This appeal from the Competition Appeal Tribunal (the Tribunal) is a sequel to a decision of the Court of Appeal (Waller LJ, Richards LJ and myself) [2009] EWCA Civ 434, given on 22 May 2009, by which we held (reversing the decision of the Tribunal) that proceedings brought by the Claimants under section 47A of the Competition Act 1998 (the 1998 Act) had been brought out of time. The Claimants had applied in the alternative to the Tribunal for such an extension, in case of need, and they then pursued that application. The Tribunal (differently constituted) refused the extension sought. The Claimants appeal with permission granted by Etherton LJ. The Defendants have served a Respondent’s Notice contending that the Tribunal had no power to extend the time.

The European Commission investigated a cartel in relation to vitamins for use in animal foodstuffs. On 21 November 2001 it published a decision as a result of the investigation, in which it held that, among others, the First Defendant had infringed article 81 of the EC Treaty by participating in agreements affecting the Community markets for various vitamins. It was ordered to bring the infringement to an end at once, if it had not already done so, and it was subjected to a fine. The decision was announced in a press release issued on the day, but the full text was not generally available until its publication in the Official Journal (OJ) on 10 January 2003. The cartel had operated for 10 years from 1989. The aggregate fines imposed by the Commission were the largest up to that date. The Tribunal referred to the terms of the Commission’s decision as being trenchant and to the conduct of the cartelists as egregious.

BASF appealed to the Court of First Instance against the amount of the fine, but not against the finding of infringement. The fact of the appeal was published in the OJ on 4 May 2002. Eventually the appeal was successful and the fine was reduced (from €296 million to €248 million) on 15 March 2006. There was no further appeal to the European Court of Justice.

The Claimants brought their proceedings on 13 March 2008. The proceedings were commenced in the Tribunal under section 47A of the 1998 Act, which permits a claim to be brought for damages by a person who has suffered loss or damage as a result of the infringement of a relevant prohibition, including a breach of article 81. Such a claim may be brought in the Tribunal if, and only if, a decision, which includes a decision of the European Commission, has already established that there has been a breach of a relevant prohibition.

Because appeals are possible against a relevant decision, section 47A provides that, while a relevant appeal is possible or, if brought, is pending, proceedings may only be brought with the permission of the Tribunal. The effect of the previous decision of this court is that an actual appeal only has this suspensive effect if the appeal challenges the relevant finding of infringement, and does not do so if the appeal is only against the amount of the fine.

For the Claimants, Mr Vajda Q.C., who had not appeared on their behalf previously, contended that this court’s previous decision had been unforeseen and unforeseeable. He said that it had had the unpredictable effect of cutting the Claimants’ time for bringing their claim down by over 4 years, from May 2008 to January 2004, and of doing so at a time when the relevant period had long since expired. He argued that there was power to extend time under the Competition Appeal Tribunal Rules (the Rules), and that European Community law required that that power be exercised in his clients’ favour. Further he contended that, if the Rules did not confer such a power, the frustration of the Claimants’ legitimate expectation that the period during which they could bring their proceedings ran until 2008, together with the European doctrines of legal certainty and effectiveness, meant that the Tribunal ought to be treated as having such a power and required to exercise it in the Claimants’ favour.

Thus, the issues for decision are:

Is there a power under the Rules to extend time for bringing proceedings under section 47A on their true construction?

If there is no such power on the construction of the Rules, should such a power be treated as existing in the present case, by reason of European Community law principles?

If there is such a power, was the Tribunal’s refusal to exercise it in the Claimants’ favour a proper exercise of their discretion, either (a) as a matter of UK law, or (b) having regard to principles of European Community law?

Mr Vajda pointed out that the circumstances of the present case are unusual in that, from May 2009 onwards, parties will have known how section 47A worked in terms of the effect of an appeal against an infringement decision, and would have no good reason for not bringing a claim under section 47A within time. By contrast, he argued (not in these terms) that the goal posts had been moved by the Court of Appeal’s previous decision to the Claimants’ disadvantage and that it should be regarded as entirely unacceptable, as a matter of European Community law, even if not of UK law, that (to adopt a different metaphor, also not one used by him) the rug should be pulled from under his clients’ claim by a finding that there was no power to extend time.

I have referred to UK law, because the 1998 Act extends to the UK as a whole, as does the jurisdiction of the Tribunal. However, except when the law of another part of the UK is relevant for a particular point, I may refer hereafter to English law and procedure. Article 81 of the EC Treaty has now been replaced by article 101 of the TFEU, but I will ignore that nominal change and others arising from the Lisbon Treaty.

A person who suffers loss as a result of the operations of a cartel conducted in breach of article 81 (or, in UK terms, of the Chapter I prohibition in the 1998 Act) may recover damages on the basis of a breach of statutory duty. Before section 47A came into force (on 20 June 2003), this remedy could only be sought in ordinary proceedings in the High Court. As such, a limitation period of 6 years would apply to it, under section 2 of the Limitation Act 1980. The period would run from the date on which loss was suffered, but the start of the period would be postponed if (over-simplifying) material facts were deliberately concealed by the Defendant, as they are likely to have been in the case of a cartel, which of its nature is secret. In such a case the time runs from the date on which the Claimant knew or ought to have known of the material facts. In practice the relevant date for the start of the period may well be the date on which a decision by a regulator (such as the European Commission) about the cartel is first published.

Different periods of time for such claims apply in different systems of law within the European Union; in Scotland, as I understand it, the period is 5 years. It is for national law to specify the regime for recovery of compensation for a breach of article 81, including any time limit. European law demands that an aggrieved party be given an effective remedy but does not prescribe how that is done.

The Enterprise Act 2002 (the 2002 Act) created an alternative remedy for a Claimant seeking damages for breach of article 81, by introducing section 47A into the 1998 Act. This procedure is available when there has been a relevant decision that article 81 or the Chapter I prohibition has been infringed. A relevant decision is one reached by the European Commission, or by the OFT (or a sectoral regulator) or on appeal by the Tribunal. This type of claim is known as a follow-on claim, as distinct from the normal procedure, referred to as a stand-alone claim. It is an alternative to a stand-alone claim: subject to the relevant provisions being satisfied, a Claimant may pursue either according to its choice.

There are special rules about bringing a follow-on claim under section 47A, to some of which I have alluded. There must first have been a relevant decision. Because such a decision is subject to potential appeals, a claim may be brought, but only with the permission of the Tribunal, if the time for an appeal has not yet expired or if an appeal is pending. The time limit for such a claim is laid down by the Rules, and is two years from the later of (a) the end of the period during which the permission of the Tribunal is needed for the proceedings to be brought, and (b) the date on which the cause of action accrued. Leaving aside the case, perhaps unusual, where the damage is suffered rather late in the day so that (b) may apply, it may be a matter of chance whether the end of the two year period running under alternative (a) comes before or after the end of a period of 6 years (or 5 in Scotland) after the publication of the relevant decision. The respective limitation periods for a stand-alone action and a follow-on action apply independently of each other.

I note in passing that in 2008 the European Commission published a White Paper about damages actions for breach of the EC antitrust rules: COM (2008) 165. It set out a number of proposals to address the obstacles to effective antitrust damages actions. Paragraph 2.7 discussed limitation periods. The Commission suggested that a new limitation period of at least 2 years should start once the infringement decision on which a follow-on claimant relies has become final. The UK legislation, therefore, already meets the requirements of this proposal.

The Limitation Act 1980 does not provide for any possibility of extending the 6 year time limit for a stand-alone claim of this kind, once the period has run. It does include certain provisions under which a limitation period does not start to run until a later date than normal, such as those dealing with deliberate concealment and fraud (section 32). It also makes some provision for a general discretion to extend time, such as in section 33, but that applies to claims for personal injury only. No such provision exists for a claim of the kind with which this case is concerned.

What is said, however, is that the Rules create a power for the Tribunal to extend time for bringing such a claim. In the decision now under appeal this was assumed, because a previous decision of the Tribunal had held that there was such a power: see Emerson Electric Co v Morgan Crucible Co plc [2007] CAT 28, known as Emerson I. Mr Brealey Q.C. for the Defendants reserved the right, which he now exercises, to challenge that ruling on appeal.

The Tribunal was created by the 2002 Act. Previously there had been a similar but differently named body, the Competition Commission Appeal Tribunal, under the 1998 Act. It dealt only with appeals against regulatory decisions of the Director General of Fair Trading or other sectoral regulators.

The 2002 Act not only created the Tribunal as it now stands, but gave it jurisdiction in relation to two classes of case in addition to regulatory appeals: proceedings by way of review of merger decisions under section 120 of the 2002 Act, and follow-on claims for damages under section 47A. A follow-on claim, of its nature, depends on there having been a regulatory decision, but it is of a different nature from regulatory appeals and reviews in merger cases, because under section 47A the Tribunal has, in effect, a first instance jurisdiction similar and parallel to that of the High Court.

Schedule 4 to the 2002 Act deals with the procedure of the Tribunal. In Part 2 it deals with the content of Tribunal rules. Relevantly, in paragraph 11, it says this:

Thus the Tribunal rules may, specifically, deal with time limits applying to follow-on claims and, generally, allow the Tribunal to extend time for bringing proceedings in relation to “any particular proceedings”. The Rules could, therefore, have given the Tribunal power to extend time for a follow-on claim; a rule which did so would not be ultra vires. The question is whether the Rules as made did have that effect.

Mr Brealey showed us, as part of the context, the previous rules which had applied to the proceedings of the Competition Commission Appeal Tribunal. For my part I do not find those relevant or helpful.

The Rules are set out in five Parts. Part I is introductory and general. Part V is supplementary and general. Both apply to all proceedings before the Tribunal: see rule 3(a). Part II applies to all such proceedings except as otherwise provided. Part III applies to proceedings for a review or an appeal against penalties under the 2002 Act. Part IV applies to claims for damages.

Part II deals in terms with appeals. Rule 8 deals with the time and manner of commencing appeals. Rule 8(1) and (2) are as follows:

In relation to appeals, the timing for the defence is also separately covered by rule 14(1), by which the respondent is to send to the registrar a defence “within six weeks (or such further time as the Tribunal may allow)” of a given date.

Rule 19 is under the heading Case management, which applies to rules 19 to 24. Its own heading is Directions. Rule 19(1) and (2)(i), on which the Claimants rely, are as follows:

Rule 19(2) as a whole deals with a wide variety of case management questions. Besides paragraph (i), which I have just quoted, it covers these matters: (a) the manner in which the proceedings are to be conducted, including time limits at the oral hearing; (b) filing of reply, rejoinder or other pleadings; (c) skeleton arguments; (d) requiring persons to attend and give evidence or produce documents; (e) as to the evidence to be required or admitted; (f) as to submission in advance of witness statements or expert reports; (g) as to examination or cross-examination of witnesses; (h) “as to the fixing of time limits with respect to any aspect of the proceedings”; (j) to enable disputed decisions to be referred back to the decision-maker; (k) as to disclosure or production of documents; (l) as to expert witnesses; (m) for the award of costs and expenses; and (n) for hearing a person who is not party where an order or direction is proposed in relation to such a person.

Part III of the Rules applies to reviews of merger decisions and appeals against penalties under the 2002 Act. Parts I and II are applied to such proceedings except as otherwise provided. Rules 26 and 27 set out time limits for proceedings for a review under two different provisions of the 2002 Act: sections 120 and 197. In the case of section 120 the rule requires the application to be made:

Under rule 27 the time limit provision for section 197 is in the same terms but with a period of two months instead of four weeks. Rule 28 applies rule 8 to such cases, with necessary modifications, and applies rule 8(2) expressly to both of rules 26 and 27.

Rule 29 provides for appeals against a penalty imposed by a notice under section 112(1) or against a decision under section 112(3) of the 2002 Act. It is unnecessary to go into the details of how such an appeal may arise. The rule specifies a time limit of 28 days from a given point in time. The Rules do not apply rule 8(2) to such appeals in terms. It is unnecessary to decide the position but it seems to me that, since they do not exclude it, it seems to me that rule 8(2) would apply to these appeals as it does to those governed by Part II. The need to apply it expressly to a review under rules 26 and 27 arises because those proceedings are not appeals. Rule 14 is modified (as regards the time allowed for a defence) both for review proceedings under rules 26 and 27 and for penalty appeals under rule 29.

Part IV, applying to damages claims, starts with rule 30. This applies (among others) the rules set out in Part II to such claims with the exception of rules 8 to 16. Thus rule 19 does apply to proceedings for damages, but rule 8(2) does not.

The time limit for bringing proceedings in the Tribunal is laid down by rule 31, which is as follows:

Rule 31(4) shuts out a claim if proceedings in a court would be precluded by a limitation period which had expired before section 47A came into force.

Thus, appeals under Part II have their own time limit and an express power to extend the time under rule 8, but only in very limited circumstances. Review applications under Part III have different time limits, but also have the same power to extend time, and appeals against penalties under Part III probably do as well. Damages claims under Part IV have a quite different time limit, and no express power to extend time. Rule 8 is excluded from application to such claims. So, Mr Brealey submitted, the plain and simple position is that there is no power to extend the time limit under rule 31, any more than there is any relevant power to extend the limitation period for a stand-alone claim under section 2 of the Limitation Act.

Rule 44 of the Rules applies in terms to claims for damages, and refers specifically to rule 19:

We were shown not only the Tribunal’s decision in Emerson I but also some other cases with a less direct bearing on the point at issue. It is clear that the Tribunal has taken a restricted view of its power to extend time under rule 8(2). Thus in Hasbro UK Ltd v Director General of Fair Trading [2003] CAT 1 Sir Christopher Bellamy, the then President of the Tribunal, sitting alone on an application to extend time for an appeal against a finding of breach of the Chapter I prohibition (the two months not having quite expired), refused to grant an extension. At paragraph 19 he said:

Rule 6(3) of the previous rules was in corresponding terms to the present rule 8(2). It was no doubt foreseen that the rules about appeals would be substantially the same as the then existing rules. I do not know what was known on 24 January 2003, the date of that decision, about the terms of the proposed rules relating to follow-on claims. As a general statement of policy, I can accept that what the President said is still relevant. It is not a decision about time limits for damages claims as such.

We were also shown the decision of Sir Christopher in Prater Ltd v OFT [2006] CAT 11 and that of Mr Justice Barling, now the President of the Tribunal, in Fish Holdings Ltd v OFT [2009] 34, as examples of a strict approach to what count as exceptional circumstances. In Prater Sir Christopher said, on the subject of compliance with the rule 8 time limit: “It is imperative that the present rules be strictly observed.”

Emerson I was, as a matter of decision, a ruling on whether the time limit under rule 31 had begun to run at all. In that case there had been a Commission investigation and a ruling of infringement against a number of participants, including Morgan Crucible. Some of those appealed to the CFI against the finding of infringement, but Morgan Crucible did not. The Claimants started proceedings against Morgan Crucible, more than 2 years after the decision as to infringement against them (plus the period allowed for appealing against it), but while the appeals by others were still pending. The Tribunal held that the appeals by other parties prevented the time from starting to run under section 47A and rule 31, so that permission was required for the claim to be brought, under rule 31(3). (Permission was later given, in a separate decision.) The Tribunal also dealt with the alternative arguments, which included an argument as to whether the Tribunal had power to extend time for a claim. It held that rule 19(2)(i) conferred such a power, because of its generality, and on the basis that case management referred to all matters arising from the moment of the submission of a claim to the Tribunal Registry, and continuing until the final determination of the proceedings. The Tribunal was, of course, aware of the different provision under rule 8; Prater was cited to the Tribunal and by the Tribunal in its decision. The Tribunal commented that:

The Claimants brought proceedings in 2004 against other members of the cartel, who did not appeal against the Commission’s decision. These proceedings were eventually settled. The Tribunal thought that these proceedings had been brought after the two-year period had expired (see paragraph 23) but it became common ground before us that this was not correct. Those claims were brought just within the two-year period.

A separate point arose in those proceedings as regards the joinder of other parties as possible claimants, to whom it was said that the original claimants had sold the relevant businesses. The Tribunal ruled that the proceedings could be amended to include these additional parties, even though the time for a fresh claim on their part had expired: [2005] CAT 1. However, this was on the basis of rule 35 of the Rules which is in general terms and is not constrained by section 35 of the Limitation Act 1980 about new claims after a limitation period has run. That decision casts no light on the present issue.

I approach this point as a matter of the proper construction of the Rules. I will come to Mr Vajda’s arguments based on European law later.

In favour of a power to extend, Mr Vajda pointed to the completely general terms of rule 19(2)(i) – “any time limits” – and to the fact that the primary legislation was itself wide enough to allow an extension of time for a claim under section 47A; see paragraph 11(2)(b) of Schedule 4 to the 2002 Act.

He also submitted that a claim under section 47A is very different, and so is the regime under rule 31, from the position in respect of stand-alone claims. This is so, he argued, because of the restriction on bringing proceedings early, for which permission is required. I agree that this feature differentiates a follow-on claim from any ordinary claim. However, I do not see that the position is, in other respects, any different from an ordinary claim. It is natural that there should be a time-limit on bringing follow-on proceedings. The rules provide for it, under Schedule 4 paragraph 11(2)(a) of the 2002 Act. It operates in just the same way as a limitation period under the Limitation Act 1980.

Against such a power Mr Brealey pointed, first, to the unequivocal terms of rule 31, and the absence of any provision specific to that rule allowing for an extension. He also relied on the fact that rule 19 deals with case management and is set out in Part II, dealing with appeals. In the context of appeals he contended that it cannot operate as a general power to extend time, because a strictly limited power is created by rule 8(2). Therefore, as regards appeals, rule 19(2)(i) applies to all time limits other than that for commencing the appeal (and for filing a defence, also covered by a specific provision in rule 14). Why should it apply to a commencement time limit in Part IV when it cannot do so in its original home in Part II, he asked rhetorically. In Part II it clearly is concerned only with case management time limits. The same is true of its application by reference to Part III. Logically, he argued, it must be the same in Part IV. If a power to extend time under rule 31 was to be conferred, it would be very odd to find that it was done not by a case-specific rule such as rule 8(2) but rather by a entirely general rule, whose main purpose (and within Part II its only purpose) is plainly different, namely to deal with time limits arising once proceedings have been commenced.

It is also noteworthy that the power conferred by rule 19(2)(i) is not only to extend but also to abridge any time limits. It could not be supposed that the Tribunal could abridge the time limit under rule 31, so as to shorten the time within which proceedings can be begun. Moreover, paragraph (i) follows on immediately after paragraph (h) which gives the Tribunal the power to give directions as to the fixing of time limits with respect to any aspect of the proceedings. That, plainly, does not apply to fixing a time limit for bringing the proceedings, which is governed by rule 31.

In my judgment Mr Brealey is right in his contention that rule 19(2)(i) does not give power to extend time for the bringing of proceedings under section 47A. The time afforded by rule 31 is definitive, just as is the time afforded for a stand-alone action by section 2 of the Limitation Act, subject to the suspensive effect of section 32 in cases of concealment. There is no general power to extend time under section 2, and likewise there is no general power to extend time under rule 31.

With respect to the Tribunal which decided otherwise in Emerson I (on a point not necessary to their decision) I cannot agree with their conclusion. Although they were of course aware of rule 8(1) and (2), it does not seem from the summary of the rival submissions at paragraphs 110 to 122 that the argument was put in quite the way that it has been before us. In relation to the point made by them at paragraph 130, quoted at paragraph [38] above, I do not regard rule 8(2) as limiting the exercise of a power conferred by rule 19(2)(i), but as setting out a limited power itself.

Rule 19(2)(i) confers a power to abridge and to extend time limits. It seems to me that it does not confer a power to extend the time limit under rule 31. Three principal reasons persuade me of that position:

Although the Act does allow rules to be made which allow for extensions of time, even of the time for commencing proceedings, I would expect that a power to extend time under rule 31, which creates a limitation period for a follow-on claim which corresponds to that imposed by section 2 of the Limitation Act for stand-alone claims, would be created in express and specific terms, just as the power to extend time in rule 8(2) is specific, both in its application to rule 8(1) directly and in its referential application to rules 26 and 27. In particular, if a power to extend time after the period prescribed by rule 31 has expired was to be created, I would expect to find it in rule 31 itself. Rule 31(3) gives the Tribunal power to permit proceedings to be brought early. If the Tribunal was also to have power to allow proceedings to be brought late it would be natural and logical not only for it to be conferred in specific terms, but also to set the relevant provision out as a sub-paragraph of rule 31.

Rule 19(2)(i) does not operate so as to confer power to extend time for bringing proceedings under Part II, in which it is found. In that Part it plainly refers to case management directions, not to allowing proceedings to be brought later than they otherwise should be.

The terms of rule 19(2)(i) do not read as if they were intended to create a general power to extend time for bringing proceedings, rather than a case management power. In particular the immediate context of paragraph (h) about fixing time limits, and the reference in paragraph (i) to abridging time, both clearly point away from the possibility that rule 19(2)(i) applies to the time limit under rule 31.

For those reasons I disagree with the decision of the Tribunal in Emerson I and would hold that there is no power under the Rules to extend the time for bringing a follow-on action. Although the existence of a power to extend time was mentioned in passing in the judgment of Richards LJ in the previous Court of Appeal judgment (in paragraph 10, and in the last sentence of paragraph 31), the existence of that power had not been debated, and it was not necessary to the decision, so those observations have no binding effect. They were based on the Tribunal’s decision in Emerson I.

Preparing to make a claim under section 47A is by no means a light or easy task. It cannot be done in a very short space of time. The Claimants may very well have faced serious difficulty in preparing documents compliant with the Rules for a claim against BASF at the same time as they were preparing their other claims in 2003 and early 2004. The Rules do not provide for the issue of a generally endorsed Claim Form, with details to follow. On the contrary rule 32 is specific and demanding about what needs to be contained in it.

We had no argument about this aspect, and I would not wish to prejudge what view the Tribunal might take. However, on the basis that there is no power to extend time for a claim under section 47A, if, for legitimate reasons, a party found itself needing to bring proceedings against a cartelist with very little time in which to prepare the documents required by rule 32, it may be that the party could issue a claim form in short form, and then seek directions permitting it to be supplemented or amended. That would be somewhat comparable to the procedure envisaged by the President in Hasbro, at paragraphs 21 to 25, although of course the rules as regards appeals are different from those for follow-on claims. Moreover, with the benefit of clarification of the position in this and the previous decision of this court, parties ought not to be under a misapprehension as to how the time limits apply to their particular facts.

Mr Vajda contended that, if the Rules on their proper construction as a matter of English law do not confer a power to extend time, then they failed (at any rate in the particular circumstances of his clients’ case) to satisfy one or more of several requirements of European law or of the ECHR: that of legitimate expectation, that of legal certainty, that of effectiveness and, in terms of human rights, that of proportionality.

He asserted that the Court of Appeal’s previous finding, that an appeal only suspended the running of time if it was an appeal against a finding of infringement, not if it was only against the fine, was not foreseen or foreseeable. Accordingly, he said, the effect of the Court of Appeal’s earlier decision was retrospectively to shorten the time available for bringing a follow-on claim, with the effect that the claim, thought to be comfortably in time when commenced, was in fact out of time by more than 4 years.

As I have mentioned, the Claimants did start follow-on proceedings within time against other members of the cartel who did not appeal. The Claimants’ solicitor made a witness statement in relation to the application before the Tribunal in which he summarised the circumstances at that time. His firm was instructed late in 2003 in order to pursue claims on the part of the Claimants for losses arising from the vitamins cartel. He said at paragraph 17:

He said that the two claims they did bring required enormous effort to bring together, from a standing start in less than 2 months, including the Christmas and New Year break.

The Claimants did not disclose any more detail of the advice they received at that time as regards a claim against BASF. It seems unlikely that the advice was in fact that they could not bring proceedings against BASF at that time. For Counsel to have said that he or she would have had to have ignored the words “otherwise than with the permission of the Tribunal” in section 47A(5)(b), the words “without permission” in section 47A(7) and (8), and rule 31(3). On the basis of the Tribunal’s view of the suspensive effect of an appeal, the correct advice would have been that proceedings could be brought against BASF but that permission was required from the Tribunal for it to be done at that stage. I imagine that that is the advice that was given.

At the end of 2003 and the beginning of 2004 there was no authority as to the effect of the suspensive provisions of section 47A and rule 31. Parties and their lawyers had to work out the position by reference to the words of the legislation alone. In Counsel’s written submissions to the Tribunal on the Claimants’ application for an extension of time, it was argued that it would be just for time to be extended:

Mr Vajda (not the author of those submissions) disavowed the idea, implicit in that proposition, that two views could reasonably have been held on the point. He contended that, until the point was taken in paragraph 18 of the Defence, dated 23 April 2008, the point had not occurred to anyone, or at any rate had not been referred to in public.

He showed us the Tribunal’s Guide to Proceedings, published in 2005. In paragraphs 6.68 to 6.71 the Guide deals with the time limit for a claim under section 47A. It describes the suspensive effect of an appeal, and does not distinguish between appeals on infringement and appeal only on penalty. The Guide, published in October 2005, plainly had no effect on the decision of the Claimants or the understanding of their lawyers in January 2004 as to how the time limit provisions worked. The status of the Guide is that of a Practice Direction (under rule 68(2)) in relation to the procedures provided for by the Rules. The introduction states that it is intended to give practical guidance for parties and their lawyers as to the procedure of the Tribunal in relation to the cases which the Tribunal can entertain. It is not a substitute for the Tribunal’s rules, and, though it does not say so in terms, it is all the more not a substitute for the substantive law in the relevant Acts and the Rules. That being so, it does not seem to me that any inference can properly be drawn from the fact that the passage in the Guide to which I have referred does not go into detail as to the suspensive effect of appeals. By way of comparison, Professor Whish’s book on Competition Law, up to date as at March 2008, deals with follow-on claims at pages 307 to 309, and speaks in general terms of the “relevant date”, from which the two year period runs, as being usually the date of the decision plus any appeal period, with a footnote reference to Emerson I.

Emerson I, of course, itself held that an appeal about infringement by any of the cartelists suspended the time for bringing proceedings against any other member of the cartel even if that member had not appealed. No special point about an appeal on penalty alone arose in that case. Mr Vajda also relied on the decision of the Tribunal at the first stage of the present litigation at [2008] CAT 24, and what he said were the robust terms in which permission to appeal was refused by the Tribunal in that case: “the Tribunal unanimously reached a firm conclusion” that Mr Brealey’s arguments were not correct: see [2008] CAT 29 paragraph 9.

The previous decision of this court was reached on a reading of the 1998 Act and the Rules. The material which led us to the conclusion we reached was, therefore, available to every person interested in the point from the outset. I dare say that no-one had focussed on the potential difference between an appeal only on penalty and an appeal against the decision as to infringement. In January 2004 this was relatively new legislation, and it would not be surprising if some points were not spotted at first. I find it difficult to accept Mr Vajda’s proposition that the Court of Appeal’s previous decision was not foreseeable in late 2003 or January 2004, at the time when the Claimant had to consider (and, it seems, did consider) whether to bring proceedings against BASF as well as against others.

His first argument was that the previous Court of Appeal decision, which held that the Claimants’ proceedings brought in March 2008 were four years out of time, offended against the European principle of effectiveness. This principle is that the national law of a Member State must not make it virtually impossible or excessively difficult to exercise a right conferred by Community law. But national law in this case provides for a clear two year period from a defined date within which to bring a follow-on claim. That period is the same as that which the Commission recommended in its White Paper, referred to at paragraph [14] above. It is true that, because the 2002 Act came into force after the Commission’s decision in this case, the Claimants did not have the full two year period, but since that Act gave them an additional remedy, they can hardly complain of that. Moreover, they were in fact able to bring their proceedings against other cartelists within time. Nor is it clear from the evidence why their present solicitors were only instructed in November 2003. It seems to me that these arguments on Mr Vajda’s part on the principle of effectiveness are not justified in the facts.

In relation to the same principle, and the principle of legal certainty, he also showed us two recent decisions of the ECJ in the realm of public procurement contracts: Uniplex (UK) Ltd v NHS Business Services AuthorityCase C-406/08 and Commission v Ireland, Case C-456/08, both delivered on 28 January 2010. Directive 89/665 required Member States to take the measures necessary to ensure that decisions taken by contracting authorities in relation to public procurement contracts may be reviewed effectively and as rapidly as possible on the grounds of infringement of Community law in the field of public procurement, or national rules implementing that law. The provision to be made had to include power to award damages to persons harmed by an infringement. UK legislation intended to implement this Directive provided that proceedings under the implementing regulation must be brought “promptly and in any event within three months from the date when grounds for bringing the proceedings first arose” subject to a power to extend time.

Among the questions referred to the ECJ were two on this provision: one about the requirement to bring the proceedings promptly and the other about the power to extend time. The relevant passage in the court’s judgment is at paragraphs 39 to 50. The salient points for present purposes are these.

On the first aspect, Member States have an obligation to establish a system of limitation periods that is sufficiently precise, clear and foreseeable to enable individuals to ascertain their rights and obligations. The objective of rapidity does not permit Member States to disregard the principle of effectiveness, under which the detailed methods for the application of national limitation periods must not render impossible or excessively difficult the exercise of any rights which the person concerned derives from Community law. A national provision under which proceedings must not be brought ‘unless … those proceedings are brought promptly and in any event within three months’ gives rise to uncertainty. A limitation period, the duration of which is placed at the discretion of the competent court, is not predictable in its effects. Consequently, a national provision providing for such a period does not ensure effective transposition of Directive 89/665.

On the second aspect, the power to extend time, if the national provisions at issue do not lend themselves to an interpretation such that the period begins to run only from the date on which the claimant knew, or ought to have known, of the relevant infringement, the court is bound, in exercise of the discretion conferred on it, to extend the period for bringing proceedings in such a manner as to ensure that the claimant has a period equivalent to that which it would have had if the national legislation had provided for the period to run from that date. In any event, if the national provisions do not lend themselves to such an interpretation, the national court must refrain from applying those provisions, in order to apply Community law fully and to protect the rights conferred by the Directive on individuals.

It seems to me that the issues in the present case are altogether different from the points there discussed. There is no question here of a limitation rule which is intrinsically uncertain, such as a requirement to start proceedings “promptly”. The fact that the Court of Appeal came to a different conclusion from that of the Tribunal as to the effect of section 47A and rule 31 on the facts of this case does not make the provision uncertain, for this purpose, nor does it infringe the principle of effectiveness. Otherwise no provision would be valid unless all courts or tribunals which had to consider it always came to the same conclusion as to its meaning and effect. Nor is it open to the Tribunal or the court to refrain from applying the provisions of rule 31.

The point in Commission v Ireland was somewhat different. The Irish rules relevant to claims arising from public procurement disputes required that “an application for the review of a decision to award, or the award of, a public contract shall be made at the earliest opportunity and in any event within three months from the date when grounds of the application first arose” subject to power to extend for good reason. Two bidders for a contract, A and B, were invited to submit their best offers in August 2003. On 14 October 2003 A was told that the relevant authority had decided to designate B as the preferred tenderer, but without rejecting A’s offer. In December 2003 the authority decided to award the contract to B. On 5 February 2004 it signed a contract with B. In April 2004 A brought an action for damages, on the basis that the time ran from 5 February 2004. The High Court held that time had started to run in October 2003, and that the proceedings were therefore out of time. The Commission brought proceedings against Ireland for a declaration that, in various respects, the Irish national law did not fulfil Ireland’s obligations under the Directive. Among other points, it argued that the national legislation gave rise to uncertainty as to what decision must be challenged by legal proceedings, and that the legislation was unclear as to how periods within which proceedings were to be brought were to be determined. The ECJ held that the Directive had not been properly implemented because the national legislation did not make it expressly clear that the period laid down applied to a review of interim decisions taken by contracting authorities. The court’s conclusion on this was set out at paragraph 66:

The legislation was also found to be at fault because of the uncertainty of the time limit, corresponding to the decision about the English “promptly” requirement.

The issue in the present case is not about implementation of a Directive, but about the principle of effectiveness as regards national remedies for breach of obligations imposed by Community law. It does not seem to me that these two decisions under the Directive relating to public procurement add anything useful to the general proposition as regards effectiveness. I have already addressed that and found that the 2002 Act and the Rules do comply with the general obligations of Community law in this respect.

On a different point, Mr Vajda showed us Marks & Spencer plc v Commissioners of Customs & ExciseCase C-62/00. That case arose because of a change in the legislation as regards the liability of the Commissioners to repay VAT wrongly paid to them. Originally this liability had been subject to a six year time limit. In 1996 it was announced that this was to be changed to a three year time limit, with immediate effect and with no transitional arrangements. Thus any taxpayer who could, immediately before that announcement, have recovered overpayments made more than three but less than six years previously, at once lost the right to make such a claim. The ECJ held that a three year time limit was, in itself, reasonable and did not infringe the principle of effectiveness, but nevertheless held that the immediate change with no transitional protection infringed Community law.

The court dealt with the point at paragraphs 38 to 40. Paragraph 38 deserves quotation in full:

In paragraphs 39 and 40 the court noted that Member States are required as a matter of principle to repay taxes collected in breach of Community law, and acknowledged that, by way of exception to that principle, fixing a reasonable period for claiming repayment is compatible with Community law, this being in the interests of legal certainty. However, it said that, in order to serve their purpose of ensuring legal certainty, limitation periods must be fixed in advance. Accordingly, legislation which has retroactive effect so as to deprive individuals of any possibility of exercising a right which they previously enjoyed with regard to repayment of VAT collected in breach of provisions with direct effect must be held to be incompatible with the principle of effectiveness.

Moreover, the court also held that the retrospective change infringed the principle of the protection of legitimate expectations. Between paragraphs 44 and 46 it noted that it had consistently held that the principle of the protection of legitimate expectations forms part of the Community legal order and must be observed by the Member States when they exercise the powers conferred on them by Community directives. In particular, it had held that a legislative amendment retroactively depriving a taxable person of a right to deduction derived from the Directive as to VAT is incompatible with the principle of the protection of legitimate expectations. Similarly, in a situation such as that in the main proceedings, the principle of the protection of legitimate expectations applies so as to preclude a national legislative amendment which retroactively deprives a taxable person of the right enjoyed prior to that amendment to obtain repayment of taxes collected in breach of provisions of the Directive with direct effect.

These principles seem to me to have nothing to do with the present case. The unfortunate position in which the Claimants found themselves as a result of the Court of Appeal’s previous decision was not the result of any act by the Government or Parliament. It was the result of this court coming to what it considered to be the correct view of legislation which has been in the same terms at all material times. The risk that the Claimants faced, when they decided not to start proceedings against BASF at the same time as their proceedings against others, was identifiable from the legislation, even if it was not in fact identified. The limitation period was fixed in advance, by rule 31. I do not accept Mr Vajda’s argument that the interpretation of the rule and the section which we found to be correct in the previous decision was neither clear nor foreseeable in late 2003 and at the beginning of 2004. Nor can I accept that the Claimants had a legitimate expectation at that time that they would be able to bring their proceedings at any time up to two years after the eventual decision on BASF’s appeal. With advice they may have formed the view that this was the position, but nothing other than the words of the legislation contributed to that view at that time.

Mr Vajda mentioned the ECHR concept of proportionality in support of his submissions but did not articulate any particular separate point on this basis, and I do not need to deal with it separately.

For the reasons which I have set out above, I would reject Mr Vajda’s arguments that even if not under UK law by itself, his clients were entitled to bring their proceedings when they did (or at least to a properly considered exercise of a discretion as to an extension of time) as a result of the overriding effect of Community law.

It follows that there was and is no power to extend time under rule 31, and the claim was rightly dismissed. It is not necessary to consider the issues arising on the Appellant’s Notice, since they relate to the exercise of a discretion which I hold did not exist. I would dismiss the appeal.

I agree.

I also agree.