occasion upon which the implications of the judgment of the Supreme Court in Sainsbury’s Supermarkets Limited v Visa Europe Services LLC [2020] UKSC 24 (“Sainsbury’s”) on mitigation by off-setting have been considered at the appellate level. The issue has added significance given that a differently constituted CAT, which included the then President, in Royal Mail Group Limited v DAF Trucks Limited & Ors [2021] CAT 10 (“Royal Mail”), interpreted Sainsbury’s in a restrictive manner, with which the appellants disagree, and this was then followed by the CAT in the present case.

The present claim is a follow-on claim which rests upon findings in the Decision. That instrument is relied upon therein for certain facts, including that the defendant, NTN, has admitted unequivocally its liability for breach of the competition rules.

It is helpful to set out briefly the background relating to the regulatory proceedings which give rise to this claim. An investigation into suspected price fixing was initiated by the Japanese Fair Trade Commission (JFTC) who conducted on-site inspections on 25^th July 2011. The EC Commission also conducted inspections between 8^th and 10^th November 2011 in Europe. Applications for leniency were lodged with the Commission by various of the investigated parties (not however including NTN). Formal proceedings were initiated on 22^nd January 2013 with a view to engaging in settlement discussions with the parties. These took place and culminated in December 2013. The Commission sought “settlement submissions” from all parties, an integral component of which was:

NTN and others made this unequivocal admission. The Decision was addressed to six corporate groups based in Japan and Europe of which NTN was one: (1) JTEKT Corporation, JTEKT Europe Bearings I3.V., Koyo France SA and Koyo Deutschland GmbH ("JTEKT"); (2) NSK Ltd., NSK Emopc Ltd. and NSK Deutschland GmbH ("NSK"); (3) Nachi-Fujikoshi Corporation and Nachi Europe GmbH ("NFC"); (4) AB SKF and SKF GmbH ("SKF"); (5) INA-Holding Schaefiler GmbH & Co. KG, Schaeffler Holding GmbH & Co. KG, Schaeffler AG, Schaeffler Technologies GmbH & Co. KG and FAG Kugclfischer GmbH ("Schaeff1er"); and (6) NTN Corporation NTN Wälzlager (Europa) GmbH and NTN-SNR Roulements SA ("NTN").

In the Decision (recital [53]) the Commission found that the objective of the cartel was to limit discounts to customers and achieve prices for their products above the competitive level. All cartelists had: “…an identical anti-competitive object and single anti-competitive aim”. Within this scheme:

The Commission summarised the products covered by the agreement (recitals [3]-[5]). These were bearings for automotive applications comprising bearings supplied to automotive OEMs which were car, truck and automotive component manufacturers. Bearings are machine parts with rolling elements used in rotating parts within such cars, trucks and automotive components:

In recitals [28] – [30] the Commission described the cartel:

The issue on this appeal is whether the CAT erred in concluding that the pleaded defence did not disclose a proper averment which should go to trial. Two preliminary issues arise. First, as to the test to be applied. Secondly, as to the evidential standard that must be applied as part of the test. The CAT pointed out (correctly) in Royal Mail (ibid paragraph [32]) that these issues were not particular to competition law but were equally applicable to other contractual and tortious claims.

The basic test is that there has to be a sufficient causal nexus or connection between the steps that a defendant says a claimant took by way of mitigation (the off-setting) and the overcharge.

Recent authority, including the Supreme Court in Sainsbury’s, treats the analysis by Viscount Haldane LC in British Westinghouse Electric v Underground Electric Railways [1912] AC 673 at page [689] (“British Westinghouse Electric”) as articulating the governing principles of mitigation. These were, even in 1912, considered to be “well settled” and were based upon earlier Victorian case law of long standing. The first governing principle is that compensation is ordered by a court for “… pecuniary loss naturally flowing from the breach”. The requirement that loss must flow “naturally” introduces the need for some causative link. The right to recover such naturally flowing loss is however qualified by a second principle which imposes upon a claimant “… the duty of taking all reasonable steps to mitigate the loss consequent on the breach and debars him from claiming any part of the damages which is due to his neglect to take such steps”. There is thus a common law duty on claimants to take reasonable steps to mitigate loss. Viscount Haldane went to on to explain the limits of this duty to mitigate:

Viscount Haldane also cited by way of illustration various authorities, concerned with breach of contract, where to mitigate a loss a claimant had engaged in subsequent transactions which turned out to be very profitable. The issue in these cases was whether the benefits obtained from the subsequent transactions had to be set-off against the damages caused by the initial breach of duty. In Stanisforth v Lyall (1830) 7 Bing 169 the plaintiff, a ship owner, by taking prudent steps by way of mitigation, in fact made more money from the breach of contract by the defendant charterer, than would have been earned by due performance of the contract. The Court held that, if a subsequent (profitable) “transaction” was to be taken into account to reduce quantification, it had to be one “arising out of the consequences of the breach and the ordinary course of business” (ibid page [690]). On the facts of Stanisforth the Court held that the benefits did have to be taken into account. Viscount Haldane however cited a number of authorities which illustrated the other side of the mitigation line. These were where there was either (a) no causative link between the benefit and the breach or (b) where there was some link but the benefit was due to steps taken by the claimant which were outside the usual course of business. In these cases the claimant was held to be entitled to retain the benefit because it was collateral and it did not have to be taken into account in quantum: e.g. Bradburn v Great Western Ry (1874) L.R 10 Ex1; and Jebsen v East and West India Dock Co (1875) L.R 10 C.P. 300. According to Viscount Haldane, the rationale behind these latter types of case was that the benefit was derived not from the breach but from “… a contract wholly independent of the relation between the plaintiff and the defendant” (ibid page [690])). The implications of this line of cases is considered more fully below at paragraph [67].

It is clear from the analysis of previous authorities cited in Sainsbury’s that the Supreme Court did not cast doubt upon these governing principles and indeed was doing no more than seeking to apply them to the facts of the case before it. The italicised words set out in the quotation in paragraph [18] above, and in particular the words “arising out of the transaction”, were cited with approval by the Supreme Court in Sainsbury’s (ibid) paragraph [215] – set out below at paragraph [47]), as authority for the proposition that for a defence of mitigation by off-setting to run, there had to be a “legal or proximate connection” between the breach (the overcharge) and the act of mitigation.

The analysis of the Supreme Court was applied by the CAT in Royal Mail (ibid paragraphs [24], [25]). The CAT addressed, and rejected, an argument by the defendant that the analysis in Sainsbury’s proceeded upon the basis that causation was a “… subtle and imprecise question that can only be determined on the known facts of the case and must therefore be determined at trial”, not therefore at an interim stage, and that the judgment properly understood, amounted to a “green light” to plead a defence of mitigation by off-setting in broad and theoretical terms (ibid paragraphs [28ff]). The rejection of this “green light” approach was followed by the CAT in the present case.

This bring me to the second question which is the evidential standard that the CAT (or a court) must apply to determine whether a pleading is to be allowed to proceed to trial. There is a connection with the outcome of the first question since if all that is required is a broad brush and theoretical averment, devoid of fact or particularisation about the causative link, that the claimant did mitigate by off-setting, then it might be quite difficult to strike out a pleading, irrespective of the standard to be applied. Nonetheless, the second stage is important, especially if something more than a theoretical, bare bones, averment is required. As to this the parties in Royal Mail and in the instant case were agreed as to the test. In Royal Mail, at paragraph [22], the test was framed as follows:

In some cases, the CAT has used the expression “plausible” as a proxy for “realistic prospect” (e.g. Royal Mail ibid paragraphs [22] and [29]). There is no difference, in this context, between the two terms.

The parties have referred to a number of important policy considerations which underpin the test that the courts have identified over the years and which are not said to be modified in this case simply because the mechanism for mitigation is argued to be off-setting. These considerations are important in that they guide the exercise of determining what is “realistic” or “plausible”.

The first is a general observation about the balance to be struck between claimants and defendants: victims and wrongdoers. In case law and in commentary thereupon, reference is made to ensuring that “excessive demands” are not imposed upon claimants, that the rules should not be disproportionate, that the standard is “not a high one”, that instead of absolute precision in quantification all that is required is a “broad axe”, that the right being enforced should be made “effective”, that the rules should be applied with “pragmatism”, and that it must not be forgotten that at the end of the day the defendant is the “wrongdoer”.

All these considerations have a common core: whilst rights of defence must be observed, nonetheless where a claimant has a justiciable right the procedural and evidential rules governing the enforcement of that right must not be allowed to become so onerous that they undermine or weaken the very right itself by making it too hard to vindicate.

So, for instance, it has long been the case at common law that the courts are reluctant to impose too onerous a burden on the claimant in relation to the duty to mitigate. In the light of a settled body of case law the authors of Clerk & Lindsell on Torts (23^rd Edition, 2020) at pages [2062], [2063], state: “Judges are reluctant to impose excessive demands on claimants” (ibid paragraph [27-09]). In similar vein the authors of Chitty on Contracts (24^th Ed, 2021) at paragraph [29-09]) observe that: “the standard is not a high one, since the defendant is a wrongdoer”. There is no reason why this sentiment should be confined to the duty to mitigate and not extend to a more general understanding of how the principles of mitigation work.

Next there is the “broad axe”. In Sainsbury’s the Supreme Court emphasised the need to achieve proportionate and pragmatic justice, even if at the expense of precision. Common law instructed judges to determine compensation “… accomplished to a large extent by the exercise of sound imagination and the practice of the broad axe” (per Lord Shaw in Watson Laidlaw & Co Ltd v Pott, Cassels & Williamson [1914] SC (HL) 18, endorsed in Sainsbury’s at paragraph [218]). The Supreme Court stated:

The principle of effectiveness also applies. Under this principle, which is derived from EU law, procedural and evidential rules must not make it practically impossible or excessively difficult for a claimant to vindicate its justiciable rights (Sainsbury’s paragraph [188] and [189]). The common law is very much to the same effect; the rules should not be applied in such a way that the very right sought to be enforced is undermined.

How do these principles apply to a pleaded defence of mitigation? In ruling upon a pleading the CAT is not teeing up a trial which seeks a counsel of perfection. It is making a judgment call about what is realistic and proportionate, accepting that the adjudicatory process of quantification is a “broad axe” and that procedural rulings must not make vindication of the enforceable right too difficult. The CAT acts in a pragmatic manner consistent with the principle of effectiveness and with the common law concern to eschew artificial demands for precision.

In paragraphs [33]-[36] of Royal Mail, the CAT, taking into account that, were it to allow the defence to run to trial, this would impose a heavy burden upon the claimant, and referring to paragraph [189] of Sainsbury’s, rejected the argument that all that a defendant had to do was plead “broad economic or business theory”:

The CAT in the present case applied the principles in Royal Mail with the same end result.

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.

This is the test that I apply to the pleading in the present case.

All of the above is a precursor to an analysis of the pleadings. The essential thrust of the defence is that there was no overcharge, but that if there was, it was fully mitigated and/or passed on. The disputed mitigation by off-setting defence is contained in two documents: paragraph 41(c) of the Amended Defence and in subsequent Voluntary Further Particulars of the defence.

This is an alternative line of defence behind two prior averments, namely that the claimant passed on the overcharge to its own customers and/or that it negotiated away any overcharge through its superior negotiating power with suppliers.

The most basic line of defence is that the claimant mitigated its loss though the process of annual price reductions and other forms of negotiation and commercial pressure imposed by the claimant upon the defendant – mitigation by negotiating power. In paragraph 41(b) the following is pleaded;

This defence takes as its starting premise that the claimants had the “means” to seek reductions in price. It then proceeds to the conclusion that the “Claimants thereby mitigated their losses and/or had the means of doing so”. Whether the fact (assuming it to be true) that an OEM has the “means” of “seeking” cost reductions leads, as a reasonable inference, to a conclusion that “thereby” there was, actual, mitigation of a secret overcharge (i.e. “their losses”), is disputed by the claimant.

The pass on defence is pleaded at Paragraph 47: “…it is averred that any Overcharge (if there was any Overcharge) would have been passed on.” The phrase “would have been” is also an inference.

The crux of the present dispute lies in Paragraph 41(c) which raises the alternative defence of mitigation by off-setting:

The defendant elaborated this defence in paragraphs [3]–[9] of the Voluntary Further Particulars:

Mr O’Donoghue QC, in attractive and persuasive arguments for NTN, advances a number of points by way of criticism of the CAT. First, the present pleading meets the test in Sainsbury’s, which expressly contemplates that a mitigation by off-setting defence based upon the pleading of an ordinary cost control system can be “raised”, without more. Secondly, the pleading in dispute in fact goes beyond a bare-bones pleading because it is based upon disclosure (albeit limited) provided by FCA, from which it is possible to identify real details about the claimant’s cost controls system and the use within that system of targets. Thirdly, the pleaded inferences about the effectiveness of the cost control system in negating overcharges flow reasonably from such facts as are known and which have been particularised. Fourthly, it would be unfair if more than this had to be pleaded because, as was recognised by the Supreme Court in Sainsbury’s, these cases are characterised by evidential and informational asymmetry. A law which requires a defendant to plead relevant evidence of causality amounts to an unfair Catch 22 if the law then condemns the pleading because of the omission of evidence that can only be obtained if the pleading is allowed to run to trial and disclosure is then given by the claimant of its cost control system. Fifthly, if the CAT is correct, then it would be very hard indeed for any defendant in a price fixing cartel involving secrecy ever to mount a defence of mitigation. The application of the policy considerations described above (see paragraphs [26]-[34]) does not oust normal rights of defence. Sixthly, the issue raised is new and novel or at least relatively so and one that has not to date been aired and thrashed out at trial. The issue has important ramifications for the ability to run defences in cartel damages cases. All in all these are important reasons for not imposing an impossible burden on NTN in relation to the pleading, for allowing disclosure to occur, and for the court at trial to be in a fully informed position to assess the issue.

For the reasons set out below I do not agree.

The starting point is the Sainsbury’s judgment and the appellants’ argument that Sainsbury’s has endorsed the principle that evidence or particularisation of causation need not be pleaded because, given informational asymmetry, to require such evidence would make the defence impossible ever to plead.

In Sainsbury’s, the Supreme Court addressed issues of mitigation in a section of the judgment concerned with the “broad axe” principle. This relates to the degree of precision that is required in the assessment of mitigation of loss at trial where a defendant asserts that the claimant has mitigated its loss through the pass-on of all or part of an overcharge to its customers (see ibid paragraph [175]). The Court of Appeal had taken a view which the Supreme Court held, on analysis, was too strict upon a claimant. It adopted a more flexible and pragmatic (pro-claimant) approach which permitted a reasonably high degree of estimation – the “broad axe” (ibid paragraphs [217]-[226]). The Court was not specifically dealing with mitigation by off-setting; nor was it dealing with the standard that was to be applied to a pleading at an interim, pre-trial, stage and whether it met the threshold required to sustain an averment of causation as between the breach and the mitigation to permit it then to proceed to trial. At the point at which the Supreme Court was addressing the issue there was no dispute about the realism or plausibility of causative links between the pleaded breach and the mitigation; such matters were distant history. The Court’s observations on mitigation by off-setting must be seen in this context.

The facts of the case related to losses sustained by merchants though operation of the multilateral interchange fee (“MIF”) charged by a card customer’s bank (the “issuing bank”) to the merchant’s bank (the “acquiring bank”), which had been held to violate Article 101 of the TFEU. The MIF is passed on by the acquiring bank to the merchant as part of the bank’s merchant service charge (“MSC”). In a judgment at first instance focusing upon quantum of damage ([2016] CAT 11) the CAT held, on the facts of the case, that as a large-scale business and sophisticated retailer, a supermarket would respond to the MIF in one or more of four ways, the third of which referred to mitigation by off-setting:

The Supreme Court accepted this analysis as a finding by the CAT describing how the retail sector operated in practice. It emphasised that whether there had been mitigation by off-setting was a question of fact but that once it had been (properly) “raised” there was a burden – potentially quite heavy – upon a claimant to provide disclosure of documents relevant to its cost control system. This would reveal which, if any, of the cost avoidance mechanisms had been used. If adequate disclosure was not provided then a court might draw an inference that it had mitigated all or some of its loss. The relevant paragraphs of the judgment are as follows:

The Court did not spell out what in practical, pleading, terms it meant in paragraph [216], when it referred to the issue of pass-on having been “raised”. This was not an issue before the Court. It did not say that it sufficed merely to plead the bare assertion that there had been mitigation by off-setting and nor did it say that there was no burden on a defendant to support such an averment with sufficient particularisation and evidence, such that it surpassed a minimum threshold of realism and conviction. It was plainly relevant that, by the time the Sainsbury’s case was heard by the CAT, and by the courts on appeal, the issue of MIF pass-on generally had been live in the industry for many years as had the compatibility of the MIF with competition rules. The first EU Commission proceedings in relation to a MIF were in the early 1990s. The CAT had addressed the issue on several prior occasions, including in relation to damages and pass-on defences. It had been aired in the General Court of the EU. There was nothing secret about the imposition of a MIF. It was a transparent, known, charge and it was a recognised industry practice that acquiring banks passed it on to retailers in the MSC (see Royal Mail paragraph [25]). All of this was clearly relevant to the burden facing a defendant in this sector seeking to raise a realistic case of mitigation. The MIF was a systemic and troublesome cost that any major retailer would, inevitably, have had to confront. The facts therefore contrast with those of a typical, secret, price fixing cartel.

I agree with the CAT both in this case and in Royal Mail that the Supreme Court was not addressing in its judgment the standard that a pleading had to reach before being allowed to proceed. It follows that the Court was not saying that the pleading of a bare-bones averment that there was mitigation by off-setting would always pass muster.

I turn now to my conclusions on the pleading. Does NTN’s pleaded case on off-setting meet the requisite standard? In my view it does not establish a realistic (or “plausible”) case of mitigation by off-setting. The defence is theoretical. It is based upon successive “inferences” which, on analysis, are not inferences that properly can be drawn from primary facts. It makes assumptions about the existence of hard and certain facts which are very far from being hard or certain at all. This can be seen by a careful break down of the pleading.

When first served the amended defence was devoid of particularisation. Paragraph 41(c) pleaded that there was full mitigation by the claimant by a variety of means only one of which involved off-setting through “reducing other costs”. This is an assertion of fact. But an assertion is not evidence and no particulars were provided.

The pleading then put down a marker (derived from Sainsbury’s) as to what the claimant had to prove, which included an obligation to prove the negative including that it did not mitigate the overcharge. According to the pleading the claimant had to prove: “… not only that any alleged loss was passed on to them, but also that they did not pass on any alleged loss (or otherwise mitigate it) to their own customers or otherwise mitigate it, including through reducing their other costs.” See paragraph [40] above.

The suggestion that the claimant must establish a negative is an erroneous interpretation of Sainsbury’s paragraph [216] (supra) which makes clear that the burden lies throughout upon the defendant to prove its defence. This point is also made clear in paragraph [219]. The reference to the heavy “evidential” burden on claimants is a true description of the burden that disclosure would impose were it to be ordered, but it is not an indication that the legal burden of proof shifts to the claimant once the defendant has “raised” mitigation as a defence. The reference to the defendants being entitled to invite the court to draw an adverse inference in such a case is one which unsurprisingly arises if a claimant fails in its disclosure burden. The fact that it is the defendant who is entitled to invite the court to draw an adverse inference is not inconsistent with it being the defendant who still has the legal burden of proof, and, of course, the court is not bound to draw such an inference or at least not the inference that the defendant invites it to draw, in which case the defendant might fail to establish its defence.

The Voluntary Particulars endeavoured to add flesh to the bare bones of paragraph 41(c). They refer to various facts which, as the CAT found, were matters NTN would have known about and, in the event, are not matters FCA has taken issue with at the pleading stage. These facts relate to the existence of a cost control system involving the use of targets. Paragraphs 3(a) - (d) set out that FCA prospectively sought to control costs on inputs from suppliers by setting targets for both the total costs of a particular vehicle or part and for reducing those input costs. There is also a pleading that the purpose behind the costs targets was to provide relevant FCA procurement staff with benchmarks. Paragraphs 3(e) and 4 aver that FCA maintained systems for monitoring EEA supplier performance. In addition, the FCA Purchasing and Finance Teams measured the commercial performance of suppliers which included whether suppliers realised FCA’s targeted cost savings. It is also pleaded that the purpose behind such measures was to ensure that the total input costs for a vehicle or part of a vehicle did not exceed a specific level.

The facts set out in the pleading are limited and high level. Standing back they present no surprises since it is intrinsically likely that any OEM customer in this industry would have some sort of a system in place to endeavour to control and reduce input costs. For the purpose of analysing the mitigation by off-setting defence I will assume, for the purpose of testing the arguments, that these averments of fact are realistic and plausible.

It is that which follows which is problematic. The point of departure is that NTN’s primary case is that there is no overcharge at all and/or that if there was it was passed on. By its nature it is therefore hypothetical and conditional. The pleading does not therefore set out any particulars or evidence explaining how or why any element of overcharge would be identified and quantified by a customer such that the procurement teams could then, properly appraised, seek to off-set the overcharge by increased efforts to reduce costs elsewhere; nor how or why FCA would actually be successful in pushing though such cost reductions from other sources.

In order to fill this gap in the evidence on causation, and taking as the starting point the pleaded facts, NTN pleads a series of inferences which lead it, ultimately, to the factual conclusion that FCA would have mitigated any overcharge. These inferences concern matters important to any defence pleading causality between the breach and mitigation and cover: the internal system pursued by FCA to incentivise staff to meet targets; the utility of the targets system in setting budgets; and, whether the system was “effective” in regulating costs and in ensuring that FCA made a profit.

The pleaded inferences (set out in paragraphs [3(d)], [4] and [5]) are as follows:

FCA measured the performance of its procurement staff in meeting targets and provided incentives accordingly.

FCA’s use of targets was “effective” in controlling FCA’s costs.

Targets were “useful” in planning overall budgeting and ensuring FCA made a profit in line with plans and expectations.

FCA would set and pursue targets as part of the process of setting product prices.

Paragraph [6] then proceeds upon a further assumption, namely that the costs targeting system achieved its purpose. It is prefaced by the words “If effective”. From this conditional starting point, NTN then pleads that FCA procurement staff “would” negotiate the prices of the various inputs, so that the total costs of those inputs did not exceed the target. The use of the imperative “would” brooks of no possibility of failure to meet targets. As drafted, this is an inference based upon inferences.

Finally, paragraph [7] starts with “in the premises” and thereby takes as its starting point all the preceding assumptions and inferences. This now asserts, in equally emphatic terms, that if there was an overcharge (which of course is denied) the effect in practice of the existence of the costs regulation system “would” have been that FCA “would” have negotiated “lower prices with other suppliers to offset any overcharge (or part thereof)”. As pleaded, this is an inference upon an inference itself based upon inferences.

The pleading, as the CAT concluded having reviewed the pleading as well as the evidence NTN relied upon (ibid paragraphs [26]-[36]), is theoretical. It assumes what has to be proven and a number of pivotal links in the logic chain represent assumptions which are evidential leaps in the dark and are certainly not inferences that can be drawn from such limited facts as can properly be pleaded, concerning the mere existence of a cost control system which uses targets.

As for the inferences, for my part I do not see how these flow, naturally or causally, from the facts that NTN can properly plead. Legal dictionaries define an inference in terms of conclusions that flow logically, reasonably or rationally, through a process of reasoning, from proven or admitted facts.

Here, the mere fact that there is a cost control system which involves targets does not reasonably (logically or rationally) lead to the inferred conclusion that FCA would in fact mitigate an overcharge by obtaining better prices from suppliers of other products.

The most striking of the averred inferences – and probably the most critical – is that because FCA has a system for controlling costs involving targets it would be “effective” in off-setting overcharges by reduction in costs from other suppliers. The existence of a system to reduce costs is no guarantee of its success and it is not an inference which can therefore be inferred from the pleaded facts. There are innumerable reasons why, even with the best system in the world, a purchaser could never be certain of its efficacy in suppressing input costs to a level which counteracted supra-competitive prices charged elsewhere. Many of the factors upon which success will depend will be beyond the control of the purchaser. For example, in a cartel case the fact of the overcharge will be a closely guarded secret. If – because of the cartel - all suppliers are charging supra-competitive prices, then from the purchaser’s perspective, looking around the market, all prices will be comparable, and customers will, rationally, assume that the visible or quoted prices (in response to RFQs designed to stimulate competitive tendering) are competitive prices. Indeed, as the Decision established, it was the very object of NTN’s covert arrangement to keep input costs up across the board and for about 7 years NTN and its co-conspirators acted so that prices across the market would appear to be competitive when in reality they were not. If a purchaser does not know that prices it is charged are supra-competitive it is hard to see how that fact can then be factored into any cost control system, including targets, which then sets out to neutralise the overcharge.

Further, if the cost of the cartelised component is a small or modest portion of total costs, then there might in any event be little incentive or ability to seek an off-set. The cost of the component as a part of the total costs is not of course the relevant comparator; it is the overcharge (i.e. the increment over the competitive price) as a percentage of total costs/expenditure that matters. If the cartel is very successful and pushes prices up by (say) 5% over the competitive level on a component that itself is (say) 5% of overall costs then the overcharge, which amounts to only 0.25% (i.e. 5% of 5%) of total costs, that might still not be such as to trigger any impetus for off-setting, even in the most rigorous and challenging of cost control regimes. Moreover, NTN’s case is that because of the customer’s negotiating power there was no overcharge at all. The absence of any pleaded admission as to the level of overcharge reinforces the speculative and theoretical nature of the defence. What if the overcharge was only 0.5% or 1% instead of 5%? On the facts of the present case the CAT found that the gross amount of the overcharge as a percentage of FCA’s relevant expenditure was “extremely small” (ibid paragraph [31]). This finding has not been challenged.

And yet further, whether a cost control regime will be effective might depend upon the negotiating power of the other suppliers, who by definition are supplying different products which will have different cost structures, and whether these other suppliers had the capacity or tolerance to agree price reductions, which may be largely beyond the purchaser’s control.

There are a number sanity checks that can be applied to the reasonableness of the proposition lying behind the pleading. First, I have already explained how the premise of the defence is hypothetical and conditional (see paragraph [6] above). NTN’s pleading is not that FCA must have mitigated by off-setting since its basic case is that there was no overcharge, so nothing therefore to be mitigated. NTNs case is hence only that if – which it denies – there was an overcharge it would have been mitigated. Secondly, if however there was an overcharge, FCA was oblivious to its existence (it being a secret) such that if FCA’s costs control system did, unwittingly, negate the overcharge by the extraction of better discounts from other unrelated suppliers then FCA would, on the case law, at least have a shout at arguing that such a benefit was a collateral advantage unrelated to the breach and therefore irrelevant to quantum (see paragraph [19] above). Thirdly, NTN does not argue that, if FCA failed to eliminate the overcharge by off-setting, it was in breach of its common law duty to take reasonable steps to mitigate loss (see paragraph [18ff] above). Had it advanced such a plea it might have been met with the judicial reaction that to contend that it was unreasonable for FCA to fail to take steps to counter a secret price fixing cartel specifically designed to undermine its procurement system was far-fetched. Mr O’Donoghue QC for NTN did not demur from this proposition. But, if it was reasonable for FCA not to take steps to neutralise a secret overcharge, then by the same token it seems inconsistent, simultaneously, to argue that it should be reasonably inferred that it would have done so.

The appellant argues that FCA is a large and sophisticated entity with contract negotiating power over its suppliers. But even if this be so, it does not address an important, admitted, fact in this case. It is clear from the Decision, and its description of the cartelists and their turnovers and the international scale of their operations, that they were major corporations who designed their cartel with the specific objective of neutralising such negotiating power as customers were able to exert.

NTN also argues that the CAT ignored an important point which is that this is a novel area of law, especially following Sainsbury’s, which should be explored fully, but only at trial in the light of the evidence. With this in mind, the CAT should have allowed the pleading to stand and should have proceeded to order disclosure. I disagree. Disputes about mitigation have arisen in countless tortious and contractual cases over the decades and, as already explained (see paragraph [18] above), by 1912 the principles were “well established”. Insofar as there is novelty it is only in the application of these governing principles to the factual permutation that the mitigation is said to be by off-setting; but even this has been explored before as an issue of fact in the credit card interchange fee cases.

NTN argues that a rejection of its pleaded off-setting defence would imply that such a defence could never be raised. I accept that raising a viable case of this nature might be difficult, but it is not impossible in all cases. None of the above analysis assumes that mitigation by off-setting is always incapable of being advanced in a pleading. The Supreme Court in Sainsbury’s recognised the possibility that a defence of mitigation by off-setting might arise and could affect the quantum of compensatory damages. Whether this is so will always be a question of fact. The CAT, in Royal Mail, endeavoured to identify some examples of matters that might in a proper case justify the pleading of a viable defence. The Tribunal stated at paragraph [42]:

The Tribunal concluded at paragraph [43] as follows:

The CAT in the present case (ibid paragraphs [31ff]) referred to these examples from Royal Mail but held that, in its view, none applied on the facts. Neither judgment however indicated that, where such facts do exist, they necessarily suffice to raise a viable pleading of mitigation by off-setting. The Tribunal in both cases was doing no more than positing possible indicative factors which might, on the facts, be relevant to whether a sustainable defence could be mounted. The CAT confirmed that a realistic pleading does not necessarily have to refer to or rely upon disclosed documents. It is well established in case law that, at the interim stage, the court will make due allowances for the fact that disclosure has yet to occur. Nonetheless, there still has to be some basis for permitting the averment to proceed to trial.

Before considering these non-exhaustive illustrations of pleadable circumstances which might indicate mitigation I would add an observation. It is notable that NTN has chosen not to plead a positive evidential case and instead has advanced its case on causation solely by reference to inferences said to be reasonably flowing from high level primary facts. In the course of the seven plus years of the cartel it is at the least likely that NTN would, internally, have generated documents reviewing or reflecting the success or otherwise of the cartel and the extent to which it was effective in enabling NTN to impose an overcharge, which after all was its objective and purpose. This internal documentation might have shed light on the steps customers took to counter cost increases. It has not, for instance, pleaded that on “x” date NTN (or even some other cartel member) overcharged FCA by “y” amount and that it was observable from FCA’s response that it endeavoured to mitigate the overcharge. It is also possible that there would, over the course of the cartel, have been exchanges between the parties of a general nature evaluating or commenting upon the reactions of purchasers to the prices being charged to them which shed light on how OEM purchasers in the sector habitually reacted. NTN might have pleaded evidence to show, just as with the supermarket retail sector and its response to the MIF, that this was the sort of sector where customers had well established strategies for defeating supplier costs increases which extended to off-setting. None of this has been pleaded. The reason a defendant in a case of this sort might not advance a positive case could, of course be that, if pleaded, it risked being inculpatory and counterproductive to the core defence that there was no loss at all, and it could risk opening doors to disclosure which a defendant might prefer remained firmly locked.

I turn now to the indicia referred to by the CAT and their relevance to a case such as the present.

First, the CAT mentioned a claimant’s knowledge of the nature and amount of the overcharge in which case it might be “inherently likely” that a claimant would seek to address it. In the present case the fact of the cartel was kept secret from FCA for over 7 years so that there was nothing FCA could address its mind to as an identifiable or quantifiable cost to be countered. In any event if a customer becomes aware of a cartel the most natural response might very well be, forthwith, to issue proceedings and/or complain to a regulator either of which should suffice to bring the cartel to an abrupt end. It might not therefore be “inherently likely” that the customer’s response to learning of the overcharge would be to seek to mitigate by off-setting.

Secondly, reference was made to the case where the overcharge as a proportion of the claimant’s relevant expenditure was high such that it was more likely that some step would have been taken to mitigate the impact. Again, this might assume that the customer is aware that the overcharge is “high” whereas in a secret cartel the impression given is that (a) all prices are competitive ergo (b) there is no overcharge of any magnitude. I would add that in this case there is no pleading as to the portion of relevant total costs or expenditure that the cartelised components accounted for. I have addressed this further at paragraph [65] above.

Thirdly, the CAT referred to the relative ease with which FCA’s business could be expected to reduce certain input costs or input costs “generally”. In this case NTN does plead that, on its theory, FCA could and would have off-set costs. But, as observed, the pleading is abstract and not grounded in facts or evidence. And one must return to the point that without any knowledge that a particular cost is abnormally high there might be little impetus to seek to off-set it. The facts of the present case contrast with the facts of Sainsbury’s. There the disputed charge was transparent and retailers had well established strategies for countering it which were known to include all of the methods pleaded in the present case, including off-setting. Mitigation by these strategies was an entrenched and established part of business as usual; the Supreme Court in Sainsbury’s described the issue of causation in the sector as “straightforward” (ibid paragraph [215] cited at paragraph [47] above).

Fourthly, the CAT referred to the fact that other supplies made by NTN or its associates to FCA had been renegotiated in years following the increase in the prices alleged to have been caused by the anti-competitive conduct. That is not pleaded here and, once again, its indicative value would always be highly fact sensitive; there might be many reasons for a price renegotiation which are unrelated to overcharging.

Mr O’Donoghue QC placed considerable reliance upon the existence of targets, as part of the control system, as creating a mechanism which it was said created an inference that mitigation would necessarily flow. He criticised the analysis of the CAT who addressed this argument (ibid paragraphs [33] and [34]:

In my view the assessment of the CAT about what could reasonably be inferred from the existence of targets was one of specialist judgment and any criticisms of it are not, at base, points of law over which this court has jurisdiction. In any event, for the reasons already given, I see considerable force in the CAT’s assessment.

I return finally to the argument that the analysis of the CAT places NTN in a “Catch 22” predicament whereby the CAT accepted that in principle an off-setting defence could be pleaded but, simultaneously, made it impossible to plead sufficient facts to get such a defence off the ground by denying NTN the chance to obtain and review disclosure from the claimant. The short answer to this is that if a defendant does not have any realistic evidence of a possible defence, then it has no right to go fishing in disclosure to see if there is anything that might turn up which would help. As the CAT below and in Royal Mail observed there has to be a properly pleadable starting point before the claimant is put to the heavy burden that disclosure involves. In this case the pleading simply does not arrive at the starting point.

The CAT applied the correct test to the pleadings. It did not misconstrue the judgment in Sainsbury’s, nor did it misapply the relevant policy considerations. The CAT made no errors of law in its analysis of the pleadings and in its conclusion that they failed to meet the appropriate test. I agree that the pleaded off-setting defence lacks particularisation or evidential underpinning. It is hypothetical and theoretical and, in some respects, counterintuitive. It lacks realism. There are no policy considerations which justify permitting this defence to continue to trial.

For all these reasons, whilst I would grant permission to appeal, I would then dismiss the appeal.

I agree.

I also agree.