The Federal Deposit Insurance Corporation v Barclays Bank PLC & Ors [2024] EWHC 85 (Ch)

[2024] EWHC 85 (Ch)Case No FL-2017-000002IN THE HIGH COURT OF JUSTICEBUSINESS AND PROPERTY COURTS OF ENGLAND AND WALESFINANCIAL LIST (ChD)Venue Royal Courts of Justice, Rolls Building, Fetter Lane, London, EC4A 1NLDate 23/01/2024
Before
MR JUSTICE MILES
Between
THE FEDERAL DEPOSIT INSURANCE CORPORATIONClaimant(1) BARCLAYS BANK PLCDefendants(2) BANK OF SCOTLAND PLCDefendant(3) BBA TRENT LIMITEDDefendant(4) BBA ENTERPRISES LIMITEDDefendant(5) COÖPERATIEVE RABOBANK UADefendant(6) DEUTSCHE BANK AGDefendant(7) LLOYDS BANKING GROUP PLCDefendant(8) LLOYDS BANK PLCDefendant(9) NATWEST MARKETS PLCDefendant(10) NATWEST GROUP PLCDefendant(11) UBS AGDefendant
Marie Demetriou KC, Alex Barden & Charlotte Thomas (instructed by Quinn Emanuel Urquhart and Sullivan (UK) LLP) for ClaimantAdrian Beltrami KC & James Willan KC (instructed by Clifford Chance LLP) for First DefendantJames Duffy (instructed by Hogan Lovells International LLP) for Second, Seventh & Eighth DefendantsCharles Béar KC & Duncan McCombe (instructed by Macfarlanes LLP) for Third & Fourth DefendantsConall Patton KC (instructed by Milbank LLP) for Fifth DefendantSonia Tolaney KC & Nehali Shah (instructed by Slaughter and May LLP) for Sixth DefendantAdam Sher & Laurie Brock (instructed by Clifford Chance LLP) for Ninth & Tenth DefendantsBrian Kennelly KC, Paul Luckhurst & Andrew Trotter (instructed by Gibson Dunn & Crutcher LLP) for Eleventh DefendantHearing Hearing dates: 4 & 5 December 2023
Approved JudgmentThis judgment has been handed down remotely at 10.30 on 23 January 2023 by email to the parties’ representatives and publication on The National Archives

Introduction

[1]This judgment arises from the fourth CMC and the claimant’s (the C’s) application by notice dated 23 October 2023 for disclosure.[2]As the background is well-known to the parties I shall not set it out in any detail. In brief, the C, as the receiver of 19 closed US banks (Closed Banks), alleges collusive suppression by D1 and D2, and D5 to D11 (the bank defendants or BDs) of the USD LIBOR rate, which it says D3 and D4 (the BBA) facilitated, as well as unlawful exchange of information between all Ds, giving rise to liability under competition law and US tort law.[3]At CMC3 in December 2022 Zacaroli J set out a timetable for any applications for further disclosure to be made by October 2023 and for the C to replead its case with detailed particulars by 16 February 2024.[4]This timetable has been superseded by further disclosure being given by the BDs since 31 July 2023. The C accepts that it is not able at the moment to make granular requests for disclosure of the kind anticipated by the order made at CMC3. The C and the BDs have agreed that any such granular requests should be made once the further disclosure has been reviewed.[5]The C however contends that the court should nevertheless require the BDs to give certain disclosure (including transaction data) and to provide information to assist the disclosure process. The application notice sought five heads of relief.[6]After service of the Ds’ evidence the C stated that it intended to continue with four of these, as follows (by reference to the Updated Draft Order served with its skeleton argument): i) The Key Categories Application (paragraph 1 and Appendix 1 of the Updated Draft Order) – seeking to identify what searches the BDs have performed or are willing to perform, focused on “4 key areas” in the claim –(i) the BBA,(ii) the FXMMC,(iii) media inquiries and(iv) brokers. ii) The Voice Data Application (paragraph 2 of the Updated Draft Order) – seeking a process for identifying the speakers on voice call documents which the BDs have disclosed. iii) The Transaction Data Application (paragraphs 4-6 and Appendices 2(a) and 2(b) of the Updated Draft Order) – seeking to identify what categories of data and information the BDs hold. iv) The Refinitiv Application (paragraphs 8-12 of the Updated Draft Order) – seeking third-party disclosure.[7]The relief sought in the Updated Draft Order was in some respects different from that sought in the application notice. As explained below the C also sought further to refine the relief sought in a further draft produced at the hearing. The C has not applied to amend its application. This approach was the subject of criticism by the BDs who said that they had not had the chance to give further evidence addressing the precise form of order now sought.[8]The fifth head of relief sought in the original application notice was the Date Range Application (paragraph 3 of the Updated Draft Order). This sought to extend the date range for the Ds’ disclosure generally to cover the period 1 August 2007 to 31 December 2011.[9]The witness statements served in relation to the applications are extensive, running to several hundred pages. I shall refer to them by name and number. The C relied on Vernon 8 on the main application, Vernon 10 (in reply) and Vernon 9 on the Refinitiv Application. The Ds relied on Nicholls 1, Bickerton 2, Bullock 1, Clark 1, Stait 2, Bristow 1, Donnelly 7 and Owens 4. I have taken account of all of this evidence and of the extensive submissions made before and at the hearing even where not specifically recited in this judgment.

Summary of the background

[10]The C contends that the BDs made individual USD LIBOR submissions to the BBA (which was responsible for publishing LIBOR) which were known by them to be artificially low and not in accordance with the LIBOR definition (“lowballing”), with the effect of reducing the overall level of USD LIBOR. The C alleges that they colluded and agreed, including with the BBA, in doing so, and that they breached Art 101 TFEU and Chapter I of the Competition Act; and/or that they are liable in tort under US state law for fraudulent misrepresentation relating to USD LIBOR, and/or conspiracy. The C’s case is that the 19 Closed Banks of which it is receiver suffered loss as a result of the suppression of USD LIBOR by reducing their receipts from their banking businesses, including in relation to loans and derivatives.[11]The proceedings were commenced in March 2017, following a decision of the US Court in 2016 in litigation known as the LIBOR Multi-District Litigation (in which the C is one of a large number of plaintiffs) that it did not have jurisdiction against various non-US Panel Banks (including the BDs).[12]UBS applied (unsuccessfully) to strike out the competition claims on limitation grounds. The determination of this issue took until July 2020.[13]CMC1 was held in March 2021, before me. This considered, inter alia, the scope of the first stage of disclosure.[14]CMC2 was held in February 2022 before Sir Anthony Mann. He granted a case management stay, following a decision of the US Court of Appeals in December 2021, so as to enable the parties to consider next steps. He also ordered the C to produce certain of the further information sought by the BDs.[15]CMC3 was held before Zacaroli J in December 2022, following the lifting of the stay. This set down a timetable to a trial in February 2026 of the claims of three Closed Banks (“the Trial 1 Sample Banks”). He also ordered that the C and the BBA complete standard disclosure by 31 July 2023. While not the subject of a Court Order, the BDs also indicated that by 31 July 2023 they would give disclosure of further documents to be produced by them in the US (pursuant to orders of the US Court or compromises of applications made there).[16]The order at CMC1 required the parties to seek to agree on the disclosure each of them would give, with the matter to be referred to CMC2 if agreement could not be reached. This process was postponed by reason of the attempts to resolve the English proceedings described above.[17]The C and the BBA agreed to give standard disclosure. At CMC3 a deadline of 31 July 2023 was set. The C and the BBA met that deadline.[18]At CMC1 the order provided that the BDs were to give a first tranche of disclosure by providing, by 30 April 2021, their “US Disclosure Sets”. This was disclosure given in the US proceedings. That disclosure was provided on a “lift-and-drop” basis, as the BDs put it: they had given extensive disclosure in the US proceedings and had not further reviewed the US Productions to limit them to the specific issues in these proceedings. The BDs disclosed the First Tranche in May 2021. This approach was explained by the BDs on the basis that there was a substantial overlap in the issues in the two sets of proceedings and that to re-do the disclosure process would involve unnecessary duplication.[19]There was substantial subsequent correspondence between the parties about the BDs’ disclosure. The C sent long letters in November and December 2021 raising questions about the BDs’ disclosure, which the BDs answered in a series of letters by September 2022. The C wrote again on 31 March 2023 and the BDs responded between 16 and 29 June 2023.[20]In the meantime, in the US Proceedings the plaintiffs applied for specific further disclosure. The plaintiffs compromised with some of the BDs and the Court required some of the others to give specified further disclosure.[21]More specifically the plaintiffs made an application to the US court for further disclosure relating to six broad categories, including(a) communications with interbank brokers to co-ordinate their LIBOR submissions; and(b) the role of the BBA and related committees including the FXMMC. The US plaintiffs stated that they had spent thousands of hours reviewing the US Productions and had identified certain common gaps. These included communications between the banks and the BBA in respect of meetings and calls during the relevant period. The plaintiffs identified a large number of additional “relevant search terms that the Defendants did not search” and additional custodians. They sought the addition of more than 100 search terms and more than 40 custodians. Judge Buchwald made an order in April 2023. She applied the US test of relevance, which appears to be similar to our own, and (in doing so) considered whether the plaintiffs had established the probability of documents being found by the new terms which would not have been captured by the existing search terms. She noted the breadth of some of the existing terms including “BBA” and “LIBOR”. She decided that some but not all of the search terms sought by the plaintiffs should be applied. She also required some additional custodians.[22]The application ultimately proceeded against three of the BDs. Others, including D5 and D2/7/8, reached a compromise with the plaintiffs.[23]Some of the further documents were provided in England from December 2022 onwards, but the production took longer than had initially been expected. Between 7 July 2023 and 13 October 2023 over 69,000 documents were provided. The production process was not complete at the date of the hearing before me. D11 explained that there were over 15,000 further documents still to come.[24]The C’s position at the hearing before me was that it had been reviewing the BDs’ productions, but for the purposes of making more granular requests it was going to be necessary to look at everything together, including the more recent productions, and to compare each BD’s disclosure to that of the others and of the BBA. As a result, like the Ds, the C was not yet in a position to make granular requests at this hearing.[25]To summarise, as between the C and the BBA, standard disclosure was delivered on time in July 2023; there may be particular points arising, and the parties’ positions are reserved in that regard pending the next hearing. As for the BDs, disclosure has been made from the US Proceedings; there have been recent supplemental productions which are now (mostly) complete and, again, the parties have reserved the right to make specific disclosure requests and applications.[26]The trial is listed for 19 weeks from 23 February 2026. The key intermediate steps are as follows (the current dates having been set with a slightly earlier trial start date of October 2025 in mind):(i) the C to serve draft Re-Re-Amended PoC by 16 February 2024 - that date will need to be postponed in order to follow the provision of specific disclosure;(ii) factual evidence by 29 November 2024; and(iii) expert evidence to be completed by 27 July 2025.[27]The C and the BDs are agreed that there should be another hearing in Spring 2024 for the hearing for specific disclosure applications, with the C’s re-pleaded case to be served before the 2024 summer vacation, and other deadlines extended as a result. The BBA takes the position that there should be no further significant delays in any of the above steps.

The Key Categories Application

[28]As already explained, by the time of the hearing the C was seeking the relief in the Updated Draft Order and it is convenient to concentrate on that. By that draft the C seeks an order that the BDs are “to provide, whether by witness statement or disclosure certificate, a statement signed by a statement of truth which, for each of the categories set out in Appendix 1 hereto, explains:a. What consideration (if any) was given to the four key categories, and what searches (if any) have the BDs undertaken which they consider adequate to have captured that category.b. The number of documents disclosed overall by each BD compared to(i) its wider document universe and(ii) its regulatory productions.c. For each stage of the process – being, (a) the steps taken to identify the initial universe of potentially relevant documents; (b) the production of documents to regulators as part of the regulatory investigations; and (c) the preparation of the US Disclosure Set: i. The search terms applied to collate these documents; ii. The custodians selected, and how and why they were selected; iii. The repositories of documents that were searched; iv. The process of review implemented, explaining which issues were treated as relevant and how the review was carried out (i.e., was it a manual review or was TAR utilised); and v. Any other steps taken – e.g., use of document coding from previous reviews.d. What further searches, if any, the BD considers it could perform to capture documents in the key categories, and whether it is willing to do so, and if not the specific grounds on which it objects to doing so.” a. What consideration (if any) was given to the four key categories, and what searches (if any) have the BDs undertaken which they consider adequate to have captured that category. b. The number of documents disclosed overall by each BD compared to (i) its wider document universe and (ii) its regulatory productions. c. For each stage of the process – being, (a) the steps taken to identify the initial universe of potentially relevant documents; (b) the production of documents to regulators as part of the regulatory investigations; and (c) the preparation of the US Disclosure Set: i. The search terms applied to collate these documents; ii. The custodians selected, and how and why they were selected; iii. The repositories of documents that were searched; iv. The process of review implemented, explaining which issues were treated as relevant and how the review was carried out (i.e., was it a manual review or was TAR utilised); and v. Any other steps taken – e.g., use of document coding from previous reviews. d. What further searches, if any, the BD considers it could perform to capture documents in the key categories, and whether it is willing to do so, and if not the specific grounds on which it objects to doing so.”[29]The categories set out in Appendix 1 are:
“1. Communications between the USD Panel Banks, and with the BBA, with respect to each of the meetings and calls that took place across the Disclosure Period (as defined below), which involved discussion of USD LIBOR submissions and/or rates. 2. All Foreign Exchange and Money Markets Committee (FXMMC) meeting minutes, and all communications of each BD with respect to any FXMMC meeting or other meeting (physical or by telephone) whether internal and / or with other Panel Banks and/or with the BBA and/or the Bank of England relating to LIBOR across the Disclosure Period (as defined below). 3. Communications within each BD and between that BD and the BBA with respect to media enquiries about lowballing / USD LIBOR manipulation / USD LIBOR submissions / USD LIBOR rate movements, and communications between that BD and another USD Panel Bank and/or the BBA with respect to such enquiries. 4. All communications with brokers about the actual or proposed USD LIBOR submissions of a USD Panel Bank and/or the USD LIBOR rates as set by the BBA, along with all internal communications within that BD about any such communications with those brokers.”
[30]During the hearing the C produced a further version of the relief sought under this head, as follows (the Revised Relief):
“… the BDs are to provide, whether by witness statement or disclosure certificate, a statement signed by a statement of truth which, for each of the categories set out in Appendix 1 hereto: a. Identifies the person or persons most likely to have held relevant documents (including the name and job title); b. Explains the extent (if any) to which the disclosure provided to date in these proceedings is apt to capture those persons’ documents in the relevant category, and why; and c. Explains what further steps (if any) could be taken to identify further relevant documents, and whether it considers that such steps would be proportionate, and if not, why not.”

Outline of the C’s submissions

[31]Disclosure is governed in this case by CPR 31 and PD 31C (which governs competition cases). The touchstone is proportionality. This applies to disclosure and specific disclosure.[32]The First Tranche disclosure came entirely from that already produced in the US proceedings, which was in turn only a subset of the disclosure prepared by each BD for the purposes of their earlier regulatory investigations (the Regulatory Productions). None of the BDs have, therefore, incurred any separate cost for the purposes of giving disclosure in the present proceedings in England. Each of the BDs is a large, global financial institution with considerable resources, well able to carry out the further disclosure steps necessary to secure a fair resolution of these proceedings. The value of the claim is well in excess of USD 200m. It is complex litigation being carried out by C, a public body, concerning serious misconduct on the part of the BDs which has already been the subject of wide-ranging regulatory findings.[33]As is often the case in fraud/cartel proceedings, the defendants hold all relevant documents. Ensuring that the right range of disclosure is provided is of central importance.[34]Properly tailored disclosure in a large and complex disclosure exercise such as this requires a proactive approach on the part of litigants. In The RBS Rights Issue Litigation [2015] EWHC 3433 (Ch), at [78], the Court referred to the need for a focused approach to the disclosure process, which starts by considering “top down” where and how relevant documents which are the “real stuff of the case” might be located rather than undertaking a mechanistic search of custodians’ documents.[35]In Coll v Google [2023] CAT 72, the tribunal made orders for the defendants (which had given disclosure of documents produced in US regulatory proceedings) to give further information about the manner in which the disclosure had been given, and in particular, the approach taken to the identification of custodians and search terms and to answering questions posed by the claimant. Analogous orders are sought here.[36]The initial selection of documents by the BDs for the Regulatory Productions was, by design, not focused on the issues in these particular proceedings, but rather on the requirements of the regulators in their investigations and whatever material the BDs thought was in their interests to provide by way of exculpatory material. The BDs have refused to explain what the scope of the investigations was or the regulatory requests to which they were responding.[37]With the partial exception of D5, the BDs have not given a full and proper explanation of the second stage, namely how those regulatory searches were winnowed down to provide disclosure in the US Proceedings.[38]The BDs have not given (and say they are not required to give) any form of Disclosure Statement confirming that the searches they did were reasonable and proportionate searches covering the issues for disclosure in the present proceedings.[39]The BDs have recognised from the outset that additional disclosure might well be required. They accept that it was anticipated that there would be specific requests arising from the lift-and-drop disclosure for particular documents or categories which had not been captured. Conversely, a large volume of the lift-and-drop documents will not be relevant to the issues in these proceedings, because they were produced for the purposes of the regulatory investigations and not directed at the issues in this case. It is not clear that there has been a proper “top-down” inquiry about the matters which are of central importance.[40]By the Key Categories Application, the C seeks clarity as to what exactly the BDs have done by way of disclosure which they consider covers these Key Categories, and what more (if anything) they are willing to do – so that there can be discussions (and, if necessary, an application) about further steps in advance of making granular applications and in advance of the next CMC.[41]The C’s review of the disclosure to date has shown that the four Key Categories were insufficiently addressed by the “lift-and-drop” disclosure.[42]Specifically: i) According to the evidence of Ms Vernon, there are substantial gaps in these areas (as opposed to mere individual omissions such as the absence of a reply to a particular email). ii) The BDs have given little or no explanation as to the direction and rationale of their searches. iii) As to the information that has been provided, this is inconsistent as between the BDs. The quality of the information varies as between the BDs but some of them fail adequately to explain what happened at the various stages. iv) Though the BDs have said in their evidence in response to the application that there are further potentially relevant documents within these categories in the 70,000 or so documents provided since July 2023, the sampling carried out by the C suggests that this is wrong. v) The C’s position is not that there are no relevant documents in the disclosure on these topics – rather that(i) it is apparent that there are serious shortages which go beyond granular gaps and(ii) what is known so far about the way the BDs carried out their disclosure exercises is vague and incomplete, but suggests that these key areas were not a point of focus. What is now required is a focus on the necessary categories of disclosure that the C has identified.[43]As to the four categories, the C submitted (in summary) as follows.[44]BBA Communications. The C’s case is that it is known from the publicly available material that the BBA received many complaints about the lowballing of USD LIBOR, and was instrumental to the continued holding out of LIBOR as a genuine rate.[45]The C relies on the account given by Mr Ewan, the BBA’s LIBOR Manager, of his contacts with the BDs, in a witness statement to the SFO. The C contends that the documents actually produced by the BDs do not appear to match the nature and extent of communications described by Mr Ewan.[46]The responses of the BDs to date are vague and unhelpful. They have not said that they have disclosed all relevant documents. Rather they have (in some cases) identified some responsive documents. The right approach would be for them to identify exactly who Mr Ewan’s contacts were, and to focus specifically on communications between them and him (or internal communications referring to him).[47]I note that the C’s counsel opened the application on the basis that the BDs had not even specifically searched for documents containing Mr Ewan’s name.[48]FXMMC Meetings. Mr Ewan explained the role of the FXMMC in his statement in the SFO investigation. There were 16 known meetings in the relevant period. For most of the meetings, most of the BDs have disclosed very little, if anything at all. The evidence of the BDs is not that they have carried out specific searches relating to these meetings. Rather they appear to have carried out a retrospective application of search terms to what has been disclosed to date to suggest that they may have disclosed potentially relevant documents concerning this category. There should again be a top-down exercise. The C wants to know what searches the BDs have done (if any) which they would reasonably expect to have captured substantially all relevant documents relating to the FXMMC meetings.[49]Media Inquiries.USD LIBOR was a focus of public scrutiny in the key period, including by a journalist, Carrick Mollenkamp, in articles in the Wall Street Journal and in articles on Bloomberg. These articles were discussed within the BDs. The BDs (or some of them at least) were also approached for comment. Internal discussions, and composition of a response (or decision not to respond), will be important.[50]Some BDs, D11 and D5, have disclosed nothing about Mr Mollenkamp’s inquiries, while four others have disclosed some documents but apparently not a full chain. This indicates that none of the BDs carried out a focused search in relation to such inquiries, or followed through when (limited) documents were unearthed.[51]This should be the focus of top-down inquiries, by identifying the relevant people and the searches so far undertaken; and the further proportionate steps that could be undertaken.[52]Broker Communications. Vernon 8 explains that there were conversations about LIBOR with money brokers, who then disseminated the information to other banks. The C says that there appears to have been a process whereby brokers would frequently facilitate the exchange of information with USD LIBOR submitters, by asking them what their LIBOR submission was likely to be and giving a “run” of the other Panel Banks. Vernon 8 gives specific examples of such communications.[53]Documents relating to these communications are likely to form a key area for cross-examination at trial. It appears from the descriptions in the existing examples, and from the apparently routine nature of the communications, that such exchanges were very regular. However, what has been disclosed is incomplete even on its face, and appears to be an essentially random selection likely to be only the tip of the iceberg.[54]The BDs do not say that they have carried out targeted searches for those communications.[55]It will not be difficult for the BDs at least to identify what they have sought to do so far. The list of brokers identified by the C is modest, and each BD had only a handful of submitters.[56]Returning generally to points concerning all four categories, the evidence shows that the BDs were not in a position, even after the further productions in the US since July 2023, to say that the full disclosure to date probably includes all relevant documents to the Key Categories. The C therefore seeks a top-down inquiry of that kind. The initial broad regulatory disclosure was not conducted with the Key Categories in mind and the C’s review of that and the further disclosure shows that there remain substantial gaps on these key issues.[57]While more documents were produced as a result of the order or compromises in the US proceedings in 2023, the categories sought by the plaintiffs in the US were different from the Key Categories. Two were not identified there at all and two overlapped. The US plaintiffs were shooting in the dark in the US disclosure application. The purpose of the present application is to allow a more focused approach to specific disclosure here.

Outline of the BDs’ submissions

[58]Counsel for D11 took the lead on the Key Categories Application, with some of the BDs adding short submissions. In brief summary, the BDs submitted as follows.[59]The court has deliberately not made an order for standard disclosure in this case. Rather at CMC1 the BDs were ordered to produce their US Disclosure Sets. This reflected the fact that there were substantially overlapping proceedings here and in the US and that the BDs had carried out an enormous amount of work in producing their disclosure for the US proceedings. It has been understood throughout that the BDs and their UK lawyers have not carried out a standard disclosure exercise in the usual way. To do so would have been hugely duplicative and wasteful.[60]It was also anticipated at CMC3 (and reflected in the order made then) that once the C had reviewed all of the disclosure, including that given pursuant to the recent order of the US court (or compromises made there), the C would formulate further focused disclosure requests. That made sense because the disclosure in the US case covered the same issues of collusion and damage as this one. In a letter of 9 December 2021, the C’s solicitors said that on any view there are substantively identical issues involving many of the same parties on both sides of the Atlantic so it plainly made sense to have a single pool of documents.[61]In 2023 the US plaintiffs, including the C, raised concerns about the scope of the disclosure given in the US proceedings and sought further searches with additional search terms and custodians, designed to address a number of the same categories as those now in issue. The BDs were required by the US court’s order (or relevant compromises) to give further disclosure. Since July 2023 they have provided the same disclosure in tranches in these proceedings. As already explained, c.70,000 documents have been given and there are more to be served by D11.[62]The C should have reviewed them and formulated its requests as envisaged by the order at CMC3. Instead it is effectively asking the BDs to audit their own disclosure by reference to the Key Categories and themselves propose further searches which they might be able to undertake.[63]The Key Categories are covered by the disclosure already provided. The BDs have explained how their disclosure was gathered and there is no proper basis to require the BDs to formulate further searches. If the C wishes to do so and to seek appropriate orders that is a matter for the C, and for the court to adjudicate upon.[64]The question is not whether it can be said now that the C has everything they might ultimately need for the trial; it is whether they have sufficient material to formulate focused requests - or whether they need further information before they can do so. The answer is the former.[65]The C seeks to rely on gaps in the existing disclosure. But, first, that is premature as there has not yet been a full review of the 70,000 or more further documents. Secondly, the BDs’ evidence takes issue with the supposed gaps: the C appears not to have identified relevant documents in the further disclosure, and there are explanations for the non-existence of documents in the disclosure of particular BDs. Third, even if there turn out to be gaps, the right approach is for the C to formulate specific disclosure requests in the normal way and the C has sufficient material, or building blocks, to do that.[66]As to the process that was carried out in the original disclosure, the BDs have given a large amount of information and the C’s submissions greatly exaggerate its supposed ignorance. The BDs have each described the harvesting process in response to the original US and UK regulatory investigations, including the document custodians and types and a list of the search terms applied. Overall tens of millions of documents were searched at the cost of hundreds of millions of pounds, leading to the regulatory production of millions of documents. The search terms used by the BDs varied. But they were extensive and there was every reason to think that they would identify documents within the Key Categories. The evidence showed that there were some common search terms and very broad search terms such as BBA, FX, FXMMC and the names of various brokers. In answer to the emphasis placed by C at the hearing on documents concerning Mr Ewan, at that initial stage a number of the BDs used “Ewan” as a search term. Specifically, the name “Ewan” was a search term by D5, D6 and D2/7/8 in their original disclosure exercise. Other terms included “Wall Street Journal” and “Bloomberg” (D5); “wall” and “Wall Street Journal” (D9/10); and “Wall Street Journal” and “WSJ,” and “Media” and “Request” (D11).[67]The BDs have also each given an explanation of the roles of the custodians within their organisations (e.g. as submitters, traders, managers etc.).[68]The BDs have also described how the regulatory documents were winnowed down for the purposes of the US proceedings. Much detail has been given in the disclosure reports and EDQs but in essence the winnowing consisted of removing documents concerning LIBOR in other currencies and narrowing the date range to that in issue in the US case.[69]There have been a number of questions raised in correspondence by the C’s solicitors culminating in detailed responses by the BDs in September 2022 and June 2023. This is the right approach to disclosure in large scale litigation of this kind.[70]Ms Vernon has suggested in her evidence for the present application that the BDs had only recently (in their evidence for this hearing) provided important information. This was incorrect in material respects. The BDs had given much of it by September 2022 or June 2023 at the latest. The BDs highlighted examples of such points by reference to information provided by D5 in its disclosure report and letters of September 2022 and June 2023; and by D9/10 in the disclosure report and later correspondence. The BDs submitted that this showed that the C’s legal team had not properly reviewed the information already available about the building blocks needed for formulating specific requests.[71]Under the US Court’s 2023 order (or the relevant compromises) various further search terms were included (including e.g. the names of the brokers identified by the C for the purposes of the fourth Key Category in the present application). The negotiations that took place with some of the BDs at that time also led to the agreed introduction of further custodians and search terms. For instance D1 agreed to include “Ewan”.[72]The BDs dispute the suggestion that there are widespread gaps in their disclosure as alleged by C. They say that many of the supposed gaps are likely to be filled by the 70,000 plus additional documents. They submitted that Ms Vernon and her team appeared to have overlooked documents within the Key Categories. But that dispute cannot be determined on the current evidence. Any application for further disclosure should be made once those documents have been reviewed.[73]As to the reformulated form of the request made by the C in the course of the hearing, the C already knows(a) the identities of the BDs’ custodians and their roles within the relevant organisation; and(b) the search terms applied to these custodians. The C is therefore already in a position to make further requests – it has the building blocks. That is essentially what the C (and other US plaintiffs) have already done in the US proceedings. If it has further justifiable requests it can make an application. But the court should not effectively throw onto the BDs the task of formulating further requests for C’s benefit. That would be burdensome and costly and may turn out to be entirely unnecessary.[74]To the extent that there remain real questions about such matters as media inquiries these could and should have been addressed in correspondence (on the model of what happened in the US action).[75]Overall the BDs’ position is that they have undertaken reasonable searches for documents falling within the Key Categories. If the C is dissatisfied with this it can and should apply for specific disclosure.[76]This case is distinguishable from Coll. The C has far more information than the class representative had in Coll. In any event even in Coll the tribunal did not require the court to formulate further requests which could be undertaken.

Conclusions

[77]For the following reasons on balance I prefer the submissions of the BDs. I am not persuaded that the court should, at this hearing, make the orders sought by the C.[78]The starting point is that the C is not inviting the court at this stage to decide an application for specific disclosure of the kind anticipated by the order made at CMC3. It is common ground that there will have to be a further hearing for that purpose. Instead the C is seeking to require the BDs to undertake preliminary steps which the C says would assist it to make more specific and focused requests.[79]In my judgement the C has exaggerated the difficulties it faces in making more granular requests of the kind envisaged by the CMC3 order. The C already knows the identities of the BDs’ custodians and their roles within the BDs’ organisations; and the search terms applied to these custodians. The C and its lawyers (here and in the US) also have a massive quantity of underlying documents from which they have been able to discern information about the conduct of the BDs. The C and their lawyers have had access to that documentation for some years and the C must have a reasonably detailed understanding of the way the BDs were organised and who communicated with whom about the relevant events.[80]Counsel for the C argued that it was simply not in a position to ask for specific documents, such as (say) an email between A and B on a given date in 2010. I do not accept that the C is unable to make some requests of that specific kind – indeed it has identified some very specific gaps in the disclosure already. An example is found in a series of documents referring to calls or meetings with Mr Ewan. The C has identified certain calls or meetings by date and says that certain of the BDs have failed to disclose documents about them.[81]But more significantly, there are many degrees of granularity – and the evidence shows that the C is already able to ask for targeted, reasonably narrow, categories of documents even if it is unable to ask for single documents. The C has also done this as one of the plaintiffs in the US proceedings. I am unpersuaded by the argument of counsel for the C that it was hampered from doing that effectively because of its lack of understanding of the disclosure already given in relation to the Key Categories. As already stated, information has already been given about what was done in order to give the US Productions and the C also has a large bank of knowledge accumulated through its analysis of the very large volume of already disclosed documents. Adopting the language of counsel for the C it appears to me that the C already has the building blocks for deciding whether to make further disclosure requests. I do not accept the submission that the C will be shooting in the dark.[82]Then there is a timing point. The BDs have disclosed 70,000 documents since the end of July (and there were, at the date of the hearing, more to come from D11). Counsel for the C informed the court that most had been reviewed by the date of the hearing. The C contended that there are large gaps even taking account of these new documents. But there are disputes about that, with the BDs saying that C’s lawyers appear to have overlooked numerous relevant documents within those 70,000. The parties did not invite me to resolve these disputes at this hearing and indeed accepted that I could not do so on the basis of the evidence before me. The C submitted instead that the court can already see at a high level that there are holes in the disclosure. But until the allegations that there are gaps are resolved on full evidence, it appears to me that it would be premature to require the BDs to undertake the further steps proposed by C on the basis of the C’s assertion that there remain significant gaps.[83]I also consider that the C has understated the amount of work that the BDs would be required to undertake, and cost they would be required to expend, to carry out the steps contained in the revised proposals formulated during the hearing. I agree with the submission of the BDs that an order to carry out the steps now sought by the C would require the BDs to undertake something like an audit of their disclosure in order to say whether what they have done is apt to capture all relevant documents and why; and to formulate what if any further requests might be made. It appears to me that such an exercise would not be the “modest” one claimed by C. In order properly to comply with the proposed order the BDs and their lawyers would have to undertake a careful and comprehensive review. I have no confidence that the court is properly armed, on the evidence before it, to reach a sensible cost-benefit analysis of the steps proposed by C. I am not persuaded in the current state of the evidence that it is enough to say that what is proposed may help focus the debate at the next hearing. That is too broad and tenuous a basis for requiring the BDs to take steps which, to my mind, are likely to be extensive and costly.[84]I do not of course suggest that, on the intended further disclosure application, the court may not ultimately decide that further searches are indeed required. But the court should be armed with far more detailed evidence about the searches and/or other steps which the C proposes, the probable benefits of such steps (in light of the searches already done), and the likely costs and burdens of such steps being taken.[85]This case is far removed factually from the Coll case. Here the C has a very large amount of information about the disclosure given by the BDs and has also reviewed the great bulk of it. It is not in a position analogous to the class representative in that case (which had very limited understanding of the case). Moreover, the relief sought by the C includes an extensive review of the disclosure which has already occurred and an order that the BDs should state what further steps if any could be taken to identify further relevant documents and whether they consider such steps to be proportionate and, if not, why not. That goes well beyond anything ordered in Coll.[86]I also take account of the duty of parties to litigation of this kind to co-operate to promote the overriding objective. I see no reason, on the basis of the correspondence and evidence I have been shown, to suppose that the BDs and their representatives will not properly co-operate in relation to any granular requests that may be made. Counsel for various of the BDs made the point that requests for information about, e.g., the identity of media contacts, could and should be addressed further in correspondence. I have given significant weight to these assurances that there will be proper co-operation and engagement with reasonable requests for information about custodians and searches, including by reference to the Key Categories. The BDs will be required to engage substantively – it is unlikely to be enough, for instance, for them simply to say that the C already has lists of the functions of the existing custodians. The court hearing any substantive disclosure application will expect the parties to have shown a high degree of co-operation in respect of such requests. I also emphasise that what is said in this paragraph is to be taken as serious guidance by the parties and is not to be taken merely as a high-sounding, well-meaning exhortation. It may also be relevant to the court’s future adjudication of specific disclosure requests. Putting it another way, ultimately I accept the submission of the BDs that the C should explore its requests in respect of the Key Categories further in correspondence and, if necessary, apply to the court, once it has fully reviewed all of the further disclosure given since 31 July 2023. It is an important element of that submission that the BDs will properly engage in the correspondence process.

The voice data application

[87]The BDs have disclosed a total of 56,200 audio files, containing c.19,800 hours of audio. The voice data application seeks further details about the audio files.[88]In the application notice the C sought an order that by 26 January 2024 each BD (except D11) was to provide a disclosure list which sets out, for all audio data provided within their production set, the names and job titles of any participants to any telephone calls contained within that BD’s disclosure.[89]In the Updated Draft Order served with its skeleton argument, the C sought more limited relief, i.e. that each BD, in respect of any voice call disclosed by that BD and identified by the C as potentially relevant at any time between now and the start date of the trial, was to use best endeavours to identify the names and job titles/external organisations of any participants to that call.

Outline of the C’s submissions

[90]These files record conversations which are highly material to the dispute. The C says that some of the disclosed audio files show that the BDs were communicating with each other about their proposed LIBOR rates each day, before submitting them, directly or indirectly (through brokers). The C wishes to identify the individuals acting for the BDs who had those conversations. This is crucial evidence of manipulation and the unlawful sharing of information.[91]The information so far given about the voice files is limited and the C’s lawyers are not able to identify the participants in many cases. While some metadata has been given, this is often limited to the date and time of the calls and the identity of the person at the relevant defendant whose number was used and the number called (outward calls) or the number calling that number (inward calls).[92]As an example, D1 has explained that it has been able to identify the participants for 809 transcripts using a manual review but that it has disclosed 30,000 files where there is no metadata which would allow the identification of individuals other than the custodian for those files (i.e. the person(s) with the relevant recorded line for the call).[93]Another example is D7/8 where the audio files contain telephone numbers of non-custodian participants for only 10% of the disclosed recordings.[94]Hence the metadata so far given is of limited assistance to the C. The C has no personal knowledge of the identity of the participants in the calls and has no way of knowing who they were.[95]It will be necessary for participants to relevant calls to be identified before the trial to enable proper cross-examination.[96]The C relies on Practice Direction 31D para 31(1) which requires a party to give Disclosure Data for Electronic Documents (which include audio files). Disclosure Data includes the type and date of the document and its author/sender and recipient. The C contends that for voice records the author/sender and recipient means the participants.[97]In any case the court has the power to require a party to give information necessary for the fair and efficient determination of the dispute.[98]The C has so far identified 181 calls from the 56,200 calls. It seeks an order that the BDs undertake reasonable steps to identify the participants on those 181 calls.[99]The C submitted that the timing of the application was justified. The C’s review of the BDs’ disclosure was ongoing. The speakers involved in the relevant calls needed to be identified. It was partly because the BDs were not themselves prepared to carry out a separate relevance review of the US Productions that the C was having to do so. The process of reviewing the documents identified as relevant by the C was unlikely to be unduly burdensome. In any case nothing the BDs said explained why they could not at least review the 181 calls and seek to identify the speakers. The BDs have not sought to argue that these calls are not relevant. If it turns out that the BDs are unable to identify the speakers (having made reasonable efforts) they can say so. As to other calls to be identified by the C hereafter, the BDs should have liberty to apply to say that the process was proving disproportionate.

Outline of the BDs’ submissions

[100]This part of the argument was led by counsel for D5. The BDs submitted that the original application was for an order that the BDs identify every speaker on each of the 56,000 audio files. Even before that application was made the BDs pointed out that it would be disproportionate. The BDs repeated their points about proportionality and the impracticability of complying. These points, which the BDs say were self-evident, led to the C’s revised form of relief (in the Updated Draft Order).[101]As to that form of relief, it is based on a misconception of the requirements under PD31B at para 31. Where a party does not have the necessary disclosure data it is required to say so. Here the evidence shows that the BDs often have limited data about the custodian, in the sense of the date and time of the call and the person whose line was being recorded and, sometimes, the incoming line or outbound line. But even that information is not available for all of the BDs and all calls. There is nothing in PD31B to suggest that it is incumbent on a party to listen to audios and state who the participants are where the data is not immediately available. The rule does not require a party to create information which does not exist or cannot be reasonably ascertained.[102]As to the revised relief the BDs submitted as follows. The C’s proposal was wrong in principle. The obligation on the BDs would be open-ended, there would be no limit to the number of documents to review, and the C would be allowed to keep on adding to the list until the start of trial. This is potentially highly disproportionate. The only threshold for throwing the obligation onto the BDs would be the C’s (subjective) assertion that a call is potentially relevant. That would take the issue out of the control of the court and give it to the C. The best endeavours requirement does not affect the concern that identifying the speakers is likely to be difficult or even impossible; it will still require substantial work to be undertaken to enable the BDs and their solicitors to confirm that they have taken reasonable steps.[103]The C should instead go through the tapes and identify the material ones and should then explain what information if any it requires about the identity of the participants. The C should also have to explain the relevance of the file and why it is relevant.[104]Counsel for the C was wrong to criticise the BDs for now saying that most of the audio tapes were irrelevant. Instead the BDs were making it clear that they had not explicitly or implicitly accepted that they were all relevant to the issues in the case.[105]Counsel for the C was also mistaken to say that the BDs were saying that they would not say who the speakers were. Rather the BDs’ complaint was that there had not been a targeted approach, identifying the files which were material to the case.

Conclusions

[106]The C no longer sought at the hearing the very broad relief in its application notice. For what it is worth I am not persuaded that such relief would have been granted. There may be cases where an order of the kind sought would be simple. Here however I do not consider (in the light of the evidence) that it would be simple and straightforward for the BDs to identify all of the participants to the calls. The calls occurred many years ago and the evidence shows that the great bulk of the relevant employees have since left the BDs. The evidence also shows that there is limited metadata available to permit easy identification. It seems to me that there is force in the BDs’ arguments that an order of the kind sought in the application notice would have been disproportionate.[107]I also do not consider on the present material that the court should make the revised order sought by the C. I accept the BDs’ submission that it would effectively give the C an open-ended option to require the BDs to undertake potentially highly burdensome work seeking to identify participants to an unknown quantity of calls. On the C’s revised application that process would continue until the start of the trial and the selection of calls to be reviewed would be solely for the C with no threshold other than the C’s view that they are potentially relevant. That is to impose an order by reference to a subjective standard determined by the C. I do not accept the C’s argument that this problem would be cured by allowing the BDs to apply to the court to be released from the obligation if it turned out that it was becoming disproportionate. I do not think it is right to impose such a prima facie obligation, which may well be disproportionate, and then place the burden on the respondent to apply to vary or be released from the obligation.[108]Nor do I think that the obligation is cured by the imposition of the best endeavours standard. As counsel for the BDs observed, in order to decide who was on call A would the BDs be required to listen to calls B, C, D or E to see whether the same voices appear? It is far from clear where the obligation to use best endeavours would stop.[109]On the other hand I do not accept the BDs’ argument that the C must in every case explain the specific relevance of an audio file before it is able to seek some assistance from the BDs about the identity of the participants. It seems to me that that would be to introduce an unnecessary and potentially burdensome hurdle.[110]It also appears to me that there is some force in the C’s argument that the BDs may be able to do more than they have to date to identify participants. As the C submitted the BDs may still have access to former employees who are able to assist with this process. I also agree with the C that the voice records are likely to be important documents to the finding of facts at the trial and that it would be highly disruptive (and possibly inconclusive) if the identification of participants in calls only took place during oral evidence. If at all possible, the speakers should be identified before the trial starts.[111]This is an area where the overriding objective calls for co-operation and pragmatism. It seems to me that it would assist the court at the next hearing to have the benefit of some concrete experience, rather than simply hearing argument on the basis of general statements in evidence. The C has identified 181 calls which it says are relevant. I shall not make an order, as it seems to me unnecessary and (as explained above) potentially difficult to police, but I shall indicate (by way of guidance) that the BDs should co-operate and make reasonable efforts to identify the names and job titles/external organisations of the participants on those calls. They should also be in a position at the next hearing to explain what steps they have taken to seek to achieve this. It also seems to me that the C should explain which parts of the relevant conversations they are interested in where there are more than two speakers. At the next hearing the parties will be able to make focused submissions based on the actual experience of examining those 181 calls. This will provide the court with concrete evidence. It may be that there is some limited number of further calls which the C is able to identify as relevant in the same way as the 181 calls. The BDs should take the same approach to those as outlined in this paragraph. I also repeat my earlier comments about the duty to co-operate.[112]In short I conclude on this part of the application that the parties should follow the guidance set out in the previous paragraph and that, in the light of this, the court should not make the order sought by the C.

The Transaction Data Application

[113]In the application notice the C sought an order that by 26 January 2024 the BDs should conduct a reasonable search for and provide disclosure of certain categories of data, to the extent not already disclosed. There then followed nine categories. The first category was broken down into eight sub-categories of USD-denominated money market borrowing data and interbank lending and borrowing data covering the period from 1 January 2002 to 31 December 2016 (called the Data Request Period). The data was to cover 10 “data fields” (some of which were further broken into sub-fields). The BDs were to provide a disclosure statement containing a statement of truth certifying that a reasonable search had been carried out.[114]By the Updated Draft Order the C sought, in respect of the same categories, an order that each BD should provide a witness statement explaining(a) whether it considers that it has already disclosed a substantially complete set of the data falling within that category (and if so, explaining briefly the source);(b) whether it holds additional data falling within each category which have not been previously disclosed (and if not, why not);(c) for any data falling within sub-paragraph (b) whether it agrees to provide such data and (i) if so, its proposals as to the timing and form of such disclosure and (ii) if not, the specific grounds on which it objects to giving such disclosure (including any relevant estimate of cost). The explanation was to take place by reference to the same data fields as set out in the application notice. The proposed experts and the BDs were to meet to seek to agree the data (including the date range and the form of the data) and prepare a joint report of the agreed and disagreed categories of data.[115]The BDs’ broad position (set out in correspondence) was that this application should be held over to a further specific disclosure hearing next year to enable the parties, with the assistance of their experts, to discuss what further disclosure (if any) is required and can be given in a proportionate manner. They also invited the court to rule now (if it reached a sufficiently clear view) that categories 3-9 were entirely unjustified.

Outline of the C’s submissions

[116]The transactional data will be needed for the experts to be able to give their opinions about, first, the extent of US-denominated LIBOR manipulation (including the extent of any losses suffered by the Closed Banks) by reference to counterfactual, unsuppressed, rates; and, secondly, the incentives that the BDs had to manipulate LIBOR.[117]Categories 1 and 2 contain data needed for a regression analysis for the first of these areas. Categories 3 to 9 are data relevant to the incentives the BDs had to suppress LIBOR.[118]In competition cases data requests of this kind are typically expert-led. The C’s requests here are based on discussions with its own experts. It has explained the relevance of the categories of data in its evidence, and there is much missing material of the kind that its experts would expect to see.[119]It is common ground that there should be discussions about (at least) categories 1 and 2 informed by the views of the experts (the C in fact says that the experts themselves should lead this process). The orders now sought are intended to ensure that the experts should be able to have properly focused discussions where they know what data is likely to exist within the various categories and the practical consequences (including the costs) of production. Otherwise the experts (or solicitors informed by them) will have a merely theoretical discussion, uninformed by concrete information about what data is available and how difficult it will be to retrieve it. The purpose of this relief is to give the experts the building blocks for discussions to take place.[120]As to any possible dispute about the date range, again an expert-informed discussion is likely to be more focused and fruitful if the experts know what data is available for various dates than if they are working in a vacuum.[121]Specifically as to category 1, where there is no dispute about the likely relevance of at least some of the data (though there is a dispute about some of it and the relevant data period) the C has carried out an analysis (in Annex 4 to its skeleton) which shows that the different BDs have made differing disclosure of data within one or more of the sub-categories contained in the draft order. The BDs also appear to have used different date ranges for the data they have disclosed.[122]The court cannot properly conclude at this stage that categories 3-9 are obviously unjustifiable. The C’s experts consider that documents within these categories are needed for their incentives analysis. This is a question which should be informed by the experts. The C submitted that the court does not have the necessary competing expert material to reach a determination on these categories at this stage.

Outline of the BDs’ submissions

[123]Counsel for D1 led on this head, submitting in summary as follows. There is no adequate expert evidence justifying the various categories. Although the original application was for disclosure of documents, the revised application before the court is for information about documents. Since the C has not even established the relevance of the categories the court should not order the BDs to give information about them.[124]The process should be different. The experts should engage about the data needed for their evidence, but the present application (in both its original and revised forms) is premature.[125]The BDs have already provided huge amounts of data. For instance, D1 provided data concerning over 1m unique trades, D9/10 over 2.5m, and D11 over 1.3m, covering (in each case) at least six and a half years.[126]This application was foreshadowed in a long letter sent on 6 October 2023. That was over two and a half years after the transactional data was disclosed to the C. The BDs responded in a letter from Clifford Chance dated 1 November 2023 which proposed that the parties should discuss the requests further in correspondence. The BDs suggested that that should take place before an application was made so that disputed categories of transaction data could be determined on a granular basis. The present application was issued shortly afterwards.[127]The BDs’ position is, in short, that they have already disclosed vast amounts of transaction data. There should be discussions involving the experts to determine whether more is really needed. This goes both to the categories of data and the date range – at the moment there is a major difference between the 15 years proposed by the Cs and the 6.5 years of data already given.[128]The experts should have such discussions before the court should rule about the categories or the date range. Moreover the parties should be able to provide properly informed evidence explaining the positions of the experts about any remaining disputes. The court does not have a proper basis for deciding which of the categories will be required or the date range.[129]Some of the categories are widely drawn and would potentially involve a mass of material. They are extravagant, unmanageable and in some cases incoherent. Counsel made detailed submissions on some of these which (for reasons set out below) I do not need to summarise. There is no proper expert evidence justifying them, but even in their own terms they are impossible to justify.[130]The process of providing the information sought in the revised relief is likely to be costly and difficult. The evidence shows that for some of the categories the information was not centrally held within the BDs.[131]In order to state what there is within the categories (and provide the information sought by the C) the BDs would have to undertake full searches. Hence to comply with the revised relief the BDs would in effect have to undertake the disclosure exercise. This is the wrong way round. There should first be properly informed discussions, assisted by the experts, to seek to agree if possible what data is really needed. If agreement cannot be reached the court will have to adjudicate, but it will be asked to do so on the basis of proper evidence (informed by expert analysis) which explains the potential benefits and costs of any disclosure being requested.[132]The BDs’ overarching submission was that the application was premature. They did not seek to say that if the present application were to be dismissed this would (without more) prevent the C from returning to court to seek disclosure of further transactional data.

Conclusions

[133]I am not persuaded that it is appropriate at this stage to order the BDs to give the information sought.[134]On the evidence before the court it appears to me, first, that there are potentially substantial disputes about (at least) a number of the categories of information. It appears to me that there was little focused evidential support for many of the categories and that the categories may consist of little more than a wish list of information which the C’s expert would like in an ideal world. While Vernon 8 and 10 referred to the requests being informed by discussions with C’s experts, there was little detailed analysis of the way in which a large number of the categories are alleged to contain data which is really needed for either of the two purposes for which it is sought (regression analysis and incentives).[135]Second, it seems to me that there is bound to be a hierarchy of relevance in relation to the data, with the information in some of the categories being of substantially less materiality than others. Some may, on analysis, indeed have no real materiality at all. Moreover it appears to me that information in the lower relevance categories may only be required if there is a nil or inadequate return of data in other, higher relevance, ones. The court does not have the necessary evidence to reach proper conclusions on these points.[136]Third, there is likely to be a significant dispute about the date range of the data. The date range in the Updated Draft Order is the one selected by the C’s experts. But very little justification has been advanced for that in the evidence. There will be a dispute about this even in relation to categories 1 and 2, which are relevant to the regression analysis. The C says that its experts require a clean period on either side of the alleged period of suppression. The length of that period is debatable.[137]It appears to me that it is still more debatable whether a clean period is required for the other main purpose of proposed expert evidence, the BDs’ incentives (categories 3-9). The underlying allegation is that the BDs had profit-based incentives to lowball during the suppression period. It is far from obvious to me at the moment why there is any real evidence in their incentives outside that period and the C was unable to provide any particularly convincing explanation. I am not reaching a final conclusion on this point, but it is enough to say at this stage that there are likely to be serious arguments about the date range.[138]The selection of the date range could, of course, have a significant impact on the costs and management time involved in complying with any disclosure requirements.[139]Fourth, I am unable to accept the submission of the C that its proposal is a modest one. It appears to me that the process of providing the information sought by the Updated Draft Order is likely to be burdensome (and may indeed be very costly in costs and management time). Moreover some of the BDs have said in evidence that they do not maintain some data of the form assumed by C’s categorisations, so that providing the data would probably involve a process of reconstruction. It seems to me that the parties should be entitled to put in further evidence on the question of the burden of providing the data once expert-informed discussions have narrowed or refined the requests. Again I do not think that the evidence before the court enables the court to undertake a realistic analysis of the costs and benefits of the disclosure of the data.[140]I also accept the submission of the BDs that an order to give the information sought by the C at this stage would effectively require the BDs to undertake much of the work required for the disclosure searches themselves.[141]In this regard I accept the argument of the BDs that a number of the categories are widely drafted. Category 7 is an example. It is for “[a]ll profit and loss statements for all trading desks of each BD dealing in USD LIBOR-related products, as well as any and all underlying data on which those statements were based (including documents detailing the profit and loss attribution of each source of trading gain and loss) during the Data Request Period [sc. 1 January 2002 to 31 December 2016].” The request would require a search of profit and loss statements for all trading desks, and all underlying data and all documents detailing the attribution of profit and loss on which they are based over some 15 years. That is a potentially vast exercise. It is also open to the objection that it is not really clear what is being sought. It is worth recalling that the relief sought is for each BD to state whether it has already disclosed a substantially complete set of the relevant data and to say whether it holds any additional data within the category. The BDs have not sought to provide the category 7 data. In order to comply with the proposed order they would therefore have to conduct a search of a potentially huge data set.[142]Similar comments may be made about most if not all of the other categories.[143]Fifth, I am not persuaded that the experts (directly or indirectly) are unable to have properly detailed and informed discussions without the provision of this further information in advance. Such discussions will be iterative, with each side explaining its position, and with the experts being able to discuss the nature and extent of the information likely to be available, its likely relevance, and the extent to which it is covered by other data. The parties to these discussions do not have only one shot. They will refine their positions in the light of the information provided. I do not accept the C’s contention that they will be shooting in the dark. It seems to me that the C overstated the problems the experts would face in having such discussions in the absence of the explanations being sought in the proposed order.[144]Sixth, it appears to me that it is more in keeping with efficient and proportionate case management for the expert-informed discussions to occur before any application is made to the court for disclosure. It may be that the parties are able to reach agreement on an appropriate date range and data set. But if they are not, the court will be in a far better position than it is now to adjudicate on the disputed categories and date range if it is armed with evidence which takes into account the views of the parties’ experts. As things stand all the court has is broad assertions in Vernon 8 and 10 which have been informed by discussions with C’s experts, but which do not provide focused or concrete justifications for the data being sought. Nor does that evidence carry out the hierarchical analysis of the categories of data I have described above.[145]Part of the C’s submissions was that the court lacks the necessary expert evidence at this stage to determine the disputes about the relevance of the data. The C said that when arguing that the court could not properly dismiss the requests regarding categories 3-9. But it seems to me that the point applies generally to the state of the evidence concerning all of the categories and their materiality.[146]In short I agree with the submission of the BDs that the application is premature and that there should first be discussions involving the experts; if those discussions do not resolve matters the parties will then be in a position to inform the court with proper evidence about the benefits of getting the data and the attendant costs.[147]Seventh, as discussed with counsel for the parties at the hearing, the court expects and indeed requires a high measure of co-operation between the parties in seeking to resolve disputes about which data is to be provided. Counsel for the BDs agreed that the parties would co-operate. If there are disputes about transactional data which require adjudication at the next hearing, the court will have concrete evidence of the extent to which the BDs have reasonably provided information and will no doubt take account of this in framing any orders. The court is likely to be unimpressed if at that stage there has not been a reasonable degree of open engagement.[148]Eighth, as already explained, in reply the C concentrated some of its fire on category 1 and said that what it sought there was particularly modest. In relation to category 1, I am not persuaded that it would be appropriate to make the order sought: i) It would require each of the BDs to confirm that it has reasonably searched each of the 8 sub-categories within category 1. However some of these are contentious. Some (but not all) of the BDs have disclosed no documents in some of the sub-categories. And in others (“puttable instruments,” “callable instruments” and “other short term debt securities”) it appears from Annex 4 that none of the BDs have made any disclosure. For the same reasons as given more generally above I am not persuaded that data for these sub-categories (or at least all of them) will ultimately be required. ii) There was no detailed expert evidence to justify them and they may be more in the nature of a wish-list than information which is really material and needed. iii) It also appears to me that there is again likely to be a ranking of relevance, so that if documents in one sub-category are disclosed there may be little need for those in another (and vice versa). Debate on this is again likely to be informed by expert evidence. iv) There is also a difference between the parties about the applicable date range. Again that should be the subject of expert informed discussions. v) I am not persuaded by the C’s submission that the requirement to provide information about the missing boxes in Annex 4 is modest or would not involve the BDs in substantial time and costs. Expanding the date range on its own would require the BDs to carry out further searches. Moreover, as just explained there are some sub-categories where there has been no disclosure so entirely fresh searches would be required. Again the parties should be entitled to put in further evidence on the question of the burden of providing the data once expert-informed discussions have narrowed or refined the requests. vi) I am not persuaded by the C’s submission that expert-informed discussions would be fruitless without further information being ordered now. On the contrary, it appears to me that reasonable and constructive discussions should be possible at this stage to seek to flush out what further data (if any) is needed by the experts and the costs of retrieving it. The court at the next hearing will then be in a far better position to adjudicate on any remaining disputes about category 1. vii) I repeat my earlier comments about the extent to which the parties are required to co-operate with a view to ensuring that adequate data is provided - and about the view the court is likely to take at the next hearing if this does not happen.[149]While there is some force in some of the BDs’ arguments about categories 3-9 on the current information, I am not satisfied at this stage that categories 3-9 can be dismissed once and for all as obviously unsustainable. Expert-informed discussions should first take place.

The Date Range Application

[150]The C did not move this application before me and said it should be adjourned until after the result of the first trial. The Ds said it should be dismissed. I shall not say more about it in this judgment. There may be an issue as to whether it should be dismissed or adjourned until after the first trial (and the terms of any such order) but that can be determined in relation to the appropriate order.

The BBA’s timing points

[151]It will be seen from the earlier sections of this judgment that (as well as saying that the applications lacked merit) the BDs submitted that any applications for information or disclosure were premature and that there should be a further hearing as envisaged by the order made at CMC3. Counsel for the BDs agreed with the C that this should take place in April 2024 if possible.[152]The BBA took a different position. They submitted (based on judicial statements made at CMCs 1, 2 and 3) that the pleadings were conclusory and under particularised and that if the disclosure hearing only happens in April 2024 there will be inadequate time for them to digest and address any further pleadings by the C before the trial (given all the other steps required). While there is some force in the BBA’s concerns, I have concluded on this point that the position of the C and the BDs is to be preferred to that of the BBA parties. It appears to me that there is adequate time within the timetable if the disclosure hearing happens in April 2024. Moreover the alternative would be to determine now that the C cannot seek further disclosure at all – which would be at odds with the position taken by the BDs. However in order to ensure that the case proceeds efficiently towards the trial date in February 2026 the parties are required to continue to co-operate to enable a further disclosure hearing to be properly effective.

The Refinitiv Application

[153]I was not addressed about this at the hearing. I shall hear further from counsel about their proposals in this regard.

Conclusion

[154]I thank all counsel for the high quality of their arguments at the hearing. They should seek to agree an order giving effect to this judgment.

Cited in 1 later judgment