Gemalto Holding BV & Ors. v Infineon Technologies AG & Ors.

The central question in this appeal is whether Gemalto’s claims for loss arising from an unlawful cartel in which the defendants, Infineon and Renesas (together the defendants), participated are statute barred. The cartel related to the supply of Smart Card Chips (“SCCs”). The question turns on the proper approach to be adopted to the application of section 32(1) of the Limitation Act 1980 (section 32(1)). Mrs Justice Bacon (the judge) identified two possible tests for the determination of whether the claimants’ knowledge, more than 6 years before the commencement of the proceedings, was sufficient to amount to discovery for the purposes of section 32(1)(b).

The first was the so-called statement of claim test of whether the claimant had, or could with reasonable diligence have, obtained such knowledge as would allow it and its professional advisors properly to plead a claim that would not be liable to be struck out as unarguable or lacking a sufficient evidential basis (see the judgment at [50]). The statement of claim test was applied by the Court of Appeal inArcadia Group Brands v. Visa [2015] EWCA Civ 883 (Arcadia), DSG Retail v. Mastercard [2020] EWCA Civ 671 and by Foxton J in Granville Technology Group v. Infineon Technologies [2020] EWHC 415 (Comm) (Granville) (which was appealed, but not on that point, in OT Computers v. Infineon Technologies [2021] EWCA Civ 501 (OT Computers)).

The second possible test was the so-called FII test adumbrated by the Supreme Court in a case concerning a mistake of law (Test Claimants in the Franked Investment Group Litigation v. HMRC [2020] UKSC 47 (FII)). The FII test suggests that time should begin to run from the point when the claimant knows, or could with reasonable diligence know, that it has a worthwhile claim. The claimant must know about the mistake (or in this case the concealment) with sufficient confidence to justify embarking on the preliminaries to the issue of proceedings, such as submitting a claim to the proposed defendant, taking advice and collecting evidence. The Supreme Court in FII adopted that formulation from Lord Donaldson MR in Halford v. Brookes [1991] 1 WLR 428 at page 443 (Halford).

In this case, the judge decided to apply the statement of claim test (which she acknowledged might in some cases allow time to start to run later than the FII test), because neither side had identified any practical difference that they said would result from the application of one or other of the two tests.

The judge concluded, having applied the statement of claim test, that the limitation period in respect of Gemalto’s pleaded claims had started to run under section 32(1)(b) at least from the end of April 2013. The European Commission (the Commission) had issued a press release on 22 April 2013, referring to having sent a Statement of Objections to participants in the alleged cartel. Gemalto had previously received two Requests for Information (RFIs) from the Commission on 3 July 2012 and 25 September 2012, identifying a time period for their enquiries concerning the alleged cartel between 2003 and 2006. Gemalto had responded to the first RFI on 24 July 2012 and 20 August 2012, and to the second RFI on 9 October 2012. Accordingly, the judge determined that Gemalto could legitimately have pleaded a cartel covering 2003–2006 from the end of April 2013. Since the proceedings were not issued until 19 July 2019, more than 6 years after 22 April 2013, they were statute barred.

Gemalto had made a follow-on claim based on the Commission’s infringement decision (the decision) in Case AT.39574 Smart Card Chips, finding that between 2003 and 2005 various suppliers of SCCs had unlawfully coordinated their pricing behaviour and exchanged competitively sensitive information. The decision was dated 3 September 2014, some 16 months after the date upon which the judge decided that the limitation period had begun. Gemalto submitted to the judge and to this court that the limitation period began to run on the announcement of the decision. If that were right, Gemalto’s claim would have been brought in time.

It is against this background that Mr Jon Turner QC, leading counsel for Gemalto, submitted that the judge was wrong to rely on the Statement of Objections, which Gemalto did not see at the time, as providing a basis for a reasonable belief that there may have been participation in an unlawful cartel. The Statement of Objections was, Gemalto submitted, not, without more, a fact from which it could reasonably infer that a defendant to whom it had been sent was a party to an unlawful cartel that caused it loss. In addition, Gemalto could not have legitimately pleaded a cartel between 2003 and 2006 on the basis of the RFIs alone.

Infineon’s skeleton suggested that FII cast some doubt on the application of the statement of claim test in cases of this kind, because time may start to run when the claimant appreciates it has a worthwhile claim or sufficient confidence to justify embarking on the preliminaries to the issue of proceedings. Infineon did not, however, in its skeleton, object to the judge’s conservative adoption of the statement of claim test. Renesas’s skeleton also did not object to the application of the statement of claim test. Both Infineon and Renesas supported the judge’s reasoning without filing a respondent’s notice.

When the argument in the appeal began, the court queried with counsel whether it was actually appropriate for it to apply the statement of claim test, if, on a proper analysis, the Supreme Court’s reasoning in FII indicated that a different test was appropriate to the proper construction of section 32(1). We mentioned that, whilst FII was a tax case in which the Supreme Court had declined to express a view on cases of concealment, the reasoning of the majority construed the words: “the period of limitation shall not begin to run until the plaintiff has discovered the fraud, concealment or mistake (as the case may be) or could with reasonable diligence have discovered it”. Those words at the end of section 32(1) applied as much to section 32(1)(c) (where the action is for relief from the consequences of a mistake) as to section 32(1)(b) (where any fact relevant to the claimant’s right of action has been deliberately concealed from it by the defendant).

On the first day of argument, it seemed that there was nothing between the parties as to the correct test following FII. Mr Turner submitted that limitation would begin to run only when the claimant recognised that it had a worthwhile claim, and that a worthwhile claim arose only when a reasonable person could have a reasonable belief that (in this case) collusion had occurred or that there was a cartel (see [196] in FII). The respondents accepted that to be the test, but submitted, in effect, that once the Commission had issued its Statement of Objections, it was obvious that a reasonable person could have a reasonable belief that there was a cartel, even if there was always the possibility that the Commission would withdraw its Statement of Objections or change its mind as to the participants. The Statement of Objections was only issued after a detailed investigation by the regulators and constituted the charges to which the respondents were required to produce any answers they might have.

On the second day of the hearing, Gemalto slightly changed its position, arguing in reply that four propositions emerged from FII as follows: (i) the facts which the claimant must discover are those which include each of the essential ingredients of the cause of action (including in this case when the claimant bought the goods in question at distorted prices), (ii) the claimant only discovers these essential facts when it recognises that it has a worthwhile claim in damages and has sufficient confidence to embark on the preliminaries of making a claim, (iii) having sufficient confidence to embark on the preliminaries of making a claim did not mean investigating whether the essential elements of the claim existed; that already had to have been discovered (see (i) above), and (iv) the concept of recognition that there is a worthwhile claim implies that the claimant has a reasonable belief as to these facts, which means that a reasonable person could conclude on the basis of available material that it was more likely than not that the essential ingredients of the tort could be pleaded.

Against that background, I intend to deal with the following issues: (i) what, after FII, is the applicable test to determine when time begins to run in a case where any fact relevant to the claimant’s right of action has been deliberately concealed by the defendant, (ii) whether the judge was right to place reliance on Gemalto’s knowledge of the existence of the Statement of Objections, and (iii) whether Gemalto had sufficient knowledge of the period of the alleged cartel to allow time to run at the end of April 2013.

Before turning to these issues, I will set out section 32, and deal with the common ground between the parties and the basic facts that the judge found. I will also attempt a brief analysis of the essential reasoning in FII.

I should, however, say at once that I have concluded that, applying the FII test relating to the discovery of a mistake to the discovery of deliberate concealment, the conclusion the judge reached was right and that the appeal should be dismissed.

Section 32 provides as follows:

The judge recited the common ground between the parties at [23]-[27] substantially as follows:

The judge summarised the material available to Gemalto at [30]-[39] substantially as follows.

On 7 January 2009, the Commission published the following press-release:

That announcement was picked up by major news outlets, and quickly came to the attention of Gemalto. Internal emails were sent on 7 January 2009, including to and from Gemalto’s CEO, Executive General Counsel and Deputy General Counsel. They forwarded the Commission’s press release or news articles reporting on the announcement, including some that named Infineon as confirming that it had been among the companies that had been raided by the Commission. One email asked senior staff to “use your and KAB [key account buyer] network and fish for serious information on this”, to which the reply was that Infineon had already been caught “red-handed” (“la main dans le sac”)for the same kind of collusion in the DRAM sector. The emails later identified Renesas as having confirmed that it too had been raided by the Commission.

By 16 January 2009, Gemalto considered consulting a law firm with “heavyweight” competition law experience, in case the Commission investigation “turns sour and we consider suing our suppliers”.

The opening paragraph of the first RFI received on 3 July 2012 read:

The second RFI received on 25 September 2012 asked specific questions about particular SCCs supplied by Philips, Samsung, Renesas and Infineon, again covering the period 2003–2006.

The Statement of Objections was announced on 22 April 2013. The Commission’s press release read:

This announcement was reported in the press and discussed in internal Gemalto emails. Bloomberg reports, which were forwarded within Gemalto, on the same day said that Infineon and Renesas had confirmed that they had received the Statement of Objections. On 24 April 2013 the minutes of a Gemalto board meeting recorded:

The Commission’s summary decision and press release were published on 3 September 2014. The press release named the addressees of the decision and the precise duration of the infringement. The summary decision stated that Infineon, Renesas, Samsung and Philips had bilaterally discussed pricing and pre-pricing components for SCCs (such as production capacity and capacity utilisation), future market conduct, and contract negotiations with common customers, and had exchanged competitively sensitive information. The infringing contacts were found to have taken place from September 2003 to September 2005, with Infineon’s participation ending in March 2005, 6 months before that of the other participants.

Some time then elapsed during which the Commission considered the redactions that would be necessary to publish a non-confidential version of the decision. The full non-confidential version was eventually published on 16 December 2016. The decision was the subject of appeals to the European Court of Justice by Philips and Infineon on both liability and the amount of the fine. The findings on liability were upheld, but Infineon’s fine was reduced by the General Court.

The Supreme Court dealt with the section 32 question in FII by reconsidering the correctness of the House of Lords’ decision in Deutsche Morgan Grenfell [2007] 1 AC 558 (DMG). It is not necessary to rehearse the details of either decision, but for these purposes it is sufficient to observe that much of the Supreme Court’s reasoning in FII is expressly and directly applicable to cases of fraud and concealment, even though both DMG and FII themselves were cases about a mistake of law.

At [167], Lords Reed and Hodge cited the key passage from Lord Hoffmann’s speech at [31] in DMG to the effect that the reasonable diligence proviso in section 32(1) depended upon the true state of affairs being there to be discovered. Lord Hoffmann had said:

At [173], Lords Reed and Hodge noted the logical paradox that a claimant could be unable to discover the existence of his cause of action even after he has brought his claim: he cannot discover it until his claim succeeds. At [177], they explained why that could not be right by comparison to the position of claimants in other kinds of case:

Lords Reed and Hodge concluded that part of the discussion at the end of [178] by saying that “[t]his line of thought suggests that the focus of attention under section 32(1) … should not be on judicial decisions, but on the claimant’s ability to discover that he had a worthwhile claim”. This was supported at [179] by the Supreme Court’s view that Lord Hoffmann’s construction of section 32(1) defeated Parliament’s intention in enacting limitation periods. It was a result that Parliament could not have intended.

At [180]-[186], Lords Reed and Hodge explained why Lord Hoffmann’s approach was also inconsistent with the position where there was fraud. They cited [165] of Lord Brown’s dissenting judgment in DMG as follows:

Lords Reed and Hodge cited, in support of their view, Lord Hoffmann’s own words in the Hong Kong Court of Final Appeal in Peconic Industrial Development Ltd v. Lau Kwok Fai [2009] HKCFA 16 at [56] to the effect that a claimant only had to know facts in a fraud case which amounted to a prima facie case in order to set the clock running. They recognised at [185] that the approach in fraud cases differed from Lord Brown’s approach “only in its focus on the date when the claimant … had sufficient material properly to plead a claim in fraud”. Lord Brown’s approach was consistent with that adopted authoritatively in analogous contexts where fraud was not in issue, and was also in accordance with principle.

Lords Reed and Hodge then noted that Lord Brown’s approach was also consistent with the cases concerning other postponements to the running of the limitation period, particularly in relation to personal injury claims. They cited Lord Nicholls at [9] in Haward v. Fawcetts [2006] 1 WLR 682 (Haward), where he approved Lord Donaldson MR’s guidance in Halford at page 443 where he had “noted that knowledge [did] not mean knowing for certain and beyond possibility of contradiction. It [meant] knowing with sufficient confidence to justify embarking on the preliminaries to the issue of a writ, such as submitting a claim to the proposed defendant, taking advice and collecting evidence”. Lords Reed and Hodge noted that that placed the commencement of the limitation period slightly earlier than the fraud cases, which identified the point in time when the claimant could plead a statement of claim. They said that FII was “not the occasion on which to review the formulation used in the fraud cases, which [reflected] the special standards applicable to the pleading of fraud”. They concluded at [191], however, that the formulation in Halford and Haward (and AB v. Ministry of Defence [2013] 1 AC 78) was consistent with Lord Brown.

There followed at [191]-[196] of Lords Reed and Hodge’s judgment an important passage which explained why Lord Brown’s view was consistent with principle. In my view, this part of the judgment applies as much to cases of mistake as it does to cases of concealment. They said this:

That compelling example was followed at [193] by the comment that “[t]he purpose of the postponement effected by section 32(1) is to ensure that a claimant is not disadvantaged, so far as limitation is concerned, by reason of being unaware of the circumstances giving rise to his cause of action as a result of fraud, concealment or mistake”. That purpose was achieved if time runs from the point in time when the claimant knows, or could with reasonable diligence know, that it had been mistaken “with sufficient confidence to justify embarking on the preliminaries to the issue of a writ, such as submitting a claim to the proposed defendant, taking advice and collecting evidence”, or it “discovers or could with reasonable diligence discover [its] mistake in the sense of recognising that a worthwhile claim arises”. There was no substantive difference in the formulations: “each of which is helpful and casts light on the other”.

Mr Turner placed particular reliance on [196] where Lords Reed and Hodge explained that only that approach was consistent with the rationale of limitation periods, because it was in the nature of litigation that facts and law were commonly disputed. Until the court has resolved those disputes “the parties can, at best, have only a reasonable belief that their assertions are correct”:

At [197] to [198], Lords Reed and Hodge also approved Lord Brown’s analogy with the meaning of the word “discovery” in revenue cases. In Earl Beatty v. Inland Revenue Commissioners [1953] 1 WLR 1090 (Earl Beatty) at page 1095, the court had held that “discovery” in that context “need not be a complete and detailed or accurate discovery”. At [201], admittedly in the context of section 32(1)(a) and (c), Lords Reed and Hodge said that, in general terms “the question is not whether the claimant could have established his cause of action more than six years … before he issued his claim, but whether he could have commenced proceedings more than six years before he issued his claim”. That approach was also consistent with the test for whether the claimant “could with reasonable diligence” have discovered a fraud (Paragon Finance plc v. DB Thakerar & Co [1999] 1 All ER 400 (Paragon), Millett LJ at page 418: “[t]he question is not whether the plaintiffs should have discovered the fraud sooner; but whether they could with reasonable diligence have done so”). There is an assumption that the claimant desires to discover whether or not there has been a fraud (Law Society v. Sephton & Co [2005] QB 1013, Neuberger LJ at [116] (Sephton)). Molloy v. Mutual Reserve Life Insurance Co (1906) 94 LT 756 (Molloy) also supported the same approach. Sir Richard Collins MR at page 761 held that the limitation period ran from the time when the plaintiff discovered the facts essential to his cause of action.

Lords Reed and Hodge then summarised their reasoning at [213] before deciding that DMG had been wrongly decided. At [213(14)], they summarised the point they had made at [196] as follows:

In my view, it is undesirable and inappropriate to seek to interpret Supreme Court decisions as a deed. We need, however, to take proper account of the reasoning of the Supreme Court in considering the proper construction of section 32, even if we are concerned with deliberate concealment rather than mistake. Indeed, the parties all accepted that to be the case. Moreover, this court must adumbrate a test which is consistent with the clear reasoning of a majority of the Supreme Court.

I have set out the reasoning of the majority in FII because it is relevant to the question of whether or not the word “discovery” in the proviso to section 32(1) is to be given the same meaning in relation to both sections 32(1)(b) and 32(1)(c). It may be noted at the outset that, whilst Lords Reed and Hodge said that they were not determining the position under section 32(1)(a) in relation to concealed fraud, part of their reasoning alluded to such cases.

The premise of the reasoning in FII was that there was a paradox on the basis of the law in DMG that a claimant might be unable to “discover” the existence of his cause of action under section 32(1) until long after he has brought his claim and succeeded in it. That paradox suggested that the focus of attention under section 32(1) should be on the claimant’s ability to discover that he had a worthwhile claim.

The Supreme Court then went through a further 8 reasons why that was a consistent and appropriate test. Lord Hoffmann’s approach had paid insufficient regard to the principle that legislation should be given a purposive construction. The legislative construction in question was, as I have mentioned, the meaning of the word “discovery”, which defined when the limitation period would start to run. Lords Reed and Hodge regarded Lord Brown’s “worthwhile claim” formulation as consistent (a) with the approach in fraud, even if it was necessary in that situation to be able to plead a statement of claim [185]-[186], (b) with other postponements to the running of the limitation period, particularly in relation to personal injury claims (see Haward and Halford), (c) the purpose of the postponement effected by section 32(1), which was to ensure that a claimant is not disadvantaged, so far as limitation is concerned, by reason of being unaware of the circumstances giving rise to its cause of action as a result of fraud, concealment or mistake, and (d) the rationale of limitation periods that the parties can, at best, have only a reasonable belief that their assertions are correct.

This reasoning and that which I have summarised at [37]-[38] above led to the adoption of the FII test for mistake cases, namely that time runs from the point in time when the claimant has discovered, or could with reasonable diligence have discovered (a) that it had been mistaken with sufficient confidence to justify embarking on the preliminaries to the issue of a writ, such as submitting a claim to the proposed defendant, taking advice and collecting evidence, or (b) its mistake in the sense of recognising that a worthwhile claim arises. The question before us is whether an analogous test is applicable to deliberate concealment cases.

Undoubtedly, the reasoning adopted by Lords Reed and Hodge applies equally to fraud, mistake and deliberate concealment. If there is a difference, it would have to be found, as they have said, in the stricter rules of pleading in fraud, something that is not applicable in competition cases (or to cases of deliberate concealment). Gemalto did submit that an analogy was to be drawn between pleading a cartel case and pleading fraud, such as was found by Snowden J in Federal Deposit v Barclays Bank [2020] EWHC 626 (Comm) about an alleged LIBOR fixing cartel. I do not believe the analogy is a good one on the facts of this case; if anything, the secret nature of a cartel leads to a more liberal approach to pleading in advance of disclosure of the defendants’ materials or the results of the regulators’ investigation (see Nokia v. AU Optronics [2012] EWHC 731 (Ch) at [62]–[69], Bord Na Mona Horticulture v. British Polythene Industries [2012] EWHC 3346 (Comm) at [30]-[31], Granville at [33]–[34], and Males LJ in OT Computers at [60]). The other possible difference is that the event relevant to a section 32 postponement in mistake is based on discovering something affecting how the claimant has acted, whilst a section 32 postponement in deliberate concealment and fraud cases is generally based on discovering conduct by the defendant. That seems to me, however, to be a distinction without a real difference insofar as the meaning of the word “discover” is concerned.

In my judgment, the parties were right to submit that, after FII, limitation begins to run in a deliberate concealment case when the claimant recognises that it has a worthwhile claim, and that a worthwhile claim arises when a reasonable person could have a reasonable belief that (in a case of this kind) there had been a cartel. Gemalto’s four propositions overcomplicate the position. The FII test must be applied with common sense. As the judge held, there is unlikely in most cases, as in this case, to be a real difference between the application of the statement of claim test and the FII test. Indeed, the statement of claim test is, perhaps, little more than a gloss on the FII test. It is also worth noting that competition cases are not to be treated differently from other cases under section 32 (see Arcadia at [51]).

First, the FII test makes clear that the claimant is not entitled to delay the start of the limitation period until it has any certainty about its claim succeeding. So, whilst in a fraud case, if there were an essential fact about the fraud that the claimant had not discovered, without which there would have been no fraud, it would make sense to say that the claimant had not discovered the fraud. But in concealment, what needs to have been discovered is just that, the concealment. Once the claimant knows objectively that a cartel has been concealed, it does not need to have certainty about its existence or about the details of that cartel. That is why the Supreme Court made clear that the claimant needs only sufficient confidence to justify embarking on the preliminaries to the issue of a writ, such as submitting a claim to the proposed defendant, taking advice and collecting evidence. The term “worthwhile claim” is also not to be construed as a deed. It requires a common sense application. A claim in respect of a concealed event would not be a worthwhile one if it were pure speculation, but it would be if, as in this case, an authoritative regulator had thought it sufficiently serious, having investigated all the evidence available, to lay charges or issue a statement of objections.

Secondly, the test adumbrated by the Supreme Court must be intended to operate in all situations in which there has been mistake, fraud or concealment, and to be consistent with the Limitation Act more generally. It would make no sense for the limitation period for a road traffic accident to start running when it happens (at which point the victim may know nothing about the circumstances of the accident that, for example, rendered them unconscious), but for section 32 to allow a claimant a lengthy period of investigation before it is said to have “discovered” that the facts relating to its claim have been concealed. The person who is run down knows that they have a worthwhile claim, even if they may eventually be shown to have been responsible for the accident by running in front of the vehicle. The claimant cannot postpone the start of the limitation period until it has had the time to investigate the details of the claim and the possible defences and to evaluate its prospects, any more than the road traffic victim is able to do so. That is what the 6-year limitation period is for. The question of whether a claim is worthwhile is not a complex balance of the chance of success as Mr Turner suggested. The limitation period is not postponed until the claimant can show that it is more likely than not to succeed. Of course, if the putative claim would be struck out as not disclosing a cause of action, it would be right to say that the claimant had not discovered that it had a worthwhile claim (see the comparisons with Earl Beatty, Paragon, Sephton and Molloy above at [37]). That is why I say that I am far from sure that there is a real difference between the statement of claim test and the FII test so far as concealment cases are concerned.

I should say something about the authorities, upon which both sides relied, that deal with the detail of the statement of claim test. I did not find much assistance from these cases, not because I doubt their correctness, but because FII explained the rationale of section 32(1) in a new and comprehensive manner. In those circumstances, as I have said, we would be failing in our respect for the doctrine of precedent if we did not start from a consideration of its reasoning.

In these circumstances, perhaps the most difficult part of this aspect of the case is really the question of whether, in a concealment case (and perhaps in a fraud case too), the FII test requires that the claimant has discovered every essential element of the claim that has been concealed. The pre-FII cases made clear that that was necessary. In my view, however, post-FII, that can no longer be necessary at least in a concealment case.

It makes no sense to say that the test for whether the limitation period has begun to run is when the claimant recognises that it has a worthwhile claim, and then to say that it does not have a worthwhile claim when it knows there may have been a cartel, but did not know, for example, the period during which the cartel operated. The formulation for the necessary knowledge is “knowing with sufficient confidence to justify embarking on the preliminaries to the issue of a writ”. One can embark on the preliminaries to the issue of a writ once one knows that there may have been a cartel without knowing chapter and verse about the details. That is what one either finds out when making investigations or will only find out upon disclosure within the eventual proceedings.

If, for example, claimant householders know that their jewels have been stolen in a burglary, they would know that they might have a claim against the burglar in wrongful interference with goods (conversion). But until the burglar is arrested and charged or otherwise identified, they cannot know that they have a worthwhile claim against anyone. Once, however, the victims know of the arrest and charge, they cannot wait for the trial and the conviction to discover details of how the crime was committed before time begins to run. Once the burglar is charged, they know that they have a worthwhile claim. They must use the limitation period to undertake the preliminaries to the issue of proceedings.

This example shows that there is reasonable consistency between the running down cases and the position under section 32(1). In section 14A cases, time does not begin to run until the claimant knows the identity of the defendant under section 14A(5), (6)(b), and (8)(b), but it is not postponed until every detail of the claim is available.

To summarise, therefore, the position after FII is that the proviso to section 32(1) has to be construed consistently as between mistake and deliberate concealment cases. Time begins to run in a deliberate concealment case when the claimant recognises that it has a worthwhile claim. In a case of this kind, a worthwhile claim arises when a reasonable person could have a reasonable belief that there had been a cartel. The claimant can embark on the preliminaries to the issue of a writ (and therefore the limitation has begun) once it knows that there may have been a cartel and the identity of the participants, without knowing chapter and verse about the details. It would not, however, know that it had a worthwhile claim if a claim pleaded on the basis of the details it knew would be struck out.

In the light of the analysis in the previous section, this question admits of a simple answer. The judge’s conclusions at [79] and [85] were, in effect, that Gemalto could properly have pleaded a claim following the announcement of the Statement of Objections, because the content of that announcement combined with the other material then available to it allowed it to identify the essential elements of the cartel, legitimately relying on inferences available to it from the materials in its possession. Gemalto submitted, as I have said, that the Statement of Objections did not provide a basis for a reasonable belief that there may have been participation in an unlawful cartel.

In my judgment, the judge was right to say that what Gemalto knew, taken together, provided it with an objectively reasonable belief that there had been an unlawful cartel in which the defendants had participated.

The judge also correctly summarised the status of the Statement of Objections at [59]-[63] as follows:

This answers the ground of appeal raised by Gemalto to the effect that the judge had been wrong to rely on the Statement of Objections as founding a reasonable belief in the existence of a cartel.

It is, however, worth saying something briefly about the position of the Statement of Objections following FII. It is, in my judgment, obvious that, once the regulator publicises the fact that it believes, subject to defences, that there is prima facie case that certain persons have participated in an unlawful cartel, a claimant knows that it has a worthwhile claim. A claim pleaded on the basis of that information and inferences drawn from it would never be struck out without the court being able to see the Statement of Objections itself, which would provide many of the details that a claimant from whom the cartel had previously been concealed would be lacking.

In the course of argument, Green LJ explored the question of whether or not a Statement of Objections was protected from disclosure in national proceedings. It appears that it was not at the time relevant to this case, but has been since 7 August 2015 (see Article 16a(3)(b) of Commission Regulation (EC) No 773/2004 of 7 April 2004 as amended by Commission Regulation (EU) 2015/1348 of 3 August 2015). It seems to me, however, that the point is not significant, because any court knowing the bare bones of what was reported about the Statement of Objections would find it almost impossible (save in circumstances not contemplated in argument before us) to strike out without being shown the document itself.

The FII test does not mean that the claimant must have everything it needs to succeed in its claim on the day that the limitation period begins to run. No ordinary claimants with claims under other torts that have been concealed from them have such an advantage. They need to have discovered the facts relevant to their right of action which have been deliberately concealed from them by the defendant, to adapt the precise wording of section 32(1)(b). To do so, they need objectively to know they have a worthwhile claim. A public statement by a regulator after a four-year investigation (as in this case) that they think there is evidence that there may have been a cartel must be powerful evidence in support of that knowledge.

In this case, the additional evidence of what Gemalto knew was also important. I shall not repeat it (see [17]-[24] above). I would, however, expressly endorse what the judge said at [90] in relation to Infineon’s denials of wrongdoing. FII pointed out at [196] and [202] that it was in the nature of litigation that facts and law are disputed. Such disputes as to an element of a cause of action do not mean that the commencement of the limitation period is postponed until that dispute has been resolved.

I should emphasise, however, that there is no universally applicable rule. Cases will turn on their facts. Here, the judge was right to find as she did in relation to the significance of the Commission’s announcement of the Statement of Objections.

This question too answers itself once one understands the significance of FII. The judge held at [85] that Gemalto could legitimately have pleaded a cartel covering 2003 to 2006, on the basis of the period identified in the two RFIs. I agree. That would, of course, have required Gemalto to draw an inference from the RFIs that this was the period during which the Commission suspected an infringement had taken place.

As I have already said, the limitation period starts to run under FII when the claimant can embark on the preliminaries to the issue of a writ. That is when it knows that there may have been a cartel and the identity of the participants, without knowing chapter and verse about the details. A claim in respect of the cartel pleading that it took place “in or around 2003 to 2006” would have been a worthwhile claim and would not have been struck out for all the reasons I have already given. Gemalto had sufficient knowledge of the period of the cartel to allow time to run at the end of April 2013.

The position after FII is that the proviso to section 32(1) has to be construed consistently as between mistake and deliberate concealment cases. Time begins to run in a deliberate concealment case when the claimant recognises that it has a worthwhile claim. In the case of a deliberately concealed unlawful cartel, a worthwhile claim arises when a reasonable person could have a reasonable belief that there had been a cartel. The claimant can embark on the preliminaries to the issue of a writ (and therefore the limitation has begun) once it knows that there may have been a cartel and the identity of the participants, without knowing the details, including the precise period of the cartel. A claimant would not, however, know that it had a worthwhile claim if a claim pleaded on the basis of the details it knew would be struck out. The claimant can draw inferences from the public statement of the regulator announcing the issue of a Statement of Objections.

I would dismiss this appeal.

I agree for the reasons given by the Master of the Rolls that this appeal should be dismissed. In particular I agree with the conclusion that the principles articulated in FII apply. I do not propose to add to the analysis in his judgment. The only issue I address concerns the application of the FII test to competition law cases where, as here, there are on-going regulatory proceedings likely to result in an infringement decision having a binding effect upon a court hearing parallel civil proceedings. In this regard the rules in relation to prescription in competition law cases changed in 2017. Following the oral hearing of the appeal the parties made submissions on the implications of this rule change for the appeal. I found the arguments generally to be relatively finely balanced albeit that ultimately I came round to the view of the Master of the Rolls. In these circumstances I have set out why I did not find the ultimate conclusion to be straightforward.

The limitation rules changed in 2017 with the implementation into domestic law of Directive 2014/104/EU (Footnote: ¹)(“theDamages Directive”). This was implemented into domestic law via Regulations (Footnote: ²) which amended the Competition Act 1998 (“CA 1998”) by inserting a new Schedule 8A (“the Damages Regulations”). The facts of this case occurred whilst the UK was a member of the EU. The Statement of Objections (“SO”) was issued in April 2013. The final decision was adopted in September 2014. On any analysis the cause of action also accrued before 2017, when the new law came into being.

Paragraph 19 of Part 5 of Schedule 8A CA 1998 makes specific provision for the commencement of limitation in a competition damages case in relation to any cause of action accruing after 9^th March 2017. Paragraph 19 sets out the basic position:

Paragraph 21 entitled: “Suspension during investigation by competition authority” addresses the situation that arises when there is an extant investigation by a competition authority:

The parties advanced competing contentions upon the back of these provisions. Gemalto observes that these measures were introduced to exclude for limitation purposes the period of investigation by a competition authority, which includes the period between issuance of an SO and adoption of a decision. It is argued that this reflects policy objectives which should apply equally to the application of section 32(1)(b). The respondents disagree arguing that the measures have no relevance. They amount to new provisions aligning UK law with the Damages Directive and create a bespoke limitation regime for competition law claims only and which, for the future only, suspend time for the entire period of the competition investigation regardless of whether time would have run at any point in that period under the Limitation Act 1980.

In applying the FII test to the facts an understanding of the practical implications is important. This was made clear in FII at paragraph [210] which explained that the application of the test was fact and context specific. There are certain preliminary observations I would make.

The first is that following the rule changes there is now a specific prescription regime for competition law such that the Limitation Act 1980 and the FII test will have an ever-diminishing role to play in cases such as the present. The second, as the Master of the Rolls has emphasised, is that under the Limitation Act 1980, the test in a case of concealment focuses, for the reasons set out in paragraphs [193] and [195] of the judgment in FII, upon discovery of the circumstances giving rise to the cause of action as a result of the concealment. The central ingredients of the cause of action are: (i) the existence of an agreement or concerted practice which affects trade between the Member States of the EU, and which has as it object or effect the restriction of competition; (ii) the existence of an identifiable defendant; and (iii), loss and damage (including causation). In a cartel case it is the agreement or concerted practice to which the putative defendant was a party that is concealed. It is therefore the revelation of sufficient facts about the cartel and its parties that matter. The other ingredients of the cause of action (effect on trade, and object or effect on competition) are consequences of the existence of the cartel, not discrete facts that are, in a practical sense, capable of being independently concealed. Once revealed that a defendant was party to a cartel these matters are consequential.

Next, applying FII (paragraphs [193] and [195]), limitation runs from the point in time when the claimant either (i) knows, or could with reasonable diligence know, of the concealed facts “with sufficient confidence to justify embarking on the preliminaries to the issue of a writ, such as submitting a claim to the proposed defendant, taking advice and collecting evidence” or (ii), (applying earlier dictum from case law) “discovers or could with reasonable diligence discover” the concealed facts “in the sense of recognising that a worthwhile claim arises”. According to FII paragraph [193] there is no difference in substance between those two formulations. The language used makes clear the fact sensitive nature of the analysis. In paragraph [196], in the context of the application of these tests, the Court confirmed that for time to run all that was required was a “reasonable belief” in the relevant facts, not the truth of them, that being something which is established at trial, not at earlier stages of litigation. The Master of the Rolls in his judgment has addressed the distinction between knowing a fact and a reasonable belief in a fact and I will not repeat what he has addressed.

My initial concerns can be summarised as follows.

First, the information found in this case to be sufficient to trigger limitation flowed from a disparate and quite fragmented combination of: (i) a press release announcing that the Commission had issued an SO; (ii) press coverage indicating that the defendants were addressees of the SO; and (iii) earlier RFIs issued to the claimant seeking certain factual information about the market and the place of the claimant in the market. However, as the judge recorded in [54] of her judgment, the defendants had at various points in time argued that time ran from the emergence of different permutations of these facts which highlighted a lack of certainty in the law as to what sorts of facts individually or in combination might be relevant to the commencement of time.

Secondly, the judge relied upon inferences about facts drawn from these revealed documents but no actual facts about the concealed cartel. The inferences were that a third party, the Commission, considered that there was a case to answer on the part of certain undertakings in the chip market that they had entered into some form or type of horizontal, price related, agreement. Mr Turner QC for Gemalto argued that no sensible corporation, properly advised, would treat this as giving rise to a “reasonable belief” upon which to form any sort of a “confident” conclusion that a claim was “worthwhile”.

Thirdly, a regulatory decision finding an infringement can be relied upon as dispositive of liability under the CA 1998. The fact of the decision becomes a proxy for the usual liability components of the cause of action and the existence of the decision is hence a matter relevant to the cause of action under section 32(1)(b). It is argued that this means that in follow-on cases time cannot run until the decision had been published since only then is a key ingredient of the cause of action known and pleadable.

Fourthly, in practice neither the claimant nor the court has access to the SO because according to normal Commission practice and procedure it is treated as a strictly confidential document usable only in the Commission’s administrative proceedings. The appellants argued that this means that a claimant is unable to plead the cartel with the high degree of particularity that the law requires, by reference to pleadings of causes of action based upon unconscionability (such as fraud). When the SO was issued (in 2013) the UK was a member of the EU and in fulfilment of the duty of cooperation which existed between a Member State (including its courts) and the EU institutions (which included the Commission) the domestic courts would not, in practice, therefore order disclosure of an SO. Our attention was drawn to various documents emanating from the Commission and a variety of decisions of the domestic courts reflecting this more or less invariable practice. The respondents cited for example a Commission Opinion of 2014 in the context of the domestic litigation concerning interchange fees (Footnote: ³). There the Commission adopted an infringement decision (December 2007) which was under appeal before the General Court. The claimants’ application for disclosure of the SO in the High Court was refused (see Morrison v Mastercard [2013] EWHC 2500). The Commission Opinion at paragraph [15] records the objection of the Commission to disclosure of documents prepared for an ongoing investigation in view of Article 16a(3) Commission Regulation (EC) No 773/2004 of 7 April 2004 (as amended by Regulation 2015/1348) which precludes permitting national courts to use documents held on the Commission file, which would cover an SO. It is unnecessary to go into the law more deeply save to say that whether as practice or as a binding rule the national courts have never granted access to the SO, at least pending completion of a Commission procedure. This places a claimant upon the horns of a dilemma. An SO sets out the detailed evidence relied upon by the Commission to prove the existence of the cartel. Armed with such information a claimant is equipped to know whether it has a cause of action. In contrast, when precluded from access to the SO a claimant is at risk of being able only to plead the barest of bare bones which, as history has shown, has induced defendants to seek to strike out the pleaded claim upon the basis that it failed to meet due pleading standards and was exiguous: see for example Nokia. These problems however fall by the wayside once the decision is published which is a transparent instrument setting out all hitherto concealed facts. There is at this juncture no continuing prejudice flowing from the inability to access the SO. Mr Turner QC argued that this was a further reason why time should therefore run only from the date of the decision.

Fifthly, on one view of FII there is an underlying premise that the point at which time runs is the point at which a claimant can “confidently” embark upon a “worthwhile” litigation strategy which, it is assumed, will be capable of being pursued to a conclusion through to a final judgment. That was certainly the case on the facts of FII. However, in cases where there are parallel regulatory proceedings there is a near inevitability that the domestic court will stay proceedings pending the decision. Stays can last for years if the regulatory proceedings drag on and if a resultant adverse decision is then appealed to the European courts. If time runs from knowledge that an SO has been issued against a defendant then a claimant cannot exercise the normal right of access to a court to push forward its claim.

Sixthly, there is also a concern as to the implications for other types of claim, for instance brought by consumers (whether in collective or standalone proceedings) lacking the resources or sophistication of corporate claimants. Would a rule such as that articulated by the judge become an unfair trap for the uninitiated and unwary consumer?

Finally, many advisers have considered that time runs in a concealed cartel case from the date of the decision that being the first occasion upon which the full details of the cartel are laid bare. This was a proper basis upon which to advise. Any other solution risked uncertainty and wasteful litigation costs. The deferment of limitation until publication of the decision overcame these problems. Such an approach accorded with the concern of the Supreme Court in FII to match the commencement of limitation with knowledge on the part of a claimant sufficient to convey requisite confidence in embarking upon the preliminaries to litigation of a worthwhile claim all of which could be based upon a reasonable belief about the facts. It also accorded with the policy which governed limitation for causes of action accruing after 2017.

There is force in these points. However, as the argument unfolded, I came to the conclusion that an interpretation of the legislation which deferred the running of time until the publication of the decision would be inconsistent with the FII test. The judgment of the Master of the Rolls focuses upon the inferences that can be drawn from knowledge about the issuance of an SO to an identified defendant alleged to be party to a cartel. He points out that from this point onwards a claimant has six years in which to decide to commence an action. When one cuts through the detail of much of the argument this must be right, at least when viewed in light of a recognition that if a claimant acts upon what is, by its nature, limited information to issue proceedings the courts will not, having held that time is now running, then inconsistently turn around and criticise the claimant for not being able to advance full particulars. Further, even if I were of the view that such deferment amounted to sound policy, reflecting the fact that a legislative decision was taken in 2017 which did defer the running of time, it is still not open to this Court to substitute the due application of the statute as interpreted by the Supreme Court in FII, for its own view of a desirable policy outcome.

First, as to legal uncertainty there is in my view no real risk that a rule which centres upon knowledge of the issuance of the SO will create significant legal uncertainty. The revelation of the fact that an SO has been addressed to an identified defendant is a pivotal moment. From this stage onwards all possible victims of the defendant know that the regulator has reached the point when it considers that there is a real case to answer on the part of a defendant and that this implicitly covers (i) the existence of an agreement or concerted practice; (ii) effect on trade; and (iii) restriction of competition. There is a good deal of law on the status of an SO (which it is not necessary to go into). At its core lies the proposition that issuance of an SO reflects a considered decision taken by the Commission, as the lead regulator in the EU, that it possesses sufficient inculpatory evidence to justify initiating a formal procedure against an undertaking. There is a serious case to answer. The test in FII is based upon reasonable belief, not actual knowledge. A claimant will not know for certain that a defendant has violated the competition rules because the SO is merely a provisional document triggering rights of defence and any final decision is subject to appeal but given the status of the Commission and the formal nature of an SO there will still be a “reasonable belief” that a violation might have occurred. These facts go a long way to address the “confidence” and “worthwhile” thresholds set out in paragraphs [193] and [195] of FII. The fact that an SO has been issued will not, standing alone, suffice to trigger the running of time because a claimant needs also to know (i) the broad nature of the subject matter of the alleged violation described in the SO and (ii) the identity of the undertakings to whom it has been addressed. Without this a claimant could never know that any violation by a possible defendant had any relevance to the claimant and its business. In practice, however, the minimum details are usually covered in the press release announcing the issuance of the SO and if not are routinely available in the specialist and trade press. For example, in the present case the Commission press release summarised the nature of the alleged violation (a horizontal price related agreement in the smart card chip market “to keep prices up”). On this occasion it did not identify the suppliers concerned but their identity was confirmed in the press shortly afterwards. In this connection knowledge that prior investigatory steps are being carried out by the Commission (such as the conducting of dawn raids and/or the issuance of RFIs) does not bear the same significance as the issuance of an SO. These are merely procedural steps along the way to the conclusion of the Commission that the case against a defendant is sufficient to justify the initiation of formal proceedings by the issuance of an SO. As to the argument that all of this arose only by inference, this is without significance. It is trite that facts can be inferred from other facts. Here the fact that the Commission has issued an SO is itself a fact from which a great deal can reasonably be inferred about the seriousness of the allegation.

Secondly, the fact that a modest basket of inferred facts might serve to start time running is not a problem provided that if the claimant then decides to issue proceedings the Court does not summarily dismiss such a pleading upon the basis that it is premature or exiguous. As to this no legal representative seeking to strike out or obtain reverse summary judgment could, in accordance with that person’s overriding and primary professional duty to the Court, properly advance an argument that the pleading was premature or inadequate knowing that its client, the defendant, was in receipt of an SO that it would not, for whatever reason, disclose to the claimant or the court. Further, in cases where the detailed facts relevant to the cause of action have been, and remain, largely concealed even after issuance of an SO the courts will, in accordance with the overriding objective, bend in favour of a relaxed approach to the pleading requirements to avoid an otherwise obvious risk of injustice. The Courts have made this point in a variety of different ways in the past (eg Nokia at paragraph [67]).

Thirdly, Mr Turner QC sought to argue by parity of reasoning with fraud cases that the courts demand a high standard of particularity in relation to the pleading of cartel cases alleging unconscionable and clandestine behaviour placing the defendant in jeopardy of reputational harm and regulatory, quasi-criminal, fines if found to have violated the law by the court. He contended that this being so a worthwhile claim could only confidently be pleaded when the final Commission decision was published. I am not persuaded by this. Even if the argument was generally correct (a point it is not necessary to explore) the benign approach the courts have taken to bare bones pleadings described above is an answer. The point at which a court might demand greater pleading rigour would be when the decision was published when the full details of the cartel would be revealed. But, at this stage the court is unlikely to require further pleading particularity because the Commission decision is binding, so that any pleading on such matters may be largely otiose for liability purposes. And as to the exposure to incremental reputational or penal risk, as of the date of the decision that risk has eventuated; the defendant has been fined and its reputation sullied by the facts recorded in the decision – the worst has already happened.

Fourthly, as to the suggestion that since virtually all such cases are stayed pending the regulatory decision and the claimant is prevented from exercising normal rights of access to a court, this overstates the difficulty. It ignores the fact that the parallel regulatory proceedings might be dispositive of liability which would otherwise be for the claimant to prove so the claimant can sit back and allow the Commission to undertake the heavy lifting. In addition, victims are not precluded from the parallel Commission proceedings since under relevant Commission rules a complainant has the right to apply to the Commission to obtain standing in the Commission proceedings to make submissions, submit evidence and attend oral hearings. To facilitate this the Commission will provide to the complainant a non-confidential version of the SO.

Fifthly, as to the ability of advisors to proffer sensible advice, this in large measure has been rendered academic by the changes to the law. So far as legacy cases are concerned where the FII test applies, the effect of the judgment below, and the confirmation in this judgment, is that advisers will henceforward take the view that a press release indicating that an SO has been served on a defendant might start time to run. Claimants have six years in which to formulate their litigation strategy.

Finally, in relation to a risk that this rule will prejudice unwary consumers I do not think that the Courts would permit this to happen. The Supreme Court in FII was clear that the application of the test was fact and context sensitive and took into account the nature of the claimant (corporate or otherwise) and its resources including access to legal advice and corporate priorities: see paragraph [210(2)]. The position of a consumer relative to that of a corporate litigant might be very different. Time might well not begin to run from the same point as it did in relation to a corporate claimant. A typical consumer might be oblivious to the facts of the unlawful conduct, or as to its concealment, or as to the existence of an SO even if the fact of the SO was in the public domain, or as to its significance even if known of. A typical consumer might also be unlikely to have the resources and funds to instruct lawyers capable of providing specialist advice or embark upon preliminary exploratory work even if they could obtain advice, especially if the claim was modest compared to the costs. On reflection I do not consider that the FII test risks leading to injustice where the claimants are consumers.

For all of these reasons, I agree with the conclusion of the Master of the Rolls that application the FII test will not lead to impracticability or legal uncertainty even if it results in time running from about the date of revelation of the existence of an SO.

I agree that, for the reasons given by the Master of the Rolls, this appeal should be dismissed.