Property Alliance Group Ltd v The Royal Bank of Scotland Plc

This judgment deals with disclosure of documents and privilege. It follows from a number of previous judgments of my own in the same case ([2014] EWHC 4308, [2015] EWHC 321, and [2015] EWHC 322).

Briefly, the claimant (PAG) is a property developer with a portfolio worth about £200 million. It entered into four interest rate swaps with the defendant (RBS) between 2004 and 2008. Each swap employed 3 month GBP LIBOR as a reference rate.

LIBOR rates are determined by a form of averaging based on rates submitted by banks which are members of a panel. As is now well known, a number of panel banks, including RBS, have been found to have been manipulating LIBOR. RBS has reached settlements with regulators including the US DoJ and CFTC and the UK Financial Services Authority (as it then was). RBS has admitted misconduct to the regulators relating to Japanese Yen and Swiss Franc LIBOR.

PAG alleges that RBS made misrepresentations about LIBOR which induced PAG to enter into the swaps or that the swaps themselves contains implied terms relating to the conduct of RBS relating to LIBOR. Essentially PAG’s case is that by proposing LIBOR as a reference rate in the swaps, RBS represented that it was not rigging the rates for its own ends. In Graiseley Properties v Barclays Bank [2013] EWCA Civ 1372 the Court of Appeal held that allegations based on implied misrepresentation relating to LIBOR of the same general scope as the ones made by PAG in this case were properly arguable and should not be dismissed summarily.

In its Defence in these proceedings, RBS has formally admitted misconduct relating to Japanese Yen and Swiss Franc LIBOR and has denied any other misconduct relating to LIBOR. In particular RBS denies misconduct in the setting of GBP LIBOR, including 3 month GBP LIBOR.

At the hearing on 24^th November a major issue was the scope of the disclosure obligation applicable to RBS. I decided ([2014] EWHC 4308) that given the wide scope of PAG’s pleaded case, the disclosure obligation included all LIBOR currencies and tenors and was not limited to 3 month GBP LIBOR or to GBP LIBOR generally. Further consideration of disclosure was adjourned to a hearing to be fixed in early February 2015 with directions for RBS’s legal team to carry out various tasks relating to disclosure in the meantime. The key difficulty appeared at that stage to be that RBS has in its possession a very large number of documents (about 25 million) which would need to be reviewed if one was to carry out a full scale standard disclosure exercise on the relevant issues. A full standard disclosure exercise appeared to be necessary and proportionate focussed on 3 month GBP LIBOR, since that was the rate and tenor to which each of the swaps was referenced and in practice that exercise would cover GBP LIBOR generally (i.e. all tenors). However neither party nor the court had any appetite for requiring RBS to embark on a similar exercise dealing with all the other currencies.

It seemed in November that a way forward would be for RBS to disclose internal reports, reviews and summaries relating to the allegations of LIBOR misconduct. There was no doubt these so called “high level documents” existed. They would allow the parties and the court to direct a more focussed disclosure exercise in the light of the information revealed in them. A disclosure exercise could be properly tailored to the case in a proportionate manner.

RBS carried out the necessary review and produced a disclosure list signed by Paul de Gruchy, Senior Legal Counsel of RBS on 12 January 2015. This list includes a number of high level documents of the kind envisaged. However the only documents which RBS has produced for inspection by PAG from that list are ones which do not include any statement summarising the nature and extent of any manipulation of LIBOR. There are other high level documents in the disclosure list but they are all covered one way or another by objections to inspection taken by RBS. The objections include various forms of privilege: legal advice, litigation, and without prejudice.

The matter came back to court on 11^th February 2015. That hearing did not address the privilege claims. PAG reserved its rights to challenge them. RBS was directed to file a Further and Better List addressing privilege. The disclosure list before the court in February referred to regulatory investigations but only in an unspecific way. The references needed to be more specific in order for the court and PAG to understand the nature of the claims to privilege (cf. West London Pipeline v Total [2008] EWHC 1729 (Comm) at paragraph 86, particularly 86(1)). RBS was free to supplement any other aspects of its privilege claims if it wished.

The matter came back again on 8^th May to deal with privilege and case management. The parties filed very full skeleton arguments, seven lever arch files of papers and two bundles of authorities. The hearing lasted two days. This is the judgment from that hearing. The first part deals with PAG’s challenge to the privilege claims made by RBS. The second part deals with case management directions concerning LIBOR disclosure generally.

At the heart of PAG’s objection to the approach taken by RBS are some simple points. There is an obvious and significant imbalance between the information available to PAG about LIBOR manipulation as compared to that available to the bank. If the LIBOR scandal had not come to light at all, PAG would be none the wiser. Aside from RBS, the only sources of information available to PAG are public findings of regulators. These show that RBS did manipulate Japanese Yen and Swiss Franc LIBOR. They show that RBS has been fined. The public regulatory findings show that other currencies were investigated. Two currencies which have been investigated, at least to some extent, are USD and GBP. These were both mentioned in court at the hearing before me without objection.

PAG is not a banking regulator and naturally enough it would far prefer not to have to pay its legal team to trawl through thousands of documents trying to work out what did or did not happen. PAG argues, attractively, that by now, some years after the relevant events the bank must know what happened. That must be so given the highly regulated environment in which banks operate and it must be so as a matter of common sense. Those managing an organisation like this bank in these circumstances will naturally have wanted to have been informed about what happened and what went wrong in some detail.

Thus PAG argues there must be documents in existence which summarise and report on what happened. It argues that these documents cannot all be privileged. Surely, submits PAG, the Board of Directors or some sub-committee of the Board must have received information about what happened in order to perform its function. That reported information would be in summary form.

Accordingly, submits PAG, one way or another there is something wrong with the claim to privilege by RBS. Perhaps a wrongly narrow view has been taken of the scope of the documents which were the subject of the January disclosure list so that the claims to privilege are well taken but there are other, non-privileged, summary documents in existence and which ought to be disclosed. The problem with that is that RBS does not suggest that there are any documents which include summaries or reports of what happened in which it does not claim privilege. So, PAG contends, the only alternative is that the privilege claim is wrongly made somehow.

It is against this background that the detailed privilege issues fall to be decided. There is no doubt that access to privileged summaries would enormously simplify PAG’s task in these proceedings and would reduce costs substantially. However while this background may explain PAG’s motives in bringing this application, it is important to state that these case management considerations do not give PAG any right to see privileged documents.

The claims to privilege by RBS

The law is as follows. A party disclosing documents is entitled to take proper objections to inspection on the grounds of privilege (see CPR r31.3 and r31.19(3)-(6)) but one way or another such objections are open to scrutiny by the Court in appropriate circumstances.

Beatson J summarised the principles governing the approach to challenging claims to privilege in West London Pipeline v Total at paragraph 86. The key points are as follows. The burden is on the party claiming privilege. The court must be careful to consider how the claim is made out. Affidavits should be as specific as possible without making disclosure of the very matters that the claim for privilege is designed to protect. An assertion of privilege in an affidavit is not determinative but it is difficult to go behind an affidavit at the interlocutory stage of proceedings. The affidavit is conclusive unless it is reasonably certain something has gone wrong. Where the court is not satisfied that a right to withhold inspection has been established, it has four options, three of which are relevant to the case before me. The three relevant options are an order for inspection, an order for a further affidavit dealing with the deficiencies or an order whereby the court may inspect the documents itself.

In paragraph 86(3)(a) to (c) of his judgment Beatson J dealt with the exceptions to the conclusive nature of the affidavit which I have summarised above as something going wrong. He said:

“[…] The affidavit is conclusive unless it is reasonably certain from:

(a)

the statements of the party making it that he has erroneously represented or has misconceived the character of the documents in respect of which privilege is claimed: Frankenstein v Gavin's House to House Cycle Cleaning and Insurance Co, per Lord Esher MR and Chitty LJ; Lask v Gloucester Health Authority.

(b)

the evidence of the person who or entity which directed the creation of the communications or documents over which privilege is claimed that the affidavit is incorrect: Neilson v Laugharane (the Chief Constable's letter), Lask v Gloucester HA (the NHS Circular), and see Frankenstein v Gavin's House to House Cycle Cleaning and Insurance Co, per A L Smith LJ.

(c)

the other evidence before the court that the affidavit is incorrect or incomplete on the material points: Jones v Montevideo Gas Co; Birmingham and Midland Motor Omnibus Co v London and North West Railway Co; National Westminster Bank plc v Rabobank Nederland.”

PAG’s application is for an order that RBS produce the various documents for inspection or alternatively that the court should inspect the documents itself. It is supported by a witness statement of Philip Rubens of Cooke Young & Keidan dated 2^nd March 2015. The bank’s evidence on this application is a witness statement of Samuel Coulthard of Dentons dated 16^th March. There is also a statement in reply by Mr Rubens dated 30^th March.

The claims to privilege are made in three parts of Appendix B to the Defendant’s Disclosure list. Part 1 identifies various documents and claims the right to withhold inspection of them based on legal advice privilege. Part 2 identifies documents and claims the right to withhold inspection of them based on litigation privilege. Part 3 identifies documents and claims the right to withhold inspection of them based on without prejudice privilege.

The live challenges to privilege relate to some documents within part 1 known as the ESG documents, and to the whole of parts 2 and 3. I will address them in turn.

Legal advice privilege and the ESG documents

There was no dispute before me about the scope of legal advice privilege itself. PAG referred to Three Rivers District Council v Bank of England (No.6) [2005] 1 A.C. 610 and in particular to Lord Scott’s paragraph 38, which in turn referred to Taylor LJ in Balabel v Air India [1988] Ch 317 at 331-332.

The Executive Steering Group (“ESG”) was a special RBS committee formed in 2011. It has been at the centre of the various RBS investigations into the extent of its own LIBOR misconduct. The bank’s external legal advisers, Clifford Chance, played a key part in the work of the ESG and the meetings were attended by Clifford Chance.

The evidence on this issue is Mr Rubens’ 2^nd March witness statement at paragraphs 22-34, Mr Coulthard’s answer at paragraphs 60 – 66 and Mr Rubens’ reply is at 31-39.

RBS has confirmed that a number of documents meeting the definition of documents to be disclosed in the January list have been identified as part of the process of searching for ESG documents. This is not surprising given the manner in which these high level documents were defined and given the task apparently undertaken by the ESG. RBS claims to be entitled to withhold all of them from inspection on the grounds of legal advice privilege. Thus they are within Part 1 of Appendix B. RBS also argues that part of PAG’s complaint is based on a false premise. The premise is the assertion that RBS has claimed privilege in all ESG documents of any kind or at least all which could be called “minutes, papers and agendas” of the ESG committee. RBS has not done this. At this stage the only documents which have been disclosed are high level documents and thus it is in respect of high level ESG documents only that the claim to privilege is made.

Part 1 claims legal advice privilege in “original and copy documents passing between the defendant and its external and internal legal advisors for the purpose of taking and receiving legal advice on the regulatory investigations into the Defendant’s setting of LIBOR.” This category is then exhaustively defined in five sub-categories. All of the sub-categories define documents prepared by the defendant’s legal advisors and so, despite the generality of the first paragraph, all the documents in Part 1 are in fact documents prepared by those persons. The five sub-categories are (i) advice notes on the regulatory investigations, (ii) memoranda and summaries advising on periodic progress of the reviews and findings into the defendant’s setting of LIBOR, (iii) memoranda advising on the production of documents and information to certain regulators, (iv) documents advising in relation to meetings with certain regulators and (v) memoranda and advice notes prepared on (a) the outcome of investigations and findings, (b) discussions with regulators and (c) the implications for actual and contemplated civil proceedings brought against the bank.

Mr Rubens raised the role of ESG and the role Clifford Chance played in it in his evidence on 2^nd March. He questioned RBS’s solicitors’ statement that Clifford Chance were retained to provide legal advice and argued that the information provided in these proceedings so far showed that the law firm’s role extended more widely than that in such a way that no blanket privilege could arise. Secondly Mr Rubens pointed out that ESG documents such as minutes of meetings are obviously not communications between RBS and Clifford Chance. They may record legal advice, in which case that could properly be redacted, but that would not justify a claim to privilege in such documents as a whole.

In reply Mr Coulthard explained that RBS had not claimed privilege in all ESG documents as a category for the reason I have mentioned already. He pointed out that disclosure of ESG documents which are not high level documents is not yet due. At paragraph 65 (b) he addressed ESG documents which are within the January list and expressly set out, as he put it, to reassure the Court, that claims to privilege have been made after an individual assessment of each document, each of which fell within one or more of the sub-categories identified in Part 1.

In answer Mr Rubens maintained the concerns raised on the basis that a claim to privilege in relation to minutes of the committee seemed too extensive in nature, that RBS had still not explained with any clarity what the ESG documents within the disclosure list actually are in order to allow the claim to privilege to be understood properly. He pointed out that the ESG appears to have been a high level committee of senior RBS employees set up for the specific purpose of overseeing and coordinating the various internal and external investigations into LIBOR misconduct. He submitted that it could not seriously be suggested that the documents and papers of the ESG should not be disclosed.

A significant development came on 23^rd April in a letter from Dentons to CYK. At paragraph 13 Dentons wrote:

“As we have previously explained, the purpose of the ESG meetings was for our client to receive legal advice from its external legal advisers (principally Clifford Chance) on the regulatory investigations. Those legal advisors attended the ESG meetings for the purpose of imparting this legal advice, or receiving instructions from our client for the purpose of providing further legal advice.”

On receiving this letter PAG’s solicitors indicated on 28^th April that their client would no longer pursue the issue of ESG documents. They did not accept that the purpose had been “previously explained” that way and pointed to earlier wider descriptions, but CYK accepted (in my judgment rightly) that this description of the purpose of ESG supported the claim to legal advice privilege. On this basis it seemed as though the ESG issue would not be live before me (save for costs). However it was re-ignited when, from PAG’s point of view, the defendant’s skeleton argument characterised the purpose of ESG in a different way. The skeleton argument characterised the ESG as “formed … to oversee the investigations and potential litigation concerning LIBOR”. PAG therefore maintained its application after all.

A key question in getting to grips with this dispute about privilege is to identify the precise nature of the ESG and the role undertaken by it. In terms of its nature, PAG submitted it was a sub-committee of the Board of Directors, relying on the way the ESG is described in the defendant’s Disclosure Report dated 12^th January 2015. The purpose of PAG’s characterisation of the position of the ESG was to then rely on the statement by David Richards J in Financial Services Compensation Scheme Ltd v Abbey National Treasury Services Plc [2007] EWHC 2868 (Ch) at [17]-[18] that Board minutes are a “common example” of a document which is “clearly not privileged”. RBS denied that the ESG was a sub-committee of the Board and it is fair to say that the Disclosure Report is capable of being read either way. I will assume for present purposes that the ESG was not a sub-committee of the Board. Even if it was not, it was plainly an important committee operating at the highest levels.

As to the role of the ESG, PAG is entitled to be concerned that the description has changed. On the wider description of the role, it is hard to believe that the totality of its meetings can have been for the purpose of imparting legal advice. On the contrary it seems much more likely that one purpose of the meetings (or part of them) would have been to inform the bank factually about the outcome of the investigations the ESG was set up to oversee. In other words one natural purpose of the meetings of the committee set up to oversee investigations concerning LIBOR would be to tell the bank what has been found to have happened. No doubt another purpose was to give legal advice but it is hard to credit that that could be the sole purpose.

Mr Rubens drew together quotes from the disclosure list, correspondence and witness statements in this case which described the tasks Clifford Chance had undertaken at the ESG and which appeared to be much more administrative in nature than merely providing legal advice. They included acting as the secretariat for ESG and assisting in an internal investigation which the CFTC had requested RBS to undertake.

Before me Mr Railton did not seek to contend that the purpose of the ESG was as narrowly drawn as is stated in the letter of 23^rd April. His objections to this part of the application were as follows. First that there was no application for specific disclosure of ESG documents. That is correct but is focussed on a different point, concerning the balance of non-high level ESG documents. Second he submitted that there was no case advanced by PAG that ESG documents falling within the description in Part 1 of Appendix B were not privileged documents. This is the real point taken by RBS. Mr Railton also referred to a further letter from Dentons dated 8^th May which made essentially the same point, stating that regardless of the purpose of the ESG, all the ESG high level documents fall into four categories. The four categories in the letter correspond to sub-categories (i), (ii), (iii) and (v). There are differences in wording between the categories defined in the letter and those in the disclosure list. I will refer to the wording of the disclosure list. If the difference in wording was important then that should have been pointed out.

I will therefore return to the claim to privilege in Part 1 of Appendix B. Beyond the 8^th May letter which rules out sub-category (iv), RBS has not specified which sub-categories any particular ESG documents fall into. In that respect the evidence from RBS is not as specific as it could be (cf West London Pipe para 86(1)). I will take sub-categories (ii) and (v)(a) as an example. They include memoranda prepared by the bank’s legal advisors on the progress and outcome of reviews, investigations and findings. One would naturally have assumed there were high level ESG documents of this type in any case and the 8^th May letter seems to confirm that. Whether this sort of material is privileged depends critically on whether the ESG’s role was solely to provide legal advice or not. If it was then I can see the documents would in all probability be privileged but if not, not. If part of the ESG’s role included the task of overseeing investigations and reporting to the bank then I cannot see how the fact that a high level document is as described in sub-category (ii) and/or (v)(a) means it is privileged. Perhaps the document contains legal advice in which case that legal advice is no doubt privileged, but it does not follow from this that any factual summary in the document is privileged. Perhaps the correct approach is redaction, but I am having to speculate because RBS has not addressed this with any specificity.

Neither the information in the Disclosure List itself nor Mr Coulthard’s witness statement nor the correspondence allows the court to understand the basis on which this claim to legal advice privilege is made. Mr Coulthard’s statement does not articulate the point about the ESG’s role made in the 23^rd April letter. Assuming it was prepared on the same basis as the letter then I am not satisfied that basis is correct given the wider way in which RBS now describes the role of ESG. If it was prepared on a different basis then it does not answer the point made by PAG. RBS has not satisfied me that the claim to legal advice privilege in relation to high level ESG documents is correctly made. That applies to all the high level ESG documents in Part 1.

PAG’s representatives were very suspicious about the approach to ESG documents adopted by the bank’s representatives and submitted that given the history of this part of the dispute, RBS had had enough chances. PAG submitted I should not direct RBS to provide a further statement and the time had come to order the bank to produce the documents to PAG because the claim to privilege has not been made out.

I agree with PAG that there is no justification for the bank to be ordered to produce yet another statement of its claim to privilege. RBS was given a specific opportunity to review its claim knowing that a challenge was likely and although that review was triggered by a different point, RBS was given specific permission to consider the issue generally. However I will not simply order inspection of the documents, mainly having regard to Mr Coulthard’s explanation in his witness statement that an individual assessment of each document had been made.

The fall back proposed by PAG was inspection of the ESG documents by the court. Albeit a solution of last resort, that seems to me to be the practical approach. Given that I have been docketed to deal with this case counsel suggested that if this was to take place perhaps another judge should undertake the task. I will order that the ESG documents disclosed in the January list be produced to the court pursuant to CPR r31.19(6). I will hear the parties about whether that judge should be myself or one of my colleagues.

Appendix B Part 2 – litigation privilege

In Part 2 RBS claims to be entitled to withhold inspection of “original and copy documents brought into existence for the dominant purpose of actual or contemplated litigation” which are then further defined in four sub-categories of proceedings against the defendant. These are (i) documents relating to civil proceedings brought by the US CFTC, (ii) documents relating to criminal proceedings brought by the US DoJ, (iii) documents relating to civil and/or criminal proceedings brought by the Attorneys General of various US States, (iv) documents relating to civil proceedings brought by customers of the defendant or third parties in England and the United States.

The general principles applicable to litigation privilege were not in dispute. PAG referred to Lord Carswell at paragraph 102 of Three Rivers (No 6) as follows:

“The conclusion to be drawn from the trilogy of 19th century cases to which I have referred and the qualifications expressed in the modern case law is that communications between parties or their solicitors and third parties for the purpose of obtaining information or advice in connection with existing or contemplated litigation are privileged, but only when the following conditions are satisfied: (a) litigation must be in progress or in contemplation; (b) the communications must have been made for the sole or dominant purpose of conducting that litigation; (c) the litigation must be adversarial, not investigative or inquisitorial.”

A distinction exists between adversarial litigation (in relation to which the privilege may be claimed) and other kinds of proceedings or dispute in which it may not. PAG referred to re L (a minor) (Police Investigation: Privilege)[1997] AC 16, where Lord Jauncey (with whom Lord Lloyd and Lord Steyn agreed) described litigation privilege as "essentially a creature of adversarial proceedings" and held that the privilege could not be claimed in respect of a report prepared for use in non-adversarial care proceedings under Part IV of the Children Act 1989. PAG also referred to the decision of the Competition of Appeal Tribunal in Tesco Stores Ltd v the Office of Fair Trading [2012] CAT 6, where Lord Carlile expressed the obiter view at [44]-[46] that Tesco would be entitled to claim litigation privilege in respect of notes of interviews with potential witnesses that had been made in the context of an OFT investigation under the Competition Act 1998 but submitted that the key point about Tesco v OFT was that a Statement of Objections and Supplementary Statement of Objections had already been issued by the time that Tesco began to contact potential witnesses and so the fact-finding administrative stage of the OFT investigation had concluded and the proceedings had transformed into an adversarial contest.

PAG submitted that the claim to litigation privilege in Part 2 of Appendix B could not be correct in the light of these legal principles because at the time at which the relevant documents are said to have been prepared, the regulatory bodies identified in the disclosure list were (at most) merely in the process of conducting an inquiry to get to the bottom of the facts and so there were no adversarial proceedings in respect of which a claim to litigation privilege could be made.

PAG also challenged the claim to litigation privilege on two further grounds. First that RBS had not established litigation was reasonably in prospect (referring to United States of America v Philip Morris Inc[2004] 1 C.L.C. 811 and to AXA Seguros SA De CV v Allianz Insurance Plc[2011] Lloyd's Rep. I.R. 544). Second that RBS had failed to establish the dominant purpose for which the documents were created (referring to Guinness Peat Properties Ltd v Fitzroy Robinson Partnership[1987] 1 W.L.R. 1027; Westminster International v Dornoch Ltd [2009] EWCA Civ 1323; and West London Pipeline [2008] 2 CLC 258 at [53]).

RBS submitted that all of PAG’s arguments about litigation privilege submissions were moot and irrelevant because all the documents which formed the subject of the challenge were also covered by a claim by RBS for legal advice privilege and that claim was not challenged by PAG. RBS pointed out that the additional claim to legal advice privilege for these documents was set out expressly in Part 2 of Appendix B.

PAG’s response was as follows. First it submitted that it was not possible for a document to be protected by both legal advice privilege and litigation privilege. Thus the privilege claim was flawed. It should be rejected or at least this showed that the privilege claim was sufficiently incorrectly based, having regard to paragraph 86(3) of West London Pipeline, to justify inspection of the documents by the court. Second it submitted that any privilege in some of the documents in Part 2 had been waived because those documents had been shown to or in one case actually provided to regulators. RBS did not accept that the privilege had been waived and I need to address that “non-waiver” point separately below.

I am not going to devote further time to considering the claim to litigation privilege because the effort is not worth the return. RBS is right that it has claimed legal advice privilege for all the documents as well. Thus establishing that they are not covered by litigation privilege does not directly lead to an order for inspection. At best it is one of PAG’s West London Pipeline points. That is to say a submission designed to show that the approach to privilege taken by RBS is flawed in a particular way so as to support an attack on the claim to privilege generally.

The problem with PAG’s West London Pipeline point is that a review of the cases and textbooks cited to me shows that it would not be unreasonable for a lawyer to approach claims to privilege on the basis that the litigation privilege and legal advice privilege overlap. Some textbook authors say they are mutually exclusive while others do not treat the two concepts that way. Compare Thanki, The Law of Privilege 2^nd Ed paragraphs 1-11 to 1-13, (citing Waugh v British Railways Board [1980] AC 521, In Re L and Three Rivers (No 6)) with Hollander, Documentary Evidence 12^th Ed para 18-01. The textbook Privilege by Passmore (3^rd Ed) points out the clear distinction between the two drawn by Lord Edmund-Davies in Waugh v BRB but notes that others have not drawn such a clear distinction, referring to Lord Scott in Three Rivers (No 6) at paragraph 27 and other judgments. If it mattered I would of course consider the primary authorities and not the commentaries. However I do not have to go that far because it is plain that even if, on analysis, the approach of RBS was wrong, the existence of such an error cannot establish the existence of a wider flawed approach which might have an impact on the other privilege claims in issue.

PAG also submitted that legal advice privilege necessarily cannot be claimed in documents which represent communications with third parties, whereas the documents in Part 2 are such documents. This is in fact tied up with the non-waiver point. Part 2 is not defined so as only to relate to documents which represent communications with third parties, but all the documents in Part 2 which were shown to third parties are also the ones the subject of the non-waiver point.

Finally I should note that in the course of argument there was some confusion about exactly how many documents were being referred to in Part 2 (and Part 1) as being documents shown to third parties. It seemed to be very unclear and PAG submitted that that lack of clarity was another factor to be taken into account as part of its West London Pipeline argument. In fact what seemed to be a major ambiguity did not exist when the matter was looked at carefully. There was a minor point about which documents fell into which groups but that was clarified in correspondence over the weekend. There are six documents in the “non waiver” category in Part 1 of which five are also in Part 2. I will address “non-waiver” below.

Appendix B Part 3 – Without prejudice

Following the British Bankers Association review of LIBOR in 2008 which led to an article in the Wall Street Journal on 16^th April 2008 and an investigation by the CFTC which started in 2010, the FSA engaged with panel banks on the LIBOR issue in 2010. On 10 February 2011 the FSA asked panel banks including RBS about the adequacy of LIBOR systems and controls. The FSA’s formal enforcement investigation began in October 2011. RBS states that very extensive material was produced to the FSA in the course of this investigation including millions of pages of documents. The FSA Final Notice was issued on 6^th February 2013. It finds that RBS had undermined the integrity of LIBOR and summarises and explains its conclusions with a full analysis in 139 paragraphs. The Final Notice imposes a penalty on the bank of £87.5 million. Because RBS had agreed to settle at an early stage of the FSA investigation, this penalty was at a discount.

In Part 3 of Appendix B RBS claimed the right to withhold inspection of communications passing between RBS and the FSA (and/or their lawyers) dated between December 2012 and January 2013. The documents were marked “without prejudice”. They represent negotiations with the FSA in connection with the FSA Final Notice.

PAG submits the claim cannot be sustained. Its point is that without prejudice privilege is really concerned with settling civil litigation whereas these communications arose in the context of a regulatory investigation resulting in a finding of misconduct and the imposition of a penalty. Those circumstances are very far removed from a genuine negotiated settlement in which parties to an ordinary civil claim seek to compromise their disputes.

The FCA (the organisation which has replaced the FSA) supports the without prejudice claim in relation to these documents. The FCA did not seek to intervene in the proceedings but asked for a four week grace period along the same lines as the ones I provided for on previous occasions (see [2015] EWHC 321 and [2015] EWHC 322) if I was going to order inspection.

The basic legal principles were common ground. The words of Lord Griffith in Rush & Tompkins v GLC [1989] AC 1280, at 1299 are regularly cited on this topic and I will do so again. The without prejudice rule is:

“…a rule governing the admissibility of evidence and is founded upon the public policy of encouraging litigants to settle their differences rather than litigate them to a finish. It is nowhere more clearly expressed than in the judgment of Oliver L.J. in Cutts v. Head [1984] Ch. 290, 306:

‘That the rule rests, at least in part, upon public policy is clear from many authorities, and the convenient starting point of the inquiry is the nature of the underlying policy. It is that parties should be encouraged so far as possible to settle their disputes without resort to litigation and should not be discouraged by the knowledge that anything that is said in the course of such negotiations (and that includes, of course, as much the failure to reply to an offer as an actual reply) may be used to their prejudice in the course of the proceedings. They should…be encouraged fully and frankly to put their cards on the table.... The public policy justification, in truth, essentially rests on the desirability of preventing statements or offers made in the course of negotiations for settlement being brought before the court of trial as admissions on the question of liability.’”

Other important points are the following. The passage cited above concerns public policy but there are in fact two possible justifications for the rule: one based on implied contract between the negotiating parties and one based on the public policy of encouraging parties to negotiate and settle their disputes rather than fight (Muller v Linsley and Mortimer [1996] 1 PNLR 74 at 77). The rule applies not only against the other party to the “without prejudice” communication, but also entitles disclosure to be withheld from third parties in any subsequent litigation connected with the same subject matter (Rush & Tompkins at 1301). The mere fact a document is headed “without prejudice” is not sufficient to confer the benefit of the privilege (Dixon Stores Group v Thames Television [1993] 1 All ER 349). The rule is not absolute and “without prejudice” communications may be relied on for a variety of reasons (Rush & Tompkins at 1300 and also Unilever v Procter & Gamble [2000]FSR 344 at 353/4).

PAG made two further submissions of law which I should record. It submitted that the rule is a rule of admissibility rather than a substantive right like legal professional privilege, referring, in addition to Rush & Tompkins itself, to a discussion on that topic in Hollander (12^th Ed) at 20-02. Second, PAG submitted that bearing in mind Art 10 ECHR, the rule should be applied with restraint and only when the public interests underlying the rule are plainly applicable, referring to the judgment of the Vice Chancellor, Sir Andrew Morritt in Prudential Assurance v Prudential Insurance of America No 2 [2002] EWHC 2809 (Ch). Nothing turns directly on either of these points and RBS did not dissent from either of them. It is not necessary for me to examine them in any depth. I will take them as correct.

At one stage it seemed to RBS that PAG was taking a timing point about whether a dispute had really arisen. RBS submitted that the need for “a dispute” capable of settlement does not mean there is a need for litigation to be ongoing for the rule to apply and referred to Bradford & Bingley v Rashid [2006] UKHL 37 and Barnetson v Framlington Group [2007] EWCA Civ 502. However I did not understand PAG to put its case that way. PAG’s case is more fundamental, that the sort of regulatory context in which the communications with the FCA took place was not within the rule at all.

PAG submitted that there was no reported case in which a party subject to an investigation by the FSA/FCA which results in the issue of a decision or final notice has sought to invoke the without prejudice privilege. PAG also submitted that the leading cases proceed on the basis that the rule applies to negotiations aimed at the settlement of civil litigation (citing Rush & Tompkins and adding Ofulue v Bossert [2009] 1 AC 990 and Oceanbulk v TMD [2011] 1 AC 622). PAG is right but from the point of view of the problem before me, all one can say is that those cases were in civil litigation. They were not concerned with the problem in this case.

PAG also submitted that if it was possible to take such a point then it would have been taken in two earlier cases. First was Real Estate Opportunities v Aberdeen Asset Managers Jersey [2007] 1 All ER 791 which concerned correspondence between the defendants and the FSA, and other material provided to the FSA by the defendants, in the course of an investigation by the FSA into ‘split capital trusts’ which resulted in the defendants’ agreement to a compensation scheme devised by the FSA. The second was the National Grid Electricity Transmission v ABBlitigation, which concerned inter alia material provided to the European Commission in connection with a ‘leniency application’ (see e.g. [2011] EWHC 1717 (Ch) and [2012] EWHC 869 (Ch)). Yet without prejudice privilege does not appear to have been asserted in either case. I cannot say whether a claim to without prejudice privilege might have been available on the facts in either case. Neither the judgments of Arden LJ (with whom Tuckey and Lawrence Collins LJJ agreed) in the former nor of Roth J in the latter are concerned with without prejudice at all.

In order to decide the issue, it is necessary to examine the FCA enforcement process in some depth. The Financial Services Authority was set up by s1 of the Financial Services and Markets Act 2000 (FSMA) in its unamended form. The FSMA was amended by the Financial Services Act 2012 with effect from 1^st April 2013 and by s1A of FSMA in its amended form, the Financial Services Authority was renamed as the Financial Conduct Authority. Although the communications in issue took place prior to the setting up of the FCA, both sides argued the case on the basis of the statutory framework and guidance in place today. From now on the references to the FSMA are to the Act in its amended form. The persons subject to regulation may be an individual or firm. I will refer to that person as a firm.

The FCA has a number of operational objectives including a consumer protection objective at s1C FSMA and an integrity objective at s1D FSMA. The former is defined in the Act as securing an appropriate degree of protection for consumers and the latter as protecting and enhancing the integrity of the UK financial system.

It has the power to gather information and carry out investigations. If it finds cases of misconduct and it finds it is appropriate to take action then the FCA has the power to take various forms of action including imposing a financial penalty (s206 FSMA) and imposing public censure (s205 FSMA ).

The FCA has the power to issue Warning Notices (s207 and s387 FSMA) and after that to issue Decision Notices (s208 and s388 FSMA). A person to whom a decision notice is given has the right to refer the matter to the Tribunal. The “Tribunal” is the Upper Tribunal (Tax and Chancery Chamber). If the FCA has given a person a decision notice and the matter has not been referred to the Tribunal, the FCA must issue a Final Notice (s390 FSMA). A Final Notice about a penalty must state the amount and provide certain other information (s390(5)).

Section 413 FSMA provides that a person may not be required under the Act to produce, disclose or permit inspection of certain “protected items”. The protected items are defined in such a way as to create a form of legal advice and litigation privilege as between the firm and the regulator.

Details of the enforcement process are found in a Decision Procedure and Penalties Manual (DEPP) and an FCA Enforcement Guide, which have been produced pursuant to requirements in the FSMA that the FCA should prepare and publish statements of policy or procedure on the exercise of its enforcement and investigation powers. RBS also referred me to an outline of the FCA enforcement process in the textbook Financial Services: Investigations and Enforcement by Freshfields Bruckhaus Deringer (3^rd Ed) paragraphs 10.8 - 10.17. This text helpfully summarised the process, noting at paragraph 10.16 that very little of the process is set out in the FSMA .

The process consists of a number of phases starting with a fact finding investigation when the regulator becomes aware that a regulatory issue has arisen. The primary purpose of the first, investigatory, phase is to give the FCA a proper understanding of the facts so that it is in a position to assess properly what regulatory action should be taken.

Following the investigation the FCA may issue a preliminary findings letter, normally annexing a Preliminary Investigation Report. Assuming the FCA staff responsible for the Preliminary Investigation Report make the appropriate recommendation then the matter is referred to the FCA’s Regulatory Decisions Committee (RDC). The RDC is a committee of the FCA but is separate from the enforcement staff who investigated the matter. It is normally the RDC which decides whether to issue a Warning Notice. If a Warning Notice is issued then, following representations, it is the RDC which decides whether to issue a Decision Notice. If the matter is not referred to the Tribunal then the Final Notice will follow at the end of the process.

Where the outcome is potentially a financial penalty, the FCA will send a letter at an early point in the enforcement process known as a “Stage 1 letter” (Enforcement Guide 5.15). This invites the firm to enter into settlement discussions. The Enforcement Guide (5.18) explains that the latest point at which the FCA will send a Stage 1 letter is when the firm is provided with the Preliminary Investigation Report. The exact timing of the Stage 1 letter varies, as follows:

“ […] Sufficient investigative work must have taken place for the FCA to be able to satisfy itself that the settlement is the right regulatory outcome. In many cases, the FCA can send out the stage 1 letter substantially before the person concerned is provided with the FCA's preliminary investigation report (see paragraphs 4.30 to 4.33). […]”

The Stage 1 letter gives the firm a period within which to reach agreement with the regulator. If agreement is reached within that so called “Stage 1 settlement period” the firm receives the maximum permitted discount of 30%. The scheme for determining the appropriate level of financial penalty is set out in DEPP Part 6.5. The settlement discount scheme is in DEPP 6.7. The stage 1 settlement discount of 30% is at DEPP 6.7.3 (3). A settlement may give rise to a Final Notice.

Paragraph 5.3 of the Enforcement Guide describes the settlements as follows:

“Settlements in the FCA context are not the same as ‘out of court’ settlements in the commercial context. An FCA settlement is a regulatory decision, taken by the FCA, the terms of which are accepted by the firm or individual concerned. So, when agreeing the terms of a settlement, the FCA will carefully consider its statutory objectives and other relevant matters such as the importance of sending clear, consistent messages through enforcement action, and will only settle in appropriate cases where the agreed terms of the decision result in acceptable regulatory outcomes. Redress to consumers who may have been disadvantaged by a firm’s misconduct may be particularly important in this respect. Other than in exceptional circumstances, FCA settlements that give rise to the issue of a final notice or supervisory notice will result in some degree of publicity (see chapter 6), unlike commercial out of court settlements, which are often confidential. ”

PAG emphasised the distinction drawn here between settlements with the FCA and “out of court” settlements and the point that any settlement is a regulatory decision.

Neither the DEPP nor the Enforcement Guide mention the words “without prejudice”. The Enforcement Guide addresses the basis of settlement discussions in paragraphs 5.9 – 5.10. These paragraphs emphasise that the negotiations do not involve the RDC and in paragraph 5.9 is the following:

“[…] The FCA would expect to hold any settlement discussions on the basis that neither FCA staff nor the person concerned would seek to rely against the other on any admissions or statements made if the matter is considered subsequently by the RDC or the Tribunal. This will not, however, prevent the FCA from following up, through other means, on any new issues of regulatory concern which come to light during settlement discussions. […]”

Clearly the first sentence is explaining that the FCA expects to hold settlement discussions on a basis which is at least similar to without prejudice negotiations in civil litigation.

In its letter of 2^nd April 2015 the FCA explained that in settlement discussions it was standard practice for the drafts of the statutory notice exchanged between it and the firm to be marked “without prejudice” and for the settlement agreement to record that the negotiations had been conducted confidentially on a without prejudice basis. The letter stated that this practice was followed in the negotiations which led to the RBS Final Notice.

Given the FCA’s explanation of its standard practice and the fact that the label “without prejudice” is in fact used, one might ask why paragraph 5.9 does not mention it. The FCA letter of 2^nd April addressed this. It stated that the reasons why paragraph 5.9 is drafted that way are set out in excerpts from a 2007 public consultation process which were in an Annex to the letter.

As I understand the Annex, the position prior to 2007 was that the then equivalent of the DEPP (called the DEC) did refer explicitly to the possibility of discussions taking place between the firm and the regulator on a ‘without prejudice’ basis (at DEC App 1.2.1). It also contained a sentence about mutual non-reliance in later regulatory proceedings which is broadly the same as the first sentence in the quotation from EG 5.9 above. The 2007 consultation document (CP 2/07) consulted on this topic. It explained that the concept of ‘without prejudice’ communications had been imported from civil proceedings and referred to the public and private interest of parties being able to negotiate by discussing their positions candidly and openly. It stated that the regulator regarded regulatory settlements as in the public interest in offering potential saving of time and costs and early redress for consumers. However given the regulator’s statutory objectives and responsibility, there could be circumstances in which it is obliged to take action pursuant to information passed to it in the course of ‘without prejudice discussions” (see the second sentence in paragraph 5.9 quoted above). The consultation proposed that DEPP should reaffirm that the regulator and the firm may agree to mutual non-reliance on admissions or statements made in settlement discussions but that while the label ‘without prejudice’ was useful shorthand in discussions, the term may carry different connotations to such an expression in civil proceedings. Thus the proposal was to replace express references to ‘without prejudice discussions’ with the more general statement in EG paragraph 5.9.

The policy statement following the consultation explained that mixed responses had been received to the proposal. It stated that one respondent contended that the change might discourage open discussions while another contended that regardless of how the discussions were labelled, the agreement not to rely would need to carry significant weight for confidence to be attached to the settlement process. Paragraph 5.9 was the result.

Thus the position today is that although the Enforcement Guide does not use the words “without prejudice”, in the context of settlement discussions the FCA does mark communications that way and accepts communications marked that way from a firm. The expectation of both sides must be that the basis of these settlement discussions is at least as set out in EG paragraph 5.9. Admissions will not be used later by the RDC or before the Tribunal. It follows that the FCA does purport to conduct a form of without prejudice settlement discussion with subject firms.

The FCA’s letter of 2^nd April explains that the FCA’s view is that the ability to conduct settlement negotiations on a without prejudice basis is “vitally important” to the success of the discussions. The FCA is concerned that firms may choose not to enter into settlement discussions with the FCA if they think that there is a risk that admissions made in those discussions could be disclosed/ inspected in separate proceedings. The FCA’s findings are public. If firms were not confident that admissions made in settlement discussions would not be used against them by the FCA or in separate proceedings, they would be less willing to enter such discussions. The FCA also stated its view that early settlement is of benefit to consumers and the UK financial system and therefore furthers its statutory objectives. It allows the FCA to send messages to markets more quickly which can lead to appropriate legislative change, it saves resources and allows consumers to obtain compensation earlier.

Against this background the FCA made the following legal submissions:

RBS supported the FCA’s submissions.

PAG contended to the contrary, as follows. First, whether the firm agreed to it or not, the Final Notice is still an administrative decision. The regulator itself has recognised in the consultation that the term ‘without prejudice’ may carry different connotations to such an expression in civil proceedings. Second, the early agreement is not a true negotiated compromise. The 30% discount is simply a function of the express terms of the DEPP. Third, the underlying public policy rationale is inapplicable in this context. Despite bare assertions by the FCA there is no reason to believe firms would be less likely to engage in discussions with their regulator if they were unable to do so on a without prejudice basis. The early settlement discount is mandatory and gives a strong incentive to enter discussions anyway. Fourth, Principle 11 of the FCA Principles provides that firms must deal with the FCA in an open and cooperative way and disclose anything the regulator would reasonably expect. Failure may lead to enforcement action and may be a criminal offence under s398(1) FSMA. Fifth the discussions are not inviolable. The FCA may be obliged to act on information received. This had not deterred discussions since the FCA states that enforcement action is frequently settled. Sixth, the FCA can publish such information about the matter in the Final Notice as it considers appropriate (s391(7) FSMA). This is very far from the position in civil litigation. Seventh, the FCA has extensive powers to disclose confidential information under s349 FSMA.

These are substantial points but I am not persuaded that they justify the conclusion PAG advances. The first point ignores the fact that the legislative scheme governing investigations provides that the matter may be litigated before the Upper Tribunal. The second point is true but does not go far enough. The fact the tariff is fixed does not undermine the point that frank discussions are more likely to lead to settlement. I believe the third point is wrong. I would not go as far as the FCA in saying that its public policy rationale is stronger than the one applicable in civil litigation but I accept the FCA’s submission that the public policy justification for facilitating settlement is a powerful one. As the FCA’s letter explained, a beneficial effect of an early Final Notice arrived at following settlement is in notifying consumers of the possibility of a civil claim, perhaps before any limitation period might have expired. The fourth point (principle of openness and cooperation) is important but can be satisfied by the fact that the investigation process itself is an open one and the Final Notice is public. The fifth, sixth and seventh points do not help PAG. The fact that the FCA may be obliged to act on, publish or disclose information is clear to firms entering the discussions. The without prejudice rule already has a number of exceptions but they do not undermine the rule when they do not apply. These provisions work in the same way.

The implied contract basis for a form of without prejudice rule is also relevant. The FCA made and accepted communications marked in that way and in such a case a firm would be entitled to expect that the label meant at least what the words of EG 5.9 provide for. Although EG 5.9 does not refer to use in other civil proceedings with third parties, the protection would be undermined if it was not covered (cf Rush &Tompkins 1301 D).

The settlement discussions with the FCA which are labelled without prejudice take place along side the investigation process and along side the production to the FCA of documents and other material as part of its investigation process. In that sense there is an analogy between those discussions and normal without prejudice discussions in civil court proceedings, which also take place along side the exchange of pleadings and argument in court. The nature and extent of a without prejudice rule as it would apply to settlement discussions with the FCA is not exactly the same as the way the rule applies in court proceedings given the FCA’s position as a regulator and its obligations but that is not a reason not to apply the same principles by analogy. In my judgment the public policy on which the without prejudice rule is based is capable of applying in order to promote settlement of FCA investigations.

Finally PAG contended that the real issue here and the only basis for sensitivity of RBS against disclosure must be because the communications which are not recorded in the Final Notice must contain further admissions of misconduct. On the basis of that inference the public interest is entirely in favour of publicity relating to the manipulation of LIBOR (Graiseley v Barclays [2013] EWHC 67 (Comm) Flaux J at paragraph 61). Moreover the argument advanced by the FCA is the same as the one rejected in Kaufmann v Credit Lyonnais [1995] CLC 300. Before considering Kaufmann I will say that I do not accept the inference PAG seeks to draw. I do not believe it would be legitimate to enter into speculation of that kind in any event, but even if it were, PAG’s inference does not follow.

Kaufmann was about an attempt to prevent inspection of correspondence and material passed by Credit Lyonnais to the SFA, one of the predecessor regulators of the FSA. The material had been provided “in confidence” and “voluntarily”. In that sense Kaufmann does have parallels with this case but the basis on which inspection was resisted in Kaufmann was public interest immunity, not the without prejudice rule. Arden J (as she then was) held that public interest immunity did not apply. A key reason was that the test for public interest immunity explained in Burmah Oil v Bank of England [1979] 1 WLR 473 required that the informant providing information to the public authority had to be sure that the confidence would be absolutely respected. That did not apply in Kaufmann because the SFA might be required to disclose the information to another regulator, the SIB, which might lead to further disclosure. PAG drew an analogy with the position of RBS in this case in that the FCA’s procedures made clear that the FCA may have to act on information disclosed in the discussions. At most PAG’s point indicates that public interest immunity could not be claimed, for the same reasons as in Kaufmann. RBS submitted that it is not relevant to without prejudice and that the normal without prejudice rule already has a number of exceptions. I agree.

PAG also submitted that the wider public policy considerations addressed by Arden J in Kaufmann are relevant. The point relied on is that Arden J held that the need to promote candour in discussions with the regulator could not justify withholding the whole class of documents sought to be protected in that case. The learned judge said this:

“The communication may be untrue and designed to mislead the SFA. A class based immunity could hardly be described as necessary to the proper function of the SFA in that event. Immunity might rather have facilitated deception than deterred it. By way of further example, the attempt at compliance might have been inept. The member firm might have disclosed less than it ought to have done and failed to report matters promptly.” (p317-C)

In my judgment this is PAG’s best point. It highlights the importance of a difference between the effect of the without prejudice rule in civil litigation and the possible effect of this rule applied to FCA investigations. When a civil case settles following without prejudice negotiations, there is no reasoned judgment on which one of the settling parties might rely subsequently. When an FCA investigation is settled, the result will normally be a published regulatory finding, as in this case.

Consideration of the RBS Final Notice in this case makes the point. The RBS Final Notice makes findings of misconduct relating to CHF and JPY LIBOR. It does not make a finding of misconduct relating to GBP LIBOR. It would be tempting to draw the conclusion RBS had therefore been exonerated in relation to GBP LIBOR, which after all is the reference rate in the PAG swaps. Tempting but wrong. This Final Notice is the product of a settlement agreement. Although it was arrived at after the first investigatory phase of work by the regulator, it is not the product of a full statutory process of investigation and decision making. The communications on which the agreement is based might have been incomplete, mistaken or misleading (whether inadvertently or not).

PAG’s Particulars of Claim paragraph 77 alleges that the implied representations about LIBOR were false on the basis that RBS had in fact manipulated LIBOR. In support PAG rely on the findings of misconduct in the Final Notice, amongst other things. In its Defence RBS responds to this at paragraph 260. RBS admits manipulation of CHF and JPY LIBOR, expressly denies manipulating GBP LIBOR and makes a general denial of the allegations in paragraph 77 (at paragraph 260 (6)). However RBS also pleads positively as follows: “For the avoidance of doubt, there have been no regulatory findings of misconduct on the part of RBS in connection with GBP LIBOR” (paragraph 260(3)(b)).

RBS characterised this as a “mere mention” of the finding and as nothing more than the Defence “noting the (incontrovertible) fact that no regulator has made findings against it in respect of GBP”. That is unreal. The bank’s formal Defence in these proceedings advances a positive point that the regulators have not found misconduct relating to GBP LIBOR. In taking this course RBS has itself put in issue the basis on which the regulatory findings were made.

RBS pointed out that PAG referred to the regulatory findings in the Particulars of Claim and so naturally RBS has referred to them too. The relevant sub-paragraph in PAG’s Defence (260(3)(b)) is part of the section dealing with PAG’s paragraph 77(3) which is concerned with the regulatory findings. However this does not help. The basis on which the regulatory findings were made was not put in issue by PAG relying on the findings of misconduct because RBS admitted those findings of misconduct. The problem arises because RBS goes further and positively relies on what is not in the Final Notice as supporting its denial. Even if paragraph 260(3)(b) of the Defence is read in a narrow way focussed only on admissions made by RBS to regulators (and I do not believe it can be), the point holds good.

If the communications on which the Final Notice was based were false, then to allow RBS to rely on what is absent from the Final Notice but at the same time to withhold inspection of those communications would compound the falsehood. That will not do.

The question which troubles me is whether this reasoning undermines the FCA’s approach as a whole or whether it shows that while there can be a right to withhold inspection, it cannot be maintained in the circumstances of this case given the stance adopted by RBS. If it is the latter then I need to deal with the submission by the FCA (and RBS) that by analogy with without prejudice privilege in civil litigation in which both parties have the right to prevent inspection, the FCA in this case had a separate right to object to inspection, regardless of the stance adopted by RBS.

Most of the argument before me relating to without prejudice focussed on the issue of whether a right to withhold inspection exists at all rather than whether RBS’s stance had led to a waiver. The focus of this waiver argument (which RBS called a “mass waiver”) was on the so called “non-waiver” point addressed next. Nevertheless PAG advanced the submission in general terms and RBS replied in kind.

The conclusion I reach as to the law is the following. In my judgment a firm the subject of an FCA investigation has the right to withhold inspection of communications which were part of genuine settlement discussions between that firm and the FCA. That right applies before the FCA, the Upper Tribunal and in civil litigation with a third party. Application of that right before the FCA and the Upper Tribunal is presumably only likely to matter if settlement is not reached. Application of that right in civil litigation may arise whether or not settlement was reached. The fact a Final Notice is issued does not mean the right is lost. The right arises by analogy with the without prejudice rule and it is convenient to use that expression to refer to it but it is not identical with the normal rule in civil litigation. The right does not prevent the FCA from acting on information received in those discussions, for example by following up new issues which come to light. The right cannot be maintained in civil proceedings if the basis on which a Final Notice was decided is itself in issue in the proceedings. In such a case the content of the settlement discussions will be admissible evidence. If the basis of such a decision is in issue then the FCA has no ground on which to object.

I have already considered the substantial public policy justification for a form of without prejudice rule in promoting settlement as well as the implied contract justification. As long as the basis on which the Final Notice was reached is not in issue in court, there is no reason why the without prejudice communications should not remain covered by the rule. A subject firm need not purport to rely on what is not in a Final Notice. I am not concerned with a firm making public statements about a Final Notice. However if the firm puts in issue before a court the basis on which the Final Notice was produced, then in those circumstances justice demands that the communications which led to the Final Notice be disclosed. The FCA is not in the same position as a counterparty to without prejudice negotiations in civil litigation. The FCA could not object to a firm relying on a Final Notice and putting the basis on which it was produced in issue. In such a case I do not see how the FCA could claim the right to prevent inspection when the firm had taken that course.

I will make an order for inspection of the documents in Part 3, subject to a four week grace period.

Non waiver

Six documents are identified in the Supplemental List as either having been shown to regulators or in one case handed over to regulators. They are documents in which legal advice privilege is claimed and in some cases litigation privilege. PAG contends that any such privilege has been waived by the acts of showing or handing over. RBS contends that privilege was not waived because every document was shown or provided on a confidential “non-waiver” basis. There are non-waiver agreements with the CFTC, SEC, DoJ and the Attorneys General of various US states in evidence. The only case for which an express non-waiver agreement does not exist is relating to the provision of one document to the Japanese regulator, the JFSA. However Mr Coulthard’s evidence confirms that it is was provided on the same agreed basis as the others. I was not persuaded by PAG’s argument that the position of the JFSA is different.

PAG also relied on the same kind of waiver resulting from the bank’s reliance on the regulators findings as I have addressed above in the without prejudice section. RBS did not agree, submitting that the argument that the mere mention of a regulatory finding could lead to “mass waiver” is irreconcilable with basic principles.

I turn to the legal principles. Confidentiality is an essential pre-condition for privilege: see Lord Scott in Three Rivers (No 6) at paragraph 24. Confidentiality can sometimes be maintained in a document despite the fact that it has been disclosed to a limited number of third parties: Gotha City v Sotheby's (No.1)[1998] 1 W.L.R. 114.

RBS submitted that a distinct but related doctrine permits waiver to be made for a limited purpose. I agree. This doctrine prevents even the person to whom the document was disclosed from using it in some circumstances, if they are outside the limited purpose for which privilege was claimed. The cases which make good this submission are British Coal Corporation v Dennis Rye [1988] 1 WLR 1113 (CA), B v Auckland District law Society [2003] UKPC 38, and Berezovsky v Hine[2011] EWCA 1089 (CA).

RBS submitted that it can maintain its claim to privilege because two conditions are satisfied. First, the agreements with the US authorities explicitly state that the documents are being provided on the basis that confidentiality and privilege would be preserved as against third parties. Second, the documents were provided for the limited purpose of the ongoing investigations. Therefore there has been no waiver.

PAG submitted that the agreements in fact contain what RBS called “carve-outs”. PAG’s case was that the agreements do not completely prohibit further disclosure of the documents by the regulator. Each agreement gives the relevant regulators carte blanche to share documents with other third parties (such as other governmental or regulatory agencies) and/or to make that material public or to disclose it further. Thus in each agreement RBS relinquished control of the documents.

RBS submitted that these carve-outs are there simply to preserve the rights of the regulator to make further disclosure in performance of its statutory duties or otherwise required by law. The carve outs cannot convert an agreement which explicitly states that privilege has not been waived vis-à-vis third parties and that confidentiality will be preserved, into an agreement with theopposite effect. And this is all the more so in the cases in which the regulators were not even handed a copy of the document but only shown it (as happened for all the US regulators).

RBS submitted that there was no English authority on the point but that other common law jurisdictions had come to a conclusion upholding a limited waiver despite the presence of the equivalent of carve-outs. The issue arose in the Supreme Court of Ireland in Fyffes v DCC [2005] IESC 3 (Geohagen, Fennelly and McCracken JJ) where a company (DCC) had provided privileged material to the stock exchange concerning insider dealing. The confidentiality agreement pursuant to which they were provided obliged the stock exchange to keep the documents confidential but contained an express carve-out for disclosures obliged by law and in respect of the DPP (to whom the stock exchange had a statutory duty to pass on information). DCC’s purpose of providing the information had been in the hope that the stock exchange would influence the DPP not to prosecute. Another company subsequently sought disclosure of these materials in a civil action against DCC and argued that privilege had been waived on the grounds that “any disclosure leads to loss of the privilege”.

One of the issues before the Irish Supreme Court was whether to follow British Coal line of cases (including a Northern Irish decision Downey v Murray [1988] NI 600, which also applied British Coal) or whether to follow a decision of the High Court of Australia in Goldberg v Ng 137 ALR 57. In the Goldberg case the Supreme Court of New South Wales (at [1994] 33 NSWLR 639) had disagreed with British Coal and ordered inspection of documents which had been provided to the Law Society despite an express reservation of confidentiality, applying “considerations of fairness”. The High Court of Australia upheld the decision.

The Irish Supreme Court preferred the reasoning in British Coal to Goldberg. McCracken J made the point (p19 of the report) that there may be many situations in which it is desirable or even mandatory that privileged documents be disclosed to a third party for a limited purpose. He was concerned that the lack of a clear rule which permitted limited waiver would lead to a risk that “serious injustices could occur because a party might fail to make a disclosure due to the fear that they might lose the benefit of a legal privilege to which they were entitled”. I also note, as PAG emphasised, that McCracken J did not rule out the fact that there might be cases in which the question of fairness arises and the general rule that disclosure for a limited purpose does not amount to waiver may have to give way where a serious injustice might result.

RBS also referred to Citic v Secretary of State for Justice [2012] HKCA 153 in the Hong Kong Court of Appeal. Here the Hong Kong Securities and Futures Commission (SFC) had served an order requiring documents to be delivered to it. Amongst the documents provided were six documents subject to privilege (which Citic was not under a duty to provide). At the time the documents were handed over, there was no express statement as to the limited basis on which they were provided. The police subsequently launched an investigation and Citic learned that the SFC had passed the documents to the (Hong Kong) Department of Justice, and that the police were also seeking the documents. Citic sought to have the documents returned to it on the basis that its waiver of privilege did not extend beyond the limited purpose of the SFC investigation. Citic’s claims were upheld, following British Coal and Auckland District Law Society and even though the limited circumstances of the waiver were only expressed some timeafter the documents were handed over.

By applying the existing English law and supported by the decisions of the courts in Ireland and Hong Kong, I hold that RBS were entitled to maintain the claim to privilege of documents which were only shown or provided to regulators on a limited basis and despite the existence of legal rights or duties on the part of the regulators to use, act on or even publish the documents pursuant to their regulatory powers. In this case the agreements between RBS and the regulators expressly provide that privilege and confidence are maintained. That should not be undermined by the existence of the carve outs. The fact that the carve outs recognise the regulator’s rights and obligations to take a step, which might go as far as even publishing the information in the document, makes no difference if that has not happened. Until they do, I fail to see why the confidentiality and privilege would not be preserved. I reject PAG’s case that the terms of the agreements pursuant to which RBS showed or provided documents to regulators mean that those acts led to a waiver of privilege.

I turn to the argument about waiver resulting from the bank’s reliance on the regulatory decisions. This covers the same ground as before in relation to the FCA final notice and the without prejudice rule. This time however the regulators are different although nothing turns on that. Another point was that the privileged documents were “unidentified” but I fail to see what difference that makes. The very same logic which I have applied relating to the without prejudice rule applies here too. The problem does not relate to the public nature of the regulatory findings, the problem arises from what RBS seeks to do with them in these proceedings. RBS really cannot have it both ways. It cannot on the one hand rely on absences from the regulators’ findings as indicating the limits of its misconduct and yet on the other hand seek to maintain as privileged what it put to them.

I will order inspection of the six documents, subject to a four week grace period.

Part of PAG’s application was for an order seeking to deem all ESG documents to be high level documents. There is no point in making this order. Any ESG document which is not a high level document but which ought to be disclosed, should be disclosed in due course anyway. There is no reason to bring that forward.

Case management – the scope of further disclosure

An important general issue remains to be resolved about LIBOR disclosure. RBS accepts that it must give standard disclosure relating to GBP LIBOR and that is underway. I am told that it involves some 50,000 documents. There is also some further common ground about certain classes of documents which I do not need to address. The question I need to decide is what to do about the bank’s general obligation to give disclosure relating to LIBOR for all the other currencies and tenors. The parties’ rival proposals are as follows.

RBS’s proposal is based on a collection of 2,687 individual documents which have been identified as being of particular interest by RBS’s legal advisers as a result of work over a number of years. RBS proposes that the reasonable search it should carry out is to review those documents and make its disclosure from them. This collection of documents has been called “the Regulatory Review Documents”.

PAG’s proposal is for an order that the bank shall carry out a reasonable search for documents in the following category:

“Any internal contemporaneous documents which are adverse to its case or which evidence the actual or attempted manipulation if any LIBOR currencies and tenors (including, in particular, any GBP tenors) and:

(a)

which are referred to in any High Level Reports contained in its Original List of its Supplemental List of Documents; and/or

(b)

which it has otherwise identified as such:

(i)

in the course of carrying out its own internal investigation(s) into LIBOR misconduct ; and /or

(ii)

for the purpose of preparing any High Level Reports; and/or

(iii)

for the purpose of preparing for, participating in or making submissions to any external investigation into LIBOR misconduct.”

In fact PAG’s proposal as drafted includes GBP LIBOR but that is an artefact. The issue is non-GBP LIBOR. The “High Level Reports” are the same as the high level documents referred to in this judgment.

Each side criticises the proposal of the other. PAG contends that the proposal from RBS is much too limited. RBS contends PAG’s proposal is far too wide and disproportionate.

RBS also contends that PAG’s proposal is unfair as it highlights the adverse documents and does not cater for the disclosure by the bank of documents supporting its case, but I do not regard this as a good justification for refusing PAG’s proposal if it was otherwise justified.

There is no difficulty about part (a) of PAG’s draft because as I understand it RBS has confirmed that the Regulatory Review Documents include all the documents referred to in the High Level Reports. Equally RBS says that part (b) (ii) of the definition covers the same ground as the Regulatory Review Documents. The problem for RBS is the open ended nature of part (b) (i) and (iii). It is not clear to RBS how it could comply with an order couched in those terms without re-reviewing all 25 million documents. However PAG’s argument is that if that is what needs to be done, then it should be done. PAG refers to the judgment of Rix LJ in Nichia v Argos [2007] EWCA Civ 741 at paragraph 72 in which he referred to the interests of justice in this way:

“Above all, it would be against the interests of justice if documents known to exist, or easily revealed, which would harm a party's own case or assist another party's case need not be disclosed because of a blanket prima facie rule against any standard disclosure. Once such a principle of disclosure were known to hold sway, dishonest or cavalier litigants would reap an unmerited advantage, contrary to the interests of justice.”

If I may say so I agree with Rix LJ, but the point he was making was in a different context from the question I have to decide. Nichia v Argos was concerned with whether to dispense with standard disclosure altogether on a particular issue in a particular kind of case (obviousness in patent cases). The proposal was to replace it with a rule for no disclosure on the issue at all, albeit that was to be prima facie and could be open to exceptions. That is very different from the situation here. A vital point, which PAG’s arguments about disclosure sidestep, is that RBS has accepted that it must give standard disclosure relating to GBP LIBOR (all tenors). Since sterling was the currency to which the swaps in this case were related that is significant. Another important point is that RBS has admitted misconduct in relation to CHF and JPY.

Thus PAG can already establish that a general representation that RBS was not engaged in manipulating LIBOR at all is false. It does not need more disclosure for that. PAG is also entitled to and will be given disclosure relating to the relevant currency (GBP). It is against this background that the question of the correct approach to disclosure addressing the nature and extent of any misconduct arising from other currencies has to be judged. I have no doubt that to absolve RBS of any obligation to give disclosure in relation to the other currencies would be wrong. I decided that issue in November. However, while I do not diminish the public significance of LIBOR manipulation, in this action the issues to which that wider disclosure would be addressed are likely to be of lesser significance than the matters already covered. The question is about the appropriateness, practicality and proportionality of the two rival proposals.

If the extent of the manipulation of LIBOR by RBS was significantly wider than has currently been admitted, then the combination of full disclosure relating to GBP LIBOR and the disclosure proposed by RBS is likely to reveal it. If it exists then it is likely to be revealed directly or at worst this disclosure will set off a train of enquiry leading to focussed specific further requests (which I am not to be taken to be encouraging).

Bearing in mind the overriding objective to deal with cases justly and at proportionate cost, I prefer the proposal from RBS. An order to review the Regulatory Review Documents and disclose what is relevant from them (perhaps all of them) is likely to produce appropriate disclosure of the documents PAG is most interested in. The marginal cost, both in terms of money and time, of the alternative order requiring RBS to go over the store of 25 million documents, even electronically, is very significant indeed. I am not satisfied it is worth the likely return.

RBS also proposed an order that following disclosure, PAG should provide a more focussed pleadings of its case on LIBOR manipulation. PAG resisted this and sought to shift the criticism onto RBS for a lack of specificity about LIBOR. I am not going to make an order about pleadings at this stage because the issue needs to be looked at in the round once disclosure has been given.

In summary I will: