[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales High Court (Queen's Bench Division) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Queen's Bench Division) Decisions >> Crossley & Ors v Volkswagen Aktiengesellschaft & Ors [2019] EWHC 698 (QB) (06 March 2019) URL: http://www.bailii.org/ew/cases/EWHC/QB/2019/698.html Cite as: [2019] EWHC 698 (QB)

[New search] [Printable PDF version] [Help]

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

		Royal Courts of Justice, Rolls Building Fetter Lane, London, EC4A 1NL
		6th March 2019

MR JUSTICE WAKSMAN

1. This is the first substantive CMC in this GLO case, where the claimants are all VW owners or ex-owners or, in some cases, owners of Audi, Seat or Skoda cars. The defendants are VW manufacturers, dealers or finance companies. The common theme is that all the claimants have or have had VWs or the other cars with a common engine, which is a diesel type EA189. At the moment there are about 117,000 claimants on the Group Register.
2. On 18 September 2015 the US Environmental Protection Agency issued a Notice of Violation of the Federal Clean Air Act on the basis that VW had dishonestly cheated the US emissions testing regime.
3. As set out in paragraphs 10 and 11 of the Particulars of Claim, VW then acknowledged the deceit and the existence and operation of a particular device or software in a series of public statements and in evidence given on 8 October 2015 by Michael Horn, the President and CEO of Volkswagen Group in America, to the US House Committee on Energy and Commerce Subcommittee on Oversight and Investigations.
4. Following developments in the United States -- and this much is common ground -- it was established that there are approximately 1.2 million affected vehicles in this jurisdiction.
5. On 1 October 2015 VW Group UK, acting on behalf of VW, Audi, Skoda and Seat, sent four formal notifications to the UK's Driver and Vehicle Standard Agency, admitting that, in respect of all affected vehicles, the NOx (ie nitrous oxide or dioxide) emission levels of the vehicles did not meet relevant emissions limits and did not meet the regulatory requirements for registration of the vehicles. The VW defendants now say that these were incorrect admissions.
6. A central feature of this case is the use by Volkswagen of software within the engine's Engine Control Unit ('ECU') which has been referred to as cycle recognition software.
7. Put broadly, it works as follows. It programmes the engine so that if it detects parameters showing that the car was being operated in a testing environment, as opposed to being driven on the road in the usual way, it would cause the engine -- using the term widely -- to produce sufficiently low levels of NOx emissions to satisfy the tests for emissions. When not in the test mode, the engine would operate so as to produce an increased level of NOx.
8. All of that is effectively admitted in paragraphs 8B and C of the Generic Defence. That this was not disclosed to any relevant authority until September 2015 is also admitted in paragraph 37C of the Generic Defence.
9. The allegation of dishonesty or deceit underlying this claim is that, in broad terms, that same ECU software, operated in the same way here as in the US -- unsurprisingly, since it is the same engine type -- and in so doing cheated the testing system which is relevant here, which is based on a European standard, NEDC, the New European Driving Cycle.
10. The claimants say that the software used by the defendants constituted a defeat device within the meaning of article 3.10 of the EU Emissions Regulations 715/2007.
11. That provides as follows, that a defeat device is:

"any element of design which senses temperature, vehicle speed, engine speed (RPM), transmission gear, manifold vacuum or any other parameter for the purpose of activating, modulating, delaying or deactivating the operation of any part of the emission control system that reduces the effectiveness of the emission control system under conditions which may reasonably be expected to be encountered in normal vehicle operation and use".

12. Thus far, and prior to the selection of any test cases to be pleaded out hereafter, there have been generic statements of case in the forms of Particulars of Claim, a Defence and a Reply.
13. The claimants make five different types of claim against some or all of the defendants.
14. First, a claim for breach of statutory duty, the statutory duty being the regulation I have just referred to, which the claimants say confers private rights of action on individual claimants. In this regard the claimants also rely so as to prove breach, a decision binding on this court or at least a decision of great weight, that being the decision of the KBA, which is the German Federal motor transport authority. As pleaded, that report included the details of investigations by the KBA; and that determination said that this was a defeat device within the relevant regulation. It said that on 23 September 2015, really concomitant with events in the US, VW admitted that prohibited defeat devices were to be found in certain diesel type approved EU, which was recorded at the KBA report, though that now is also in issue.
15. The KBA made a decision on 15 October that affected vehicles which had been granted approval certificates included the defeat devices. It referred to it as a system of illegal manipulation. It then exercised its powers under article 30 of the Framework Directive to require VW to recall the relevant models and remove the defeat devices so as to ensure such vehicles conformed with the relevant type approval, which they did not prior to the removal of the defeat devices. I infer that it did not when the engine was running in something other than test mode.
16. It was in these circumstances that what the claimants have referred to as a "fix", but which the defendants have referred to as "technical measures", were instigated free of charge for all relevant VW owners. It is claimed in paragraph 23.7 that the KBA was acting as the relevant approval authority and its decisions bind regulatory authorities of the other Member States. Following on from that, it is alleged that such a decision would be binding on courts, including this court.
17. Secondly, a claim in the tort of deceit. This is based on certain alleged representations made to all potential purchasers of the relevant VWs on the basis of public statements and advertising to the effect, expressly or impliedly, that the cars conformed to all UK and EU standards with regard to emissions; that there had been no dishonesty used in passing the relevant tests; that there had been no defeat devices used; and that all the cars had relevant EU approvals. The claimants say that those representations were false and knowingly so.
18. Thirdly, breach of contract. As against the VW dealers who sold the cars to individual claimants here, and VW Financial Services, who supplied them on finance terms, breach of contract, this is on the basis that they were not of satisfactory quality either at the outset or because, once VW had the relevant work to remove the offending software, the cars then underperformed in various ways.
19. Fourthly, and as against all defendants, where cars had been purchased by the claimants as consumers after the introduction in 2014 of the relevant part of the Consumer Protection from Unfair Trading Regulations 2008, a claim that that the regulation was violated by the installation and operation of the relevant software. About 23.5% of the current claimants fall into that class.
20. Finally, as against VW Financial Services only, a claim that, by reason of installation of the software, there was an unfair relationship between the claimant as a consumer and the defendant as a provider of finances within the meaning of section 140A of the Consumer Credit Act 1974 as amended.
21. The defendants resist all of these claims. In particular, for present purposes, they deny that the software here amounted to a defeat device within the meaning of article 3.10. They also deny that the KBA decision to the contrary is of any binding or other real force. They further deny that if there was a breach of the regulation it did not give rise to individual claims for breach of statutory duty in favour of each individual claimant.
22. I now have to decide, as the first matter in this CMC, the broad shape of this litigation going forward. The first question here is whether there should be a trial of preliminary issues in the near future, as the claimants contend, or rather, as the defendants contend, that there should simply be a selection of lead cases which will form the core of a number of test claims to be decided at trial, in due course, in the usual way, when all questions of liability and quantum will be decided.
23. I should add the defendants do not exclude the possibility of one or more preliminary issues arising appropriately down the line, but they say that it is premature to do so now. I approach this question without assuming that either course is prima facie the correct one where a GLO is involved. That would be far too simplistic.
24. The court already has the power to order the trial of one or more issues under CPR 3.1(2); but in CPR 19, governing GLOs, there are the additional powers to order the trial of certain lead or test cases which, by definition, usually only arise where there is a multiplicity of claimants and therefore where there is a GLO.
25. That said, paragraph 15.1 of Practice Direction 19B adds that the management court may give direction for the trial of common issues or for the trial of individual issues. The rubric to that then refers back to CPR 3.1 and states that 15.1 makes it clear if necessary that the court can give such direction in relation to issues arising not in a single claim but in several claims. It refers here to GLO issues as being "common or related issues of fact or law arising in several claims".
26. There is, in fact, no reason in principle, and in an appropriate case, not to order trial of a preliminary issue at an early stage, as well as providing for the ultimate trial of selected lead cases, although, of course, in that event, the relationship between the preliminary issues and the ultimate trial needs to be considered carefully, just as it would be in a non-GLO situation with only one claimant and only one unitary trial.
27. Indeed, the claimants do not now baulk at the notion that, if preliminary issues were to be ordered, work could still progress and should progress at the same time towards the ultimate trial, but most proximately the selection of the relevant test cases.
28. The three issues which the claimants say should be tried as preliminary issues need to be set out in full.

(1) Issue 1: Is the High Court of England and Wales bound (having regard to the terms and operation of the EC Type Approval legislation and pursuant to its duty of sincere cooperation) by the finding of the competent EU type approval authority (the KBA, in this case) that a vehicle contains a defeat device in circumstances where that finding could have been, but has not been, appealed by the manufacturer; and/or is it an abuse of process for the Defendants to seek collaterally to attack the KBA's reasoning or conclusions by denying that the affected vehicles contain defeat devices?

(2) Issue 2: Where a vehicle's engine control unit is capable of identifying the New European Driving Cycle test and operates in a different mode during the test by altering the rate of exhaust gas recirculation to reduce NOx emissions, does the vehicle contain a "defeat device" within the meaning of Article 3(10) of Regulation 715/2007/EC?

(3) Issue 3: Does the full effectiveness of the prohibition against defeat devices in Article 5(2) of Regulation 715/2007/EC require Member States to ensure the availability under domestic law of civil proceedings for damages than can be brought against the manufacturers of vehicles fitted with defeat devices, by the purchasers of such vehicles upon proof that such breach has caused them loss?

29. I should add that, in the course of argument yesterday, Mr de la Mare QC for the claimants, suggested amendments to the scope of issues 2 and 3, in order to deal with points made by the defendants that there were yet further issues related to but not quite the same as issues 2 and 3 that would be going to trial anyway. I refer to these supplemental suggested issues as 2A and 3A and will deal with them later.
30. So far as the law is concerned, I bear in mind the obvious starting point which, of course, is the overriding objective, of which I have been reminded by Mr de la Mare QC. In particular, within the context of trying cases justly and at proportionate costs, there are the objectives of ensuring the parties are on an equal footing, saving expense, dealing with the case in ways which are proportionate, having regard to the importance of the case, the money involved, the complexity, financial position of each party, ensuring they are dealt with expeditiously and fairly and allotting an appropriate share of the court's resources.
31. So far as the power of the court in relation to ordering issues and lead cases, I have already described those.
32. As for preliminary issues, a well known starting point is the list of nine factors which the court should take into account when coming to the ultimate decision which is whether it is just to order the trial of preliminary issues, as explained by Neuberger J (as the then was) in Steele v Steele [2001] CP Rep. 106:

(1) whether the determination would dispose of the case or at least one aspect of the case;

(2) would the determination significantly cut down cost and time involved in pre-trial preparation or in connection with trial itself?

(3) if it is an issue of law, how much effort will be involved in identifying the relevant facts?

(4) if it is an issue of law, to what extent it is to be determined on agreed facts? The more the facts are in dispute, the greater the risk the law cannot safely be determined until those facts are resolved.

(5) where the facts are not agreed the court should ask to what extent that impinges on the value of a preliminary issue.

(6) whether the determination of a preliminary issue may unreasonably fetter either or both parties or the court.

(7) to what extent is there a risk of the preliminary issue determination increasing costs or delaying the trial?

(8) to what extent the determination of preliminary issue may be irrelevant.

(9) to what extent there is a risk determination of preliminary issue could lead to application for amendments.

(10) then the ultimate question: having taken into account all the previous points, which overlap to some extent, is it just to order a preliminary issue?

33. Some cases will not be suitable for preliminary issues, especially where the scope of a relevant factual dispute which relates to that issue is not clear or where there are no agreed facts or where there are hypothetical or assumed facts which may, in time, prove not to be the only ones. That uncertainty poses real problems for preliminary issues: see, for example, the cases of Tilling v Whiteman [1980] AC 1, McLoughlin v Grovers [2001] EWCA Civ 1743 and indeed Steele itself.
34. As for preliminary issues in the context of a GLO, they are hardly unknown. They tend to concentrate on points of law of a wide variety but which can include limitation. In that regard see the cases of Deep Vein Thrombosis Litigation [2002] EWHC 2825, Carey v HSBC [2009] EWHC 3417,Poole v HM Treasury [2006] EWHC 2731 (the Lloyds Names litigation) and Dobson v Thames Utilities [2007] EWHC 2021 (TCC) and Bodo Community v Shell [2014] EWHC 958 referred to in paragraph 31 of the claimants' skeleton argument.
35. Indeed, as we will see, there is an issue between the parties here as to whether issues 1 to 3, or any of them, are truly issues of law or involve factual or possibly expert questions which may be disputed and which render them inappropriate for trial as preliminary issues. At the end of the day, as the case-law makes clear, the question whether or not to order preliminary issues, because it is so multi-faceted, is highly case sensitive.
36. That said, although test cases have not yet been selected here, although the procedure for selecting them is largely agreed, I see nothing wrong, in principle, in ordering preliminary issues based on the existing generic pleadings, especially if the statements of case relating to any selected case thereafter will add nothing to the debate already pleaded. Equally, if a preliminary issue was otherwise suitable the fact that it is ordered now, at the first CMC, is not itself objectionable. Indeed, if it is appropriate, it can be said that the sooner the better.
37. Let me now turn to the issues in more detail. The defendants' defence to the claim that the software was a defeat device is as follows:

(1) The way in which NOx emissions are reduced when the software operates is through the recirculation of exhaust gases within the engine itself. So that when they leave the engine but before entering other emission controls in the car and, in particular, the exhaust system, the NOx has already been reduced, and conversely if the mode is changed.

(2) However, the "thing" which must be modified by the defeat device in order to render it such under article 3.10 is the Emission Control System ("ECS"); this is something separate from and technically posterior to what goes on in the engine, ie it is essentially the exhaust system;

(3) Since the software here does not alter that but only what goes on in the engine, so, it is argued, the regulation is not engaged. The claimants' response to that is to say that this is an artificial divide; that the regulation must be constructed functionally, so that, for its purposes, the ECR includes anything, wherever located, whether in the combustion chamber of the engine or the exhaust or otherwise, which has the purpose or effect of altering or reducing the NOx emissions. By way of example, the claimants rely upon an expansive definition of the expression "system" to be found in the EU Framework Directive which relates to the regulation.

(4) The second of the defendants' two principal lines of defence itself consists of two parts:

(a) First, they say it has to be shown that the effect of the ECS is reduced under normal vehicle operation and use. However, they say, the claimants' comparator is simply between mode 1, which operates in test conditions, and mode 2, which operates in non-test, in normal road conditions. That is not the right comparator; the defendants say the true comparison is between mode 1 in normal road conditions and mode 2 in normal road conditions; and then to show a reduction in emissions or an increase in emissions as between mode 1 and 2 in that environment. And that exercise has not been done;

(b) Further or alternatively, the defendants say that, even if the correct comparator has been used here, the claimant does not and cannot show that the effectiveness of the Emission Control System as a whole has been reduced. Put very simplistically, if that results in a higher NOx emission than in the test conditions or mode 1 conditions there will be corresponding reductions of other relevant emission elements, for example diesel particulates, so that it cannot be shown overall that the ultimate collection of emissions, or adverse emissions if I can call them that, are any the worse. I refer to this argument as the holistic argument.

(5) The claimant rejects both strands of this second defence, largely on what it says is the proper interpretation of the regulation.

38. The defendants' defences, in the way that I have just described them, are set out briefly, but clearly and succinctly, in paragraphs 19 to 23C of the Generic Defence.
39. It is important to note here that the claimants pin their colours very firmly to the mast of interpretation. They say the true comparator for the purpose of 3.10 is what has been established, that is the emission level in mode 1 under test conditions and the higher level of NOx under normal use, using mode 2. It says that it needs no evidence on this point because the defendants have effectively already admitted this or it is obvious. It does not intend, if it is wrong on its interpretation, to adduce evidence to show different results on the basis of mode 1 and mode 2 both operating on normal road conditions. Nor would it adduce evidence in the alternative that on a holistic basis there is still an overall reduction in the efficiency of the emission control system. In short, for the claimants' case a change in NOx will do.
40. It is common ground that issue 2, as with issues 1 and 3, will have to be decided at some point or other. The only question is at what stage. They can never be rendered academic by the determination of facts in a particular way: compare the case of Tilling. Indeed, issue 2 has already been tried as one of a number of preliminary issues last year in an equivalent class action against VW in Australia. Judgment is awaited and it is not certain when it will be delivered. But part of the Australian emission regime imports the same provision as article 3.10, as is plain from the Australian papers before me, and it is not suggested otherwise. For a very clear and comprehensive exposition of how VW argued the defeat device point there, see section 5 of its lengthy closing submissions at paragraphs 108 to 210. The introductory paragraph at 109 very closely approximates the way that it has been put in paragraphs 19 to 23C of the Generic Defence here; and that is hardly surprising.
41. It is material to note, first, that the defendants in Australia did not call any expert evidence on this point. The claimants called some, but it was limited. But, for the most part, both sides relied on an agreed technical document which has been called here the Australian Technical Document ("ATD"). This explains, very clearly, how Volkswagen's engine gas recirculation system works, how it operates within the combustion chamber of the engine and not further down the line in the exhaust system.
42. I have read large parts of that document. At the risk of, perhaps, overextending myself, it does not appear to me to be rocket science; and I believe I understand the distinction made by VW as to the location of the emission reduction, whether or not that is correct for the purposes of article 3.10.
43. What is much more difficult to understand is why VW seem now reluctant to agree the ATD document for the purpose of this litigation. It does not appear to be suggested it is wrong in any way; and that would be surprising if it was. Mr Gibson QC says that a proper resolution of issue 2 will require expert evidence to deal with what he calls the "topography" of the engine. I do not see why. It was not in Australia; and the point is clear from the ATD. In my judgment, the debate over the application of article 3.10 in the Australia case is highly informative as to how it would play out here whenever it is tried.
44. Mr Gibson QC says I should be careful about drawing too much on the Australian case because the lawyers there may have played it differently from here for other reasons. I do not understand that submission, put in such broad terms; and I do not give it any real weight. There is no reason to suppose a class action and preliminary issues tried within it is in any less important to VW than here, albeit there is an EU dimension. This is all large scale litigation, but it is essentially fought on the same basis.
45. It may be VW's prerogative now not to agree the ATD - although of course it could be the subject of a Notice to Admit Facts hereafter - but I do not see why its reluctance now to agree a document that it had previously agreed, entitles it to suggest that the resolution of the 3.10 or issue 2 here is more complicated than it really is; especially given the claimants' reliance on its points of interpretation without any alternative evidential case.
46. In my judgment, whatever else can be said about issue 2, it is or is essentially a pure question of law. No doubt a certain amount of EU material may be referred to in order to inform the interpretative debate, but a question of law it remains .
47. I should stress the fact that this case is unusual in the sense, as already noted, that issue 2 has already been argued out at length. It is difficult to see how VW now could be any better prepared to deal with that issue than it already is.
48. Let me then turn to issue 1. This is really an adjunct to issue 2, because if it is correct it either decides issue 2 or it very much helps it along the way. It is hard to see why, as a matter of English law, the KBA decision could not be of at least some assistance to the English court on issue 2 in any event. But, as to its binding or presumptive effect, the claimants say, again, it is really a matter of law. Was it the relevant approval authority? (Although admittedly not so for Skoda and Seat). And, if so, given its actions, what is the effect of that legal decision on a court here?
49. Again, the defendants say it is more complicated than that. There may need to be evidence as to why VW did not appeal the KBA decision. One suggestion floated yesterday was that there may not have been a route of appeal available under German law for some reason; although I have to say I find that rather unlikely.
50. But, in any event, the lack of an appeal is really only relevant to the question of the fact and the permanence of the KBA decision. Since there has been no appeal, it is not going to change and it is not going to be reversed at some later date. Why VW decided not to appeal is actually neither here nor there. It is possible, I suppose, that VW's dealings with the KBA prior to the decision are relevant; although I suspect not much. But, if so, they will all be documented and I am told that VW has now agreed to give disclosure of its dealings with the KBA and other authorities -- the VCA and the Spanish authorities -- so that takes care of that.
51. It is also said that there would need to be extra factual evidence on the question of whether the challenge to KBA decision here indirectly or re-litigating here is an abuse of process under English law. I do not see why. It seems to me that is likely to be an objective analysis on facts which are not likely to be disputed.
52. Accordingly, and again, I consider issue 1 is essentially a point of law or a point of law and a relatively confined analysis of documents as to the nature and the true extent of the decision of the KBA outside of the particular ordinance which it issued to VW, essentially to remove the offending software.
53. Before turning to issue 3, let me turn to issue 2A. This arises as follows: paragraph 24, as presently drafted, of the Generic Particulars of Claim says that even if the devices were not prohibited defeat devices within the meaning of the regulation, their purpose and effect were the same as those of defeat devices as so defined, namely to cause NOx emissions in testing to be reduced to levels which would not be replicated in normal road use; to conceal their use and operation from regulators and others; and to cheat the emissions testing regime. The follow-up to paragraph 24 is in the deceit claim at 63.4 on the question of representations. It says there was a representation that there were not devices incorporated preventing proper and accurate testing; and, 63.5, the vehicles did not require modification to meet emissions standards. Those particular allegations do not depend, as such, in showing that there is a defeat device as properly so-called.
54. Then, on falsity, 66.5 says the representations were false because the defeat device, whether or not covered by the regulations - in other words using the term more broadly - prevented the proper and accurate testing and recording of emissions.
55. And 66.7 says the affected vehicles would not have satisfied EU emissions testing if operating in mode 1 which is mode 2 for the purposes that we have been referred to (normal road use), which apparently was admitted in Australia.
56. On the face of that series of allegations, it is not actually suggestive of another allegation of breach of regulations other than 3.10. This is one element of what Mr de la Mare QC called his cheat device claim. In other words, an alternative to a claim based on the notion that the device was a defeat device within the precise terms of 3.10.
57. If that is all there is, there is probably not much of a dispute, save on the question of representations. I say that because, if one looks at paragraph 24 of the defence in answer to paragraph 24 of the particulars of claim, it deals with it in a somewhat oblique manner. It says:

"In so far as if the unidentified devices were not defeat devices, their purpose and effect cannot be the same as a defeat device within the meaning of the article. If a party wishes to seek to install a defeat device that intention does not make a device a defeat device. It can only be a defeat device" -- and this is important -- "if it is a defeat device within the meaning of 5.2 and 3.10."

58. This is the beginning of a theme which one sees running through the defence, which is an underlying contention made by the defendants, right or otherwise, that if it does not fall into the category of a defeat device within the article, then, whatever else you call it, whatever else it does is simply irrelevant.
59. I understand in some way why paragraph 24 was pleaded as it is, because, in paragraph 24 of the particulars of claim it says that the purpose and effect were the same as defeat devices as so defined.
60. If you took those words out the allegation is the purpose and effect of whatever they were was to cause NOx emissions to be reduced to levels which would not be replicated in normal road use, concealing them. Despite asking what the defendants' position on that allegation was, I am afraid to say that I received no satisfactory answer. It seems to me it would be quite hard to see what the answer is, given the admissions that have already been made, but that, perhaps, is for another day.
61. So far as 66.5 and 66.7 are concerned, they are dealt with in paragraphs 74 and 75 of the defence, which, again, make the point that this is not a defeat device within article 3.10. Though they also say, for example, the vehicles did not require modification to comply with the Emissions Regulations.
62. But on this first limb of the alternative cheat device case, as presaged by paragraph 24, I think it should be left alone at this stage. In my judgment, it does not negate the significance of dealing with the defeat device point as a preliminary issue if that were otherwise appropriate.
63. As with the other element of the alternative cheat device claim, this really only emerged again in the course of argument yesterday. I do not think it is appropriate simply to add in new issues on the hoof like that, however easy, according to Mr de la Mare, it would be to "fold them in".
64. The same applies to the other part of the alternative cheat device claim, which, in fact, only became clear to me late in yesterday's argument. Here we have to go to paragraph 71 of the particulars of claim. That alleges that there were breaches under article 4.1, 4.2, 5.1, 5.2 of the Emissions Regulations by, 71.1, manufacturing them with a defeat device; but then, point 2, failing to ensure they were designed, constructed and equipped so they would, in normal use, comply with the emissions limits; failing to see they were properly tested; failing to see that they had adequately and lawfully demonstrated that they met the emissions limits before being put on the market.
65. I should interpose that the round up paragraph is 72, which says that these are actionable private law individual breaches under 72.1. 72.2 says that there is a sui generis tort, but Mr de la Mare QC made it clear yesterday that that was not intended to add anything.
66. The scope of the arguments here are much less clear to me at this stage than the defeat device point, no doubt because the latter had been very clearly signposted as a preliminary issue before yesterday.
67. I appreciate that Mr de la Mare QC wanted to add them because he says (a) they are actually quite straightforward; and (b) it can then be said to be another strand of his case that could be closed off, or, alternatively, definitively established, after the trial of preliminary issues. But I am not persuaded that this justifies the addition of any part of an alternative cheat device claim at this stage and in this way to the preliminary issues.
68. That should be seen in the context of the fact that Mr de la Mare QC, nonetheless, made it plain that the principal argument here remains that the software is indeed a defeat device within article 3.10. So I am not going to countenance 2A as a preliminary issue.
69. Let me then turn to issue 3. It is common ground that there is no yet reported decision as to whether a breach of 3.10 gives rise not just to obligations on manufacturers, which are then to be enforced by the competent EU authorities in Member States, but additionally confer an individual right to sue for compensation on the individual claimants.
70. The test for whether such an individual right does exist, however, it is well known, it is all about whether making the relevant regulations fully effective require an individual right to sue as well: see how the Advocate General put it in paragraphs 24 to 26 of his opinion in Munoz v Frumar [2003] Ch 328. That was a case about a regulation dealing with food standards and descriptions as they applied to grapes. The nature of the inquiry by the Court of Justice was at a relatively high level: see paragraphs 30 to 32 of its judgment.
71. I accept that the inquiry here on our issue 3 will involve a factual inquiry, although, to judge from Munoz, it may be quite limited and at quite a high level. The defendants say otherwise. Mr Kennelly QC took me to United Road Transport Union v Secretary of State for Transport [2014] 1 CMLR 25. This involved a question whether the Road Transport Regulation or the Road Transport Working Time Directive conferred individual rights on workers where broken. The first instance decision was given by Mr Justice Hickinbottom and is repeated in the judgment of the Court of Appeal. I just need to refer to two parts of it.
72. In his judgment as to why there was not any such right, first of all at (vii), he said:

"Given the obligations imposed on mobile road transport workers for breaks and rest periods, it is difficult to envisage circumstances in which a worker would have a civil claim against his employer, other than where he himself would be guilty of an offence of infringement. That is materially different from the scheme for general workers. It is unsurprising the relevant regulations do not envisage a right of claim that could only be exercisable, in practice, by an employee on the basis of their own criminal act."

73. In (viii) he says:

"These are matters of law. In respect of effectiveness, there has been no evidence that the system of enforcement relating to breaks and rest periods are not effective...".

74. Insofar as the defendants rely on this case to show that issue 3 is likely to be resolved in their favour, first, the United Road Transport case is on very different facts, as paragraph (vii), in particular, shows. But, more importantly, just because the defendants think they will win is not a reason why it is not suitable as a preliminary issue.
75. If one looks at the reasoning of Mr Justice Hickinbottom, in overall terms approved by the Court of Appeal, again it was at quite high level, and most seems to have been drawn from documents or agreed facts. It is hard to see how it would be very different here. After all, in this particular case, it is very much a matter of record as to what the authorities could do, because whatever they could do they did do in the context of this very alleged breach. One has examples from the US as well. To that extent it could be said there is a full picture of what can be done when assessing the question of effectiveness.
76. However, and at least at this stage, there are, in my view, several real potential problems with having issue 3 as a real preliminary issue now.
77. First, if it is concerned only with a private law claim based exclusively on 3.10 and we do not deal with a private law claim based on the other regulatory breaches on emissions (see paragraph 71), which I have already ruled out, you then have the potential for related arguments but at two different stages of the case. That seems to me to be undesirable, particularly on a question as important as the existence or otherwise of a private right claim.
78. Second, while one can see how well or not consumers have been served by action on the part of the EU enforcement authorities on this very question, because they have been at it since 2015 and dealt with by now, so it is a developed landscape, I can see that the debate might look at the other remedies available to the consumers and here another four distinct claims have been pleaded out. Whether it is straightforward, on a preliminary and early basis, to make a consideration of the adequacy or sufficiency or otherwise of those remedies before they have been tried is something I cannot be sure about at this stage.
79. Fourthly, while I think the question of appeals has been somewhat overstated, as far as issues 1 and 2 are concerned, I can see that any issue 3 decision might end up with a reference to the Court of Justice; and I am just not confident that all of that can be resolved pre-trial in such a way as to mean that, for example, the claimants will really be in a position to say now they are going to drop all of their other claims or, in some other way, radically shorten the trial or not. Indeed, Mr de la Mare QC said he might keep the CPUT claim going because of a more generous measure of damages.
80. As for the point made to me, particularly on issue 3, that if issues 1 and 2 are in as well, then if they all succeed for the claimants, then most or all of the claims can then safely be dropped, I think is too speculative at this stage.
81. I do not think that the analogy of follow-on damages claims in competition cases is particularly useful here. That is because in such cases the quantum trial will happen anyway, on the basis of the original finding of liability. So that even if, after the trial is finished, an appeal from CAT or some other body means that there is now no underlying liability after all, it does not affect the fact of the trial going ahead.
82. A further point made by Mr de la Mare QC was that there was a pressing need for the question of private claims to be ventilated as early as possible in order to have the best chance of obtaining a binding Court of Justice decision on a reference, which might not be available later because of the effect of Brexit. I am afraid I am not going to speculate on that. In any event, it does not seem to outweigh the other points against issue 3 so far mentioned.
83. For those brief reasons, I do not consider it is appropriate to have issue 3 tried as a preliminary issue, at least not as I sit here today. It is possible that the arguments and the impact may become clearer later on, so that it can be revisited, as might also be true with issue 2A, but that is all for the future.
84. Accordingly, the new 3A, which would just add to the scope of private law claims by reference to paragraph 71, falls away as well, as would a further "folded in" addition, being a consideration of whether any particular head of loss which is claimed would fall within the scope of any private law right to compensation if that existed.
85. The effect of all that is that I now have to consider the utility and the other features of issues 1 and 2 standing by themselves because issue 3 will not be a preliminary issue.
86. On that basis it can be obviously said that, in that case, there is not even, in theory, a simple route to success on liability for the claimant via breach of statutory duty. So there is bound to be a trial about something or other. I agree. But that does not necessarily rule out issues 1 and 2 as preliminary issues. It remains the case that the defeat device allegation appears in many places in the particulars of claim.
87. Just to give a flavour of that: apart from the breach of statutory duty itself, where it is a necessary condition, in relation to deceit I have already referred to 66.1. The first way in which falsity is put, 66.4 specifically says that there was an illegal component which is the defeat device. In the breach of contract claim at 77.1 the satisfactory quality is said not to be there because they were fitted with prohibited defeat devices.
88. Under the consumer protection claim at 92.4, the defeat devices meant that the vehicles were not manufactured to a high quality standard which would meet consumer expectations because they could only pass emissions by cheating and did not comply with the emissions requirements.
89. Then, finally, on unfair relationship, at paragraph 103.1, in entering a relationship they were misled by the representations which were false. So that feeds into that particular allegation, as did the underlying points about satisfactory qualities. And 103.2 makes reference to the unlawful defeat devices.
90. So the allegation of defeat device, so far as the claimants are concerned, percolates well beyond the foundation for the breach of statutory duty claim. It remains the case that the defendants appear to rely on the fact it is not a defeat device so as to answer a number of allegations that go somewhat wider. I have already given an example of that in paragraph 24 of the defence.
91. It remains the case that if the claimant loses on the defeat device issue, it will adversely affect the other claims and the breach of statutory duty will go altogether and with it, actually, any need to decide issue 3 at all so far as 3.10 is concerned. If the claimants win on issue 2, with or without issue 1, then I accept that its claims will be more straightforward and this will have a real potential to shorten the trial.
92. I am of the clear view that success or failure for the parties on issue 2 is important and, in any event, will clear away some matters other than those to be tried. Of course matters like quantum will remain, but that is inevitably the case in a situation where you are having preliminary issues, otherwise the issues would not be preliminary.
93. Various features of the other claims will remain as a whole, but, in my judgment, there is a real prospect of a shorter trial. If VW loses it will not be able to rely on the fact that it is not a defeat device; it is going to have to reshape its approach significantly to the way in which the matters have been pleaded against it so far, because of its reliance so much on the fact of it not being a defeat device and other matters are then simply irrelevant.
94. In GLO cases, as Mr Gibson QC says, you can have a staged approach for many things. I would include within that particular issues. You do not have to show a real prospect of a radically different or a radically shorter trial or no trial before ordering them. It is enough, as here, that they are disputes of a central feature of the litigation and they are at present a huge bone of contention between the parties.
95. In my judgment, there is real utility in dealing with that question now, rather than leaving it for trial. It is the case, as I have already explained, that issues 1 and 2 are, in essence, matters of law. They are neat. They are confined and they are manageable in scope.
96. In other words, I can well see how either side, but perhaps more likely the claimants here, since it is the party seeking the preliminary issues, could have applied for summary judgment on both of these questions and asked the court to decide them one way or the other, as so often happens on points of law and construction in a commercial case. We are in truth, in my judgment, not very far from that situation here.
97. There remains of course the prospects or the possibilities of appeal and perhaps a reference on issue 2, but, in my judgment, much less so than on issue 3. Once one moves away from the notion that the determination of issues 1 to 3 means that the trial is going to be radically different or radically shortened or disappear then the impact of the appeals is less important.
98. The claimants, in arguing for all of issues 1 to 3, say that determination may encourage settlement. I agree that the mere fact that the parties say: well, they will not settle, they will fight to the bitter end, is not much of an answer, even in very large scale litigation like this, because we know, in practice, that settlements do occur, even in very large cases, even between protagonists in bitterly disputed litigation. The prospect of settlement needs to be looked at more objectively.
99. More important here is the fact that these two issues will not themselves preclude any trial; but because of their obvious importance to both sides, I do not discount the possibility that their determination one way or the other will have an impact on settlement or a consideration of discussing settlement.
100. I see no reason why the selection of cases should not start now as contemplated by the defendants. That is because the preliminary issue trial of issues 1 and 2 will not affect it or delay it. That is especially so because the basis of selection of the test cases, first down to 80, and then down to 20, is based on information about individual cases which will vary, none of which relate to issues 1 and 2. That is not surprising because they feature in every single case. They are true common issues.
101. Let me deal with some residual points:

(1) I agree with Mr Gibson QC that the claimants came to proposing the preliminary issues late in the day, only at the end of January and before then they seemed content with the principle simply of test cases. But that does not seem to me to be relevant if, on the merits, a proposal for the issues as preliminary ones is sound, just as it appeared to be to the court in Australia.

(2) Secondly, I have been shown Table A to the claimants' skeleton, showing what would be left after issues 1 to 3 have been decided, and a modified version produced yesterday afternoon by the defendants to show a multiplicity of points. I have also seen the defendants' own Table 1. I follow all of that, but it does not seem to me to affect the utility of ordering issues 1 and 2 to be tried as preliminary issues now;

(3) A point arose about dealers. The claimants are happy to have the dealers in the preliminary issue trial if they wish. They are happy to have a case against a dealer, as opposed to one merely against the finance company, as one of the test cases. None of that seems to affect my deliberations for present purposes.

(4) Finally, on the point of timing, in my judgment issues 1 and 2 can be tried properly within about two working weeks. That is a modest and a proportionate use of court time. It will not adversely affect or detract from the progression of the test cases; and the decision is likely to be given a long stage before trial.

102. So, having made all of those observations, I then, finally, return to the ten factors referred to by Mr Justice Neuberger (as he then was).
103. Is the determination disposing of the case or at least one aspect of the case? It certainly is disposing of one aspect of the case, the defeat device issue, which I regard as important, for the reasons already given.
104. Secondly, would it significantly cut down the cost and time involved in pre-trial preparation or in connection with the trial itself? Yes, in the sense that there is likely to be some narrowing of issues or, if the claimant loses, the removal of a claim altogether and the removal of significant strands of its other claims.
105. Third, how much effort will be involved in identifying the relevant facts? No real effort is required here because the facts are either agreed or largely uncontroversial.
106. Fourth, if it is an issue of law, to what extent it is determined on agreed facts. Well, that is rather similar to point three. Effectively, that, if the facts are not agreed, they are uncontroversial or actually common ground. Indeed, it was not submitted, on the findings that I make the question of law does not turn on serious issues of disputed facts.
107. Fifth, where the facts are not agreed, how that impinges on the value of the preliminary issue. That does not arise.
108. Sixth, would it be unreasonably fair to either or both parties or the court in achieving of a just result? No, for the reasons I have given.
109. Seventh, is there a risk of determination increasing costs or delaying the trial? No, in my judgment, for the reasons I have given. This is a short, discrete matter, which can run perfectly sensibly in parallel.
110. Eighth, Will it be irrelevant? No, it is obviously relevant.
111. Ninth, the issue of amendments does not arise.
112. I should make it clear, as Mr Justice Neuberger made clear, the nine factors are not each threshold conditions which have to be satisfied before the court can order a preliminary issue. What the court has to do, as factor 10 makes clear, is to ask itself whether, taking into account all the previous points, it is just to order a preliminary issue, and the nine specific tests overlap to some extent. Having taken all of those factors into account, I have no doubt it is just to order a preliminary issue trial on issues 1 and 2.
113. So far as the overriding objective is concerned, I consider that such a trial is fair to both sides. It helps to deal with the case overall expeditiously and proportionately. That is particularly so where issue 2 has already been ventilated extensively and remains important to both sides.
114. I am therefore going to order a trial of those preliminary issues. I will consider afterwards, either today or on another occasion, what direction to give; but my provisional view is that this matter can be tried quite conveniently and properly within the autumn term.