The Federal Deposit Insurance Corporation v Barclays Bank Plc & Ors

This is an application by the Eleventh Defendant (“UBS”) to strike out the claim of the Claimant (“FDIC-R”) and/or for summary judgment to be granted in UBS’s favour on limitation grounds.

FDIC-R is an independent agency of the US government which acts as regulator, insurer and now as receiver for 39 failed US depositary institutions (the “Closed Banks”). FDIC-R’s claim relates to the alleged manipulation by the Defendants of the United States Dollar London Interbank Offered Rate (“USD LIBOR”) benchmark from about August 2007 to at least the end of 2009. At the time, USD LIBOR was one of the most widely used international interest rate benchmarks.

At the relevant times in and from 2007, UBS and the other financial institution Defendants (the “Bank Defendants”) were members of the contributing panel of 16 banks (the “Panel Banks”) who were instrumental in the process by which the British Bankers’ Association (the “BBA”) set and published its daily USD LIBOR benchmark for various maturities (“tenors”). The Panel Banks were required, in accordance with a definition published by the BBA (the “LIBOR Definition”) to make daily submissions to the BBA of the interest rates at which they believed that they could borrow United States Dollar funds from other banks in London for various tenors. After excluding the higher and lower quartiles of submissions, the mean average of the two central quartiles of the range of submissions then formed the published USD LIBOR rate.

The LIBOR Definition required the Panel Banks to answer the question,

In outline, FDIC-R contends that the Bank Defendants colluded with each other and with the BBA to suppress the level of USD LIBOR during the relevant period from August 2007 by making artificially low submissions to the BBA of the rate at which

So far as relevant to the present application, FDIC-R’s claim is based on alleged breaches of Article 101 of the Treaty on the Functioning of the European Union (the

FDIC-R’s central allegations are summarised as follows in sub-paragraphs 3(4)-(9) at the beginning of the Particulars of Claim,

In this regard, many hundreds of examples of manipulation across several currencies and tenors have been identified, including in relation to USD LIBOR. This has led to extensive regulatory sanctions, as part of which, findings and admissions have been made that Panel Banks (including the

Such findings and admissions demonstrate that Panel Banks in general, and the Bank Defendants in particular, had the motive, opportunity and willingness to disregard their obligations and to manipulate LIBOR in their own interests. Several ex-Panel Bank employees have also been criminally convicted as a result of such wrongdoing.

FDIC-R’s claim arises out of a particular form of the manipulation of LIBOR by the Bank Defendants, namely the sustained and material suppression of USD LIBOR from

The Bank Defendants had incentives to Lowball their USD LIBOR submissions, including incentives which arose from or were enhanced by the financial crisis. These incentives included a desire to present a false picture of the financial health of the banks, and of the financial system, and to distort competition between at least [the] Bank Defendants and nonPanel Banks in relation to their trading portfolios. These incentives were only capable of being realised by the Defendants or at least the Bank Defendants, if they acted collusively or in concert.

The Bank Defendants’ collusive and/or concerted suppression of USD LIBOR during the Suppression Period was participated in and/or facilitated by and/or directed by their trade association, the BBA, and by a committee of the Panel Banks, known as the Foreign Exchange and Money Markets Committee (the “FXMMC”), which purported to regulate LIBOR.”

In paragraph 4 of its summary, FDIC-R claims that the material suppression of USD LIBOR during the “Suppression Period” was either the result of an agreement between the Defendants or the Bank Defendants (the “Agreement”), or if there was no such agreement, that the suppression,

The more detailed pleading by FDIC-R in the body of its Particulars of Claim includes a Section D1 in which it is alleged that USD LIBOR was materially suppressed on a sustained basis throughout the Suppression Period, and a Section D2 which alleges that this was the result of deliberate Lowballing by at least the Bank Defendants. Importantly for present purposes, that section D2 includes the following allegation at paragraph 52(4),

In support of that contention, Section D2 of the Particulars of Claim first refers (at paragraphs 53-55) to the regulatory findings concerning the Bank Defendants and the evidence from criminal trials. Section D2 then continues at paragraph 56 to allege that the Bank Defendants had a number of “individual and collective incentives” to engage in Lowballing and to lower USD LIBOR. These are then set out in paragraphs 57 to 60 and include the following,

Presentation of individual financial health To varying degrees, each Bank Defendant had an individual incentive to materially Lowball its own USD LIBOR submissions … in order …to present to other participants in the Interbank Market(s), to counterparties … to governments and regulators, to consumers and to the world at large that it was in a better financial position than was in fact the case …

Collective incentive to prevent scrutiny of individual banks by promoting the impression of collective financial health and stability The Defendants had a collective incentive for the Bank Defendants to collude on the level of their USD LIBOR submissions in order to present to the world at large a picture of collective health and stability of the Bank Defendants (and more generally the Panel Banks)….

Moreover, the Defendants also had a collective incentive in the Bank Defendants materially Lowballing their USD LIBOR submissions and USD LIBOR in order to present to the world at large as a group they were in better financial health, and were more stable, than was in fact the case….”

Section D2 of the Particulars of Claim then alleges,

Section D3 of the Particulars of Claim contains the core allegations of collusive behaviour against the Defendants – i.e. the alleged “Agreement” or “Concerted Behaviour”. Paragraph 70 of the pleading indicates that, pending disclosure, FDIC-R relies on a number of matters to establish that such collusion took place, the first two of which are as follows,

The material suppression of USD LIBOR could not have taken place had only a single Panel Bank suppressed submissions.

The Lowballing was in fact extensive and appears to have been pervasive across at least the Bank Defendants such that it is implausible that it would occur in the absence of the Agreement or the Concerted Behaviour and it cannot be explained as coincidental parallelism.

There was significant clustering of submissions by Panel Banks.

The third matter relied upon by FDIC-R is the fact that the Defendants had the opportunity to collude, and the fourth matter is said to be a series of particular emails and other communications derived from regulatory investigations and the evidence given at criminal trials which are said to show collusion and the exchange of confidential information between some of the Panel Banks and the BBA.

FDIC-R also pleads an alternative case on Concerted Behaviour,

FDIC-R alleges that such Agreement or Concerted Behaviour constituted an infringement of Article 101 TFEU or section 2 CA 1998 and thus an actionable breach of statutory duty. FDIC-R contends that the Closed Banks relied on the published USD LIBOR benchmark as a reference rate for their financial products and other transactions such as interest rate swaps, including with the Bank Defendants. It is claimed that each of the Closed Banks thereby suffered loss and damage.

UBS has not filed a defence and does not admit collusive suppression of USD LIBOR by Lowballing. However, assuming for the purposes of this application that there is an arguable case of such collusion under the TFEU or the CA 1998, UBS nonetheless contends that it has a clear limitation defence because the primary limitation period for breaches of the TFEU or the CA 1998 is six years, and FDIC-R’s claim form was issued on 10 March 2017, which is more than six years after the conduct complained of.

This application has been made because UBS also contends that FDIC-R has no real prospects of overcoming that limitation defence by satisfying the requirements of section 32(1)(b) of the Limitation Act 1980 (“Section 32(1)(b)”). Section 32(1)(b) provides that if any fact relevant to the plaintiff’s (claimant’s) right of action has been deliberately concealed from him by the defendant, the period of limitation shall not begin to run until the claimant has discovered the concealment or could with reasonable diligence have discovered it.

UBS contends that irrespective of whether deliberate concealment of collusive Lowballing took place, there were sufficient facts in the public domain which could have been discovered by FDIC-R with reasonable diligence which would have enabled FDIC-R properly to plead a complete cause of action against UBS (the socalled “statement of claim” test) by a date six years before the claim was issued, i.e. by 10 March 2011. UBS therefore contends that it can clearly be seen at this stage of the proceedings that FDIC-R cannot rely on Section 32(1)(b) and the claim against UBS in relation to infringements of competition law before 10 March 2011 ought to be struck out or dismissed summarily.

In response, FDIC-R argues that it neither discovered, nor could with reasonable diligence have discovered, sufficient facts properly to plead a statement of claim alleging collusive Lowballing of USD LIBOR submissions by UBS prior to 10 March 2011. FDIC-R contends that, pending disclosure, its claim is necessarily one of inference drawn from various strands of evidence. It argues that although there were some earlier pieces of economic analysis and speculation in the financial media to the effect that USD LIBOR was consistently too low, that was not a universal view, and not one that provided a proper basis for an inference to be drawn that there had been unlawful collusion between some of the Panel Banks, or which involved UBS in particular.

FDIC-R contends that the critical evidence which tipped the balance and enabled it properly to plead that the Panel Banks including UBS had been colluding to suppress

regulatory findings and admissions in relation to LIBOR manipulation by individual Panel Banks which first began to be published in June 2012 in respect of Barclays, and were published by the FSA in the UK, the Commodity Futures Trading Commission (“CFTC”) and Department of Justice (“DOJ”) in the US, and the Swiss Financial Market Supervisory Authority (“FINMA”) in respect of UBS on 19 December 2012; and

evidence from criminal trials of traders and brokers concerning manipulation of LIBOR which began to come into the public domain in 2015.

FDIC-R therefore resists the application to strike out or for summary dismissal of its claim against UBS and contends that the matter should go to trial.

The general principles to be applied by a court to an application to strike out or for summary judgment are not in dispute. They have been set out in a series of decisions including in particular EasyAir Ltd (trading as Openair) v. Opal Telecom Ltd [2009] EWHC 339 (Ch.) at [15] per Lewison J, TFL Management Services Ltd v. Lloyds TSB Bank plc [2014] 1 WLR 2006 at [26]-[27] per Floyd LJ, and Global Asset Capital v. Aabar Block [2017] EWCA Civ. 37 at [27] per Hamblen LJ (as he then was). The principles were also summarised by Cockerill J in Daniels v Lloyds Bank [2018] EWHC 660 (Comm) at [48].

Pursuant to CPR r.24.2 the Court may give summary judgment to a defendant on a claim or on a particular issue if it considers that the claimant has no real prospect of succeeding on the claim or issue, and there is no other compelling reason why the matter should be disposed of at trial.

The test of whether there is “no real prospect” of success connotes an absence of reality, and the burden of showing that is so rests upon the defendant applying for summary determination. In the instant case it therefore rests upon UBS to show that there is an absence of reality in FDIC-R seeking to rely upon Section 32(1)(b) at trial.

I have set out the essential elements of Section 32(1)(b) above. In Arcadia Group Brands v Visa Inc [2015] Bus LR 1362 Sir Terence Etherton C (as he then was) considered the relevant authorities on the meaning of that provision, including Johnson v Chief Constable of Surrey (Court of Appeal, 19 October 1992, The Times, 23 November 1992) (“Johnson”), C v Mirror Group Newspapers [1997] 1 WLR 131 (“Mirror Group”) and AIC v ITS Testing Services [2007] 1 All ER (Comm) 667 (“The Kriti Palm”).

In The Kriti Palm, Rix LJ said, at paragraph [307],

The shorthand expression “statement of claim test” appears to originate from the decision in Mirror Group, in which Neill LJ referred to Johnson and commented, at page 137,

In The Kriti Palm, Rix LJ then continued, at [323]-[324],

In The Kriti Palm, agreeing with Rix LJ, Buxton LJ commented, at [453],

In Arcadia Brands, after referring to those passages, Etherton C therefore concluded, at [49],

Although the parties were broadly agreed as to the test that I should apply, there was disagreement as to precisely how it is to be applied in a competition case and to the nature of the allegations made in this case.

For UBS, Mr. Kennelly QC suggested that there is a “generous” or “relaxed” approach in competition cases to the standard of pleading required of claimants in cases involving secret cartels. He submitted that as a consequence it would have been open to FDIC-R to plead a case against UBS at an early date. That proposition was essentially based upon the dictum of Sales J (as he then was) in Nokia Corp v AU Optronics [2012] EWHC 731 (Ch) (“Nokia”).

In Nokia, the claimant had issued a claim form supported by an initial particulars of claim which alleged that the defendants had been parties to secretive anti-competitive agreements or concerted practices in respect of the supply of LCDs. The claimant was unable to specify when the arrangements had begun or exactly which of the defendants had been parties to them, but relied upon the fact that the US DOJ and the EU Commission had publicly announced that it was investigating the defendants who had either pleaded guilty or admitted being under investigation. After disclosure in various proceedings in the US, the claimant produced a far more extensive draft amended particulars of claim and sought leave to amend.

Sales J identified the main issue as being whether the original claim form and particulars of claim pleaded the same causes of action against the defendants with proper particularity as did the amended particulars of claim, so that the defendants would suffer no prejudice if the amendments were to be allowed after the expiry of the limitation period. He held that the original claim form and pleadings were in proper form, and allowed the amendments. In explaining his approach, he stated, at paragraph 62-67,

A claimant's counsel is subject to professional obligations in relation to what case may be pleaded (thus, e.g., a claim in fraud can only be pleaded in certain well-known circumstances, where there is sufficient material available to the pleader to justify such a plea)….

An application to strike out or for summary judgment may be made where, on the evidence about the facts, there is no reasonably arguable case on which the claimant could succeed. In the present case, none of the defendants put in evidence to demonstrate that this was the case.

Requests for further information may be put forward by a defendant to clarify exactly what case is being made where a general pleading is put forward….

If it became clear at some stage in proceedings that a claimant had further information available to him but failed to provide it when he ought to do so to clarify his case on the pleadings, it would be possible for the defendant to apply to strike out the claim on the grounds of abuse of process or to obtain an order (ultimately an unless order, threatening dismissal of the claim) for provision of particulars in response to a request for further information….

In my judgment, the availability of such procedural protections for a defendant to ensure that a claim is fully and properly explained in good time before trial (as against the possible loss to a claimant of an entire, potentially meritorious claim), indicates that in resolving the tension referred to above and determining whether a cause of action has been sufficiently pleaded in a statement of case (particularly in the claim form and/or the particulars of claim when an action is commenced), the balance is to be struck by allowing a measure of generosity in favour of a claimant. Such an approach is appropriate and in the overall interests of justice and the overriding objective set out in CPR Part 1.1…”

In Nokia, Sales J was not dealing with the same issue with which I am concerned, because a claim had been issued before the expiry of the limitation period. It is also apparent that although Sales J contemplated that a measure of generosity concerning pleading should be given to claimants in competition cases where some covert or secret behaviour is alleged, at the end of paragraph 62 and in paragraph 63 of his judgment, he nonetheless made clear that this generosity depends upon certain procedural protections being available to defendants. In that regard he identified, in particular, the requirement that defendants should not be subjected to allegations of fraud or dishonest conduct without sufficient material being available to those who plead statements of case.

In Playboy Club London v Banca Nazionale Del Lavoro [2018] EWCA Civ 2025 at [46], in the context of an argument about whether, following revelations in the course of evidence in a negligence trial, it was an abuse of process for a subsequent claim in deceit to be pursued, Sales LJ (as he then was) explained the restrictions on pleading fraud or deceit in greater detail,

The reason why an allegation of dishonesty must be pleaded with sufficient particularity was explained by Lord Millett in Three Rivers District Council v. The Governor and Company of Barclays of England (No 3) [2003] 2 AC 1 at [186],

That statement of principle was considered by Flaux J (as he then was) in JSC Bank of Moscow v Kekhman [2015] EWHC 3073 (Comm) who set out the specific requirements of pleading a case of dishonesty based on inference at [20]:

From these authorities, I believe that the protections to which Sales J was referring in Nokia are that parties should be reticent about pleading allegations of fraud and deceit, and should not do so without a “solid foundation” in the evidence, which Sales J contrasted with “speculation and inference”.

Such principles are, in my judgment, relevant in the instant case, because although dishonesty is not a necessary component of a claim under Article 101 TFEU or Chapter 1 CA 1998, it is difficult to see how an allegation that Panel Banks colluded to engage in Lowballing for the purposes of suppressing USD LIBOR over a sustained period could be anything other than an allegation of dishonest conduct. The essential requirement for each Panel Bank was to submit a statement of the rate which it believed it could borrow USD on the inter-bank market. It is inherent in the very concept of USD LIBOR as a benchmark, and (from mid-2008) was explicit in the BBA Guidance in particular, that the rate submitted had to be that which the individual Panel Bank genuinely believed that it could borrow US dollars inter-bank in London at the stated time.

Drawing these threads together, the question on this application for summary determination is whether the facts in the public domain which FDIC-R could, with reasonable diligence, have discovered prior to 10 March 2011, provided a solid foundation for the inference that UBS had colluded with some or all of the other Defendants to suppress USD LIBOR by Lowballing over an extended period.

Before I turn to consider the facts in greater detail, I should make two preliminary observations.

The first follows on from the legal requirements to which I have already referred and the chronology of the case. Although FDIC-R additionally relies in its pleaded case on the regulatory findings which were first made public in 2012 and the evidence from the criminal trials from 2015, those matters all occurred after the critical date for limitation purposes of 10 March 2011. UBS’s argument is that this later information added nothing new, so that the earlier information must alone have been strong

In assessing that argument, care must be taken not to overstate the inferences that could legitimately have been drawn from earlier materials by a subconscious use of hindsight drawn from the later materials. The analogy frequently used in cases in which inferences are required to be drawn from circumstantial evidence is that of a cord consisting of a number of strands, any one of which may be inadequate on its own to sustain the weight of a particular finding, but where several strands taken cumulatively will justify the inference: see e.g. Arif v HMRC [2006] EWHC 1262 (Ch) at [22] per Lewison J. In the instant case, the cord which supports the pleaded inference of collusive Lowballing consists of strands of evidence which became public knowledge at different times; the issue which I have to decide is whether the strands of evidence available prior to 10 March 2011 were strong enough on their own to support such an inference.

The second observation concerns the need to bear in mind the conceptual difference between the conduct which is alleged to form the basis of FDIC-R’s cause of action, which is collusive Lowballing between UBS and other Defendants with a view to suppressing USD LIBOR over a sustained period, and two other types of conduct in relation to LIBOR.

The first type of other conduct is where a Panel Bank would, independently from other Panel Banks, make lower LIBOR submissions than it genuinely believed reflected its costs of borrowing in order, for example, that it would not appear a weaker credit risk than other Panel Banks when the submissions were subsequently published by the BBA. Even if other Panel Banks might, for the same or similar reputational reasons, independently adopt a similar approach (so-called “parallel” conduct) this would not be actionable under the TFEU or the CA 1998 due to the absence of an agreement or concerted practice between the Panel Banks.

That said, Mr. Kennelly drew attention to FDIC-R’s alternative case, set out in paragraph 76 of the Particulars of Claim (and repeated in paragraph 58 of the Reply). In those paragraphs, FDIC-R in essence alleges that it would also be “concerted practice” so as to be actionable under the TFEU or the CA 1998 if a Panel Bank was deliberately Lowballing and came to know that another Panel Bank or Panel Banks were also deliberately Lowballing for the same or materially the same reasons, but the first Panel Bank nonetheless carried on acting in the same way. Mr. Kennelly submitted that UBS could not have struck out such allegations on the basis that they were an incorrect statement of the law based upon EU competition cases.

The second other type of conduct is what was referred to in the evidence as “trader based (mis)conduct”. This involved ad hoc collaboration between derivatives traders from different Panel Banks to move a LIBOR rate up or down artificially to suit their particular trading positions from time to time. UBS contends that the regulatory findings and criminal trials from 2012 onwards all related to this type of trader based conduct, could not support any inference of collusive Lowballing, and thus added nothing to the information available prior to 10 March 2011. FDIC-R acknowledges that such trader based misconduct did not amount to collusive Lowballing to suppress USD LIBOR of the type of which it now complains. FDIC-R’s argument, however, is that publication of such regulatory findings and admissions made a critical difference

I have set out the key elements of FDIC-R’s claim from its Particulars of Claim above. The economic commentaries to which UBS draws attention, and from which it contends that FDIC could have pieced together sufficient information to be able to plead a case against UBS by 10 March 2011 include the following.

On 10 April 2008 Scott Peng, a research strategist at Citigroup Global Markets and two of his colleagues published a paper under the heading “Is LIBOR broken” which stated the belief that the current liquidity crisis had created a situation in which LIBOR might be understating actual interbank lending costs by 20-30 bp. The paper drew attention to the disparity between LIBOR and the yields produced by a recent TAF auction by the US Federal Reserve. The TAF auction had produced a rate 10 bp higher than LIBOR, and the Citigroup paper pointed out that since the TAF was secured borrowing, it was counterintuitive for banks to pay a higher rate to borrow from the TAF than to borrow from the interbank market.

The Citigroup paper concluded,

That paper did not suggest that there had been any collusion between the Panel Banks in fixing LIBOR. Instead it suggested that there was an individual incentive on each Panel Bank to ensure that its submitted rate was no higher than the rates submitted by its competitors. It also made the point that all LIBOR postings were publicly disclosed, which suggested a route other than collusion by which Panel Banks could see (albeit after the event) how they compared to the rates submitted by their competitors and modify their submissions accordingly.

On 16 April 2008, the Wall Street Journal picked up the point in an article by Carrick Mollenkamp under the headline “Bankers Cast Doubt On Key Rate Amid Crisis”.

The article summarised what it characterised as “growing suspicions about LIBOR’s veracity” as follows,

The Wall Street Journal article returned to the Bank for International Settlements’ report in its final paragraph stating,

Mr. Mollenkamp was also a joint author of a further article for the Wall Street Journal the next day, 17 April 2008, which reported that the BBA had fast-tracked an enquiry into the accuracy of the LIBOR rate. The article commented,

In a further article the following day, Mr. Mollenkamp also reported that there had been a sudden jump in the USD LIBOR rate which could have been a sign that the Panel Banks were responding to increasing concerns that the rate did not reflect their actual borrowing costs.

The following week, on 21 April 2008, the Financial Times published an article which attributed the problems with the LIBOR rate to the fact that there was a shortage of long-term money in the market, resulting in the Panel Banks increasingly resorting to “intelligent guesswork” in submitting quotes. The article commented,

On 16 May 2008 JP Morgan Chase published a research note of a number of its analysts, in essence responding to Scott Peng’s question of whether LIBOR was “broken”. The headline points of that article were as follows,

Similarly, on 20 May 2008, the Federal Reserve Bank of New York published a paper by its Markets Group which stated,

Banks quote the rate at which they could borrow funds and these rates are published. This may lead to some deliberate misreporting designed to avoid the stigma of revealing high funding costs.

The panel is asked to provide quotes that are subject to ambiguity along at least two dimensions. First the transaction size is not clearly specified. Second, quotes are often given for maturities (e.g. 7-month LIBOR) or in market conditions (e.g. now) in which there is little or no actual interbank term activity. The lack of clarity result in panel members using dissimilar methods for determining quotes.

Many market participants feel the BBA does not currently have sufficient monitoring mechanisms in place to ensure the quality or validity of the quotes.

On 29 May 2008, Mr. Mollenkamp and Mark Whitehouse co-authored an article in The Wall Street Journal which picked up the thread of the earlier Journal articles. It presented an economic analysis showing that a number of the Panel Banks, including UBS, had reported significantly lower borrowing costs for LIBOR than the rate which might be suggested by the credit default insurance market. The article stated,

The Wall Street Journal article went on to suggest that the LIBOR rates submitted by the Panel Banks during the first four months of 2008 had clustered together and were far too similar to be believed. The article also drew attention to the disparity between the submitted rates and other market indicators. So, for example it stated,

On 29 May 2008, Bloomberg reported that a strategist at Barclays Capital had admitted in a television interview that the rates that banks submitted to the BBA had become “a little bit divorced from reality”. He reported that when the bank had quoted the “right rates”,

In my judgment, the theme which appears to run through these articles is that there were a variety of views on the accuracy of the LIBOR rates and the structure and robustness of the BBA’s system. A number of commentators had drawn attention to

However, no commentators had provided any evidence by this stage to suggest collusion between the Panel Banks. Indeed, to the extent that such a possibility was raised in the articles, the reported views of the BBA and other market participants either firmly rejected it, or considered that it was unlikely.

On 10 June 2008 the BBA published a consultative paper relating to LIBOR. It did not accept that there had been Lowballing, still less collusion between any of the Panel Banks in fixing the LIBOR rate. However, to address concerns as to the accuracy of the LIBOR benchmark, the paper set out steps proposed by the BBA to strengthen the governance by the FXMMC of LIBOR submissions, to widen the membership of that committee and the number of Panel Banks, and to investigate whether steps could be taken to avoid the potential stigma for a Panel Bank arising from making a LIBOR submission which stood out, whilst still maintaining transparency.

On 4 August 2008, a number of bankers and economists including Abrantes-Metz and Kraten published a paper, extending the study which had been reported by the Wall Street Journal on 29 May 2008. The paper analysed and compared LIBOR with other market data, including in particular credit default spreads. The abstract for the paper concluded,

The conclusion of the paper was as follows,

Again, I do not read this paper as providing solid support to a thesis that the Panel Banks had been colluding to Lowball LIBOR. On the contrary, it indicated that the evidence was thought by the authors either to be inconsistent with, or at least to be inconclusive as to whether there had been collusion or manipulation of LIBOR.

A number of further articles and academic papers appeared in 2009 debating the accuracy and reliability of LIBOR. None, however, pointed towards a conclusion that there had been collusive manipulation of LIBOR by the Panel Banks. Instead, they largely focussed on the economic incentives upon individual banks to report rates similar to other banks in order to avoid giving the appearance of financial distress.

In 2010 two more detailed economic analyses were published. The first, by Snider and Youle, dated 2 April 2010, presented statistical evidence that supported the view that LIBOR did not accurately reflect the actual borrowing costs of the Panel Banks. It also challenged the view that this was the result of Panel Banks attempting to portray themselves as less risky. Instead, the authors postulated that the incentives to make inaccurate LIBOR submissions arose from the positions of the derivative portfolios of the particular Panel Banks.

Moreover, referring back to the Abrantes-Metz paper of August 2008, Snider and Youle appeared to dismiss the possibility of collusion during the period after August 2007, reporting that the variance of submitted rates was actually lower prior to August

The conclusion of the paper was as follows,

As I read that paper, Snider and Youle were confirming that the Panel Banks had been submitting figures that understated their true borrowing costs, but they were not suggesting that this was the result of collusion between Panel Banks to Lowball USD LIBOR in the manner alleged by FDIC-R. Instead, they were propounding a theory that Panel Banks had incentives to push LIBOR in the direction favourable to their own portfolio positions – i.e. conduct which was more likely to involve trader-based misconduct of the type to which I referred above.

Abrantes-Metz returned to the issue of LIBOR in a study with Villas-Boas which was published in July 2010. The paper used a statistical tool called Benford’s law to analyse LIBOR data, and concluded that for about 20 years LIBOR had followed Benford’s law, but that there had been significant departures for a period both before and after August 2007. The authors concluded that “biased signals coming from the individual banks (agent aggregation bias), rate manipulation or collusion appear as one likely answer” (my emphasis), but noted that their analysis was “exploratory”. Whilst indicating that the authors thought that something was awry with LIBOR, this paper did not clearly indicate what that might be. Certainly it did not suggest that collusive Lowballing between Panel Banks was the only, or even the likely, explanation for the observations which they had made.

Commencing in April 2011, a number of class action lawsuits were commenced in the US against Panel Banks by various complainants. They each relied (among other things) on the economic analyses and articles to which I have referred.

The first such complaint by FTC Capital GmbH was filed on 15 April 2011, which is shortly after the relevant date for present purposes. It was then subsequently amended on 30 April 2012. The central section of the FTC complaint alleged that the defendants suppressed LIBOR during the relevant period, and supported that allegation by reference to (i) the defendants’ “powerful incentives to mask their true borrowing costs and to reap unjustified revenues by setting artificially low interest rates in LIBOR-based financial instruments that investors purchased”; (ii) a comparison between LIBOR submissions by individual Panel Banks and other indicators of the probability of those institutions defaulting; and (iii) the behaviour of LIBOR when compared with other market indicators. The FTC Capital complaint was then amended in 2012 to refer in addition to the regulatory findings and admissions which had been made.

The matters under (i)-(iii) above were consistent with parallel behaviour rather than being collusion. So, for example, the FTC Capital complaint alleged that,

That was also the stance taken by UBS, which applied to strike out the various complaints against it in the US, contending that the plaintiffs had built a case of “parallel conduct” by the Panel Banks to suppress LIBOR, motivated by their individual economic interests, but which failed to allege any anti-trust conspiracy so as to be actionable under the US Sherman Act. That contention was never tested, because the regulatory findings (particularly in relation to Barclays) were published before it could be ruled upon. Accordingly, when the US court came to consider the matter, it did so on the basis of the amended pleading which also relied upon regulatory findings and admissions; and the court eventually dismissed the antitrust complaints on other grounds.

UBS’s stance in contending that the US complaints filed in April 2011 did not contain allegations of collusion and were unsustainable is not a promising backdrop for its contentions in this case. Mr. Kennelly QC submitted, however, that a direct comparison should not be made between the positions advanced by UBS in the US and in the instant application. He contended that there is a critical difference between the lack of any proper support for the allegations of collusion in the US complaints which UBS sought to strike out, and the way in which FDIC-R has used, and could have used the same pre-March 2011 materials in its claim in these proceedings.

That critical difference, Mr. Kennelly QC submitted, was that in contrast to the US complaints, FDIC-R has made the additional allegations in paragraphs 52(4), 61 and 71 of the Particulars of Claim to which I have referred in paragraphs 9, 11 and 12 above. Those allegations are to the effect that it is “implausible” that “the material and sustained suppression” of USD LIBOR throughout the Suppression Period could have occurred in the absence of the Agreement or the Concerted Behaviour (and hence that it cannot plausibly be explained as coincidental parallelism); that the incentives which it is alleged that the Panel Banks had to Lowball LIBOR “were only capable of being realised if the Panel Banks acted collusively or in concert”; and that there was significant clustering of the Panel Banks’ submissions.

At this stage it should be pointed out that FDIC-R’s contentions in these respects are disputed by those Defendants who have filed defences. So, for example, Barclays contends in answer to paragraphs 52(4) and 71 of the Particulars of Claim that instructions that were given to its LIBOR submitter that its submissions should not be conspicuously higher than those of the next highest Panel Bank were an “independent reaction by Barclays to the prevailing market circumstances”. Barclays further contends that even if other Panel Banks were making artificially low submissions, this could be explained by parallelism and not collusion.

UBS’s stance on these allegations is not known because UBS has not filed a defence. However, Mr. Kennelly QC argued forensically that FDIC-R could have applied the same logic now set out in the paragraphs of its Particulars of Claim to which he referred, and drawn the inference prior to 10 March 2011 that there must have been collusion between Panel Banks. He submitted that on the basis of the allegations which FDIC-R now makes, this would follow from the facts (i) that USD LIBOR had been materially lower for a sustained period from August 2007 than other market indicators would suggest was appropriate, (ii) that Panel Banks stood to benefit from a low USD LIBOR and thus had incentives to bring such a situation about; and (iii) that there had been clustering of USD LIBOR submissions by the Panel Banks.

There are, I consider, a number of answers to that argument.

The first is that on the materials that I have summarised above, none of the financial journalists, academics and other commentators who had considered the position prior to March 2011 had come to the conclusion that there must inevitably have been collusion between the Panel Banks on the basis of such an argument. Those journalists and others were well aware of the system for setting LIBOR, they plainly had a considerable degree of concern and scepticism about the accuracy of USD LIBOR, and the possibility of collusion had been raised. However, collusion was not seen by any of the writers as the most likely, still less the only logical, explanation for the sustained low levels of USD LIBOR, and the clustering of submissions by the Panel Banks that had been observed. Nor, indeed, had the US attorneys acting for FTC Capital.

Moreover, as Ms. Demetriou QC stressed in argument, even if the Panel Banks had financial incentives to benefit from a low USD LIBOR rate, together with the opportunity to manipulate the system to that end, FDIC-R could not simply assume that it was inevitable, or even likely, that the Panel Banks had taken that opportunity to behave improperly.

Secondly, the remaining Bank Defendants who have put in defences all deny the premise of Mr. Kennelly QC’s argument, namely that a sustained low rate of USD LIBOR, clustering of submissions and the ability to realise the incentives said to have existed could only be explained by collusion. It is also notable that at no time did I understand Mr. Kennelly QC to concede this point on behalf of UBS.

Thirdly, the allegations made by FDIC-R upon which Mr. Kennelly QC relies should not be read in isolation, but must be read in the context of FDIC-R’s pleading as a whole. So, for example, in the summary which I have set out in paragraph 7 above, the allegation in paragraph 3(8) of the Particulars of Claim that the Bank Defendants had incentives to Lowball their USD LIBOR submissions and that such incentives could only be realised by collusion, comes in a single sentence after pointed references in paragraphs 3(4) to 3(6) to the regulatory findings and evidence from criminal proceedings which are said “to demonstrate that Panel Banks in general, and

Likewise, sub-paragraphs 71 (1)-(3) of the Particulars of Claim are said to be “further to” reliance on previously pleaded matters, and paragraph 71 comes in the middle of a rather convoluted section D3 which contains numerous internal and external crossreferences to the pleading of regulatory findings and evidence from criminal proceedings.

The same point applies to FDIC-R’s alternative case that even if there was no concerted practice initially, by 2007 or 2008 there was such actionable conduct because the Bank Defendants carried on Lowballing in the knowledge of, or whilst turning a blind eye to, the fact that other Panel Banks were also Lowballing for the same or materially the same reasons. In support of that allegation, paragraph 76 of the Particulars of Claim simply refers generically to all of the other matters set out in the pleading, but adds, “including in particular the discussions at the FXMMC and the BBA.” That generic reference includes matters both before and after March 2011, and the particular reference is to a series of communications within the BBA and FXMMC, which Mr. Kennelly QC accepted only became public knowledge after March 2011.

Fourthly, there is force in the point made by FDIC-R that until the regulatory findings against Barclays and then UBS began to be published in 2012, although there had been questions asked in the US and UK about the reliability of the LIBOR benchmark, there was nothing to make even well-informed observers think that any defects were the result of widespread dishonesty. In this regard, Ms. Demetriou QC drew attention in particular to evidence given to the House of Commons Treasury Select Committee on 17 July 2012 by the then Governor of the Bank of England, Sir Mervyn King (now Lord King), and the then Deputy Governor, Paul Tucker (now Sir Paul Tucker).

The hearing before the Select Committee took place about two weeks after publication of the regulatory findings in the US and UK in relation to Barclays in June 2012. In the course of searching questioning from the members of the Committee as to whether the Bank of England had suspected any dishonest manipulation of USD

Later, Sir Mervyn King was again questioned in similar vein by Mr. John Mann, “Q112 John Mann: … Despite the fact that there are serious commentators in the US questioning the viability of parts of the London market, you appear to be still in denial that it was known that LIBOR rigging was going on—the low-balling— when it was patently obvious to everyone that it was known. Why are you still in denial over that? Wouldn’t it help the situation if the Bank of England, along with the FSA, recognised that low-balling had failed to be spotted because you had other priorities because of the economic crisis? That honesty might give us some credibility in going forward and dealing with this crisis.

Ms. Demetriou QC made the point that such evidence emphasises how significant the publication of the regulatory reports concerning Barclays in the US and UK in 2012 was in marking a fundamental shift in attitude concerning the problems with LIBOR. The point made repeatedly by Sir Mervyn King and Mr. Tucker was that until that point there was no material available, even to the Bank of England, to support a conclusion that the difficulties with LIBOR had been the result of Panel Banks acting dishonestly in relation to LIBOR submissions rather than being the product of a flawed system in dysfunctional markets.

I consider that this point has real force. It is of course true, as Mr. Kennelly QC observed, that it was the FSA and not the Bank of England which had regulatory responsibility for LIBOR. Mr. Kennelly QC also suggested that the comments from Sir Mervyn King and Mr. Tucker were defensive and that the Treasury Committee had found the Bank of England to have been “naïve” in its earlier view that there was

As such, I consider that the evidence that senior figures at the Bank of England did not draw the inference until after the publication of the regulatory findings against Barclays in June 2012 that there had been dishonesty on the part of the Panel Banks or their employees in making their LIBOR submissions, supports FDIC-R’s argument that it would be unreasonable to have expected FDIC-R (which was an outsider to the UK banking system) to have reached that conclusion over a year earlier on the basis of the then publicly available material.

In addition to his analysis of the pre-March 2011 documents, Mr. Kennelly QC also submitted that the detailed contents of the subsequent regulatory findings (against Barclays and UBS in 2012 and thereafter against Lloyds/HBOS, Deutsche Bank, Rabobank and RBS) could not have made the difference, and FDIC-R could have pleaded its case against UBS on the basis of the pre-March 2011 materials alone.

The outline of the first set of regulatory findings against Barclays can be seen from the following extracts from the summary of the CFTC Order published on 27 June 2012,

During the period from at least mid-2005 through the fall of 2007, and sporadically thereafter into 2009, Barclays based its LIBOR submissions for US dollar (and at limited times other currencies) on the requests of Barclays’ swaps traders, including former Barclays swaps traders, who were attempting to affect the official published LIBOR, in order to benefit Barclays’ derivatives trading positions;…

During the period from at least mid-2005 through to at least mid-2008, certain Barclays euro swaps traders, led by a former Barclays senior euro swaps trader, coordinated with, and aided and abetted traders at certain other banks to influence the EURIBOR submissions of multiple banks, including Barclays, in order to affect the official published EURIBOR, and thereby benefit their respective derivatives trading positions; and

During the volatile, global market conditions of the financial crisis of late August 2007 through early 2009 … Barclays lowered its LIBOR submissions in order to manage what it believed were inaccurate and negative public and media perceptions that Barclays had a liquidity problem based in part on its high LIBOR submissions relative to the low submissions of other Panel Banks that Barclays believed were too low given market conditions.”

The findings of trader-based misconduct were made more explicit in the Appendix to the non-prosecution agreement between Barclays and the US DOJ published on the same date,

The regulatory findings in relation to UBS were published simultaneously in the US, the UK and Switzerland on 19 December 2012 and described what the CFTC termed

As regards collusion, the CFTC stated,

The extraordinary nature of this trader-based misconduct can be illustrated from the following extract from the CFTC report and non-prosecution agreement between the

As regards the false reporting to protect UBS’s reputation, the CFTC findings also included extensive quotations from communications between USD traders and USD LIBOR submitters in London discussing the implications of the two directions from within UBS to “err on the low side” and to be “in the middle of the pack”. These quotations made it clear that at least within UBS, the view was that,

Moreover, when discussing a possible response to the Wall Street Journal on 21 May 2008 in response to a question why UBS was paying 12 basis points for commercial paper more than it was posting as a LIBOR quote, a UBS manager stated that,

Mr. Kennelly QC made two main points in respect of regulatory findings such as these. The first was to rely on the fact that none of the regulatory findings against UBS (or indeed any of the other Bank Defendants) included findings of collusive Lowballing to suppress USD LIBOR throughout the Suppression Period as now alleged in this case. He submitted that in relation to UBS, the regulatory findings related only to (i) trader-based misconduct (primarily in relation to Yen LIBOR and other benchmarks apart from USD LIBOR) so as to improve UBS’s derivative trading positions, and (ii) unilateral Lowballing by UBS’s submitter for USD LIBOR to protect UBS’s financial reputation. Hence, said Mr. Kennelly QC, the regulatory findings against UBS in 2012 could not logically have been seen by FDIC-R to be relevant to the existence of a cause of action against UBS for collusive Lowballing to suppress USD LIBOR, but was merely something which improved its prospects of success in such a case.

The second point was that when FDIC-R pursued anti-trust proceedings in the US, it did so both against Panel Banks which had not been the subject of any regulatory findings, as well as those which had been the subject of regulatory findings. Mr. Kennelly QC argued that this showed that regulatory findings against any particular Panel Bank could not have been essential to establishing FDIC-R’s cause of action against that bank.

I do not accept either submission. As to the first, I consider that it takes too narrow a view of the regulatory findings to concentrate simply on the particular type of manipulation of a particular benchmark or benchmarks that they revealed. As I have explained, a key issue in the claim now made under TFEU or the CA 1998 is whether there has been an agreement or concerted practice. Prior to the regulatory findings there was no solid basis in the materials that would have justified the inference that dishonest collusion had occurred between the Panel Banks. There were other entirely plausible explanations for the low level of LIBOR and the other statistical observations that had been made.

What the regulatory findings against Barclays and UBS in 2012 showed, for the first time, and very strikingly, was not isolated instances of misconduct by traders in the market, or even isolated failures of management at some Panel Banks. Instead they showed that there had been widespread and systematic misconduct by employees at two leading banks, together with their network of contacts at other Panel Banks, to manipulate many of the benchmarks upon which much of the world’s financial system relied. In relation to trader misconduct this had involved extensive collusion between traders and submitters in a material number of the institutions; and in relation to both types of misconduct which had been identified, it was apparent for the first time (certainly in relation to UBS) that the activities of the traders and submitters had been known to, and participated in or directed by, senior management.

I do not think that it is at all obvious that such revelations can be surgically separated and compartmentalised by type of LIBOR and type of misconduct, and then discounted as irrelevant in the way that Mr. Kennelly QC suggests. It is true that there was no express finding in relation to collusive suppression of USD LIBOR on a long term basis in the regulatory sanctions and orders against Barclays and UBS. But there were new findings and admissions that manipulation and disregard of the fundamental principles underlying the process of setting the LIBOR benchmark had been widespread, and had involved numerous other traders and submitters in other institutions, together with the management of UBS.

I consider that it is difficult to overstate the enormity of those findings and the seachange which they represented in the material that I was shown. Put shortly, the regulatory findings were of a wholly different nature to the speculation and theorising which had gone before. As I see it, it is entirely realistic for FDIC-R to contend that prior to publication of the regulatory findings, the missing element which prevented it from pleading a cause of action under TFEU and the CA 1998 was any solid evidence to displace the natural assumption (shared by commentators and industry figures) that the sustained low level of LIBOR had other, innocent explanations arising from market dysfunction or parallel conduct: and that it was the regulatory findings that tipped the balance, making collusion the most likely explanation for those observations.

On a point of detail, although Mr. Kennelly QC pointed to FDIC-R’s alternative case on concerted behaviour, I was not referred to any material in the public domain prior to 10 March 2011 which could specifically have supported such a case of actual or blind-eye knowledge and continued Lowballing. The only solid evidence which might be said to indicate that particular type of concerted behaviour on the part of UBS was the comment that “the whole street was doing the same” from a UBS member of staff to which I have referred in paragraph 104 above. But that was only revealed in late 2012.

I am also not persuaded by Mr. Kennelly QC’s second point. It does not logically follow from the fact that FDIC-R chose to sue a Panel Bank in the US against which there had been no regulatory findings, that an antitrust case was properly maintainable against that Panel Bank. Nor was I told that any court in the US ever determined that this was so. Conversely, the fact that there were no specific regulatory findings in relation to a particular Panel Bank does not mean that the other regulatory findings, which made it clear that misconduct and collusion was widespread, did not tip the balance and make it likely that the Panel Bank in question was also involved.

For these reasons I am not persuaded by UBS that it would be unrealistic for FDIC-R to run a case at trial in reliance on Section 32(1)(b) that its claims are not statute barred. As I indicated at the outset, the onus lies on UBS to demonstrate on the evidence that the limitation issue should be determined summarily, and having considered all of the materials placed before me, I am of the view that it has failed to discharge that burden.

If and to the extent that UBS wishes to advance a limitation argument when it files its defence, it will be for FDIC-R to raise Section 32(1)(b), and then it will be for the trial judge to determine the issue.

I will therefore dismiss UBS’s application.