Stellantis Auto SAS & Ors v Autoliv AB & Ors

This is a cartel case. On 2 November 2023 the Competition Appeal Tribunal (CAT or the Tribunal) ruled that the defendants’ economic expert evidence should be given by a single expert shared between the three (now two) groups of defendants involved. The two questions on this appeal are whether the CAT was right to hold that there was no relevant conflict of interest in relation to that evidence and, more widely, what principles are to be applied when a court or tribunal is considering such a direction in future.

The claimants are companies in the Stellantis group of car manufacturers. Their brands include Vauxhall, Opel and Maserati. The defendants are manufacturers of occupant safety systems (“OSS products”) such as seat belts and airbags. The first to fifth defendants are members of the Autoliv group and the sixth to tenth defendants are members of the ZF TRW group. A separate company, Tokai Rika, was the eleventh defendant but the claim against it was withdrawn by consent in December 2023.

In two decisions dated 22 November 2017 and 5 March 2019 the European Commission found that the defendants, and others, had been involved in cartels in respect of sales of OSS products to motor manufacturers in the EEA, in breach of Article 101 of the TFEU and Article 53(1) of the EEA Agreement. The decisions are settlements in which the addressees admit the breaches. Neither decision concerns sales to the claimants.

The first decision (OSS1) relates to sales to certain Japanese car manufacturers (Toyota, Suzuki and Honda) between 2004 and 2011. Autoliv is an addressee of OSS1 and is identified in three of the four cartels. Its conduct started in March 2006. ZF TRW is not an addressee of OSS1.

The second decision (OSS2) relates to sales to Volkswagen and BMW from 2007 to 2011. Both Autoliv and ZF TRW are addressees of OSS2 and are identified in the two cartels in that decision.

All six cartels identified in OSS 1 and OSS2 are directed at an individual car maker. In OSS1 the cartels are also product specific whereas in OSS2 each of the cartels covers all the relevant products.

The claimants bring this claim for cartel damages (over €734 million) for the supply to them of OSS products by the defendants. The claimants plead a single cartel, involving both Autoliv and ZF TRW, directed at manufacturers including the claimants and covering at least the period addressed by the two decisions. Alternatively if there were individual cartels, then the claim is that these had the same membership and concerned supplies to the claimants. The third way the case is put is as an “umbrella” claim. Even if there were no cartels directed at the claimants, it is alleged that the effect of the cartels established by the Commission decisions was to increase prices charged to others, such as the claimants.

The claimants’ case is that Autoliv and ZF TRW are jointly and severally liable for the damages claimed. The damages sought do not differentiate between the two sets of defendants. No contribution notices have, as yet, been served by either set of defendants.

The claimants’ evidence includes factual evidence, such as emails, said to demonstrate a cartel between Autoliv and ZF TRW operating against the claimants. In addition the claimants rely on expert economic evidence to show the existence and extent of an “overcharge”. That is the difference said to exist between the prices the claimants actually paid for the relevant goods and the prices which would have been paid had the infringement not occurred. This is done using a regression analysis to identify the relationship between prices and a set of variables that may have a bearing on price, such as relevant costs. The argument is that once all the variables are accounted for, the existence of a remaining overcharge is something from which the defendants’ unlawful conduct can be inferred, in the absence of any other explanation.

At a CMC in March 2023 the Tribunal raised for the first time, and of its own motion, the possibility of directing that the defendants should call a single joint expert in the field of competition economics. The Tribunal’s view was that it was highly unsatisfactory that the claimants or the Tribunal should have to consider what would then have been three different economic models from what were then three distinct groups of defendants. The Tribunal appreciated it had not heard full argument on the issue and gave the defendants liberty to apply.

The Autoliv and ZF TRW defendants made that application, which was heard at a CMC on 20 October 2023. The application failed. The Tribunal’s ruling was given on 2 November 2023. One aspect of the applicants’ case was a submission based on the rights of the defence in Art 48 of the Charter of Fundamental Rights of the European Union, citing Emerald Supplies v British Airways [2015] EWCA Civ 1024. The Tribunal held that the order for a single joint expert would not breach Art 48. Permission to appeal on that ground (which also included Art 47 of the Charter and Art 6 ECHR) was refused here and below.

The main focus of the Tribunal’s decision was on case management. The Tribunal noted the governing principles specified in Rule 4 of the Competition Appeal Tribunal Rules 2015 (the CAT Rules). The rule provides that the Tribunal shall “seek to ensure that each case is dealt with justly and at proportionate cost”. The Tribunal also noted Rule 53 of the CAT Rules and a statement at paragraph 7.65 of the CAT’s 2015 Guide to Proceedings that the Tribunal will take into account the principles and procedures in CPR Part 35 on expert evidence. I will come back to this material below.

Having been taken to two cases under the CPR, ES v Chesterfield and North Derbyshire Royal Hospital NHS Trust [2003] EWCA Civ 1284 and Yearsley v Mid Cheshire Hospitals NHS Trust [2016] EWHC 1841 (QB), the Tribunal made the point that unlike those cases, it was not contemplating a single joint expert to be shared between the claimants and defendants. The Tribunal held it was difficult to identify any general point of principle from those cases which provided relevant guidance in the current situation.

The Tribunal referred to UK Trucks Claim Limited v Stellantis NV & Others [2023] EWCA Civ 875 in which the Court of Appeal identified a conflict of interest between two classes of representative claimants and held that separate experts would be appropriate for the two classes. I will address UK Trucks below. The Tribunal rejected the submission that there was an established practice in the CAT of defendants instructing individual experts in cartel cases given how few cartel cases had gone to trial and that UK Trucks was the only fully reasoned decision.

The Tribunal held that justice would be best served in this case by having a single joint expert in the field of competition economics shared by the defendants.

After the ruling a number of things happened. In December 2023 Tokai Rika withdrew and the claimants amended their pleadings. On 6 February 2024 permission to appeal was given by Popplewell LJ. In March or early April 2024 the claimants served their expert’s report (from Mr Hughes of Alix Partners) dated 28 March 2024.

After these events, on 15 April 2024, the remaining two groups of defendants Autoliv and ZF TRW applied to the Tribunal to revisit the expert question, asking for permission for them to have separate experts on the issue of overcharge. The Tribunal’s ruling rejecting that application was given on 22 April 2024. I will refer to paragraphs of that additional ruling as “A[1]” etc. The Tribunal gave permission to appeal and the two appeals were heard together.

The Tribunal’s consistent reasons for the two decisions of 2 November 2023 and 22 April 2024 can be summarised as follows. The court has power to make such an order if it is in accordance with the governing principles of dealing with the case justly and at proportionate cost ([19(1)]). The question of what is just has two particular dimensions in a case like this. One is whether there is a risk of a conflict of interest relevant to the matters the expert evidence is concerned with ([19(2)]). If there is such a conflict then it will not ordinarily be appropriate to order joint experts ([19(2]). The other is to consider the challenges the complexity of the proceedings places on the Tribunal ([19(3)]). It may be that the challenges of reconciliation are magnified when multiple unreconciled positions are advanced by different defendants, over and above the inevitable challenges of dealing with the evidence from a single claimant expert and a single defendant expert [19(3)]. Applying these principles to this case, given the size of the claim instructing multiple experts would not have been disproportionate. However no relevant conflict of interest has been identified, and the arguments to the contrary are theoretical and do not impact on the expert evidence ([24]-[29] and A[10] – A[18]). For both defendants to rely on their own individual expert would introduce unnecessary complexity, magnifying the challenges faced by the Tribunal and impacting on the quality of justice. The Tribunal would be faced with the task of resolving different methodologies using different data sets from the various defendants ([32] and A[19]-A[20]). Therefore the right order is for a single economics expert on behalf of the defendants.

Permission to appeal the first ruling was sought by the defendants on two grounds. Ground 1 was the ECHR / Charter rights point, which was refused. Ground 2 was put as an error of law in this way:

“Having rightly held that if there is a conflict of interest in relation to matters to which the expert evidence is directed, it will not ordinarily be appropriate to order joint experts (Ruling, paragraph 19(2)), the Tribunal erred in concluding that there are no material conflicts of interest between the Defendant groups in relation to the proposed use of expert evidence in the field of competition economics (Ruling, paragraph 29).”

The Tribunal had refused permission on Ground 2. In this court on 6 February 2024 Popplewell LJ gave permission on Ground 2 on the basis that it was arguable that such conflicts of interest did exist.

The focus of the appellants’ case on appeal was that relevant conflicts of interest between the two groups did in fact exist, and the Tribunal was wrong to hold otherwise. Therefore it followed that separate experts would and should be permitted. A key part of that submission was the Court of Appeal’s decision in UK Trucks and in particular [96], which the appellants submitted represented authority for the proposition that the existence of a conflict of interest, at least in an econometric case of this kind, meant that separate experts should be ordered.

During the course of oral argument on the appeal, following questions from the court, the appellants clarified their case. The appellants’ case has two aspects. One is the submission that the Tribunal erred in failing to identify relevant conflicts. The other is that if such a conflict exists it follows that separate experts must be permitted as a matter of principle. Other discretionary factors do not come into the matter. When asked by Arnold LJ if the appellants’ case was in effect that the existence of a conflict was a trump card, counsel fairly agreed that that was the submission. As a fallback counsel submitted that the existence of a conflict was a relevant matter which had to be weighed in the balance.

The respondents supported the Tribunal’s conclusion that there was in fact no conflict of interest in this case on the economic evidence. On the second issue of the role conflict of interest played in the analysis, in response to the questions from the bench, counsel cited a passage from Phipson on Evidence and a judgment of the Court of Appeal in Oxley v Penwarden (21^st July 2000 unreported, 2000 WL 1274095). In summary the submission was that the existence of a conflict of interest was not enough and the test was whether a party would be disabled from presenting its case.

Although it was not identified distinctly until the hearing, this second issue is a point of law distinct from the question whether conflicts do or do not exist in this case and is also, in my judgment, an intrinsic part of the appellant’s case on this appeal. It is a matter which this court should address. Neither party took a point on the absence of a respondent’s notice.

The parties both proceeded on the basis that there is no relevant difference between the principles applicable to expert evidence in the CAT and the civil courts of England and Wales under the CPR. I think that is right but since the critical provision about expert evidence in the CPR is CPR Rule 35 and that does not have an express analogue in the CAT Rules, it is worth examining that point in more depth.

The relevant aspects of CPR Rule 35 are the following. First r35.1 (and r35.4) identify the court’s duty to restrict expert evidence to that reasonably required to resolve the proceedings. Second r35.3 provides for the expert’s overriding duty to the court, which is a duty to help the court on matters within their expertise which (r35.3(2)) overrides any obligation they have to the person from whom they have received instructions or who is paying them. Third r35.7 (and r35.8) set out the court’s power to direct evidence from a single joint expert, applying when two or more parties wish to submit expert evidence on a particular issue.

Since the CAT has its own code in the CAT Rules and since it is a UK tribunal, not limited to England and Wales, one cannot necessarily assume that the principles applied in the CAT in procedural matters are the same as those in the CPR. The CAT Guide 2015 explains the relationship between the CAT Rules and the CPR in paragraphs 3.1 and 3.2, which are worth setting out in full as follows:

“SECTION 3: THE GENERAL APPROACH OF THE RULES

3.1

The 2015 Rules seek to achieve the general objective of enabling the Tribunal to deal with cases justly and at proportionate cost, in particular by ensuring that the parties are on an equal footing, that expense is saved and that appeals are dealt with expeditiously and fairly. This is set out in the governing principles in Rule 4. The Rules will be interpreted in accordance with those principles: Rule 2(2).

3.2

The Rules pursue the same philosophy as the CPR of the High Court and many of the rules are modelled on the CPR. Where, in particular as regards private actions, a rule mirrors the CPR, the Tribunal would generally expect to interpret that rule in the same way as the High Court or Court of Appeal. However, the Tribunal’s Rules are different in various respects and parties should not assume that the approach of the CPR applies to a particular procedural issue. Furthermore, the Tribunal is a United Kingdom, not an English, tribunal and it may therefore also have regard to the procedural rules that apply in Scotland or Northern Ireland, in particular in a case where the proceedings are to be treated as proceedings in either of those jurisdictions: see Rule 18.”

[emphasis in the original]

Page 1 of the CAT Guide provides that the requirements of the Guide constitute a Practice Direction issued by the President (then Roth J), pursuant to Rule 115(3) of the CAT Rules. This applies to the passages cited above (and others below).

The relevant CAT Rules are those applicable to all proceedings, such as Rule 4, and in addition the rules for specific kinds of case, in this case claims brought under s47A of the Competition Act 1998. Those specific rules are in Part 4 of the CAT Rules (see Rule 3(a) and (d)). The particular relevant provisions in Part 4 are Rules 53 to 55. There are similar rules for other types of claim in the CAT but nothing turns on that.

Rule 4 sets out the CAT’s governing principles, which are the same as the overriding objective in Part 1 of the CPR. Rule 4(7) expressly provides that parties, together with their representatives and any experts must “co-operate with the Tribunal to give effect to the principles in this rule”.

Rule 53 sets out a non-exhaustive list of case management directions which the Tribunal can give, on the request of a party or of its own initiative, to secure that the proceedings are dealt with justly and at proportionate cost. Included in the list at r53(2)(e) is a direction “for the appointment and instruction of experts, whether by the Tribunal or by the parties”. Rule 54 relates to directions at a case management conference or pre-hearing review. Rule 54(3)(g) provides that one of the purposes of the first case management conference is “to determine whether the parties should be given permission to adduce expert evidence and, if so, to what extent”. Rule 55 sets out various directions as to evidence which may be given including at r55(1)(d) “whether the parties are permitted to provide expert evidence”.

Case management is dealt with in Section 7 of the CAT Guide. At 7.51, speaking about evidence generally, the Guide explains that the Tribunal may control the evidence in particular cases in various ways including by giving permission to provide expert evidence.

Expert evidence is addressed in more depth from 7.65 to 7.70. Paragraph 7.65, as the Tribunal noted ([14]), states that the CAT will take into account the principles and procedures envisaged by Part 35 of the CPR, “notably that expert evidence should be restricted to that which is reasonably required to resolve the proceedings”. The paragraph also makes the point that it is for the party seeking to call expert evidence to satisfy the Tribunal that it is admissible and relevant to the issues and also that it “would be helpful to the Tribunal in reaching a conclusion on those issues.” Paragraph 7.66 sets out a variety of procedures which can be envisaged. Notably these expressly include the appointment of a single joint expert or of the Tribunal’s own expert. Finally paragraph 7.67 makes clear that “as under Part 35 of the CPR,” the expert has an overriding duty to help the Tribunal.

Looking overall (and bearing in mind the Guide has the status of a Practice Direction), it is clear that the CAT approaches the issues arising in this case on the same basis as the CPR, with the same three features: a duty to restrict expert evidence, recognition that experts have an overriding duty, and a power to direct evidence from a single joint expert.

Expert evidence is of a different kind from fact evidence and is subject to different, more restrictive, rules than other evidence. Given the potential for irrelevance, cost, and the complexity which can occur with poorly thought through expert evidence, it is not hard to see why permission is required in either venue. The reason is the need to retain a power to control and restrict this type of evidence, so as to ensure it is reasonably required to resolve the proceedings. That is why both the court and the CAT approach expert evidence from the same stance, namely that there is duty to restrict it (CPR r35.1 and CAT Guide 7.65).

Turning to the expert’s overriding duty, although today it is expressed in the CPR in r35.3, it was well established at common law many years before and well before it was codified in the rule: see for example National Justice Compania Naviera SA v Prudential Assurance Co Ltd (The “Ikarian Reefer”) [1993] 2 Lloyds Rep 68 (Cresswell J). That is just as applicable in the CAT as in court and so again, while it is obviously helpful for the matter to be spelled out in the CPR and in the CAT Guide, the duty would exist in any event.

Finally neither side submitted that the Tribunal did not have the power to order a single joint expert. While there is no direct analogue in the CAT Rules to CPR Rule 35.7, the CAT Rules do provide for directions for the appointment of an expert by the Tribunal (at r53(2)(e)). The CAT Guide at paragraph 7.66 expressly provides for a general power to direct a single joint expert and acknowledges that a Tribunal appointed expert would be an example.

Therefore, recognising that this is not disputed, in any event I would hold that the approach applicable in the CAT is based on the same foundations as the approach in court, including in particular as provided for in CPR Part 35.

The rules providing for a single joint expert formed part of the then new Civil Procedure Rules when they came into force in April 1999. Often the single joint expert is one instructed by both sides in a case but another kind, as in this case, is a single joint expert instructed by distinct groups of defendants. Without attempting an exhaustive survey, I will run through what appear to be the significant cases identified on this question since the CPR came into force, refer to the other relevant material and then seek to draw the threads together and identify the principles to be applied.

A year after the CPR came into force, in Oxley v Penwarden, Kennedy and Mantell LJJ allowed an appeal from a direction for a single joint expert. Mantell LJ held:

“8.

In my view, this was eminently a case where it was necessary for the parties to have the opportunity of investigating causation through an expert of their own choice and, further, to have the opportunity of calling that evidence before the court. It is inevitable in a case of this class that parties will find the greatest difficulty in agreeing on the appointment of a single expert. That burden would then be cast upon the court and would, in turn, lead to the judge selecting an expert, if there be more than one school of thought on this issue, from one particular school of thought and that would effectively decide an essential question in the case without the opportunity for challenge.”

In 2001 Lord Woolf LCJ addressed the matter in MP v Mid Kent Area Healthcare NHS Trust [2001] EWCA Civ 1703. He emphasised at [14] that the power of the court to direct that the evidence be given by a single joint expert was unrestricted, that the court has a wide discretion, and that that discretion has to be used in order to further the overriding objective in the rules. Notably at [28] Lord Woolf LCJ went on to echo a point made in his 1996 Final Report on Access to Justice, at Chapter 13 para 11, that there should be no more than one expert in any speciality unless it was necessary for some real purpose.

In 2003, in ES v Chesterfield and North Derbyshire Royal Hospital Trust [2003] EWCA Civ 1284, the Court of Appeal considered a different although related question. This was whether to permit the claimant to call more than one expert in the same discipline (obstetrics). The problem in that case was that the defendants would be calling three senior consultant obstetricians as witnesses, albeit two would be fact witnesses and one an independent expert. The Court was concerned that anyone watching the trial would be impressed by the fact that there was only one consultant obstetrician giving evidence for the claimant, while there would be three giving evidence for the defendant (Brooke LJ [24]) and decided the claimant should be permitted to call two such experts. Brooke LJ at [17] nevertheless summarised the “governing rule” in the CPR as one which “limits expert evidence to that which is reasonably required to resolve the proceedings in issue. What is reasonable in any particular context will inevitably be fact sensitive.”

The next case to mention is Yearsley v Mid Cheshire Hospitals [2016] EWHC 1841 QB (Whipple J). It is an example, at first instance, of a case management decision in which a single joint expert was refused and separate experts ordered.

Finally I refer to the relatively recent judgment of Cotter J in Muyepa v Ministry of Defence [2022] EWHC 2648 (KB) in which he said this at 288:

“There is also often, if not usually, a very marked aversion shown by those conducting higher value personal injury or clinical negligence claims to the use of single joint care experts, despite the fact that there is often no principled reason against such an instruction. In my view the common working assumption within these fields of litigation that it is axiomatically the case that each party will have a care expert is misplaced, helps perpetuate polarised expert opinions and often greatly increases the cost of litigation.”

The CAT Guide (with its status as a Practice Direction) has been mentioned already, but also relevant are the various court guides, which demonstrate that directions for single joint experts are a firmly embedded feature of civil justice in court today. I refer to [10.43]–[10.45] of the Kings Bench Guide April 2024, and, in the Business and Property Courts: to [9.21]-[9.24] of the Chancery Guide 2023, [H.2.6]-[H.2.7] of the Commercial Court Guide 2023, and section 13.4 of the Technology and Construction Court Guide 2022.

Finally, I refer to material concerned with dispute resolution pre-action. The CPR Pre-Action Practice Direction, paragraph 7, encourages the use of a single joint expert, particularly but not exclusively in low value claims. Also relevant are the two Pre-Action Protocols for Housing Condition/Disrepair Cases in England and in Wales, each of which encourage the appointment of a single joint expert, (paragraphs 7.1 and 7.2 respectively) in a situation in which there is a plain at least potential conflict in the interests of the two parties (tenant and landlord).

Pulling this together, a direction for a single joint expert to give evidence in place of separate experts from distinct parties in the proceedings, like any other direction giving permission for expert evidence, is governed by two primary dimensions. One is the overriding objective (in court), which is the same in all material respects as the governing principles (in the CAT). These are simply that the court or Tribunal will seek to ensure that the case is dealt with justly and at proportionate cost. The other is the duty to restrict expert evidence, in other words to limit it to that which is reasonably required to resolve the proceedings in issue.

Many, but not all, of the statements about single joint experts are made in the context of low value cases. There is no doubt that this approach has a particular value in such cases, when proportionality is important. Nevertheless the principles are the same irrespective of the value at stake.

In my judgment, while the existence of a conflict of interest between the relevant parties is a material factor to take into account, the existence of a conflict of interest between the relevant parties is no trump card. That can be seen simply from the fact that the rules encompass a situation in which single joint experts are appointed in place of separate experts from the claimant and the defendant. Such a direction is frequently made in the county court in a myriad of circumstances. I have mentioned housing disrepair already in a pre-action context. An example in court is valuation of the property in dispute. The claimant and defendant have a manifest conflict of interest, not just in the case overall but in relation to the very matter on which the single joint expert will express an opinion, and yet the rules clearly contemplate that such an order can be made. The two parties when instructing the expert are at liberty to put their own, distinct, views of the property’s value to that expert. However in the end the expert, whose overriding duty is to the court, will come to their own view of the value.

A suggestion by the respondent at one stage was that the principle was that if there was a conflict on the issue upon which the expert was to express an opinion, particularly if it was an issue of liability (as opposed to quantum), then the conflict did rule out a single joint expert; but that also cannot be right. There is no indication to that effect in the rules. As I have said already, when the court orders a single joint expert in a valuation case, the value of the property in question is often the most contentious and crucial issue in the case.

This is the point to consider the UK Trucks decision, relied on by the appellants. To do so it is necessary to first touch briefly on the earlier CAT decision of Royal Mail v DAF [2023] CAT 6 (Michael Green J, Sir Iain McMillan CBE FRSE DL and Derek Ridyard). That was a trial of follow on claims relating to a cartel involving truck manufacturers. The claimants Royal Mail and BT had bought trucks from those manufacturers. There were economic experts giving evidence using a regression model to address an alleged overcharge. The CAT’s judgment describes the model and the evidence at [475] to [480]. The passage as a whole repays reading, describing the difficulties and imperfections in this evidence, its inherent limitations, the problems of spurious accuracy and the need to use a “broad axe” in reaching a conclusion. The case explains that this sort of evidence is sensitive to the assumptions on which it is based and cannot produce a single “right” answer on the overcharge.

Turning to UK Trucks itself, it was an appeal from the CAT in collective proceedings brought as a follow on claim again relating to a truck cartel. The CAT made a collective proceedings order in which a single organisation, the Road Haulage Association (RHA), was made the class representative in preference to a different organisation. The issue on appeal (as explained by in the judgment of the Chancellor Sir Julian Flaux at [1], with whom Snowden and Green LJJ agreed) was whether a single class representative could represent a class in relation to a common issue in circumstances where there is an actual or potential conflict of interest between two groups of class members.

There were two classes, the buyers of new trucks and the buyers of used trucks. There was a conflict of interest between these classes because of the possible pass-on to the buyers of used trucks, from the buyers of new trucks, of the overcharge the new truck buyers had paid which was caused by the cartel. A smaller pass-on to buyers of used trucks would favour the interests of buyers of new trucks at the expense of the interests of the used truck buyers (and vice versa). The difficulty was that a single organisation, the RHA, was to represent both classes. The Court of Appeal held that this conflict, provided it was managed appropriately, did not prevent the RHA representing both classes. It would require the erection of a Chinese wall within the RHA organisation for the purposes of dealing with that issue. That would need to involve a separate team within the RHA acting for each of the two sub-classes, instructing different firms of solicitors and counsel and a different expert or experts. A different funder would need to be involved for one of those sub-classes, given that the conflict potentially extended to funding.

Like the present case and Royal Mail v DAF, the UK Trucks case would involve economic expert evidence with regression modelling of overcharge. In relation to this expert evidence, at [95] the Chancellor held that the CAT had been wrong to accept the suggestion that a single economics expert (Dr Davis) could be the expert for both classes represented by RHA. At [96] the Chancellor said this:

[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]. Since there is no single, objectively ascertainable, ‘right’ answer to the overcharge pass-on issue, and the decision of how to advance an argument on this issue in the proceedings will inevitably involve some strategic considerations, it cannot be sufficient for the divided loyalty which the RHA owes to the two groups of PCMs to be resolved by a vague promise that the RHA will decide how to act on the basis of advice from Dr Davis.

The appellants submit that this passage shows that that the existence of a conflict of interest between purchasers of new trucks and purchasers of used trucks necessitated the appointment of separate experts. However that is too compressed as a summary of paragraph [96]. It is clear that a fundamental problem identified in [96] was that the expert would be instructed by a single group (RHA) with divided loyalty. In other words the conflict of interest inside RHA was a key reason why a single expert instructed by that single organisation could not give evidence for both classes. The case is not authority for a proposition of the breadth submitted by the appellants.

The appellants also suggested, by reference to the subjectivity mentioned in [96], that the nature of the expert evidence in these regression cases supported the idea of separate experts in this sort of case. Reading Royal Mail v DAF and UK Trucks together, it is clear that what are referred to in [96] are the inherent limitations of this kind of regression analysis, its sensitivity to assumptions and the absence of a single “right” answer. So one expert might well propose a certain value for an input variable, whereas another individual expert might propose a different one. Assuming both are within the range of reasonable opinions, one cannot say simply that one is objectively right and the other is objectively wrong. This is the sense in which I believe the Chancellor was using the terms objective and subjective in [96]. It is a feature of a great deal of expert evidence, particularly in valuation cases, and it would not, on its own, provide a reason to have separate experts. However as I have explained already, in any event [96] does not reach a conclusion of that kind.

In the present case, in the decision under appeal at [19(2)], the Tribunal held that if there is a conflict of interest relevant to the matters the expert evidence is concerned with then it “will not ordinarily be appropriate to order joint experts”. I sympathise with the Tribunal, since that proposition as a minimum was not in issue, however having reviewed the matter and having had the benefit of more wide-ranging submissions in this court than occurred below, I do not believe it is correct. The stance of the court and the CAT is always that it has a duty to restrict expert evidence to that reasonably necessary to decide the case. It may well be appropriate in the interests of justice to order a single expert even if there is a conflict of interest between the instructing parties on the matter to which the expert evidence is directed. A deliberate caricature example given by my Lord, Arnold LJ, in argument makes the point. One could imagine a circumstance, which is not this case, in which the wishes of two antagonistic defendants to put forward their own experts was driven by a desire to obfuscate, by putting up two witnesses with different methodologies and approaches, their purpose being to confuse what might otherwise be a clear case against them, and to throw dust in the eye of the court. In such a case the court would be perfectly entitled to direct a single joint expert irrespective of the conflict between those parties. The example also illustrates that the court or Tribunal is right and entitled to look at the matter from the point of view of the judges who will try the case, and not simply from the point of view of the parties.

In summary therefore the principles applicable to single joint experts do not involve any gloss related to the existence of conflicts of interest. They amount to no more and no less than the conventional case management approach of applying the overriding objective (in court) or governing principles (in the CAT) in all the circumstances to deal with cases justly and at proportionate cost, conditioned only but importantly by the duty to restrict expert evidence to that reasonably required to resolve the dispute.

There are five further points to mention before turning to the specifics of this case.

First, in taking all the circumstances into account, there is a difference between having a single joint expert as the sole expert in the case, instructed by both claimant and defendant, and a case like this one, in which the single joint expert is on the defendants’ side. In the latter the court will still have a contest of expert evidence to resolve between that and an expert called by the claimant, which will be a relevant consideration.

Second, the respondent’s submission in the end was that if a single expert would effectively disable a party from prosecuting or defending its case, then separate experts would be required. I can see that in a case in which the circumstances really were that extreme, then there would be a strong case for a separate expert, but this situation will not be a common one. The fact a party requires expert evidence to advance its case does not justify separate experts. The disability referred to has to relate to the difference between separate experts and a single expert. The expert’s overriding duty to help the court means that experts are required to, and do, express views on matters which the party calling them would rather were put in a different way or not put at all. That is why the duty is an overriding one. It is not a justification for separate experts.

Third, at paragraph 35.7.1 of the White Book, the MP v Mid Kent Area Healthcare NHS Trust case is identified and the view is expressed that the power to order a single joint expert “will more usually” be capable of being exercised when it appears to the court that “the issue falls squarely within a substantially established area of knowledge and where it is not necessary for the court to sample a range of opinion or where the issue is uncontroversial.” Each of these three circumstances is capable of being relevant, but this is not and does not purport to be an exhaustive list. The list is also, I think, focussed on a single expert overall, and not on a case like this one. It is worth making clear that the utility of single joint expert evidence is not confined to uncontroversial matters. Moreover if an expert is not giving an opinion based on an established area of knowledge, one might question the basis for the admissibility of that opinion, but that is for another day.

Fourth, based on Oxley v Penwarden, the suggestion was made that if there were more than one school of thought, then separate experts may be permitted. I agree with this provided it is clear that the existence of multiple schools of thought is a quite specific state of affairs, very different from the kinds of differences of opinion one sees with expert evidence generally. The fact the experts here might use different regression models or disagree about the input variables does not demonstrate the presence of multiple schools of thought.

Fifth, proportionality, which is one aspect of the overriding objective and governing principles, is not the only consideration. Therefore, as the Tribunal found in this case, the fact that the value of the case means that the cost of separate evidence would not be disproportionate to what is at stake, does not on its own rule out a direction for a single joint expert. The just disposal of the case is a vital consideration.

Finally, a passage in Phipson at 33-60 on single joint experts was cited. One aspect referred to the schools of thought point identified in Oxley v Penwarden. The authors also state that:

“Indeed single joint experts are the norm in cases allocated to the small claims and fast tracks. In contrast in heavy and complex cases, the use of single joint experts is quite limited. ⁴⁹⁰”

As an observation of what is happening in practice I would agree with both sentences. The citation given for the second sentence (footnote 490) is to the Commercial Court Guide and as I read the relevant passage in the current version of that Guide (11^th Ed, rev 2023), it supports the observation read that way. If the statement in the textbook about heavy and complex cases was taken to mean that the scope for using single joint experts was necessarily more limited in those cases, then I would not accept it. Proportionality considerations will differ, and that will account for differences in practice, but as the present case shows, proportionality is not the only question.

The Tribunal’s reasons in favour of restricting the expert evidence of the defendants to a single joint expert on competition economics were compelling. The judges deciding the case would otherwise be faced with a multiplicity of economic models and sets of parameters. The point, as the Tribunal recognised, was that with one expert on the defendants’ side, the Tribunal would have a single set of disputes to resolve between the approaches of the pair of modelling experts (one from the claimant and one from the defendants). Importantly, adding a further expert does not simply increase the number of sets of disputes from one to two, it increases that number threefold. There would be three pairs of experts and three pairs of rival modelling approaches. The point is that the number of pairs of rival models increases disproportionately as the number of experts increases. This is even more apparent when considering what the Tribunal was originally faced with, with three defendant groups. In that case with separate experts, there would be six distinct sets of disputes (C v D1, C v D2, C v D3, D1 v D2, D1 v D3 and D2 v D3). Once this is appreciated it is obvious that the Tribunal’s concern was justified. In my judgment it also illustrates why the existence of conflicts of interest between defendants in this scenario cannot operate in such a way as to undermine the ability of the court or Tribunal to manage the just disposal of a case like this to address the problem.

This aspect of the reasoning of the Tribunal is not in dispute on this appeal. No doubt the appellants realistically recognised that an expert Tribunal’s view of matters of this kind was unlikely to be disturbed on appeal. The point on appeal on which permission was given was whether the Tribunal erred in law in holding there were no material conflicts.

It is worth noting for a start that the overall positions taken by the appellant groups in this case on the merits are aligned. They are both bound by the cartel findings of the Commission, but then they both deny anything beyond that. In particular, they deny there was any conduct directed at the respondents or any loss. As counsel for the respondents put it, they are running mutually consistent defences rather than conflicting defences.

The first alleged conflict relates to apportionment. The best the appellants can say is that in commercial terms, if the claim succeeds, there will very likely be a conflict between the two groups in due course as to their contributions. However, as the Tribunal recognised, at the moment this has no practical consequence. The case pleaded is an undifferentiated claim in which both groups are jointly and severally liable. The Tribunal will not have to address apportionment at trial.

A point was made that Mr Hughes’ expert evidence includes tables with sales identified as made by ZF TRW or by Autoliv which could be relevant to a future apportionment. That is true, but they are matters of fact, not opinions of the expert. It does not reveal a relevant conflict. Attribution of sales between the two groups is not something that the experts will address using their expertise. The Tribunal was right to reject this submission.

The other live point relates to the umbrella damages and to causation, drawing a distinction between the positions of ZF TRW and Autoliv in the Commission decisions. ZF TRW is not named in OSS1 and so is not identified as a party to cartels prior to the start date to which OSS2 is applicable, i.e. 4 January 2007. Therefore there is a nine month period before the start of the OSS2 cartels in which the only cartels identified are the six cartels of OSS1, and ZF TRW is not named in those.

However as the Tribunal recognised in its additional ruling at A[16] and A[17], while in theory a conflict could arise, in practice in this case there is no such conflict relating to the expert evidence. The overcharge approach and the expert evidence in support of it here is not capable of making distinctions relevant to this question. All it can do is identify the existence of an overcharge. There was specific evidence about this at the additional hearing, in a letter from the claimants’ expert Mr Hughes. No contradictory evidence was provided by the defendants. Mr Hughes’ letter explained that the overcharge cannot distinguish between individual cartels in any case, and that the nine-month period is too short for the data in this case to be sufficient to separate individual effects.

The defendants submitted that they could not contradict this sort of evidence without waiving privilege, but that is not correct. At case management hearings like this parties can, and do routinely, bring to the court letters and other evidence from experts about the nature of the evidence they may or may not be able to give, in order to assist the court or Tribunal in managing the proceedings before full reports are served.

In my judgment the Tribunal was correct to reject this point and to say, at A[17(3)], that the relevant expert evidence here is a blunt tool which is not capable of distinguishing those matters.

The first Tribunal decision addressed a third alleged basis for a conflict, relating to the open-ended way the pleaded case was put at that stage. However the amendments after Tokai Rika left the proceedings dealt with that issue and it no longer arises.

Overall, it follows that the Tribunal was right that there was no material conflict of interest between the defendant groups and made no error of law in reaching that conclusion. There was nothing further to take into account in considering the justice of this case and the compelling reasons the Tribunal had already identified for having a single joint expert from the defendant groups. I would therefore dismiss this appeal.

I agree.

I also agree.