Case Nos: C3/2020/0619 C3/2020/0625 C3/2020/0627 C3/2020/0643 C3/2020/0648
ON APPEAL FROM THE COMPETITION APPEAL TRIBUNAL
(Roth J, Fancourt J, Hodge Malek QC)
[2020] CAT 7
Royal Courts of Justice Strand, London, WC2A 2LL
Date: 11 November 2020 Before :
SIR GEOFFREY VOS, CHANCELLOR OF THE HIGH COURT LORD JUSTICE FLAUX
and
LADY JUSTICE ROSE
Between :
(1) AB VOLVO (PUBL) & OTHERS
(2) DAF TRUCKS N.V. & OTHERS
(3) DAIMLER AG & ANOTHER
(4) MAN SE & OTHERS
(5) FIAT CHRYSLER AUTOMOBILES & OTHERS Appellants
- and -
(1) RYDER LIMITED & ANOTHER
(2) DAWSONGROUP PLC & OTHERS
(3) VEOLIA ENVIRONNEMENT S.A. & OTHERS
(4) SUEZ GROUPE SAS & OTHERS
(5) WOLSELEY UK LIMITED & OTHERS
(6) ROYAL MAIL GROUP LIMITED
(7) BT GROUP PLC & OTHERS Respondents
Mr Daniel Beard QC and Mr David Gregory (instructed by Travers Smith LLP) appeared on behalf of the DAF Appellants
Mr Paul Harris QC, Mr Ben Rayment and Ms Alexandra Littlewood (instructed by Quinn Emanuel Urquhart & Sullivan UK LLP) appeared on behalf of the Daimler Appellants
Ms Monica Carss-Frisk QC and Mr Matthew Kennedy (instructed by Herbert Smith Freehills
LLP) appeared on behalf of the Iveco Appellants
Mr Daniel Jowell QC and Mr Tom Pascoe (instructed by Slaughter and May) appeared on behalf of the MAN Appellants
Ms Sarah Ford QC and Mr Daniel Piccinin (instructed by Freshfields Bruckhaus Deringer LLP) appeared on behalf of the Volvo/Renault Appellants
Mr Tim Ward QC, Mr Ben Lask and Ms Anneliese Blackwood (instructed by Bryan Cave
Leighton Paisner LLP) appeared on behalf of the Respondents in the appeals in the Royal Mail Group Limited, BT Group PLC and Dawsongroup Plc actions
Mr Mark Brealey QC and Mr Derek Spitz (instructed by Ashurst LLP) appeared on behalf of the Respondents in the appeals in the Ryder Limited action
Ms Marie Demetriou QC (instructed by Hausfeld & Co. LLP)
appeared on behalf of the Respondents in the appeals in the Suez Groupe SAS, Veolia
Environnement S.A. and Wolseley UK Limited actions
Hearing dates : 6 and 7 October 2020
Approved Judgment
Lady Justice Rose:
I. INTRODUCTION
Five appeals have been brought against a judgment of the Competition Appeal Tribunal (Roth J, Fancourt J and Hodge Malek QC) (‘the CAT’) on a preliminary issue arising in seven damages claims brought in the CAT by the Respondents. The damages claims are ‘follow-on’ claims - following on, that is, from the infringement decision adopted by the European Commission on 19 July 2016 in Case AT.39824 – Trucks (‘the Decision’). The Appellants are truck manufacturers, some of which were addressees of the Decision. The Decision found that the addressees had infringed the competition rules by participating in a long running infringement of the competition rules. The Decision was adopted in accordance with the settlement procedure laid down by Article 10(a) of Regulation (EC) No 773/2004. The settlement procedure applies where parties admit the infringement alleged against them following the initiation of an investigation by the Commission and may be rewarded with a 10% reduction in their fines in return for settling the case. This reduction in fine reflects the fact that their admissions mean that the Commission can carry out a much shorter investigation and adopt a more limited decision than would be needed if the infringement allegations were fully contested by the participants. The addressees took advantage of that procedure, admitted their involvement in the infringement described in the Decision and received the 10% reduction in fine.
The Respondents, who are the claimants in the actions before the CAT, all bought or leased trucks made by the Appellants during the period when the infringement was in operation according to the Decision. They now claim compensation alleging that the prices they paid for those trucks were artificially inflated by the Appellants’ unlawful conduct. The question raised by these appeals is whether it is an abuse of process for the Appellants, in defending the follow-on damages claims brought in the CAT, to put the Respondents to proof of facts that are set out in the Decision and which the addressees admitted in order to settle the investigation into their infringing conduct by the Commission.
The CAT’s judgment on this preliminary issue was handed down on 4 March 2020 ([2020] CAT 7). The CAT held that in certain circumstances it would be an abuse of process for the Appellants to seek to contest facts set out in the Decision that the addressees had accepted in the proceedings before the Commission. Permission to appeal was granted by the CAT in a separate decision of 26 March 2020 ([2020] CAT 10). The Appellants’ primary case in these appeals is that there is no place for the English common law doctrine of abuse of process here because the question of what recitals of the Decision can or cannot be contested by the Appellants in proceedings in the national court is determined by EU law. They rely first on Article 16 of Regulation 1/2003 as ruling out the application of the domestic abuse of process doctrine in these circumstances. Article 16 imposes an obligation on national courts and national competition authorities not to take decisions which run counter to a decision either already taken by the Commission or contemplated by the Commission. The relevant part of Article 16 for our purposes provides that:
“(1) When national courts rule on agreements, decisions or practices under Article [101] or Article [102] of the Treaty which are already the subject of a Commission decision, they cannot take decisions running counter to the decision adopted by the Commission. They must also avoid giving decisions which would conflict with a decision contemplated by the Commission in proceedings it has initiated. To that effect, the national court may assess whether it is necessary to stay its proceedings. This obligation is without prejudice to the rights and obligations under Article [267] of the Treaty."
The Appellants say also that to prevent them from contesting certain facts in the follow-on proceedings would be a breach of their fundamental rights of defence which form part of the general principles of EU law. The CAT’s ruling also, they submit, operates to discourage those who have committed infringements from using the settlement procedure and so puts the United Kingdom in breach of its obligation of sincere cooperation under Article 4(3) of the Treaty on European Union (‘TEU’) by jeopardising the attainment of the objectives of that procedure.
If they are wrong in their main contention about the application of EU law and if the doctrine of abuse of process does apply, the Appellants submit that the CAT erred in the test that it applied and in the conclusions it came to on the application of that test to the facts of this case.
The Appellants (except for Volvo/Renault) submit that if this Court is not persuaded by their arguments, then a reference should be made to the Court of Justice of the European Union (‘CJEU’) for a preliminary ruling under Article 267 TFEU. Any such reference must be made before 31 December 2020 because after that date Articles 86(2) and 126 of the Agreement on the Withdrawal of the United Kingdom and Northern Ireland from the European Union and the European Atomic Energy Community as given effect in English law by the European Union (Withdrawal Agreement) Act 2020 provide that such a reference is no longer possible. I granted expedition for the hearing of the appeals by order of 15 May 2020.
The Court announced at the end of the two-day hearing that the appeals would be dismissed. These are my reasons for coming to that conclusion.
II. BACKGROUND
The settlement procedure and the Decision
The CAT’s judgment describes the settlement procedure in detail at [108] onwards so I need only summarise it here and note the most important elements for our purposes. The settlement procedure is set out in amendments to Regulation 773/2004 that were inserted into that Regulation by Regulation 622/2008. The enactment of the procedure was accompanied by the publication of the Commission Notice on the conduct of settlement procedures (2008) Official Journal C 167/01 (‘The Notice’).
When the Commission initiates proceedings pursuant to Article 11(6) of Regulation 1/2003 it considers whether the case is one where it might be appropriate to engage the parties in settlement discussions. If the parties respond positively that they might be prepared to settle, the Commission tells the parties about the allegations it envisages raising against them, and discloses the information it holds supporting those allegations, including non-confidential versions of any documents it has on file which
the parties wish to see. The Commission also indicates the size of the potential fine. This early disclosure is intended to enable each party “to ascertain its position regarding a time period or any other particular aspect of the cartel”: see Article 10a(2)(c) of Regulation 773/2004. The Notice at point (16) explains further that this early disclosure of the Commission’s case:
“will allow the parties to be informed of the essential elements taken into consideration so far, such as the facts alleged, the classification of those facts, the gravity and duration of the alleged cartel, the attribution of liability, an estimation of the range of the likely fines, as well as the evidence used to establish the potential objections. This will enable the parties effectively to assert their views on the potential objections against them and will allow them to make an informed decision on whether or not to settle.”
The parties then commit to the settlement procedure by providing their settlement submissions “reflecting the results of the settlement discussions and acknowledging their participation in an infringement of Article [101] as well as their liability”: Article 10a(2). The Notice states at point (20) that the settlement submissions should contain:
“(20)(a) an acknowledgement in clear and unequivocal terms of the parties’ liability for the infringement summarily described as regards its object, its possible implementation, the main facts, their legal qualification, including the party’s role and the duration of their participation in the infringement in accordance with the results of the settlement discussions.”
The settlement submissions must also confirm, broadly, that the parties’ rights of defence have been complied with: Notice point (20)(c) and (d). The parties may call upon the Hearing Officer at any time during the settlement procedure in relation to issues that might arise relating to due process. The Hearing Officer’s duty is to ensure that the effective exercise of the rights of defence is respected: Notice point (18). Once lodged, the settlement submissions cannot then be revoked by the parties unless the Commission’s statement of objections and the ultimate decision fail to reflect them. The decision is deemed to reflect the submissions if it “reflects their contents on the issues mentioned in point 20(a)”: see Notice point (22).
Once they receive the statement of objections from the Commission, the parties should reply to it “by simply confirming (in unequivocal terms) that the statement of objections corresponds to the contents of their settlement submissions and that they therefore remain committed to follow the settlement procedure”: Notice point (26). If the statement of objections does not reflect the settlement submissions then the normal contested procedure will apply; all factual acknowledgments by the parties are deemed to be withdrawn and cannot be used in evidence against any of the parties to the proceedings: Notice point (27).
The Commission retains the right to adopt a decision that departs from its preliminary position expressed in the statement of objections endorsing the settlement submissions. The Notice provides at point (29) that in that case the Commission:
“…will inform the parties and notify to them a new statement of objections in order to allow for the exercise of their rights of defence in accordance with the applicable general rules of procedure.”
Again, the settlement submissions would then be deemed to be withdrawn and cannot be used in evidence against them.
Point (41) of the Notice confirms that a final settlement decision is taken by the Commission under the usual powers in Regulation 1/2003 and is subject to review by the General Court in accordance with Article 263 TFEU.
Turning to the procedure adopted in this case, the Commission’s investigation into this infringement was triggered by information provided by one of the participants in the infringement, MAN SE, coming forward as the whistle-blower to disclose the existence of the arrangements. The MAN SE group applied on 20 September 2010 for immunity from fines under the Commission’s Notice on immunity from fines and reduction of fines in cartel cases. Three other Appellants later applied for and were granted more limited leniency and a substantial reduction in fines. These reductions in fines for leniency are separate from the reduction in fines for the later settlement of the investigation. Unusually, it was only after, rather than before, the sending of the Commission’s statement of objections that the Appellants asked for the case to continue under the settlement procedure. Settlement meetings between each Appellant and the Commission took place at which the Appellants were able to express their views on the objections raised by the Commission against them. The Commission also provided each Appellant with an estimate of the fine likely to be imposed on it. The Appellants then made formal requests to settle and sent in their settlement submissions.
The Decision was adopted on 19 July 2016. It describes the infringement as a single continuous infringement of Article 101 in the form of collusive arrangements lasting from January 1997 to January 2011 on pricing and gross price increases for medium and heavy trucks. The collusion also covered the timing and the passing on of costs for the introduction of new emission technologies. The infringement covered the entire EEA. Recital (3) of the Decision states:
“The facts as outlined in this Decision have been accepted by MAN, Daimler, Iveco, Volvo and DAF (the "Addressees") in the settlement procedure.”
The Decision describes the early settlement meetings that took place at which each addressee expressed its views on the objections raised by the Commission; those comments were carefully considered by the Commission and where appropriate taken into account. Settlement submissions were then provided. According to recital (43) of the Decision, those submissions contained the acknowledgements required by point (20) of the Notice.
Recitals (46) to (60) of the Decision then describe in more detail the conduct engaged in by the addressees, exchanging gross price lists and information about prices and other commercially sensitive information on a multilateral and bilateral level. It describes meetings of senior managers that took place at Headquarter-level and illustrates this with the following examples:
“(52) The following examples of meetings illustrate the nature of the discussions, in particular between the Addressees at the Headquarter-Level during the early period of the infringement. On 17 January 1997, a meeting was organised in Brussels.25 It was attended by representatives of the Headquarters of all of the Addressees. The evidence demonstrates that future gross list price changes were discussed. During a meeting on 6 April 1998 in the context of an industry association meeting, which was attended by representatives of the Headquarters of all of the Addressees, the participants coordinated on the introduction of EURO 3 standard compliant trucks. They agreed not to offer EURO 3 standard compliant trucks before it was compulsory to do so and agreed on a range for the price additional charge for EURO 3 standard compliant trucks.”
Recital (59) sets out similar details illustrating meetings and email exchanges that took place between representatives at the level of the German subsidiaries of the Appellants:
“(59) The following examples illustrate the nature of the discussions in which representatives of the German-Level took part. At the end of 2004, an employee of DAF Trucks Deutschland GmbH sent an email to, amongst others, the representatives of the German Subsidiaries requesting that they communicate their planned gross price increases for 2005. The summarised and compiled price increase information was sent back to all of the participants, including all of the Addressees, a few days later containing information on intended gross price increases.39 The Addressees attended a meeting between 4 and 5 July 2005 in Munich which was attended by both non-senior Headquarter-Level representatives and employees of the German Subsidiaries.40 It appears from the evidence that common activities and meetings were scheduled. In addition special sessions were also foreseen involving the non-senior representatives of the Headquarters and separate meetings involving the representatives of the German Subsidiaries.41 During one of these latter sessions the participants, including all of the Addressees, exchanged information about their planned future gross price increases for 2005 and 2006 as well as the additional cost of complying with the EURO 4 emissions standards.42 Further meetings involving representatives of the German-Subsidiaries continued the discussions on price increases and the price increases for Euro 4 and Euro 5 standards include the meetings held on 12 April 200643 as well as on 12 and 13 March 2008.44”
The content of each of the footnotes in those paragraphs has been redacted from the published version of the Decision.
Substantial fines were imposed although MAN’s fine of around €1.2 billion was reduced to zero under the leniency programme. A reduction of 10 per cent was granted to all the addressees in accordance with point 32 of the Notice: see recital
(134). The total value of the fines imposed was over €2.9 billion. There was no appeal to the General Court from the Decision by the addressees.
The proceedings in the CAT and the CAT’s judgment
After the adoption of the Decision, proceedings were brought by the Respondents claiming that the prices they paid when they bought or leased trucks made by the Appellants were unlawfully inflated by the infringement. In their Defences to the claims, the Appellants have pleaded that they do not admit or that they deny some of the facts pleaded by the Respondents where that pleading repeats facts recorded in the Decision. In the light of that, the CAT of its own motion ordered the hearing of a preliminary issue in each of the seven claims before it. The Respondents were directed to draw up schedules identifying which findings in the recitals in the Decision they relied on and contended could not be denied or not admitted. The Appellants then identified which findings in the recitals they accepted they could not contest or which they admitted in any event.
The preliminary issue was formulated by the CAT as follows:
“Which, if any of the recitals, or parts of recitals, in sections 3, 4 and 7 of the [Decision], which are
(a) said by the Addressee Defendants in their Joint Response to the Claimants' Recital Schedules dated 5 February 2019 not to be binding; and
(b) have otherwise not been admitted by all of the Addressee Defendants; and
(c) are set out by the Claimants in their joint schedule to be served on 19 July 2019;
are binding in these proceedings in respect of the Addressee Defendants such that they cannot be contested by (a) the Addressee Defendants or (b) any of the Defendants, and for what purposes are they or the operative part of the Decision binding?”
The sections of the Decision referred to are as follows: section 3 is the description of the conduct set out in 18 recitals, section 4 is the Legal Assessment and section 7 is the fines section.
It is important to be careful about the use of the word “binding” and “non-binding” in the preliminary issue and in this case more generally. This case concerns the content of the Appellants’ pleaded Defences and whether they should be permitted to not admit or to deny an allegation made in the Respondents’ Particulars of Claim where that pleading alleges a fact which is recorded as a fact in a recital of the Decision. If the outcome of this case is that the Appellants are not so permitted, the fact is described as binding; if they are so permitted, it is described as not binding. That is the sense in which those terms were used by the CAT and are being used in this judgment.
The CAT’s judgment on the preliminary issue falls into two parts. The CAT dealt first with the question of which facts recorded in the recitals in the Decision are covered by Article 16. Again, it is useful to pause here to explain what is meant by saying that a fact is ‘covered by Article 16’ in this context. The parties rightly accept that the starting point for ascertaining which facts can be contested by the Appellants and which cannot is by ‘reverse engineering’ from the potential application of Article 16. A national court must consider whether, if its judgment included a finding of fact that was inconsistent with a fact recorded in the Commission’s decision, that would mean that the court was taking a decision that runs counter to the Commission’s decision in a way that amounts to a breach of Article 16. If it would amount to a breach, then there is little point in allowing the party to plead such an allegation at the outset of the proceedings because that plea cannot succeed. If a fact is binding because the court’s rejection of it would put the court in breach of Article 16, then that is a fact covered by Article 16.
In analysing the extent to which the recitals in the Decision were covered by Article 16, the CAT referred to a number of cases in which the CJEU or the General Court has discussed the circumstances in which those Courts can examine the accuracy of recitals in a Commission decision. These fall broadly into two kinds of case. In the first kind, the Court is considering the admissibility of a challenge in circumstances where the applicant has in fact been successful in establishing that it was not liable for any infringement but nevertheless wants to challenge the Commission decision because the Commission has found that one or more of the building blocks for infringement were in place though not all of them. Thus, in Case T-138/89 NB and NVB v Commission [1992] ECR II-2181 (EU:T:1992:95) (‘NB and NVB’) the Commission had found that there had been no infringement of Article 101 because there had been no effect on trade between Member States but the Commission had found that the applicants’ arrangements gave rise to an appreciable restriction of competition. In Case C-164/02 Netherlands v Commission [2004] ECR I-1177 (EU:C:2004:54) (‘Netherlands’) and Cases T-15 & 16/12 Provincie Groningen and ors v Commission (EU:T:2013:74), the Commission had found that some subsidies granted by the Dutch Government did constitute State aid but went on to decide that they were compatible with the common market so that there was no infringement. The attempts by the ‘winning’ party then to bring an action in the General Court to overturn those parts of the Commission decision that it did not like were dismissed as inadmissible. The reason why such a challenge by the addressee of the decision is inadmissible was explained by the CJEU in Netherlands:
“21. … it need merely be stated that, regardless of the grounds on which such a decision is based, only the operative part thereof is capable of producing legal effects and, as a consequence, of adversely affecting its interests. By contrast, the assessments made in the recitals to a decision are not in themselves capable of forming the subject of an application for annulment. They can be subject to judicial review by the Community judicature only to the extent that, as grounds of an act adversely affecting a person's interests, they constitute the essential basis for the operative part of that act.
22. In the present case, the statement of reasons in issue does not constitute the essential basis for the operative part of a decision adversely affecting the Kingdom of the Netherlands.”
The second category of cases considered by the CAT was those where there was a mismatch between the declarations in the operative part of the Commission’s decision and the description of the infringing conduct in the recitals. For example, in Case T61/99 Adriatica di Navigazione SpA v Commission (EU:T:2003:335) the recitals stated that the applicant was involved in a single infringement covering two sets of routes whereas the operative part referred to it as having been a party only on one set. The General Court held that this should not lead to an annulment of the decision since it was the narrower infringement set out in the operative part that was important. Conversely, in Case T-9/11 Air Canada v Commission (EU:T:2015:994) the General Court did annul an infringement decision because of inconsistencies between the operative part and the facts set out in the recitals. More recently in Case T-105/17 HSBC Holdings plc v Commission (EU:T:2019:675) the General Court found that some of the factual findings made in the decision against HSBC were incorrect but that they did not affect the lawfulness of the finding of infringement because there was sufficient evidence even without the material contained in those recitals.
In most of these cases the General Court’s reasoning stresses that in general only the operative part of the decision is capable of producing legal effects and so capable of being challenged before the Court. However, as the Court put it in Case T-210/08 Coppens v Commission (EU:T:2011:288), [34] recitals can be challenged if they form part of the essential basis for the operative part:
“Although only the operative part of a decision is capable of producing legal effects, the fact remains that the assessments made in the grounds of a decision can be subject to judicial review by the judicature of the European Union to the extent that, as grounds of a measure adversely affecting the interests of those concerned, they constitute the essential basis for the operative part of that measure or if those grounds are likely to alter the substance of what was decided in the operative part
(...).”
In the Air Canada case, the Court referred expressly to Article 16 in responding to submissions from the Commission that the inconsistencies between the recitals and the operative part did not matter because only the operative part of the decision was covered by Article 16. The Court stressed that the obligation under Article 16 not to take a decision running counter to the Commission’s decision is not limited to a consideration of the operative part:
“39. … a national court would take a decision contrary to that adopted by the Commission not only if it gave a different legal classification to the anti-competitive conduct examined, but also if its decision differed from that of the Commission as regards the temporal or geographic scope of the conduct examined or as regards the liability or non-liability of persons investigated in relation to the conduct at issue and whose liability was examined in the Commission’s decision.”
The CAT drew from those cases the following principles as to when a fact stated in a recital is capable of founding a challenge before the General Court. First, recitals can be relied on as an aid to interpretation if the operative part of the decision is ambiguous. A recital which performs that function is ‘binding’ in the sense that when deciding whether a potential outcome of the litigation would run counter to the Commission decision, the operative part must be interpreted in the light of that recital. Secondly, and more significant for these appeals, a recital which affects the substance of what is decided in the operative part or which constitutes the “essential basis” or provides “necessary support” for the operative part is also capable of challenge: [58]. Those phrases, both of which are found in the CJEU’s judgments, mean the same thing. Finally, a recital can be essential or necessary if it affects the amount of the fine rather than the finding of infringement.
The CAT then considered how this case law assisted in resolving the preliminary issue before it:
“64. First, we emphasise again that the obligation resulting from Article 16 is that a national court must not make a decision that would be inconsistent with (“run counter to”) a decision of the Commission. If a judgment of the Tribunal were to be inconsistent with recitals in the Decision that were the
“essential basis” or “necessary support” for the operative part, as those terms are explained above, in our view that would, on any sensible interpretation of Article 16, be inconsistent with the Decision.”
The CAT then held that the criterion used for determining the admissibility of a challenge in the General Court was also the appropriate criterion for deciding which recitals are covered by Article 16:
“67. Secondly, if a finding in a decision cannot be challenged in proceedings before the EU Courts, then it would ordinarily be a denial of justice for that finding to be binding in national proceedings. By contrast, to the extent that it can be challenged on an application in Luxembourg, it falls within the jurisdiction of the EU regime and thus outside the realm of the national court.”
Having then referred to the judgment of the Grand Chamber of the CJEU in Case C-
199/11 Europese Gemeenschap v Otis NV (EU:C:2012:684) (‘Otis’) the CAT concluded:
“68. Accordingly, we consider that the principles which determine whether a finding in a recital to a decision is susceptible to challenge before the EU courts are appropriately applicable to determine whether a finding is binding for the purpose of Article 16: the criterion is that the finding in the recital is an essential basis or the necessary support for a determination in the operative part, or necessary to understand the scope of the operative part.”
The CAT thus found in the first part of its judgment that the test to be applied in sorting out which facts recorded in the Decision are covered by Article 16 and which are not is the same test as is applied by the European Courts in deciding which facts can form the basis of an admissible challenge to the decision under Article 263 TFEU. That test requires the sifting of facts into those which form the essential basis of or necessary support for the decision (‘essential facts’) and those which do not (‘nonessential facts’). An essential fact is therefore a fact which is covered by Article 16 and is therefore binding in the sense that, because it would put the national court in breach of Article 16 if it were to reject that fact, the party is precluded from contradicting the fact in its pleaded case. A non-essential fact is a fact which is recorded in a decision but which could properly be rejected by the national court without putting the court in breach of Article 16 so that the party is not precluded, for that reason at least, from denying or not admitting it.
Applying that test to identify the findings in the Decision that were covered by Article 16 was, the CAT said, a “fact specific exercise”: [68]. The CAT noted that in the schedules prepared for the hearing, the parties had analysed each recital in the relevant sections of the Decision, in support of submissions as to which sentences, and sometimes which phrases within sentences, were or were not binding. The CAT said:
“75. … However, we think that it is important to keep in mind throughout that the question being addressed is what in the recitals is necessary to interpret the above determinations in the operative part or provides an essential basis or necessary support for these elements, such that a contradictory finding by the Tribunal would be inconsistent with those determinations as so interpreted. In some respects, that question falls to be answered by reading several recitals together and they do not always have to be read literally.”
In [76] to [98] the CAT analysed the recitals in the Decision in detail to identify those facts recorded in them which it considered were essential and which were not. The conclusions were then summarised in [148]. For example, in relation to recital (52) describing the meetings at Headquarter-level, the CAT held (see [82(b)], [83] and [148(c)]) that the findings in that recital that there had been a meeting on 6 April 1998 attended by representatives of the Headquarters of all the addressees at which they agreed certain things were essential to the infringement decision and so covered by Article 16. That meeting could not therefore be denied or not admitted by the Appellants in their Defences to the damages claims. However, the remainder of recital (52), in particular the fact of a meeting on 17 January 1997 in Brussels at which future gross list price changes were discussed was not regarded as covered by Article 16. None of the material in recital (59) concerning the meetings at the German level was regarded by the CAT as covered by Article 16. There is no appeal from the CAT’s conclusion that that is the right test or from how it applied it to the facts in the Decision.
The CAT then turned to the question of the application of the doctrine of abuse of process to the denial or non-admission of non-essential facts that it had held were not covered by Article 16. It recorded at [100] that in its skeleton argument MAN had set out the governing principles in terms that all parties had accepted as follows:
“Where the parties to later proceedings are not the same parties (or their privies) as the earlier proceedings, it will only be an abuse of process to seek to litigate issues which arose in the prior proceedings where one of two threshold conditions is met:
… if the parties to the later civil proceedings were not parties to or privies of those who were parties to the earlier proceedings then it will only be an abuse of [the] process of the court to challenge the factual findings and conclusions of the judge or jury in the earlier action if (i) it would be manifestly unfair to a party to the later proceedings that the same issues should be relitigated; or (ii) to permit such relitigation would bring the administration of justice into disrepute.”
That test (‘the Bairstow test’) was derived from the judgment of Sir Andrew Morritt V-C (with whom Potter and Hale LJJ agreed) in Secretary of State for Trade and Industry v Bairstow [2003] EWCA Civ 321, [2004] Ch 1 at [38]. The CAT cited the well-known statement of Lord Bingham of Cornhill in Johnson v Gore Wood & Co [2002] 2 AC 1 (with which Lords Goff, Cooke and Hutton agreed). Lord Bingham referred at [31] to the abuse of process doctrine as being:
“a broad, merits-based judgment which takes account of the public and private interests involved and also takes account of all the facts of the case, focusing attention on the crucial question whether, in all the circumstances, a party is misusing or abusing the process of the court by seeking to raise before it the issue which could have been raised before.”
The CAT held that both limbs of the Bairstow test were satisfied in this case. They referred to the fact that the addressees must have been aware of the potential for private damages actions following the Decision. They received a 10% reduction in penalty, amounting to a substantial saving. They have the advantage in the damages claims of a shorter more generally worded Decision and the settlement submissions are protected from disclosure: [129] and [131]
“Given the asymmetry of information between the participants in a cartel and claimants alleging that they suffered loss as a result of the cartel, those claimants inevitably look to the infringement decision for details as to how the cartel took effect. As the present actions demonstrate, the lack of detail in a settlement decision therefore presents a handicap to claimants seeking to recover, and a corresponding benefit to the addressees of the decision.
…
The defendants would be placing on the claimants the very significant burden of proving all these facts, in circumstances where the relevant information is largely held by the defendants, some of it may be undocumented and insofar as it is documented the claimants would have to obtain it by an extensive disclosure exercise involving very expensive inspection.”
The CAT then went on to discuss instances where it would not be an abuse for an addressee to put forward a positive case that was contrary to the finding of a nonessential fact in the Decision. The CAT referred to the analogous position under CPR Part 14 and the corresponding Practice Direction governing the withdrawal of admissions with the permission of the court. It described various corrections that the Appellants wanted to make to the facts recorded in the Decision, noting that the Respondents did not object to some of them. The CAT set out the principles to be applied as follows:
“141. In our judgment, having regard to the considerations discussed above, the following principles should apply in the particular circumstances of the present proceedings:
(1) It is an abuse of process for a defendant simply to deny or not admit facts set out in a recital relied on by a claimant.
(2) Where the claimants do not object, it is not an abuse for a defendant to put forward a case or evidence inconsistent with a recital.
(3) Where a defendant seeks to put forward a contrary position to a finding in the Decision on the basis that it does not accurately reflect the underlying document referred to, whether that be a contemporaneous document or a statement in a settlement submission, it will not be an abuse for a defendant to do so. If necessary, the Tribunal would have to decide whether there is indeed an inconsistency.16
(4) Where a defendant relies on new evidence which it could not reasonably have had access to at the time of the proceedings before the Commission, it is not an abuse if it seeks to advance facts inconsistent with a recital.
(5) Where a claimant pleads facts or allegations that are more detailed than, additional to or different from the facts set out in a recital, it will not be an abuse for a defendant to plead a response in appropriate terms and detail to such facts and allegations, provided that in substance its response does not amount to a denial of a recital that is not otherwise justified by (2), (3) or (4) above.
(6) In circumstances not covered by (2)-(5), it is for the defendant seeking to put forward a positive case contrary to a finding in a recital to set out the reasons why it should be able to do so in that particular instance, and then for the Tribunal to decide.”
16
How that would be done procedurally with regard to a settlement submission can be considered should the situation arise. We do not as a matter of principle see a difficulty in the Tribunal being shown part of a settlement submission without disclosure to the claimant for this purpose: cp the procedure where an issue arises as to whether a document is a settlement submission: Tribunal Practice Direction relating to disclosure and inspection of evidence in claims made pursuant to Parts 4 and 5 of the CAT Rules 2015, para 4.”
The CAT then elaborated on the kind of situations which might be covered by those gateways (‘the [141] gateways’), concluding that it was for each Appellant, where it wanted to advance a positive case, to set out the justification for departing from a finding in the Decision that the addressees had admitted. The CAT concluded by listing those recitals, or parts of recitals that they considered bound the Appellants pursuant to Article 16, in addition to those which the Appellants accepted were binding or admitted in any event. They did not set out which facts would be binding because of the application of the abuse of process doctrine since that could only be finally determined once the Appellants had had an opportunity to consider whether they wanted to rely on any of the [141] gateways.
III. THE GROUNDS OF APPEAL
The issues in the appeals can be summarised as follows:
Is the application of the abuse of process doctrine precluded by a rule of EU law? Potential candidates for such a rule are (a) Article 16 itself, (b) the EU fundamental principle of a right to a fair trial as encapsulated by Articles 47 and 48 of the European Charter of Fundamental Rights, and/or (c) the duty of sincere cooperation owed by the Member States in accordance with Article 4(3) TEU.
If the answer on that issue is insufficiently clear, should this Court refer questions to the CJEU seeking a preliminary ruling under Article 267 TFEU?
If the Court concludes that there is no EU rule preventing the application of the abuse of process doctrine, was the CAT right to hold that the doctrine was engaged in this case when the Appellants sought to contest the truth of facts that they had admitted, as recorded in the Decision?
If the doctrine is engaged, did the CAT err in failing to set a high threshold for the application of the doctrine and in concluding that that threshold was met by the facts of the present case?
Some preliminary matters
Before turning to those issues there are some preliminary matters that can usefully be dealt with here.
The boundaries of the current debate. Both sides were careful to set the boundaries of what falls to be decided in these appeals. The Appellants fully accept that the facts recorded in the Decision will play an important role in the damages claims in a number of ways. In addition to accepting the first part of the CAT’s judgment, they also accept, following the decision of the House of Lords in Crehan v Inntrepreneur Pub Co (CPC) and another [2006] UKHL 38, [2007] 1 AC 333 that the recitals in the Decision are admissible as evidence of the facts stated in them: see per Lord Hoffmann at [69]. In addition they agree that the ordinary rules of pleading and evidence apply so that prior admissions are admissible in evidence and that a party is not permitted to plead a non-admission of a fact which lies entirely within its own knowledge. This latter point arises from the fact that CPR Part 16.5(1) provides that the defendant must state in his defence which allegations “he is unable to admit or deny” and if they are within his knowledge then he is not so unable: see SPI North Ltd v Swiss Post International (UK) Ltd and anor [2019] EWCA Civ 7, [2019] 1 WLR 2865, [3]. They also point out that they have admitted many other facts in their pleaded Defences that they are not strictly obliged to admit in accordance with those principles.
For their part, the Respondents say that the present dispute concerns only the binding nature of facts which are recorded as having been accepted by the parties in a decision following the settlement procedure. The CAT records at [125] that the Respondents expressly disavowed any suggestion that it would be an abuse of process for an addressee to contest non-essential findings outside the context of the formal settlement procedure and the Respondents maintained that position before us. There may also be contested investigations in which certain aspects of the allegations are conceded by the addressee; for example the addressee might be recorded as having accepted the facts underlying the Commission’s proposed definition of the relevant market or that there was an effect on trade between Member States. The question whether it might be an abuse of process for the addressee later to deny or not admit those facts in the follow-on claims is not a question that arises in these appeals.
Further, it was common ground that the issues in this case are different from the issues that were considered by this Court in Secretary of State for Health v Servier Laboratories Ltd [2019] EWCA Civ 1096, [2020] 2 WLR 149. The scope of the EU principles of res judicata considered in that case apply to rulings of the EU courts and are not engaged by a Commission decision.
‘Main facts’ and ‘non-main facts’. The second preliminary point is the nature of the addressees’ acceptance of the facts set out in the Decision following the settlement procedure. There was a great deal of equivocation in some of the submissions made by counsel for the Appellants, seeking to minimise the consequences of their choice to follow the settlement procedure. I consider the following propositions are clear. First, it cannot be said that the addressees of a settlement decision accept that they have committed an infringement of the competition rules without admitting the facts set out in that decision. What they admit is that they engaged in the activities that are described in the decision; that as a matter of law that conduct amounts to an infringement and that it merits the fine imposed on them. Secondly, although the Notice refers to the need for the parties to set out the ‘main facts’ in their settlement submissions, any implied distinction between ‘main facts’ and ‘non-main facts’ is nothing to do with the distinction between what I have been referring to as ‘essential
facts’ and ‘non-essential facts’. It cannot be, since at the time the settlement submissions are made neither the parties nor the Commission can know the ultimate scope of the infringement which is going to be described in the decision. The reference to ‘main facts’ indicates that the settlement submissions do not need to contain an exhaustive description of every meeting, every participant, every document and every phone call of which the party is aware. Recital (4) of Regulation 622/2008 which inserted Article 10a into Regulation 773/2004 explains that the Commission has a broad margin of discretion to determine whether a case is suitable for settlement discussions bearing in mind, amongst other things, the “extent of contestation of the facts”. The submissions set out the main facts sufficient to indicate the nature and seriousness of the infringement so that the Commission can take a view, having regard to the information it holds, as to whether the parties are serious about settling.
All those ‘main facts’ on which the Commission wishes to rely in support of its case are then what are contained in the statement of objections and ultimately in the Decision. It is because only ‘main facts’ are included that the decision is so much shorter and less detailed than a decision following a fully contested investigation. I do not accept, therefore, that the Decision contains some mixture of main facts and non-main facts and that only the former are facts which have been accepted by the addressees. The Notice does not envisage that the statement of objections and the decision will include non-main facts, added in by the Commission but not admitted by the relevant party. I recognise that the initial settlement submission is not the last word as to what the party is prepared to admit. There may be facts alleged in the statement of objections or the Decision that were not included in a party’s settlement submissions. The party will then have to decide whether it will admit those facts as well. If it does not, it will need to assert that the Commission has in effect decided, as is its right in accordance with point (29) of the Notice, to adopt a position which departs from an endorsement of the parties’ settlement submissions so that the settlement process comes to an end and the case proceeds as a contested investigation. If the party does not trigger that mechanism then it has admitted that fact just as much as it has admitted the main facts set out in its initial settlement submissions. Every fact set out in the Decision is equally admitted by every addressee unless the Decision itself draws a distinction between some addressees and others. The CAT ruled that some of the facts admitted are non-essential and therefore not binding for the purposes of Article 16. That does not make them any less admitted for the purposes of the application of the abuse of process doctrine.
Motive for admissions. In emphasising the distinction between essential and nonessential facts, the Appellants suggested that there may be many reasons why an undertaking might decide to admit facts alleged against them in the statement of objections that go beyond what they said in their settlement submissions. They may conclude following further internal investigations into their own conduct that the additional allegations are true and that their settlement submissions fell short of what they now realise was the true extent of their own employees’ participation in the infringement. They may be prepared to make the admission for pragmatic reasons even if they are unsure whether the allegation is true or are convinced that it is not true but conclude that the benefits of settlement outweigh any benefit to be derived from bringing the settlement process to an end by contesting that allegation. Those pragmatic considerations may lead to the admission of some facts that turn out to be essential and some which are non-essential facts. DAF stressed that the settlement procedure is emphatically not a forum for negotiation or the resolution of factual disputes between the settling party and the Commission: “the vicissitudes of settlement mean that in practice a party will normally accede to a decision while it does not concur with each and every fact found”: see paragraph 81 of DAF’s skeleton. That does not, in my judgment, detract from the effect of the admissions in legal or practical terms. The legal effect of the admissions does not depend on the subjective intentions or motivations of the parties and the national court cannot entertain subsequent arguments about whether the party really meant what it said or not.
Whether admissions are made for a limited purpose. The Appellants argued that the parties to the settlement proceedings accept that the information is true for the purposes of the settlement process but certainly not for all purposes. The CAT rejected such a contention at [130] and was right to do so since there is no support in the law for that proposition. There is nothing in Regulation 622/2008 or the Notice to suggest that the admissions are or can be qualified in that way. On the contrary, point (20) of the Notice expressly states that the settlement submissions must contain a “clear and unequivocal” acknowledgment of the parties’ liability for the infringement. I reject the submission made by Volvo/Renault that the fact that the addressees are not required to make any public statement or give any commitment to accept the admitted facts in subsequent damages litigation indicates that the admissions are made for a limited purpose. It indicates in my view that the effect of the admissions in national proceedings is a matter for national law to determine.
By the time the settlement regime was introduced the prevalence of follow-on damages claims in national courts was well-known and the significance of an infringement decision was also fully appreciated by the Commission and the drafters of the Regulation and the Notice. Recital (2) of Regulation 622/2008 bases the usefulness of the settlement procedure on the fact that “[p]arties to the proceedings may be prepared to acknowledge their participation in a cartel violating [Article 101] and their liability in respect of such participation, if they can reasonably anticipate the Commission's envisaged findings as regards their participation in the infringement and the level of potential fines and agree with those findings.” There is no differentiation there between different kinds of liability. If, which I doubt, the Commission might nonetheless be prepared to allow a party to settlement proceedings to qualify their admissions in that way, there is nothing to suggest that the addressees did so. The effect of the admissions in the EU sphere is a matter for EU law and the legal effect in domestic proceedings is a matter for domestic law unless or until that is harmonised, subject only to the requirement that the domestic proceedings comply with fundamental EU principles.
Does EU law preclude the application of the abuse of process doctrine?
The Appellants’ submissions on this ground were made by Paul Harris QC. He emphasised the supremacy of EU law over domestic law as established in the early judgment of the CJEU in Case 106/77 Amministrazione delle Finanze dello Stato v Simmenthal (No 2) [1978] 3 CMLR 263 and the duty of the national courts to give full effect to EU law. I entirely accept the importance of all those principles but they do not avoid the need to identify a rule of EU law that is to be treated as overriding a domestic rule and to which full effect must be given.
Article 16 of Regulation 1/2003
The first candidate for creating such a rule is Article 16 itself. As I have described, the first part of the CAT’s judgment held that, if a fact is a non-essential fact, it would be open to the CAT ultimately to reject that fact without thereby being in breach of Article 16. Article 16 does not preclude the Appellants from pleading on that basis. The Appellants submit that to give full effect to Article 16, the national court must permit them to plead on that basis and to contest the truth of those facts in the national proceedings, even if that course would amount to an abuse of process under the English law doctrine.
I reject that submission. Article 16 is directed at preventing the national court from arriving at a conclusion which runs counter to a Commission decision. It has nothing to say about how the national court should deal with allegations which do not assert something that runs counter to the Decision. It does not govern what can or cannot be pleaded in English proceedings beyond ruling out the pleading of matters which, if ultimately reflected in the national court’s judgment, would put the court in breach of Article 16. The effect of the application of the abuse of process doctrine here would be that other findings in the CAT’s final judgment in the actions would also have to be consistent with the findings in the Decision even though that consistency is not mandated by Article 16. I cannot see that it would “run counter” to the Decision if the CAT held that the non-essential facts identified by the CAT are also true. A domestic rule that has that effect cannot be contrary to Article 16.
I do not accept Mr Harris’ characterisation of the CAT’s ruling as creating a ‘clash’ between EU and domestic law. The clash was said to arise from a position under EU law where non-essential facts are, as he put it, completely non-binding because they are not capable of being appealed to the General Court but conversely are completely binding under English law by reason of the abuse of process doctrine (subject to the [141] gateways).
This submission that the non-essential facts are, as a matter of EU law, to be treated as not binding in any way is, in my judgment, misconceived. It stems from what the CJEU said in the Netherlands judgment to which I referred earlier. The Appellants rely on the statement of the CJEU that an assessment made in the recitals to a decision which are not part of the essential basis for the operative part of that decision did not adversely affect the Netherlands’ interests and had “no binding legal effect such as to affect the interests of the Kingdom of the Netherlands”. It could not therefore constitute a legal act open to challenge under what is now Article 263 TFEU. I do not accept that those statements in Netherlands and similar statements in other cases are authority for the proposition that non-essential facts set out in a decision can have no effect in domestic law. The statements were made in the context of a ruling on admissibility. What the CJEU in fact held was that not only were the recitals about which the applicant complained of no legal effect but that “the decision” itself did not bring about any distinct change in the legal position of the Netherlands and had no adverse effect on them. This was because it declared the state aids to be compatible with the common market: see the arguments of the Commission at [12] - [15] and the judgment at [20]. Since the contested decision declared the aid to be compatible with the common market, it did not bring about a distinct change in the legal position of the Kingdom of the Netherlands.
Similarly, in the earlier judgment in NB and NVB where the applicants had been granted negative clearance (Footnote: 1) for their agreements because there was no appreciable effect on trade between Member States, the General Court held that the operative part of the decision neither changed the addressees’ legal position nor adversely affected their interests. The operative part was not capable of being appealed by them. That does not mean that the operative part was not binding in any circumstances. On the contrary, the Court went on to say that a third party who had a sufficient interest could mount an admissible challenge to the decision before the Court.
I note that at [33] of NB and NVB the Court refers to an argument put forward by the banks as to why the finding of a restriction of competition might affect them adversely because a national court might take a different view of effect on trade between Member States but take into account the Commission’s assessment of the agreement’s restrictive effect. The Court said “this argument must in any case be rejected without it being necessary for the [General Court] to consider the effects of the contested decision as far as national courts are concerned.” The Court did not, therefore, reject the banks’ fears as unfounded because the recitals could have no legal effect in the domestic proceedings.
It does not appear to me therefore that when the European Courts describe part of a decision as ‘not adversely affecting a person’s interest’ or as ‘not having any legal effect’ they are using those phrases to mean that that part can be ignored for all purposes. The case law relied on by the Appellants shows that the legal effect of both the operative part of a decision and the recitals depends on the context in which the issue arises and the purpose for which and by whom that legal effect is being asserted. The effect of them has to be considered in context, whether that context is another aspect of EU law or of domestic law.
Mr Harris drew our attention to the remarks of Marcus Smith J in BritNed Development Ltd v ABB AB [2018] EWHC 2616 (Ch) (‘BritNed’). The judge there addressed the question of whether recitals in a decision are binding and stated that a recital not constituting part of the essential basis for the decision is not binding on the court. I agree with the reasons given by the CAT at [118] as to why those remarks do not take the matter further. First, the Commission decision on which the claimants sought to rely in BritNed was not a settlement decision and the non-essential facts had not been admitted by the addressees. Secondly, it does not appear that the possibility of it being an abuse of process for the addressees to contest those facts was raised at all in the proceedings.
Mr Harris also argued that to allow the application of an English law doctrine to determine which parts of a Commission decision are binding and which are not risks creating a lack of uniformity in approach among the courts of the different Member States. It would create a patchwork result with some national courts treating all nonessential facts as contestable and others treating some non-essential facts as binding on the addressee. I accept that there will be that lack of uniformity and that the CAT noted at [3] that well over a thousand claims for damages following on from the Decision have been brought in various Member States. But I agree with the reasons given by the CAT for rejecting this argument at [145]. Any lack of uniformity is a consequence of the absence of comprehensive harmonisation of the rules that national courts apply when handling follow-on damages claims. Early on in its consideration of the exercise of the victims’ directly effective rights to claim damages for competition law infringements, the CJEU recognised that the national courts are entitled to apply their own rules in areas which have not been harmonised by EU legislation: see for example Case C-453/99 Courage Ltd v Crehan [2002] QB 507 at [29]:
“29. However, in the absence of Community rules governing the matter, it is for the domestic legal system of each member state to designate the courts and tribunals having jurisdiction and to lay down the detailed procedural rules governing actions for safeguarding rights which individuals derive directly from Community law, provided that such rules are not less favourable than those governing similar domestic actions (principle of equivalence) and that they do not render practically impossible or excessively difficult the exercise of rights conferred by Community law (principle of effectiveness): see Palmisani v Istituto Nazionale della Previdenza Sociale (Case C-261/97) [1997] ECR I-4025, 4046, para 27.”
There are many areas which have since been harmonised, recently by Directive No 2014/104/EU (‘the Damages Directive’). But that Directive says that there are still areas which are left to the national court. The Appellants cannot rely on a general reference to the supremacy of EU law and the desirability of uniformity to supplement the harmonisation that has so far taken place.
I would therefore hold that there is nothing in Article 16 itself that precludes the application of the principle that it can be an abuse of process to contest a finding set out in a settlement decision even if it would not be a breach of Article 16 for the domestic court to hold ultimately that the finding was wrong.
Articles 47 and 48 of the Charter
Articles 47 and 48 of the Charter provide:
“47 Everyone whose rights and freedoms guaranteed by the law of the Union are violated has the right to an effective remedy before a tribunal in compliance with the conditions laid down in this Article.
Everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law. Everyone shall have the possibility of being advised, defended and represented.
…
48.1 Everyone who has been charged shall be presumed innocent until proved guilty according to law.
48.2 Respect for the rights of the defence of anyone who has been charged shall be guaranteed.”
These articles encapsulate fundamental principles of EU law: see Otis [46]. As to Article 48, the CJEU said in Case T-474/04 Pergan Hilfsstoffe für industrielle Prozesse GmbH v Commission [2007] ECR II-4231 (‘Pergan’) that the principle of the presumption of innocence as reaffirmed in Article 48 of the Charter is an integral part of the Community legal order and applies to the procedures relating to infringements of the competition rules. The CJEU in Pergan went on to describe how that presumption operates (citations omitted):
“76. The Court also observes that the presumption of innocence implies that every person accused is presumed to be innocent until his guilt has been established according to law. It thus precludes any formal finding and even any allusion to the liability of an accused person for a particular infringement in a final decision unless that person has enjoyed all the usual guarantees accorded for the exercise of the rights of defence in the normal course of proceedings resulting in a decision on the merits of the case (…). Furthermore, the guilt of a person accused of an infringement is established definitively only where the decision finding that infringement has acquired the force of res judicata, which implies either the absence of an appeal against that decision by the person concerned within the time-limits provided for in the fifth paragraph of Article 230 EC, or, after such an appeal, the definitive closure of the contentious proceedings, in particular, by a judicial decision confirming the lawfulness of that decision.
77 As a consequence, findings which the person charged with an infringement, even though he contests their merits, has not had the opportunity to contest before the Community judicature cannot be regarded as established in law. The fact that such findings evade any review by the courts, and therefore, in the event that they are unlawful, any correction by the Community judicature, is manifestly contrary to the principle of the presumption of innocence. Any other interpretation would upset the system of division of functions and the institutional balance between the administration and the courts, on the ground that, where findings are contested, it is for the courts alone to make a definitive ruling on the existence of sufficient evidence that an undertaking is liable for an infringement of the rules on competition.”
Mr Harris argued that these principles of absolute respect for the rights of the defence and the presumption of innocence must be applied by the national courts when determining follow-on damages claims. Recital (5) of Regulation 1/2003 states that:
(emphasis added)
“This Regulation affects neither national rules on the standard of proof nor obligations of competition authorities and courts of the Member States to ascertain the relevant facts of a case, provided that such rules and obligations are compatible with general principles of Community law.”
The Appellants’ reasoning is therefore that it is the addressees’ right to challenge any adverse findings of fact set out in the Decision before the EU judicature which protects their entitlement to be presumed innocent. Since non-essential facts, by definition, cannot be appealed to the General Court, it is an infringement of the Appellants’ right to the presumption of innocence for the non-essential facts in the Decision to be binding on them in the CAT proceedings because they have not been able to challenge those findings in the General Court. The CAT must become the independent and impartial tribunal in which the accuracy of the non-essential facts is established.
In these submissions, the Appellants thus did not acknowledge that there was anything even a little strange in them insisting on their right to be presumed innocent of having engaged in unlawful conduct, even though they freely admitted to engaging in precisely that unlawful conduct when investigated by the enforcement authority and accepted an infringement decision in respect of that conduct. It does, however, strike me as odd. The argument is that because there is no right to appeal against the finding, it is contrary to the presumption of innocence to fix them with their admissions, admissions by which they accept that they are not innocent at all. That appears to me to involve a non sequitur because the character of the admission is not affected by whether it is of an essential or a non-essential fact and hence whether it is, in theory, capable of being the subject of an appeal to the General Court or not.
There is no difficulty with the rights of the defence having been complied with in the EU sphere. The Appellants must have acknowledged in their settlement submissions that their rights to defend themselves in the face of the Commission’s allegations had been respected. They were required to do so by point (20) of the Notice. The findings of fact in the Decision are therefore confirmed as having been arrived at by a procedure that complied with those rights. I do not accept, if this is the implication of the Appellants’ case, that the confirmation that they give in the settlement proceedings should be construed as limited to confirming only that the findings as to essential facts were arrived at following a fair process but that the process by which the non-essential facts were arrived at might in some way have been unfair. That cannot be what the Appellants’ confirmation said and it is difficult to see how in practice such a discrepancy could arise.
Mr Harris relies strongly on Pergan both in the General Court and as discussed and applied by the Court of Appeal in Emerald Supplies Ltd and others v British Airways plc and others [2015] EWCA Civ 1024, [2016] Bus LR 145 (‘Emerald’). In fact it seems to me that these authorities are against the Appellants rather than supporting their arguments. The dispute in Pergan arose over which parts of the Commission’s infringement decision in the peroxides cartel case should be included in the published version of the decision and which parts should be excised from that version because they were confidential and contained business secrets. The material in contention described Pergan’s participation in the main cartel during a period which was time barred and so not included in the infringement described in the operative part. Pergan
asked that all references to it and its alleged offending conduct be redacted from the version of the peroxides decision intended for publication. The Commission rejected the request for confidential treatment and proposed including that material in the definitive, non-confidential version. Pergan brought proceedings before the General Court challenging that refusal to redact the references to Pergan in the published version of the decision.
The judgment records at [35] and [52] the Commission’s submission that the application was “based on the false premise that the disputed information is of a binding nature”. On the contrary, the Commission argued, since the findings were only contained in the recitals and not the operative part and did not constitute necessary support for the operative part, they were not binding and therefore could not adversely affect Pergan’s interests. Pergan argued to the contrary that the passages in question could provide third parties with evidence for the purpose of actions for damages against it ([20]) and that they would in any event harm its reputation.
The Court described its task as being to examine whether the disputed information was protected by the obligation of professional secrecy and “in particular, ascertain whether publication of that information is likely to cause serious harm to the applicant”: [68]. The Court held that since there was no finding of any infringement by Pergan in the operative part, any application by Pergan for annulment of the decision would have been inadmissible. At [76] the Court went on to hold that the presumption of innocence therefore precluded any formal finding “and even any allusion to the liability of an accused person” unless that person has enjoyed all the usual guarantees accorded for the exercise of the rights of defence. The Court went on:
“As a consequence, findings which the person charged with an infringement, even though he contests their merits, has not had the opportunity to contest before the Community judicature cannot be regarded as established in law. The fact that such findings evade any review by the courts, and therefore, in the event that they are unlawful, any correction by the Community judicature, is manifestly contrary to the principle of the presumption of innocence. Any other interpretation would upset the system of division of functions and the institutional balance between the administration and the courts, on the ground that, where findings are contested, it is for the courts alone to make a definitive ruling on the existence of sufficient evidence that an undertaking is liable for an infringement of the rules on competition.”
Since the Commission’s findings against Pergan were capable of infringing the principle of the presumption of innocence, they must in principle be regarded as confidential and covered by the obligation of professional secrecy. That was the case, the Court said, regardless of whether and to what extent the material had probative value in domestic proceedings: [78]
The significance of this case is that the consequence of the Court’s reasoning was that the information had to be excluded from the published version of the Peroxides decision. The public interest in publishing the disputed material could not prevail
over Pergan’s legitimate interest in having the information protected. The Court therefore annulled the Commission’s decision refusing to redact the references to Pergan from the published decision.
The scope of what is often now referred to as Pergan protection was considered by this Court in Emerald where, again, all the material in dispute had been excised from the published version of the Air Freight cartel decision because it was entitled to that protection. The first instance judge who was case managing the follow-on damages proceedings had directed that the claimants have access to the redacted material subject to safeguards he put in place to prevent further disclosure and to control the use that could be made of it. The Court of Appeal held that the material should not be disclosed even in that limited way. Significantly the Court rephrased the issue as being:
“whether a national court is obliged to afford the same protection which is afforded at Community level to the confidential version of the Commission’s decision to Pergan materials, notwithstanding that a damages claimant has made an application for disclosure in national court proceedings.”
The Court held that it was so obliged and that the judge had failed to give due recognition to the absolute nature of Pergan protection to the presumption of innocence. There was a real risk that the disclosure order would conflict with any future decisions made by the Commission as to what material needed to be redacted from the non-confidential version of the decision.
Mr Harris relies on Pergan to support the absolute nature of the presumption of innocence in respect of “adverse statements made by the Commission, which can never be tested or challenged and which “cannot be regarded as established in law.”: [82]. However, the flaw in that argument is that the upshot of Pergan and Emerald was that all the disputed material in those cases was material that had been or should have been excluded from the published versions of the infringement decisions on the basis that it was confidential. In the present case, as the parties confirmed following the hearing, all the facts which the Respondents contend cannot be denied or not admitted by the Appellants are facts recorded in the published version of the Decision. This must mean that the Commission did not regard the publication of those recitals of the Decision as conflicting with the Appellants’ entitlement to the presumption of innocence. If the Appellants regarded the publication of these non-essential facts as infringing their rights of defence because they could not challenge them before the General Court, they could have asked the Commission to redact them. If the Commission had refused, they could have challenged that refusal before the General Court as Pergan did. In the event, all this material is published and the Appellants have not asserted that it is entitled to Pergan protection in order to ensure that their fundamental rights are protected. Pergan and Emerald do not, in my judgment, support the proposition that recitals included in the published version of a settlement decision have to be open to challenge in domestic proceedings in order to uphold the presumption of innocence and the other rights of the defence.
I therefore conclude that there is no breach of the Charter provisions by the application of English law which prevents the non-essential, admitted facts published
in the non-confidential version of the Decision being contested by the Appellants as determined by the CAT.
The duty of sincere cooperation
This part of the Appellants’ case was argued by Mr Beard QC. He submitted that by enhancing the effect of the admissions made during the settlement procedure, the CAT’s ruling will deter those accused of infringements from taking advantage of the settlement procedure because of the risk that they will be fixed with admissions of non-essential facts that they can neither appeal to the General Court nor contest in the later damages action. This deterrent effect of the CAT’s ruling amounts to a breach of the duty of sincere cooperation now encapsulated in Article 4(3) TEU. That provides:
“Pursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties. The Member States shall take any appropriate measure, general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the Union. The Member States shall facilitate the achievement of the Union's tasks and refrain from any measure which could jeopardise the attainment of the Union's objectives.”
I can dispose of this point shortly because I agree with the reasons given by the CAT for rejecting it at [146]. For every statement by the EU institutions to the effect that the settlement procedure is an important weapon in the competition enforcement armoury, one can find a corresponding statement that the directly effective right of a cartel victim to claim damages in the national court strengthens the working of the competition rules and that actions for damages make a significant contribution to the maintenance of effective competition in the Community: see for example Courage v Crehan at [27].
The EU legislature has struck a careful balance between the goal on the one hand of encouraging cartelists to disclose the existence of the cartel to the Commission under the leniency regime and then to make admissions under the settlement procedure and the goal on the other hand of encouraging and facilitating damages actions in national courts. The Damages Directive is now an important instrument by which that balance is struck. Recital (6) states that both private and public enforcement are tools which are required to interact to ensure the maximum effectiveness of the competition rules. It is necessary to regulate the coordination of them in a coherent manner. Recital (7) notes that the marked differences between the rules in Member States governing actions for damages leading to uncertainty and creating an uneven playing field. How does the Damages Directive strike that balance? Article 6 requires the Member States to confer virtually complete protection from disclosure on leniency statements and settlement submissions. That is explained in recital (26) as being necessary to “ensure undertakings’ continued willingness to approach competition authorities voluntarily with leniency statements or settlement submissions”. Nothing in the Damages Directive states that the admitted facts recorded in the published decision are to be afforded any protection or, more broadly, that whistle-blowers or settling addressees are immune or protected in any way from private actions. That is how far the EU legislature thought it was appropriate to go to support the aim of the leniency and settlement regimes.
There are already many advantages for a settling addressee as compared with the addressee of a contested decision facing a follow-on damages claim, in addition to the reduction in the fine and the savings in their own legal costs and management time. The brevity of the decision in many cases creates an obstacle for future damages claimants because there is less detail about the infringement and much less information about the effects of the cartel on prices. Although cartel cases are always ‘object’ infringement cases so that the Commission does not need to establish effect for the purposes of infringement, there is often useful material in a contested decision about the effect of the cartel in the context of justifying the amount of the fine. One of the factors relevant to the size of the fine is the gravity of the infringement, including whether or not the infringement has been implemented. Because the addressees to a settlement decision have already indicated that they will accept the level of fine proposed, the decision does not include anything about implementation.
I do not agree that any rule creating a countervailing disadvantage under national law where the addressee admitted the allegations against it must be disapplied in order to avoid breaching Article 4(3). Daimler argued in its written submissions that the CAT’s approach is likely to result in settlement proceedings becoming lengthier and more protracted due to uncertainty about what may or may not be binding in later national proceedings. That seems unlikely to be a significant factor since no addressee could be confident at the time the admissions are made as to what will end up being an essential fact or a non-essential fact in the infringement decision. Daimler also point out that there is a benefit to the claimants in the addressees using the settlement procedure because the Decision is adopted much more quickly than a fully contested decision and there is unlikely to be any appeal to the General Court. That is of course true, but the factors that speed up or slow down the progress of follow-on actions once the infringement proceedings are closed are many and varied. It is not inevitable that the speed of adoption of the decision will translate into a quicker disposal of the damages claims.
Mr Beard pointed to a more specific way in which he said the CAT’s judgment cuts across the settlement process. He took issue with the CAT’s second [141] gateway. This envisages a situation in which an Appellant could argue that it should be permitted to contest a finding recorded in the Decision because that finding was inconsistent with the relevant settlement submissions. It was not clear to me whether Mr Beard was arguing that it is not open to an addressee itself to waive the confidentiality of its own submissions or whether he argued that to put a party in the position of having to disclose the submission as the price of contesting a finding was equivalent to ordering it to do so. In any event I do not regard the possibility of such a procedure as undermining the settlement process in breach of the duty of sincere cooperation.
Conclusion on the EU law ground: should the Court make a reference to the CJEU?
There was no dispute about the test to be applied in deciding whether to refer a question to the CJEU under Article 267 TFEU. It was set out in the judgment of Sir Thomas Bingham MR in R v International Stock Exchange ex parte Else [1993] QB 534, 545D-F. Mr Ward QC for the Respondents reminded us of the need for national
courts to exercise restraint in referring questions so as not to overburden the Court: see per Chadwick LJ in Trinity Mirror v Commissioners of Customs and Excise [2001] EWCA Civ 65, [2001] 2 CMLR 33, 52. Mr Harris for his part pointed out that the CAT said in its decision granting permission to appeal that the EU grounds raised questions of EU law that had never been considered either by the English courts or the CJEU. The CAT did not consider that the answers to them were acte clair and they were of wider significance. They also said, however, that they did not doubt the conclusions they had arrived at in the judgment.
For my part I am sufficiently confident in the answers to the questions to be able to dispose of the appeals without making a reference. If we were to ask the CJEU: does Article 16 mean that the addressee of a settlement decision must be permitted to contest the accuracy of a non-essential fact, I am entirely confident that the Court would answer that Article 16 is, as it says, concerned only with stopping national courts from taking decisions that run counter to the Commission’s decision. Article 16 has nothing to say about how the national court should deal with fact finding outside that constraint and it cannot possibly run counter to a decision to ensure that all the facts decided are ultimately reflected in the national court’s judgment.
We might ask whether national proceedings would infringe the presumption of innocence if the national court applied a domestic rule that had the effect of holding the addressee to the non-essential facts set out in the non-confidential version of the decision. I am sure that the answer would be that if such a rule in an entirely domestic context is not of itself a breach of the defendants’ rights of defence, then it does not become so in this context.
Finally I cannot accept that there is any possibility of the CJEU ruling that it is a breach of the courts’ duty of sincere cooperation to hold that the admissions recorded in the settlement submissions must stand in the national court proceedings even where the admissions relate to non-essential facts.
I do not therefore consider that it is necessary or appropriate in this case to make a reference for a preliminary ruling to the CJEU.
The domestic law grounds of appeal
The requirement for a ‘final decision’
The first issue raised by the Appellants is whether the ‘Hunter-type’ of abuse of process, namely a collateral attack on a final decision, is engaged at all here. Ms Carss-Frisk QC argued on their behalf that in order for the doctrine to be engaged there must be a ‘final decision’. This requirement is clear from the speech of Lord Diplock in Hunter v Chief Constable of the West Midlands [1982] AC 529. The appellant in that case had been convicted of murder at a criminal trial. He then brought civil proceedings claiming damages for physical injuries he alleged were inflicted on him by police officers while he was in their custody. The judge in the criminal trial, Bridge J, had considered and determined whether the claimant had been assaulted by the police as part of his determination on the voir dire of whether the appellant’s confession was voluntary and held that it was voluntary and so admissible. The allegations were repeated before the jury at the trial at which the appellant was convicted. The House of Lords, upholding the Court of Appeal, held that the civil damages claim later brought by Mr Hunter against the police should be struck out as an abuse of process. Lord Diplock described what is now referred to as a Hunter-type abuse in the following terms: (at p 541B)
“The abuse of process which the instant case exemplifies is the initiation of proceedings in a court of justice for the purpose of mounting a collateral attack upon a final decision against the intending plaintiff which has been made by another court of competent jurisdiction in previous proceedings in which the intending plaintiff had a full opportunity of contesting the decision in the court by which it was made.”
Ms Carss-Frisk argued that a core principle of the Hunter-type abuse of process is that it constitutes an attack on a decision which is final and binding. She noted that immediately after the passage cited above, Lord Diplock (with whom the other members of the Appellate Committee agreed) said that the proper method of attacking the trial judge’s decision on the voir dire was for Mr Hunter to include such a challenge in his appeal against his conviction to the Criminal Division of this Court. He had not done so. In those circumstances Bridge J’s ruling became final when the trial ended with the return of the jury’s verdict. His ruling that the confession had not been obtained by violence could thereafter only be upset by the Court of Appeal: 542G. By contrast here, Ms Carss-Frisk argued, the non-essential facts recorded in the Decision are not capable of being appealed and have, according to the CJEU in Netherlands, no legal effect. That means that only those facts stated in the Decision which are covered by Article 16 because they are essential and so can be challenged on appeal constitute a ‘final decision’ for the purposes of this kind of abuse. We were also referred to Smith v Linskills (a firm) and another [1996] 1 WLR 763 where a claimant, following his release from prison, sued his former solicitors in negligence for their conduct of his defence in the criminal proceedings. The negligence claim was struck out as a collateral attack on the criminal conviction which had been upheld on appeal.
I do not accept the Appellants’ submissions on this point. It is not right in my judgment to make the leap from saying that a non-essential fact cannot be the subject of an appeal to saying that it is not binding for the purposes of being part of a final decision which is protected from collateral attack in subsequent proceedings. Article 288 TFEU provides that “A decision shall be binding in its entirety. A decision which specifies those to whom it is addressed shall be binding only on them.” There is no distinction drawn there between essential and non-essential parts of the decision and no suggestion there or in any other legislative provision or judicial decision that only those parts of a decision that can be appealed should be regarded as binding. I have already explained my rejection of the Appellants’ reliance on the Netherlands case and similar judgments of the CJEU. They are not authority for the proposition that non-essential parts of a decision are to be ignored as having no effect in any circumstances.
Ms Carss-Frisk argued that the Decision should be treated not as a single entity but as a bundle of many hundreds of individual factual decisions each of which is final or not, depending on whether it is capable of being appealed. I do not find any support in the case law for such a characterisation of a Commission decision. The case law establishes that a single decision is to be regarded as a bundle of decisions in so far as
it is a separate decision taken in respect of each addressee. That is why the annulment of a decision by the General Court at the suit of one addressee does not annul its binding effect as against a different addressee who did not appeal: see Bellamy & Child European Union Law of Competition (8th edn, 2018), para. 13.164. That does not mean that every factual finding counts as a different decision either as a matter of EU law or English domestic law. I agree with the submission of Ms Demetriou QC on behalf of the Respondents that there is no need to import into the doctrine of abuse of process the rather abstruse distinction drawn in EU law between essential, appealable facts and non-essential, non-appealable facts. The Decision taken as a whole is a final decision of the kind referred to by Lord Diplock for the purposes of deserving the protection from collateral attack provided by the abuse of process doctrine where that doctrine would otherwise apply.
Finally, Ms Carss-Frisk relied on the judgment in Enron Coal Services Ltd v English, Welsh & Scottish Railway Ltd [2011] EWCA Civ 2. The issue there was which facts set out in an abuse of dominance infringement decision taken by the Office of Rail Regulation were binding on the respondent railway operator in a later damages claim brought against it. The CAT dismissed the damages claim on the grounds that causation had not been proved. The claimant appealed on the grounds that the CAT had reached conclusions on the facts which had not been open to it because they were inconsistent with the facts found in the infringement decision. That judgment is of limited relevance to the present appeals since the ORR’s decision followed a fully contested investigation. Further, the focus of the appeal was on the provisions in the Competition Act 1998 expressly dealing with the extent to which infringement decisions of the domestic authorities are binding in follow-on damages claims within the CAT’s statutory jurisdiction. According to those provisions, the CAT was bound by findings of fact made by the regulator unless the CAT directed that a particular fact should not be binding. The Court of Appeal described that regime as comprehensible because: [50]
“… the regulator may make findings which are directly relevant to a decision as to infringement, but it may also make findings of much less direct relevance. Findings in the former category should be regarded as binding, because to challenge them would be tantamount to challenging the finding of infringement. However, if the finding is peripheral or incidental, on the one hand to question it may not involve subverting the infringement finding and on the other it may be fair and sensible because the undertaking may not have been concerned, for the purposes of the regulatory proceedings, to contest such a point, whereas if the finding is relied on in proceedings for damages it may have a much greater importance.”
I do not see that this assists the Appellants. The CAT’s judgment in this case was not that non-essential facts are binding in all circumstances; if one of the [141] gateways is available, the CAT can direct that the finding is not binding just as it can under the different regime in the Competition Act. The Competition Act regime does not provide the level of protection that the Appellants are asking for in this case, namely
that a finding which is ‘peripheral or incidental’ cannot be binding in any circumstances.
In my judgment therefore, the whole of the Commission’s Decision is a ‘final decision’ for the purposes of the application of the test for a Hunter-type abuse so the doctrine of abuse of process is properly engaged.
Did the CAT apply a high enough threshold?
Mr Jowell QC made submissions on behalf of the Appellants to the effect that the CAT erred in the test that it applied to determine whether there was an abuse of process here because it failed to set a high enough threshold. If it had applied the correct test, Mr Jowell argued, the CAT could not properly have come to the conclusion that the test was satisfied in the present case.
The CAT began its discussion of abuse of process by recording the Appellants’ submission that it is wholly exceptional for it to be an abuse to advance a case which is inconsistent with a previous decision where the parties to the two sets of proceedings are not the same: [121]. At [131] the CAT said:
“131. In that context, on the basis of the English authorities on abuse of process discussed above, we apply a broad, meritsbased approach in asking whether it would bring the administration of justice into disrepute and/or be unfair to the claimants if the defendants are able simply to deny the facts which the Decision records them as having admitted, or to “not admit” those facts in their defences to these claims and thus require the claimants to prove them.”
Mr Jowell criticised this passage as suggesting that the CAT considered it was engaged in weighing up the different factors without bearing in mind the need for exceptionality. This was shown he said by the CAT’s reference to the criterion being whether it was ‘unfair’ rather than ‘manifestly unfair’ or some other qualifying adjective.
In addition to the cases cited by the CAT, Mr Jowell referred to Conlon and anor v Simms [2006] EWCA Civ 1749, [2008] 1 WLR 484. In that case the partners of Mr Simms, a solicitor who had been struck off the roll by the Solicitors Disciplinary Tribunal, sought to rely on the findings of dishonesty made by that tribunal in their claim that Mr Simms had induced them to enter into partnership with him by making fraudulent misrepresentations. They argued that it would be an abuse of process for the dishonest solicitor to contest those findings. Jonathan Parker LJ said at [146] that when considering whether the Bairstow conditions are satisfied:
“the court should be slower in preventing a party from continuing to deny serious charges of which another court has previously found him guilty than in preventing such a party from initiating proceedings for the purpose of relitigating the question whether he is guilty of those charges.”
He found that the conditions for the application of the abuse doctrine were not satisfied in that case because it was not unfair to require the claimants to prove a few examples of Mr Simms’ dishonesty rather than seeking to import the entirety of the tribunal’s findings: [149]. He also considered that “right-thinking people” would consider it unfair that Mr Simms be faced with a pleading which sought to import the tribunal findings en bloc. Mr Jowell also relied on a passage in the concurring judgment of Moore-Bick LJ at [168] stressing the need to be “particularly cautious” before finding an abuse of process where the parties were not the same in the two sets of proceedings. Moore-Bick LJ also concluded that it would not be an abuse of process for Mr Simms to contest the findings of the tribunal. Mr Jowell argued that this showed the Court of Appeal applying a more stringent test than the CAT applied in its judgment.
In my judgment there is no substance to this criticism of the CAT’s judgment. The CAT did not simply engage in a broad merits-based approach but fully appreciated that a high threshold needed to be crossed in order for the abuse of process doctrine to apply. The CAT referred to cases emphasising that the situations in which it will be an abuse to litigate an issue which has not previously been decided between the same parties will be “entirely exceptional” (see per Flaux LJ in Kamoka v Security Services [2017] EWCA Civ 1665, [119]) or “rare” (per Lord Hobhouse in In re Norris [2001] UKHL 34 at [26]).
I reject the suggestion that after citing this case law, the CAT in fact applied a lower threshold. The Respondents are right to say that the appropriate, high threshold is inherent in the Bairstow test that proceedings are only regarded as abusive where they create manifest unfairness or bring the law into disrepute. The CAT recognised at [121] to [124] that the doctrine applies only in an exceptional case and went on to consider whether this case was exceptional. There was no error in the test that the CAT applied. This is not cast into doubt by the judgments in Conlon v Simms which was simply an instance where the court arrived at a different outcome when applying the test to very different facts.
So far as Mr Jowell’s criticism of the application of the test to the facts, that is in my judgment misplaced. Ms Demetriou rightly drew our attention to what was said by Thomas LJ in Aldi Stores Ltd v WSP Group plc and others [2007] EWCA Civ 1260 about the role of the appellate court when reviewing a decision on abuse of process. The Court rejected the suggestion that the lower court’s decision was arrived at by the exercise of discretion; although the court must consider many factors, there can only be one correct answer to whether there is or is not an abuse of process: [16]. Nonetheless an appellate court will be reluctant to interfere with the decision of the judge where the decision rests on balancing such a number of factors:
“16. … The types of case where a judge has to balance factors are very varied and the judgments of the courts as to the tests to be applied are expressed in different terms. However, it is sufficient for the purposes of this appeal to state that an appellate court will be reluctant to interfere with the decision of the judge in the judgment he reaches on abuse of process by the balance of the factors; it will generally only interfere where the judge has taken into account immaterial factors, omitted to take account of material factors, erred in principle or come to a conclusion that was impermissible or not open to him.”
In my judgment, the CAT was entirely justified in deciding that it would create great unfairness to the Respondents to have to prove facts that the Appellants have already admitted in the settlement proceedings regardless of the distinction between essential and non-essential facts. The reasons given by the CAT in [128] onwards are good reasons and amply justify its conclusion.
As to whether it would be an affront to most people’s ideas of justice for the Appellants to be allowed to resile from the admissions and to put the claimants to proof of those admitted facts, I agree with the CAT’s conclusion that it would. Those facts found their way into the Decision despite all the procedural safeguards and opportunities for second thoughts that are provided to the addressees by Article 10a of Regulation 773/2004 and the Notice. In considering the answer to that question one must bear in mind also the [141] gateways which allow the Appellants to mount such a challenge to the facts admitted. The Appellants stress that this is not a case where the abuse of process doctrine is needed to prevent the claimants ‘being vexed twice in the same matter”, a factor that Lord Bingham described as part of the public interest underlying the doctrine in Johnson v Gore-Wood. That argument is, however, an aspect of the exceptionality of the application of the doctrine in a case where the parties to the proceedings are not identical. It does not of itself mean that the doctrine cannot apply in these circumstances.
I accept as Ms Ford QC pointed out, that the Commission is still not a court for the purposes of Article 6 ECHR: see Case C-501/11 P Schindler Holding Ltd v Commission [2013] 5 CMLR 39, [34]. The CJEU noted there that the Commission’s role was not itself inconsistent with Article 6 if the addressee has an opportunity to challenge the decision before an Article 6 compliant tribunal. That no doubt explains why the right of appeal exists even for settlement decisions. I do not see that any reasonable person would consider that the vestigial right of appeal that applies to some admitted facts but not others would make any difference. It does not affect the application of the doctrine.
Conclusion
I would therefore dismiss the appeal.
Lord Justice Flaux:
I agree with both judgments.
Sir Geoffrey Vos, Chancellor of the High Court:
Introduction
I agree with Rose LJ’s judgment. I add this short judgment to express in my own words the main reasons why I too would dismiss the appeal. I will adopt the abbreviations used by Rose LJ.
The CAT decided three main legal issues:-
First, that the effect of Article 16 of Council Regulation 1/2003 (“Article 16”) was that the recitals to the Decision that are binding in follow-on proceedings are those which (a) are necessary as an aid to interpretation of the operative part of the Decision, or (b) constitute the essential basis or provide the necessary support for the operative part of the Decision (the “essential recitals”).
Secondly, that the English law principle of abuse of process made the non-essential recitals in the Decision also binding because the Decision was adopted pursuant to the Commission’s settlement procedure, in which the Addressees expressly admitted the facts as outlined in the Decision.
Thirdly, that the Addressees could only contest the non-essential recitals in the follow-on proceedings on one of the grounds set out in [141] of the CAT’s judgment.
Whilst it was only the second legal determination that was the subject of most of the argument, there was some discussion of the correctness of the third point as well.
The CAT said at [67] on the first legal determination above that “if a finding in a decision cannot be challenged in proceedings before the EU Courts, then it would ordinarily be a denial of justice for that finding to be binding in national proceedings”. The Addressees relied on that conclusion in support of their appeal on the second legal determination above.
As Rose LJ has said, the appeal on the second legal determination raised, in effect, the following four issues, which this court has to determine (I have put the reference issue last):
Issue 1: Whether the CAT’s application of the English law principle of abuse of process and/or its determination that the non-essential recitals are binding are inconsistent with Article 16, the right to a fair trial under articles 47 and 48 of the European Charter of Fundamental Rights (“Article 47” and “Article 48”), or the duty of sincere cooperation under article 4(3) of the TEU (“Article 4(3)”).
Issue 2: Whether the CAT was right to hold that the English law abuse of process doctrine was engaged in this case.
Issue 3: Whether in applying the doctrine the CAT (a) failed to apply the necessary “high threshold”, and/or (b) was wrong to hold that that there would be manifest unfairness to the claimants and that the administration of justice would be brought into disrepute.
Issue 4: Whether there should there be a reference to the CJEU.
The Addressees submitted that the application of the English law abuse of process doctrine was contrary to EU law (under Issue 1) and English law (under Issues 2 and 3), and that, even if the court disagreed with their submissions on Issue 1, the question was not acte clair so that a reference ought to be made to the CJEU (under Issue 4).
The Respondents (the claimants in the follow-on proceedings) submitted that the CAT was right for the reasons it gave and that no reference was necessary or appropriate.
Recital (3) to the Commission’s Decision read “[t]he facts as outlined in this Decision have been accepted by [the Addressees] in the settlement procedure”. The question, in effect, is whether the CAT was entitled to hold the Addressees to this admission, insofar as it applied to non-essential recitals as opposed to binding essential recitals.
The CAT laid down at [141] the gateways that were applicable in the event that an Addressee sought to contest a non-essential recital in the Decision. Rose LJ has set these out at [41] above.
Issue 1: Was the CAT’s application of the English law principle of abuse of process and/or its determination that the non-essential recitals are binding inconsistent with Article 16, Articles 47 and 48, and/or Article 4(3)?
The Addressees based their argument on Article 16, which provides that: “[w]hen national courts rule on … decisions … under Article [101] which are already the subject of a Commission decision, they cannot take decisions running counter to [that] decision”.
The argument, in effect, is that because the CAT ruled that only essential recitals are binding in the follow-on proceedings as a matter of EU law, and because the CAT had acknowledged that it would ordinarily be a denial of justice for a finding that could not be challenged in the EU proceedings to be binding in national proceedings, it followed that it would be a denial of justice for non-essential findings in the Decision to be binding on the Addressees in follow-on proceedings. To make the non-essential findings binding on the Addressees would be a breach of their right to a fair trial under Article 47, contrary to the presumption of innocence under Article 48, and a violation of the duty of sincere cooperation under Article 4(3).
Whilst national courts must not violate any provision or fundamental principle of EU law, there are areas in which EU law provides specifically for national law to govern. Follow-on proceedings are such an area. This is made clear, for example, by recital (5) of Regulation 1/2003 saying that the Regulation does not affect national rules on the standard of proof, and the obligations of the courts of Member States to ascertain the relevant facts of a case “provided that such rules and obligations are compatible with general principles of Community law”. Moreover, recital 11 of the Damages Directive says that “[i]n the absence of Union law, actions for damages are governed by the national rules and procedures of the Member States”.
It is against that background that it is necessary for this court to consider whether Article 16 mandates the non-binding nature of non-essential recitals. As it seems to me, Article 16 says nothing on the point. It simply provides that “national courts … cannot take decisions running counter” to Commission decisions.
EU law could, in these circumstances, only require national courts to allow Addressees to challenge non-essential recitals if it would, in some other way, be incompatible with general principles of EU law, such as Articles 47 and 48, Article 4(3) or the principles of equivalence and effectiveness.
It must be emphasised that we are dealing with a settlement decision, where the Addressees have chosen to admit the truth of both essential and non-essential recitals. As the Addressees specifically accepted, whilst they were required in order to utilise
the settlement procedure to admit the “main facts” alleged against them (see paragraph 20(a) of the Notice), they could have refused to admit all the facts in the Decision. Instead, the Addressees signed up to a Decision that provided by recital (3) that “[t]he facts as outlined in this Decision have been accepted by [the Addressees] in the settlement procedure”.
Moreover, the argument that the Addressees had no opportunity to challenge the nonessential findings is falsified by the detailed procedural rules contained in the Notice, allowing the Addressees full access to the Commission’s file and the opportunity to withdraw their settlement submissions at any stage if the statement of objections or decision goes beyond what they are prepared to admit (see, for example, paragraphs 19, 27 and 29 of the Notice).
These points provide, in my judgment, a clear answer to the Addressees’ arguments on EU law. I say nothing about what the position might be if there had been a contested decision. But, in my view, where there is a settlement decision and the Addressees have accepted the facts found by the Commission, there is nothing in EU law that mandates the national court in follow-on proceedings to allow non-essential recitals to be challenged or re-litigated.
None of Articles 47 and 48 or 4(3) is engaged once freely-made admissions have been accepted as they have here. The Notice provides adequate safeguards for those wishing to exit the settlement procedure if there are facts which they do not admit and that they do not want to count against them in follow-on proceedings. No other fundamental principle of EU law is contravened by the national court holding the Addressees to the admissions they have made. Doing so does not, in any sense, run counter to the Decision.
Argument was addressed to the proper interpretation of recital (3) of the Decision. I do not think that the Decision should be construed as a deed. It is clear from recital (3) that the Addressees accepted the truth of the non-essential recitals in the settlement process without any reservation of the right to resile from them in later follow-on proceedings. The Addressees were well aware when they did so that such follow-on proceedings were inevitable.
I would, therefore, reject this ground of appeal.
Issue 2: Was the CAT right to hold that the English law abuse of process doctrine was engaged in this case?
As Daimler submitted, an appellate court should only intervene in a lower court’s finding of abuse of process where the court has “taken into account immaterial factors, omitted to take account of material factors, erred in principle or come to a conclusion that was impermissible or not open to it” (see Aldi Stores Ltd v. WSP Group plc [2007] EWCA Civ 1260, [2008] 1 WLR 748, at [16]). This is a high threshold.
In essence, it was said that the doctrine could not be applied to the non-essential recitals because they were non-binding under EU law, and, therefore, not a ‘final decision’ for the purposes of the doctrine as explained by Lord Diplock in Hunter v. Chief Constable of the West Midlands Police and others [1982] AC 529 (“Hunter”). If
the doctrine were applicable, the requirements for its application had not been fulfilled (see Issue 3).
It is first important to understand why the abuse of process doctrine is relevant at all. The non-essential recitals are, on the CAT’s first legal determination (which was not challenged before us) not binding as a matter of EU law. But EU law does not mandate that they may not, in any circumstances, be binding in follow-on proceedings governed by national law (as I have already held under Issue 1).
The non-essential recitals were freely admitted in the settlement process, but not in the follow-on proceedings before the English court. Had they been, they would have been covered by the strict rules as to withdrawing admissions in paragraph 7 of CPR PD14. Admissions made outside the proceedings themselves are generally admissible in English civil proceedings, but their existence does not prevent the party against whom they are adduced contesting the truth of facts they previously admitted.
Had the CAT, therefore, not held that the doctrine of abuse of process applied, the Addressees could and would have been at liberty to contest some or all the nonessential recitals admitted in the settlement process. If the doctrine applied, that route would not be open to them.
In Hunter at page 541B, Lord Diplock said that there was an abuse of process where a party in subsequent proceedings mounted a “collateral attack upon a final decision against the [party] which has been made by another court of competent jurisdiction in previous proceedings in which the [party] had a full opportunity of contesting the decision in the court by which it was made”. Where the parties to the two proceedings were not the same, “it will only be an abuse of process of the court to challenge the factual findings [in the earlier proceeding] if (i) it would be manifestly unfair to a party to the later proceedings that the same issues should be relitigated; or (ii) to permit such relitigation would bring the administration of justice into disrepute” (Secretary of State for Trade and Industry v Bairstow [2003] EWCA Civ 321 at [38]) (see Issue 3 below).
The question under this issue, therefore, is whether the CAT was right to conclude that the Decision was “a final decision” against the Addressees which had been made
“by another court of competent jurisdiction in previous proceedings” in which they “had a full opportunity of contesting the decision”.
As I have already indicated, the settlement process provides extensive procedural safeguards for the Addressees. They are not forced to make admissions and they may withdraw from the settlement process at any stage. The Addressees could bring a procedural appeal in the event that the Decision or statement of objections did not reflect the admissions they made in their settlement submissions. If the Commission refused to acknowledge this, the Addressees could revert to the contested procedure as required by paragraphs 27 and 29 of the Notice.
The suggestion that the non-essential recitals were themselves unappealable because they were non-binding overlooks the fact that those recitals followed from the Addressees’ own freely made admissions. Moreover, the distinction between essential and non-essential recitals in the Decision is a distinction that relates to the extent to which the Decision binds in EU law. It is not relevant to the English law doctrine of
abuse of process, which does not distinguish between the essential and non-essential parts of a decision. EU law does not mandate that non-essential recitals may not, in any circumstances, be binding in follow-on proceedings governed by national law (as I have already held under Issue 1).
In my judgment, therefore, for the purpose of the application of the English law doctrine of abuse of process, there is no proper distinction to be made between the essential and non-essential recitals. For that purpose, all parts of the Decision can be considered part of a final decision within Lord Diplock’s rubric.
Issue 3: In applying the doctrine, (a) did the CAT fail to apply the necessary “high threshold”, and/or (b) was the CAT wrong to hold that that there would be manifest unfairness to the claimants and that the administration of justice would be brought into disrepute?
I do not accept the submission that the CAT failed to apply the high threshold required in abuse of process cases. In my judgment, the high threshold is inherent in the requirement of “manifest unfairness”.
I am satisfied that the CAT was entitled to conclude, for the reasons it gave, that there would be manifest unfairness to the claimants and that the administration of justice would be brought into disrepute if the Addressees were entitled, in the follow-on proceedings, to contest the admissions they had freely made in the settlement process. The Addressees made their admissions with the benefit of procedural rights and with full knowledge of their potential future significance. Indeed, they had access to the Commission’s statement of objections and to the Commission’s file before commencing settlement discussions. The admissions were made to secure a 10% reduction in the fine (amounting to some €330 million) and the benefit of a shorter decision than would have resulted from a contested procedure.
The CAT was also justified in reaching the procedural conclusion that the nonessential recitals were binding on the addressees unless the [141] gateways were satisfied. Those conditions were appropriately based on the procedural rules concerning admissions found in CPR Part 14. The anomaly that the [141] conditions do not apply to the essential recitals does not invalidate my conclusions. It is a function of the CAT’s first legal determination and the proper meaning of Article 16. It would be contrary to EU law for the [141] conditions to apply so as to permit the Addressees to contest the essential recitals. The question of whether the non-essential recitals are binding in follow-on proceedings, as I have said, is a matter for national law.
Issue 4: Whether there should be a reference to the CJEU?
It follows from what I have already said that I do not think there is any need for a reference to the CJEU. The questions of EU law raised by Issue 1 are acte clair, none of the parties appealed the CAT’s first legal determination, and the main foundation of my decision is a matter of English procedural law.
Conclusion
For the reasons I have given and for those given by Rose LJ, I would, as I have said, dismiss the appeal.