Pan NOx Emissions Litigations, Re

This ruling follows a hearing which has been attended by the Claimants and the Ford, Renault and Peugeot Citroën Defendants in order to deal with one matter which there was insufficient time to consider at the five-day CMC which took place two weeks ago.

The central dispute relates to the selection process for samples and the preceding, but nevertheless interrelated, question of whether or not disclosure ought to be given in relation to firmware, so as to enable, in the Claimants' submissions, its selection of samples in the way it contends is best suited for the litigation that lies ahead.

The sub-issues are (1) what the basis of selection should be; (2) the number of samples and (3) the question of timing.

I have before me evidence presented from the Claimants and from Ford. I also read the skeleton arguments and the correspondence in so far as it has been referred to in those skeletons or in oral submissions before me.

There is, happily, agreement on the fact that the object of the exercise of selecting samples is to choose representative sample vehicles, which will knock out as many issues as possible, for as many claims as possible, at the PDD trial in October with 2025. There is not agreement between the Claimants and Defendants how best those samples should be identified. In essence the Claimants say that the selection should be by reference to and via consideration of firmware, which allows them to carry out selection by reference to the prevalence of devices within particular vehicles or models or variants of models. The Defendants say, led by Ford, that the selection should come from a number of vehicles identified by popularity within the cohort, effectively, without any reference to firmware. They also say that it is inappropriate for firmware to be disclosed at the moment, because, first, the sampling process should not require it at this stage and, second, as a result it would be unnecessary and, third, it is a fishing expedition by the Claimants in order to advance their claim on liability.

In terms of the approach to sampling, as the parties are well aware, this Court has made its view clear on a number of occasions that it is not going to engage in making decisions the effect of which is to require either the Claimants or indeed the Defendants to select vehicles, models or variants of models in a particular way. Judging the appropriateness of a particular basis of selection is for the Court to consider in due course when the substance of the technical issue problems with selection, which one party or the other says has been caused by the sample selection.

It may be, of course, that if a party to litigation, such as this, presented a wholly illogical or plainly unrepresentative sample, a court would intervene at the case management stage to prevent that procedure from going forward. By contrast, where the logic of the selection process advanced by either side is at lease comprehensible, the court will be slow to intervene. The Court will be unlikely, absent particularly good reasons - for example because it would be unworkable practically - to stop a party at the case management stage from proceeding to run its substantive case in the manner in which it thinks is most suitable.

This would seem to deal with the first and principled objection by Ford and Renault, which amount in a sense to requiring the court at this stage to prevent the claimant from selecting samples in the way it presently considers is best for it. Today Ms Nolten, on behalf of Ford, has advanced submissions in order to persuade the court that, irrespective of that issue of principle, I should nevertheless decline an order in relation to disclosure of firmware because it amounts to a fishing expedition in relation to, in effect, an unpleaded claim. The pleading against Ford, it is said, is broad and can be compared against the much more specific pleading made in the Mercedes GLO. The question, identified in broad terms in the pleading, is whether various forms of generically identified PDDs exist. Broad allegations are then made in respect of their existence across the cohort identified within the pleading. It is therefore said on that basis that the pleading was not support any requirement for the identification of firmware, which is a level of granularity that the pleading does not substantiate. It is said, therefore, that the disclosure of firmware is not necessary either at all or certainly at this stage of the overall litigation with the pleadings as they presently lie.

However, as has been made clear in a number of disclosure-related applications in the context of GLOs and in particular in relation to Dieselgate GLOs, there is an asymmetry of information and that is something that the court is entitled to take into account when considering disclosure at an early stage.

I also remind myself that the order made at the CMC, two weeks ago, was that the specific pleading that the Claimants are going to be required to provide is a pleading which (a) will be specific to PDDs and (b) is to follow disclosure. It is right, of course, that during Mercedes' application there was a greater focus on the question of heterogeneity and, indeed, it is right that the pleading against Mercedes is one that descends to greater particularity than the one against Ford. However, as a matter of approach, whilst it is right to say that it is an overstatement that early disclosure in group litigation is in any way ordinary, it always remains necessary for Claimants to justify why specific documents or classes of documents ought to be provided prior to the use of phasing. I also consider that, in general terms, the more onerous the consequences of being required to provide documents earlier than would ordinarily be the case, the more justification will be required. I also bear in mind Renault's evidence, in which it is said that disclosure is not necessary because they say the Claimants already have significant information about Renault by way of the investigations that have already been undertaken.

In my judgment I consider that disclosure of limited amounts of firmware at this stage is both necessary and proportionate.

On the information before the court today there is, in my judgment, a rational explanation from the Claimants as to how selection of samples should be based upon or take into account prevalence by way of firmware. That is the way in which samples have been chosen in the Mercedes litigation, and there will be a consistency in approach from the Claimants if they are permitted to continue their sample selection in the way they choose and have been permitted to do so in the Mercedes litigation which will be tried at the same time.

The court is in no position to gainsay the Claimants' technical submission as to the importance of sample selection, by way of prevalence in the way it puts forwards. It could not possibly do so today fairly and reliably without, in effect, conducting a mini technical trial. I am satisfied that the Claimants are not embarking on a fishing expedition in the ocean of vehicles and variants. If it is fishing, it is fishing in a very small pond and whether it produces representative fish in due course is for another day.

On the basis of the sampling process the Claimants wish to advance, it is necessary, in order that disclosure is provided of limited amounts of firmware information now to assist the claimant with a preferred sample selection, that is to happen at a point at which it dovetails into the existing directions. It is of particular importance, in any judgment, that there is nothing in the Claimants' order which prevents any of the Defendants from selecting, in the way it sees fit, those samples vehicles or variants that it wishes to have investigated at trial.

Mr Auld submits that if the sample turns out to be wholly unrepresentative, it could have enormous cost implications. This is obviously true. It is not however a risk that this court can we resolve one way or the other fairly at this point. It may, of course, be an argument for another day if in due course a selection identified by the Claimants is regarded as entirely self-serving and does little to advance the overall GLO litigation. That would be a matter, no doubt, the parties would consider in their submissions as to costs whatever the outcome of the litigation may be.

What about the number of firmware strings? The Claimants seek disclosure relating to 150 variants, by reference to a graph within their skeleton argument that covers about 80 per cent of the Ford cohort. If one looks at the graph, it is clear that that 100 variants provides 70 per cent coverage and 50 variants is 51 per cent. It seems to me, looking at the numbers, that if 100 vehicles provides around 70 per cent, and a third greater effort is required to provide 80 per cent (i.e. just 10% more), 100 vehicles is the right nswer. This, still provides coverage of 70 per cent of the cohort is the most proportionate outcome.

Based on the evidence I have about the number of variants likely to exist in the other defendant bodies, limited though it clearly is by comparison to the analysis on Ford, I consider this number is an appropriate number to be applicable in relation to each of the defendant ALGLOs.

In terms of timing, Ford have indicated in their witness evidence that 21 days would be necessary in order to carry out the provision of 150. I see no need to tinker with the suggested timetable, which provides them with three weeks even though the number has been reduced.

In relation to the other submissions before me in relation to timing, Mr Auld submitted that they may not be able to provide disclosure of firmware for 150 vehicles for some months. Ms Gibaud also suggested that it could take, on instructions, up to three months for as few as 30 variants. It seems to me that the Nissan position can, without undue prejudice and trying to balance the parties' competing positions, be aligned with the Renault position and that will give Nissan some additional time.

The Renault Defendants are not required to provide information until a date which is fixed by reference to the FBS hearing and that, therefore, will provide Nissan and Renault with the sort of periods that it seems to me are likely to be realistic, if not overgenerous if the experience of Mercedes is anything to go by and the evidence from Ford is anything to go by.

The Peugeot Citroën Defendants are not yet required in the draft order to provide the firmware information until 12 July. That is some time away. I appreciate that that is a relatively short period of time after 21 June, which is the date, because of the FBS issues, that they are going to be provided with the list of identified variants.

It seems to me that given that time is a little way away, the relatively recent introduction of Peugeot Citroën into the litigation and the limited amount of time that they had getting up to speed for this hearing, that the most appropriate order is to leave it out to 12 July. Further investigations can be undertaken as necessary by Peugeot Citroën in the teeth of this Order and they have liberty to apply, no doubt with appropriate evidence, if there is good reason why they consider, notwithstanding potentially by then the evidence from other manufactures as well, as to the ease or otherwise by which this information can be provided whether that date ought to be adjusted.

Turning then, finally, to the number of samples that should be considered ultimately, at least for the purposes of the selection of samples, irrespective of whether ultimately they are considered at trial, each of the defendants say that the 16 samples that is suggested by the claimants is disproportionate and too high. The claimant says that in the first instance 16 samples for each GLO may, in fact, not turn into 16 being investigated at trial because there may be essentially categorisation of themes that emerge. That submission is undoubtedly true, there may be such a reduction. But on the other hand it is a submission that reflects, it seems to me, an acceptance of the reality this the court would not be able to deal with 64 individual engines if that is ultimately the number that ended up being considered at the trial because there had been no categorisation or reduction.

However, balanced against that, the Claimants do make the important point that one of the reasons for the ALGLOs is to provide effectively a security blanket against settlement. This is in my mind an important factor and must go into the number of factors that need to be considered in selecting the number of samples. If Mercedes, for example, were to settle, it is right that there would remain a suitably large number of samples that can be considered within the remaining cohort. I also balance the fact that the number of Claimants within the ALGLOs are fewer by comparison to Mercedes. This goes, I accept, to the question of proportionality, but also I bear in mind that this is still a large number of Claimants in absolute terms, and the difference between either 4, 6 or 16 samples in the context of the overall number of vehicles to be considered within each GLO and the number of variants, is still small.

The best solution, it seems to me, balancing these factors, and in particular (a) the need for a security blanket; but (b) the lack of sense in identifying at this stage a number that it is clear could not be coped with without reduction in due course, is for 12 samples to be collected, six from each claimant and each defendant or defendant pair. However, the order should say "up to 12". The number of 16 came from agreement by Mercedes and Claimants in the context of the technical information that they by that stage had.

The order ought to encourage the parties to discuss the appropriate number of samples which will be up to 12. It may be that the parties, given the technical spread that they will have a better understanding of, will decide that 10 is the right number or 8 is the right number, against the general indication from the court that anything approaching 64 or even in this case 50, is going to be quite difficult to achieve in a 10-week hearing in due course.