BAILII
British and Irish Legal Information Institute


Freely Available British and Irish Public Legal Information

[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

United Kingdom Upper Tribunal (Tax and Chancery Chamber)


You are here: BAILII >> Databases >> United Kingdom Upper Tribunal (Tax and Chancery Chamber) >> Donaldson & Anor v The Financial Conduct Authority (FINANCIAL SERVICES - Metro Bank's trading update for Q3 2018 included incorrect statements about risk - meaning of "knowingly concerned") [2025] UKUT 185 (TCC) (16 June 2025)
URL: http://www.bailii.org/uk/cases/UKUT/TCC/2025/185.html
Cite as: [2025] UKUT 185 (TCC)

[New search] [Contents list] [Printable PDF version] [Help]

UT Neutral citation number: [2025] UKUT 185 (TCC) 

UT (Tax & Chancery) Case Numbers: UT-2022-000148; UT-2022-000149

Upper Tribunal

(Tax and Chancery Chamber)

At Rolls Buildings, Fetter Lane, London

 

FINANCIAL SERVICES - Metro Bank's trading update for Q3 2018 included incorrect statements about risk-weighted assets - whether breach of Listing Rules by the bank –whether Applicants were "knowingly concerned" in the breach - meaning of "knowingly concerned" - whether penalties charged are appropriate - penalties reduced

 

Heard on 31 March to 9 April 2025

Judgment given 16 June 2025

Before

 

JUDGE ANNE REDSTON

TRIBUNAL MEMBER JEAN PRICE

TRIBUNAL MEMBER JOHN WOODMAN

 

 

 

Between

 

CRAIG FRANCIS DONALDSON

DAVID THOMAS ARDEN

Applicants

and

 

THE FINANCIAL CONDUCT AUTHORITY

Respondent

 

Representation:

 

For the Applicants:       Ben Jaffey KC and Simon Pritchard of Counsel, instructed by Baker & McKenzie LLP

 

For the Respondent:     Paul Stanley KC and Ajay Ratan of Counsel, instructed by the Financial Conduct Authority

 


DECISION

Introduction

  1. Metro Bank ("the Bank") began operations in 2010 with Mr Craig Donaldson as its Chief Executive Officer ("CEO"). Mr David Arden joined in March 2018 as Chief Financial Officer ("CFO"). On 24 October 2018, the Bank published its third quarter trading update ("the Q3 Update"), which included the statements that "capital ratios remain robust"; that risk weighted assets ("RWAs") were £7,398m and that total capital as a percentage of risk-weighted assets was 19.1%. Those figures were not correct. When amended numbers were published in January 2019, there was a 39% drop in the Bank's share price.
  2. The Financial Conduct Authority ("the Authority") decided that the Bank had breached Listing Rule ("LR") 1.3.3R, because it had failed to take reasonable care to ensure that the Q3 Update was not false or misleading and did not omit anything likely to affect the import of the information contained within it. On 8 December 2022, the Authority issued a Final Notice imposing a penalty of £10,002,300 on the Bank for that breach of LR 1.3.3R. The Bank did not contest the Notice.
  3. The Authority also decided that Mr Donaldson and Mr Arden had been "knowingly concerned" in the Bank's breach, and under s 91 of the Financial Services and Markets Act 2000 ("FSMA") imposed a penalty of £223,100 on Mr Donaldson and a penalty of £134,600 on Mr Arden. The penalties were imposed on the basis that Mr Donaldson and Mr Arden had acted negligently; it was no part of the Authority's case that either lacked integrity or had been reckless.
  4. The Authority set out the Decisions in decision notices ("the Decision Notices") issued on 10 November 2022. Mr Donaldson and Mr Arden referred the Decision Notices to the Tribunal ("the References").
  5. The first issue was thus whether the Bank had breached LR 1.3.3R; if the answer to that question was "yes", the second issue was whether the Applicants were knowingly concerned in the breach, and if so, the third issue was whether to uphold the penalties.
  6. We summarise the positions taken by both parties at §§270ff. Most of their submissions concerned what happened at the time, and we therefore made detailed findings of fact. The parties also disagreed about the meaning of "knowingly concerned", see §§482ff.
  7. For the reasons given in the main body of this judgment, we decided that:
  8. (1)          the Bank had breached LR 1.3.3R;

    (2)          the Applicants were knowingly concerned in the breach; but

    (3)          Mr Donaldson's penalty was to be reduced from £223,100 to £167,325 and Mr Arden's from £134,600 to £100,950.

  9. We had the benefit of excellent written and oral submissions from Mr Ben Jaffey KC and Mr Simon Pritchard for the Applicants, and from Mr Paul Stanley KC and Mr Ajay Ratan for the Authority. When we refer to those written submissions in this judgment, for brevity we have referred to them as having been made by Mr Jaffey or Mr Stanley, but we do not underestimate the significant contributions made by Mr Pritchard and Mr Ratan. The fact that we have not referred to all the submissions a party has made does not mean that they have not been carefully considered and appreciated.
  10. The law and the regulatory structure

  11. All legislation and case law in this judgment is cited only so far as relevant to the issues being considered.
  12. The jurisdiction of the Tribunal

  13. The penalties which have been referred to us for determination were issued under FSMA s 91. 
  14. FSMA s 133 is headed "proceedings before Tribunal: general provision", and applies where a person has made a reference to the Tribunal in respect of a decision of the Authority. Subsection (7A)(g) provides that a decision by the Authority to take action under s 91 is a "disciplinary reference", and subsection (5) provides that in the case of such a reference, the Tribunal:
  15. "(a) must determine what (if any) is the appropriate action for the decision-maker to take in relation to the matter; and

    (b) on determining the reference, must remit the matter to the decision-maker with such directions (if any) as the Tribunal considers appropriate for giving effect to its determination."

  16. The hearing of a reference is not an appeal against a decision made by the Authority, but a complete rehearing of the issues which gave rise to the decision, see Burns v FCA [2017] EWCA Civ 2140 at [6].
  17. The burden and standard of proof

  18. It was common ground that the burden of proof rested on the Authority. The standard of proof is the balance of probabilities, see Re B [2008] UKHL 35, where Lord Hoffman said at [15]:
  19. "There is only one rule of law, namely that the occurrence of the fact in issue must be proved to have been more probable than not. Common sense, not law, requires that in deciding this question, regard should be had, to whatever extent appropriate, to inherent probabilities."

  20. Lady Hale added at [70]:
  21. "Neither the seriousness of the allegation nor the seriousness of the consequences should make any difference to the standard of proof to be applied in determining the facts. The inherent probabilities are simply something to be taken into account, where relevant, in deciding where the truth lies."

    The PRA and capital requirements

  22. In relation to its compliance with prudential regulation and supervision, the Bank is, and was at the time, regulated by the Prudential Regulation Authority ("the PRA"), part of the Bank of England. The PRA's role is to promote the safety and soundness of the firms it regulates, focusing primarily on the harm that firms can cause to the stability of the UK financial system.
  23. At the relevant time, the UK had adopted the EU's prudential rules, in accordance with the Capital Requirements Regulation No. 575/2013 ("the CRR") and the Capital Requirements Directive 2013/36/EU ("the CRD"). Banks were required to satisfy the requirements of the three Pillars in the CRD as follows.
  24. (1)          Pillar 1 set out the minimum capital requirements firms were required to meet for credit, market and operational risk; these requirements were binding in nature.

    (2)          Pillar 2 related to the supervisory review process, and required the PRA to take a view on whether a firm needed to hold additional capital to protect against risks not adequately covered in Pillar 1. A firm's Pillar 2 requirements therefore involved  supervisory judgment and depended on the circumstances of the individual firm.

    (a)          As part of the Pillar 2 requirements, banks were required to carry out Internal Capital Adequacy Assessments ("ICAAP"), while the PRA carried out a supervisory review and evaluation process, which included consideration of the bank's ICAAP.

    (b)         The PRA could impose a Pillar 2A capital requirement so as to cover risks which were (a) not adequately addressed by the Pillar 1 requirements and/or (b) not addressed at all under Pillar 1. Pillar 2A requirements varied from bank to bank, and a bank's combined Pillar 1 and Pillar 2A requirements represented its total capital requirement.

    (c)          In some cases, the above "standardised" approach to calculating Pillar 1 capital requirements overestimated the overall level of capital required, given the risks of the particular bank. Any overcapitalisation in the calculation of Pillar 1 requirements might be used to reduce a bank's Pillar 2A requirements; this was called the "unders/overs principle". The PRA set out its approach to allowing that type of offset in a Policy Statement issued in October 2017, known by its reference number PS22/17.

    (3)          The PRA also determined each bank's Pillar 2B requirement: this was the amount of additional capital a bank had to maintain in order to absorb losses that might arise in a severe but plausible stress scenario.

    (4)          Pillar 3 related to market discipline: firms were required to publish certain details of their risks, capital and risk management in order to improve market discipline. Those details included each banks' Pillar 1 and 2A requirements, but not Pillar 2B.  

  25. A bank had to meet a minimum "capital ratio", being the amount of its regulatory capital divided by its RWAs. Regulatory capital is the sum of its ordinary shares, retained earnings, perpetual subordinated debt instruments and unsecured subordinated debt. RWAs are a bank's assets and credit exposures, weighted (i.e. multiplied by a percentage factor) according to the potential to suffer loss. Safer assets were attributed a lower allocation of capital, while riskier assets were given a higher risk weight. In other words, the riskier its assets, the more capital a bank had to set aside in relation to the risk of loss.
  26. One of the types of risk which a bank had to consider was "credit risk", namely the risk that a borrower fails to repay a loan. There were two approaches to calculating RWAs for credit risk:
  27. (1)          A standardised approach, under which banks allocated a prescribed risk weight percentage to their assets.

    (2)          An internal ratings-based ("IRB") approach, under which banks used their own internal models for calculating the credit risk in their portfolios (i.e. using internal estimates of borrower creditworthiness), but with certain risk parameters determined by the PRA. The advanced internal ratings-based ("AIRB") approach allowed a bank to estimate those risk parameters.

  28. The IRB/AIRB approach could result in lower capital requirements, but Article 143(1) of the CRR provided that it could only be used if the PRA had given permission. At the relevant time, the Bank used the standardised approach for the calculation of its RWAs, but was hoping to move to an IRB or AIRB approach.
  29. For financial institutions using the standardised approach to credit risk, such as the Bank, Article 112 of the CRR set out the different asset or exposure classes and the risk weights to be assigned to each. The risk weighting requirements in that Article were directly applicable in the UK, so the PRA did not have the discretion to amend them.
  30. Article 501 of the CRR related to small and medium sized businesses ("SMEs") and was headed "Capital requirements deduction for credit risk on exposures to SMEs"; it was commonly referred to as "the SME supporting factor", and included the following provisions:
  31. "1. Capital requirements for credit risk on exposures to SMEs shall be multiplied by the factor 0,7619.

    2.  For the purpose of this Article:

    (a) the exposure shall be included either in the retail or in the corporates or secured by mortgages on immovable property classes. Exposures in default shall be excluded;

    (b) an SME is defined in accordance with Commission Recommendation 2003/361/EC of 6 May 2003 concerning the definition of micro, small and medium-sized enterprises. Among the criteria listed in Article 2 of the Annex to that Recommendation only the annual turnover shall be taken into account;

    (c) the total amount owed to the institution and parent undertakings and its subsidiaries, including any exposure in default, by the obligor client or group of connected clients, but excluding claims or contingent claims secured on residential property collateral, shall not, to the knowledge of the institution, exceed EUR 1,5 million. The institution shall take reasonable steps to acquire this knowledge."

    The Bank's lending

  32. The Bank lent money secured on commercial and residential property; the categories of loan which are relevant to this decision were known as CRE loans, CLIP loans, PBTL loans and residential loans. We explain those terms and their risk-weighting below.
  33. CRE loans

  34. Commercial Real Estate ("CRE") loans were secured by mortgages on commercial property which were to be repaid out of the income arising from that property (such as rental or trading income). Article 126 of the CRR required that these loans be given a RWA of 100%.
  35. CLIP loans

  36. Commercial immovable property ("CLIP") loans were also secured by mortgages on commercial real estate; however, they were not to be repaid out of the income from that property but instead from some other source. On occasion, the acronym "CRE" included CLIP loans, but in this judgment we have sought to distinguish between the two.
  37. Article 126 of the CRR provided that CLIP loans be assigned a risk weighting of 50%, but national regulators had the discretion under Article 124(2) to impose a higher risk weighting. On 1 January 2014, the PRA exercised that discretion and imposed a 100% risk weighting on CLIP Loans, unless the bank's "annual average losses stemming from lending secured by mortgages on commercial property in the UK did not exceed 0.5% of risk-weighted exposure amounts over a representative period". In that situation, the risk-weighting was 50%, see Rule 4.1 of the Credit Risk section of the PRA Rulebook.
  38. Residential loans

  39. Residential loans were, as the name indicates, secured on residential property. Article 125 of the CRR required that they be given a 35% risk weighting.
  40. PBTL loans

  41. Loans secured on professional buy-to-let mortgages were known as PBTL loans. Depending on the circumstances of the particular loan, they were classified as:
  42. (1)          residential mortgages under Article 125, with a 35% risk weighting;

    (2)          loans secured on immovable property other than residential under Article 124,  with a 100% risk weighting; or

    (3)          CLIP loans under Article 126, also with a 100% risk weighting.

    COREP reporting

  43. The Bank was required to complete quarterly confidential reports to the PRA under the "common reporting framework" or "COREP". COREP returns had to include information about a bank's capital requirements, including its credit risk, operational risk, own funds, capital adequacy ratios and RWAs.
  44. The Authority

  45. The Bank is regulated by the Authority in relation to conduct matters, including the market disclosure regime set out in the Listing Rules and also in Regulation 596/2014 (the Market Abuse Regulation or the "MAR"). The MAR placed requirements on listed companies regarding disclosures to the market about matters that would be relevant to investment decisions. The following provisions of the Listing Rules and the MAR were of particular relevance to this case.
  46. Listing Rule 1.3.3R

  47. LR 1.3.3R provided that an issuer, such as a bank, "must take reasonable care to ensure that any information it notifies to a RIS [Regulatory Information Service]...is not misleading, false or deceptive and does not omit anything likely to affect the import of the information". A Regulatory Information Service is a service approved by the Authority for disseminating regulated information.
  48. The MAR

  49. Article 7(1) of the MAR defined "inside information", and subparagraph (a) provided that this comprised:
  50. "information of a precise nature, which has not been made public, relating, directly or indirectly, to one or more issuers or to one or more financial instruments, and which, if it were made public, would be likely to have a significant effect on the prices of those financial instruments or on the price of related derivative financial instruments."

  51. Article 7(2) read:
  52. "For the purposes of paragraph 1, information shall be deemed to be of a precise nature if it indicates a set of circumstances which exists or which may reasonably be expected to come into existence, or an event which has occurred or which may reasonably be expected to occur, where it is specific enough to enable a conclusion to be drawn as to the possible effect of that set of circumstances or event on the prices of the financial instruments...In this respect in the case of a protracted process that is intended to bring about, or that results in, particular circumstances or a particular event, those future circumstances or that future event, and also the intermediate steps of that process which are connected with bringing about or resulting in those future circumstances or that future event, may be deemed to be precise information."

  53. Article 17 was headed "public disclosure of inside information" and para 1 begins:
  54. "An issuer shall inform the public as soon as possible of inside information which directly concerns that issuer. The issuer shall ensure that the inside information is made public in a manner which enables fast access and complete, correct and timely assessment of the information by the public..."

  55. Para 4 of that Article began:
  56. "An issuer...may, on its own responsibility, delay disclosure to the public of inside information provided that all of the following conditions are met:

    a.       immediate disclosure is likely to prejudice the legitimate interests of the issuer;

    b.       delay of disclosure is not likely to mislead the public;

    c.       the issuer...is able to ensure the confidentiality of that information.

    In the case of a protracted process that occurs in stages and that is intended to bring about, or that results in, a particular circumstance or a particular event, an issuer...may on its own responsibility delay the public disclosure of inside information relating to this process, subject to points (a), (b) and (c) of the first subparagraph."

    The evidence

  57. FSMA s 133(4) provided that on a reference the Tribunal may consider any evidence relating to its subject matter, whether or not that evidence was available to the decision-maker at the material time.
  58. We were provided with a main bundle of 6,143 pages and a core bundle of 445 pages (together "the Bundle"). In addition, we had witness evidence from Mr Donaldson and Mr Arden; from Ms Aileen Gillan, Ms Joanne Roberts and Mr Michael Brierley on behalf of the Applicants, and from Mr Mark Somers, Mr Phillip Dransfield, Mr Guy Sutherland and Mr John Lane, on behalf of  the Authority.
  59. Approach to the evidence

  60. In Roberts v FCA [2015] UKUT 408 (TCC) at [36] the Tribunal said:
  61. "We bear in mind the dangers of hindsight, which include analysing each conversation or note line by line, and attributing greater significance to such matters in the light of subsequent events, instead of considering matters as participants saw them as they occurred, or assuming that what happened subsequently was bound to happen."

  62. In reliance on that passage, Mr Jaffey submitted that "caution should...be exercised" when considering contemporaneous documents. However, in the well-known judgment of Gestmin SGPS SA v Credit Suisse (UK) Ltd [2013] EWHC 3560 (Comm) ("Gestmin"), Leggat J (as he then was), said:
  63. "[17] ...psychological research has demonstrated that memories are fluid and malleable, being constantly rewritten whenever they are retrieved. ...External information can intrude into a witness's memory, as can his or her own thoughts and beliefs, and both can cause dramatic changes in recollection.

    [18]  Memory is especially unreliable when it comes to recalling past beliefs. Our memories of past beliefs are revised to make them more consistent with our present beliefs....

    [19]  The process of civil litigation itself subjects the memories of witnesses to powerful biases. The nature of litigation is such that witnesses often have a stake in a particular version of events...

    [20]  Considerable interference with memory is also introduced in civil litigation by the procedure of preparing for trial...The effect of this process is to establish in the mind of the witness the matters recorded in his or her own statement and other written material, whether they be true or false, and to cause the witness's memory of events to be based increasingly on this material and later interpretations of it rather than on the original experience of the events.

    [21] ...

    [22] In the light of these considerations, the best approach for a judge to adopt in the trial of a commercial case is, in my view, to place little if any reliance at all on witnesses' recollections of what was said in meetings and conversations, and to base factual findings on inferences drawn from the documentary evidence and known or probable facts."

  64. In Simetra Global Assets Ltd v Ikon Finance Ltd [2019] EWCA Civ 1413, Males LJ gave the only judgment with which Jackson and McCombe LJJ both agreed. He said at [48]:
  65. "In this regard I would say something about the importance of contemporary documents as a means of getting at the truth, not only of what was going on, but also as to the motivation and state of mind of those concerned. That applies to documents passing between the parties, but with even greater force to a party's internal documents including emails and instant messaging. Those tend to be the documents where a witness's guard is down and their true thoughts are plain to see. Indeed, it has become a commonplace of judgments in commercial cases where there is often extensive disclosure to emphasise the importance of the contemporary documents. Although this cannot be regarded as a rule of law, those documents are generally regarded as far more reliable than the oral evidence of witnesses, still less their demeanour while giving evidence."

  66. Two months later, in Kogan v Martin & Ors [2019] EWCA Civ 1645, Floyd LJ, giving the judgment of the Court, said at [88]:
  67. "...Gestmin is not to be taken as laying down any general principle for the assessment of evidence. It is one of a line of distinguished judicial observations that emphasise the fallibility of human memory and the need to assess witness evidence in its proper place alongside contemporaneous documentary evidence and evidence upon which undoubted or probable reliance can be placed....a proper awareness of the fallibility of memory does not relieve judges of the task of making findings of fact based upon all of the evidence. Heuristics or mental short cuts are no substitute for this essential judicial function. In particular, where a party's sworn evidence is disbelieved, the court must say why that is; it cannot simply ignore the evidence."

  68.  In Tui v Griffiths [2023] UKSC 48 ("Tui"), Lord Hodge, giving the only judgment with which the other members of the Supreme Court agreed, said at [70]:
  69. "(i) The general rule in civil cases, as stated in Phipson, 20th ed, para 12-12, is that a party is required to challenge by cross-examination the evidence of any witness of the opposing party on a material point which he or she wishes to submit to the court should not be accepted. That rule extends to both witnesses as to fact and expert witnesses.

    (ii) In an adversarial system of justice, the purpose of the rule is to make sure that the trial is fair.

    (iii) The rationale of the rule, ie preserving the fairness of the trial, includes fairness to the party who has adduced the evidence of the impugned witness.

    (iv) Maintaining the fairness of the trial includes fairness to the witness whose evidence is being impugned, whether on the basis of dishonesty, inaccuracy or other inadequacy...

    (v) Maintaining such fairness also includes enabling the judge to make a proper assessment of all the evidence to achieve justice in the cause. The rule is directed to the integrity of the court process itself.

    (vi) Cross-examination gives the witness the opportunity to explain or clarify his or her evidence. That opportunity is particularly important when the opposing party intends to accuse the witness of dishonesty, but there is no principled basis for confining the rule to cases of dishonesty.

    (vii) The rule should not be applied rigidly. It is not an inflexible rule and there is bound to be some relaxation of the rule...Its application depends upon the circumstances of the case as the criterion is the overall fairness of the trial. Thus, where it would be disproportionate to cross-examine at length or where, as in Chen v Ng, the trial judge has set a limit on the time for cross-examination, those circumstances would be relevant considerations in the court's decision on the application of the rule.

    (viii) There are also circumstances in which the rule may not apply: see paras 61-68 above for examples of such circumstances."

  70. Most of the "circumstances" referred to in paras 61-68 of Tui are specific to expert evidence, but they also include matters "to which the challenge is directed is collateral or insignificant and fairness to the witness does not require there to be an opportunity to answer or explain", or where the "evidence of fact may be manifestly incredible, and an opportunity to explain on cross-examination would make no difference".
  71. In assessing the evidence in this case, we have sought to follow the guidance set out in the above case law, noting that most of the events with which we are concerned took place in 2018, seven years before this hearing, and that there is contemporaneous evidence relating to most of the events in question, including emails and internal reports.
  72. Mr Arden

  73. Mr Arden provided two witness statements, was cross-examined by Mr Stanley, re-examined by Mr Jaffey and answered questions from the Tribunal. He was often reluctant to give straightforward answers on key evidential points. For example, a paper circulated before a meeting on 22 October 2018 stated that £574m of commercial mortgages had been incorrectly risk-weighted at 50% instead of 100%. Three other attendees at the same meeting confirmed that there had been no dissent from that proposition, but when Mr Arden was asked the same question three times in succession, he failed to answer. The dialogue went as follows:
  74. "Mr Stanley: There was no suggestion at this meeting of any dissent
    from the decision to reclassify risk-weighted assets, was there?

    Mr Arden: There was no decision to reclassify the assets at this stage...the committee explicitly didn't approve the paper. It noted the paper.

    Mr Stanley: Yes, it noted the paper, but nobody suggested that that reclassification shouldn't happen, did they?

    Mr Arden: No, because I knew that Deloitte were coming in at that stage.

    Mr Stanley: Nobody suggested that there was any error - that there hadn't been an error in relation to how the assets were classified.

    Mr Arden: At this stage I was waiting for Deloittes to come in."  

  75. To give another example out of many, the exchange about Mr Arden's understanding of the RWA position went as follows:
  76. "Mr Stanley: Did you think you had any idea what the bank's risk-weighted assets actually were at 30 September 2018?

    Mr Arden: That's a good question. We were carefully considering and doing work at pace to understand what the exact bank's RWAs were [sic].

    Mr Stanley: Does that mean you didn't actually know what they were?

    Mr Arden: As you know, we were doing work to get to the bottom of the issues that we were facing."

  77. The overall focus of Mr Arden's written and oral evidence was to support his case rather than to describe events in a straightforward manner. For example, it was a key part of his case that he did not know there was a material error in the Q3 Update figures until January 2019. Mr Arden said:
  78. (1)          he "considered it very unlikely that any number was accurate"; and had "no basis on which to conclude that the present estimates were reliable";

    (2)          the estimate had been based "on high-level sampling";

    (3)          he "understood there were differing views" as to how to interpret the regulatory requirements;

    (4)          the PRA had told the Bank that "all discussions" between the PRA and the Bank "should remain confidential"; and

    (5)          Linklaters had advised that the Bank did not need to disclose the CLIP loans error at the time of the Q3 Update.

  79. As is clear from our findings in the main body of this decision, none of those statements was correct. At the time of the Q3 Update:
  80. (1)          Mr Arden did not express doubt about the estimated figures, see §§365-380;

    (2)          no-one considered that there was any doubt about the regulatory requirements, see §§364-381;

    (3)          the estimate had not been based on sampling, and Mr Arden knew this was the case, see §547;

    (4)          the requirement as to confidentiality related only to discussions about the future exercise of the PRA's discretion, and was in any event not followed by the Bank, see §§278-285; and

    (5)          the purpose of the meeting with Mr Lane was to ask whether the Bank needed to make an immediate disclosure under the MAR and did not relate to the Q3 Update, see §331-335.

  81. In oral evidence, Mr Arden went beyond the matters set out in his witness statement. He stated that he doubted the Bank's financial data generally, saying "I'm not sure at this stage that I believed anything that was being presented to me"; and "I didn't quite believe anything that the internal team were telling me". Mr Stanley described these statements as "hyperbolic", submitting that they did not reflect Mr Arden's position at the time, and in any event did not help his case. We agree, see §§373-375.
  82. For all those reasons, we did not accept some of Mr Arden's evidence. However, we also find that he was not deliberately attempting to mislead the Tribunal. Following the events in question, he had been required repeatedly to justify and explain what happened, and as Leggatt J said in Gestmin, his memory had been "revised" to make it "more consistent" with his present beliefs, rather than reflecting the position at the time. The reason he was reluctant to give straightforward answers when cross-examined was because he was being forced to confront contemporaneous evidence which was inconsistent with his memory (as revised), not because he was being obstructive.
  83. Mr Donaldson

  84. Mr Donaldson provided a witness statement, was cross-examined by Mr Stanley and re-examined by Mr Jaffey. His statement, like that of Mr Arden, was largely focused on emphasising that he was uncertain about the materiality of the CLIP loan issue, but his answers under cross-examination were more straightforward. For example, his witness statement said that (a) the estimated CLIP loan error had been based on sampling, and (b) there was uncertainty as to the regulatory position. However, he agreed under cross-examination that:
  85. (1)          he had confused the process used to estimate the errors in relation to PBTL and CLIP loans; and

    (2)          he had been told KPMG had confirmed the regulatory position, and had no reason to think that KPMG were wrong.

  86. Where we have not accepted Mr Donaldson's recollections, we have explained our reasoning in the main body of this decision. As with Mr Arden, we find that his memory as reflected in parts of his evidence had been "revised" to make it "more consistent" with his present beliefs, instead of reflecting the position at the time, see Gestmin.
  87. Ms Gillan

  88. At the relevant time, Ms Aileen Gillan was the Bank's Chief Risk Officer. She provided two witness statements, was cross-examined by Mr Stanley and re-examined by Mr Jaffey. Her second statement said, in support of the Applicants, that in October 2018 "the quality of the data was so poor that it would have been impossible, without a loan-level review of non-retail lending, to have had certainty around the number".
  89. Under cross-examination she gave straightforward answers to Mr Stanley's questions, while continuing to emphasise the degree of uncertainty. For example, although she agreed that it was known by July 2018 that (a) the CLIP loans had been wrongly risk-weighted at 50% rather than 100%, and (b) this affected £1.1bn of loans, she rejected Mr Stanley's suggestion that the error was therefore "going to make a difference of hundreds of millions".
  90. However, she also accepted that the figure of £574m was the best estimate the Bank had at the time, and that the Bank had also understood that "any further adjustments to calculations are not expected to be material". By the conclusion of her oral evidence, we accepted that she was distinguishing between that best estimate and the remaining doubt as to the precise value of CLIP loan adjustment required.  
  91. Ms Roberts

  92. At the relevant time, Ms Joanne Roberts was head of the Bank's Investor Relations. She provided a witness statement and was cross-examined by Mr Ratan. Part of her witness statement concerned a meeting with Linklaters which took place on 5 October 2018, about which Ms Roberts had sent an email. In her oral evidence she said she now had no independent recollection of the advice given in that meeting, other than what was in the email. However, the Authority did not ask us to reject the evidence in the remainder of her witness statement and we have taken it into account in making our findings of fact.  
  93. Mr Somers and Mr Dransfield

  94. In May 2018, Ms Doriana Iovino, the Bank's Director of Credit Risk and Analytics, went on maternity leave and her role was filled on an interim basis by Mr Somers and Mr Edgardo Ferrari Costa on a job-share basis, with each doing three days a week. In September 2018, Mr Phillip Dransfield replaced Mr Costa.
  95. Mr Somers provided two witness statements, was cross-examined by Mr Jaffey, re-examined by Mr Stanley and answered questions from the Tribunal. Mr Dransfield provided a witness statement and was cross-examined by Mr Pritchard. We found both to be straightforward and credible witnesses.
  96. Mr Sutherland

  97. At the relevant time, Mr Guy Sutherland worked for the PRA. He provided two witness statements, was cross-examined by Mr Jaffey and re-examined by Mr Stanley. We found him also to be a credible and straightforward witness.
  98. Mr Lane

  99. Mr John Lane was a partner at Linklaters LLP. At the relevant time he was the lead corporate partner responsible for Linklaters's relationship with the Bank. He provided a witness statement but was unable for health reasons to attend the hearing. In consequence, he could not be cross-examined on his witness evidence. Mr Jaffey submitted that where there was a difference between Mr Lane's evidence and that of Mr Arden, the Tribunal should prefer the latter. As is clear from the main body of this decision, we have considered all relevant evidence when making our findings, and have also taken into account that Mr Lane could not be cross-examined.
  100. Mr Brierley

  101. Mr Michael Brierley was the Bank's CFO from its inception until 2013, when Mr Arden took over that role. His witness statement was unchallenged, and he therefore did not attend the hearing to give evidence. However, some passages in his statement concerned his view as to what Linklaters "would" or "should" have done when giving advice, and more generally what he "would expect any lawyer on whom the Bank relied" to have done. That is not evidence of facts and we have disregarded those passages.
  102. Individuals who were not called as witnesses

  103. Mr Jaffey submitted that the Tribunal should draw adverse inferences from the absence of the following individuals who had not been called by the Authority to give evidence:
  104. (1)          The PRA employees who were supervising the Bank after Mr Sutherland moved to another role in mid-October 2018. We discuss Mr Jaffey's particular submissions at §404.

    (2)          Mr Woods, a Director of the PRA, with whom Mr Arden had a conversation in January 2023. We set out the relevant extract at §435, and discuss it at §440.

    (3)          Mr Vernon Hill, the Bank's Chairman, and the non-executive members of the Board, see further §443ff.

    (4)          Ms Sally-Ann James, the Bank's General Counsel. We discuss her position at §§467-479.

  105. In considering Mr Jaffey's submissions about adverse inferences, we have followed the approach set out in Efobi v Royal Mail Group Ltd [2021] UKSC 33, [2021] 1 WLR 3863, where Lord Leggatt, with whom Lord Hodge, Lord Briggs, Lady Arden and Lord Hamblen agreed, said at [41]:
  106. "The question whether an adverse inference may be drawn from the absence of a witness is sometimes treated as a matter governed by legal criteria, for which the decision of the Court of Appeal in Wisniewski v Central Manchester Health Authority [1998] PIQR P324 is often cited as authority. Without intending to disparage the sensible statements made in that case, I think there is a risk of making overly legal and technical what really is or ought to be just a matter of ordinary rationality. So far as possible, tribunals should be free to draw, or to decline to draw, inferences from the facts of the case before them using their common sense without the need to consult law books when doing so. Whether any positive significance should be attached to the fact that a person has not given evidence depends entirely on the context and particular circumstances. Relevant considerations will naturally include such matters as whether the witness was available to give evidence, what relevant evidence it is reasonable to expect that the witness would have been able to give, what other relevant evidence there was bearing on the point(s) on which the witness could potentially have given relevant evidence, and the significance of those points in the context of the case as a whole. All these matters are inter-related and how these and any other relevant considerations should be assessed cannot be encapsulated in a set of legal rules."

    Findings of fact

  107. On the basis of the evidence summarised above, including our findings on credibility, we make the findings of fact in this decision. We have taken an essentially chronological approach and made detailed findings because the parties relied in their submissions on numerous factual points spread throughout the relevant period.
  108. The next following parts of our decision concern the three issues in dispute: whether the Bank breached the Listing Rules; whether the Applicants were knowingly concerned in that breach, and the penalties. In considering each of those issues, we link our chronological findings to the submissions made, and also make further findings.
  109. The early years

  110. The Bank began operations in 2010, with Mr Hill as the Chairman and Mr Donaldson as CEO. It is known as a "challenger" bank because it seeks to compete directly with older, more established banks. It was listed on the main market of the London Stock Exchange in March 2016. By 2018, it was the seventh-largest bank in the UK.
  111. Linklaters

  112. Linklaters provided the Bank with legal advice related to the IPO and from then on was the Bank's corporate counsel of choice. Mr Lane was Linklaters's relationship partner; he was supported by Mr Jonathan Sadler (who became a partner before 24 July 2018) and Mr Rupert Cheyne (who was not a partner) and by teams of specialists in (a) Corporate, Debt and Capital Markets led by Mr Jason Manketo and (b) regulatory matters, led by Mr Umesh Kumar. Both Mr Manketo and Mr Kumar were partners in the firm. More generally, as Mr Brierley said, Linklaters also had "a strong undercast of UK-based lawyers".
  113. Key personnel during the period from March 2018

  114. As noted above, Ms Jo Roberts was the Bank's Director of Investor Relations, and Ms Gillan was the Chief Risk Officer with overall responsibility for risk management throughout the Bank.
  115. Mr Dave MacLean was the Bank's Finance Director, reporting to Mr Arden. One of his responsibilities was the oversight of regulatory reporting, while Mr Mark Stokes was the director of commercial lending, reporting to Mr Donaldson.
  116. Ms Iovino, the Director of Credit Risk and Analytics, reported to Ms Gillan. As already noted, in May 2018, she went on maternity leave and her role was filled on an interim basis by Mr Somers and Mr Costa on a job-share basis, with each doing three days a week. Mr Dransfield replaced Mr Costa in September 2018.
  117. Ms Suzie Orrell, Mr Dave Richardson and Mr Olivier Baixas were members of the Credit Risk team led by Mr Somers and Mr Costa/Mr Dransfield. That team was responsible for credit risk modelling and monitoring, including (a) the building and maintenance of models necessary for the Bank's ICAAP reporting, and (b) dealing with the Bank's possible move from a standardised approach for calculating RWAs to an AIRB approach: that project was headed by Mr Richardson.
  118. Relationship with the PRA and the Authority

  119. The Bank was supervised by the PRA's New Bank Supervision Team; Mr Sutherland was a Senior Supervisor in that Team. From early 2016 until October 2018 he was the Bank's lead contact and supervisor; he had monthly "catch-up" meetings with Ms Gillan and Mr Arden, and also had contact with Mr Donaldson. On 1 January 2019, responsibility for supervising the Bank moved to a new team comprising Mr Nicholas Elvidge, Mr Richard Beckett and Ms Christina Fritz (who was shortly afterwards replaced by Ms Holly Jones).
  120. The Bank made frequent requests for the Authority similarly to identify a dedicated supervisory team, but the Authority said that it did not have the time and where necessary would continue to get information from the PRA. When the Bank needed to contact the Authority, it had to do so via the FCA's Supervision Hub, known at the time as the Contact Centre.
  121. 2016 and 2017

  122. In the first quarter of 2016, the Bank raised £400m of capital. Its trading update for that quarter showed that deposits had increased by 75% over the previous year and lending by 125%.
  123. On 26 October 2016, the Bank announced its first quarterly underlying profit before tax of £0.6m. On 22 February 2017, it issued its 2016 full year results, reporting that deposits had increased by 56%; lending by 66%, and that it had raised further capital of £400m.
  124. In June 2017, the Bank completed a further capital raise, this time for £278m. Its 2017 Annual Report included the statement that "as our balance sheet growth continues, we will need to raise additional qualifying regulatory capital".
  125. At some point before March 2018, the Bank purchased a book of mortgages which were internally called "London" and in that month purchased another book, internally called "Canberra" for £530m.
  126. The COREP audit and the CRE loans

  127. Meanwhile, on 9 September 2016, Mr Sutherland informed the Bank that it had been selected by the PRA for inclusion in an audit of COREP returns. Metro Bank's Internal Audit team was required to review and assess whether the Bank had effective procedures and controls in place to produce those returns, and to assess a sample of submitted returns. 
  128. The audit was carried out between January and March 2017; the work consisted in assessing whether:
  129. (1)          the numbers had been properly extracted from the firm's books and records; and

    (2)          the returns had been completed in accordance with the definitions, calculations and application of methods as defined in the CRR and in the "Implementing Technical Standards" published by the European Banking Authority.

  130. A draft report was provided to management in May 2017 and the final report was sent to Mr Sutherland in July 2017. The findings were summarised as follows:
  131. "Audit reviewed the June and September 2016 returns and identified a number of errors (e.g. omissions, inconsistent interpretation of rules) which are summarised below. The Regulatory reporting team had subsequently rectified and corrected the majority of the errors in the March 2017 COREP returns, the details of which are attached in Appendix 1. The errors in the June and September returns resulted in a net understatement of required capital in the range of £0.95m to £12.59m (0.26% to 3.50% of total required capital) for June 2016 and £0.8m to £11.46m (0.21% to 2.97% of total required capital) for September 2016 respectively..."

  132. The report explained that "the most significant impact of the errors" was due to the incorrect RWA being applied to CRE loans; these should have been risk weighted at 100% but had been risk weighted at lower percentages. The report continued:
  133. "This is due to a lack of information in the systems to allow proper classification. Since the identification of this error by audit, the Regulatory Reporting team has started categorising (and calculating the impact of) the CRE loans manually until an automated solution is in place. Commercial Lending have a project underway to assign a more granular classification to loans which will allow Regulatory Reporting to automate the classification of loans. As at 8 May 2017, the estimated understatement of capital requirement is in the range of £11.1m to £21.7m for the quarter to September 2016 and £9.12m to £20.76m for the quarter to June 2016. A remaining portion of CRE loans amounting to £223m (36% of approx. total CRE loans) is still being worked on to establish the appropriate risk weights to be applied."

  134. The detail of the report included a section headed " Treatment of exposures secured by commercial immovable property", which read:
  135. "In case of Exposures secured by Commercial immovable property the Bank has been prudent and applies a 50% risk weight to that part of the Exposure  which has an LTV<50%. However, as per the CRR rules exposures with LTV <60% are applied a Risk weight of 50%. Management should  reconsider if this prudent approach is still appropriate."

  136. It was common ground that this was incorrect. Although Article 126 of the CRR did provide that CLIP loans be assigned a risk weighting of 50%, the PRA required 100% risk weighting, see §25.
  137. By the end of 2017, over 50% of the portfolio had been reviewed and classified by the Commercial Lending team, as the result of which £500m assets had been reclassified as CRE and the Bank's RWA increased by £237m. However, as is clear from the above, there was a misunderstanding as to the correct treatment for CLIP loans and these were still incorrectly risk-weighted.
  138. Mr Arden, the Board and the committees

  139. In March 2018, Mr Arden joined the Bank as CFO and Company Secretary, replacing Mr Brierley. He had previously been the CFO of Sainsbury's Bank Plc, where he had worked since March 2012. Sainsbury's Bank was not a listed company, so Mr Arden did not have experience of the Listing Rules or the MAR. On 16 May 2018, Mr Lane and Mr Sadler gave him training on both.
  140. Mr Arden and Mr Donaldson were the only executive directors on the Bank's Board. Apart from Mr Hill, the other non-executive directors ("NEDs") were Mr Stuart Bernau, Mr Keith Carby, Mr Ben Gunn, Mr Roger Farah, Lord Howard Flight, Mr Gene Lockhart, Ms Monique Melis and Sir Michael Snyder. Most of the NEDs were experienced in financial services and many had worked for other listed entities: in particular, Mr Bernau had been a specialist adviser to the Treasury Committee in relation to the review into the failure of the HBOS group, and Mr Lockhart was formerly the CEO of Midland Bank and MasterCard.
  141. One of the sub-committees of the Board was the Risk Oversight Committee. It was chaired by Mr Lockhart, while Mr Snyder, Mr Bernau, Mr Gunn and Ms Melis were members. Mr Donaldson and Mr Arden regularly attended as guests, as did Ms Gillan and others.
  142. The Bank's Audit Committee was chaired by Mr Bernau, while Mr Carby, Mr Lockhart and Mr Snyder were members. Mr Donaldson and Mr Arden attended meetings as guests, as did Ms Gillan and Ms Yeoh, the head of Internal Audit, as well as one or more partners from PwC, the Bank's auditors.
  143. The Credit Risk, Policy and Appetite Committee ("CRPAC") was responsible for oversight of credit risk policies, reviewing proposals on risk appetite; and monitoring portfolio performance against risk appetite. Members included Mr Donaldson, Mr Arden, Ms Gillan and Mr Stokes. Before matters were referred to the CRPAC, they were reviewed by the Portfolio Quality Review forum ("PQR"), made up of members of the Bank's Finance and Risk team.
  144. The Bank also had a Disclosure Committee (sometimes called the "DisCo"); its members were Mr Donaldson, Mr Arden, and the Bank's General Counsel, Ms James. Its Terms of Reference provided that the committee's role was to "draw up and maintain procedures, systems and controls for the identification, treatment and disclosure of inside information and for complying with other disclosure obligations which apply to the Bank" and to carry out related tasks, including to:
  145. "ensure that all regulatory announcements, shareholder circulars,  prospectuses and other documents issued by the Bank under any legal or regulatory requirement are scrutinised in order to ensure that they are accurate and not misleading and do not omit anything material and comply with applicable legal and regulatory requirements."

  146. We agree with Mr Jaffey that this reflected the requirements of LR 1.3.3R.
  147. KPMG appointed

  148. The Bank had developed a plan to move from standardised basis to IRB over five years, beginning with residential mortgages, followed in turn by buy-to-let mortgages, revolving unsecured mortgages, and finally by commercial loans. The implementation of each stage would be dependent on PRA approval. The Bank began to carry out the plan in February 2018, by applying to the PRA for residential mortgages to move on to the IRB basis.
  149. At the end of March 2018, the Bank began discussions with KPMG about assisting with its preparation for its commercial loan IRB application. On 29 March 2018, Ms Iovino emailed Mr Hall at KPMG saying:
  150. "I just wanted to clarify our starting point to avoid any underestimation of the task or build false expectation. Our commercial portfolio is £2bn and made of circa 3k loans, we have no classification and very few data points, consequently our RWA assessment will not be extremely precise (especially in relation to the CRE exposure). For this reason, whilst we will happily provide access to the documentation, you will not be able to find a lot in it. The same applies to the data captured by the front end..."

  151. On 10 April 2018, the Bank formally engaged KPMG to produce decision trees to support the correct classification of loans. Key elements of the task were as follows:
  152. (1)          obtain a high level understanding of the data that would be available for  classification purposes;

    (2)          review the potential exposure classifications that are available for such lending  under both the Standardised and IRB approach; and

    (3)          develop decision trees [1] (sets of questions) that will allow new lending to be  mapped into the various exposure classifications (one for standardised and one for IRB).

  153. The decision trees were designed following a "desktop review of key documentation"; onsite meetings to discuss actual business and data structures, and analytical work in KPMG's offices. One of the factors in the decision trees was whether the SME supporting factor applied. The purpose of the decision trees was to provide a tool which the Bank's staff could apply to individual loans to determine the asset classification of each.
  154. April to June 2018

  155. On 24 April 2018, the Bank published its Q1 results for 2018; the announcement included a statement that a further debt issuance was anticipated. After the results, the Bank's share price fell 12%.
  156. On the following day, Mr Sutherland held his regular monthly catchup call with Mr Arden and Ms Gillan, and asked for their view as to the reasons for the share price fall, and Mr Arden said that investors had not expected the Bank to need more capital. Mr Sutherland also asked for an update on the CRE remediation work, and was told that this was in progress; that KPMG were helping, and that the Bank would update the PRA at the end of May.
  157. On 16 May 2018, Mr Lane, Mr Sadler and Mr Cheyne of Linklaters gave an Update Presentation on the MAR to Mr Arden, Ms Roberts, Ms James and Ms Conway. It included an explanation of what was meant by "precise" and "significant effect on price" in Article 7, see §§31-32.
  158. On 18 May 2018, Ms Iovino emailed Mr Maclean, Ms Orrell and others, and copied Mr Arden, Ms Gillan, Mr Richardson, Mr Somers and Mr Costa. She began by saying that following a meeting with KPMG, "some potential gaps in the RWA calculation" had been identified. She attached a paper written by the Credit Risk and Analytics team led by Mr Somers and Mr Costa, which referred to certain types of PBTL loans, and said (where T24 is a reference to the Bank's main computer system):
  159. "Credit Risk and analytics have performed some analysis and, based on the requirements set out by CRR [Article] 125, we propose the following to be applied and communicated to the teams with immediate effect:

    1. The following collaterals need to be classified as commercial property in T24 (currently they are classed as retail property):

    o Housing Association Properties

    o Houses in Multiple Occupation

    o Student Accommodation."

  160. The paper went on to say that the above types of mortgages were currently risk weighted as residential at 35%, but should have been risk-weighted at 100% as commercial property. At or around the same time, KPMG produced drafts of their decision trees, which stated that "exposure secured by mortgages on immovable property" were required to be given a risk-weighting of 100%, and this included PBTL.
  161. On 31 May 2018, Mr Richardson emailed Mr Costa, saying:
  162. "Talking to Suzie Orell and KPMG yesterday, it has come to light that we  don't think we are using the right standardised risk weight for commercial mortgages...in effect all loans secured on commercial property should have a 100% RW - we are still using 50% for our 'trading business' loans."

  163. Mr Richardson said his "quick and dirty estimate" suggested there was "something like £250m of extra RWA required (=£20m of extra capital)".
  164. July 2018

  165. On 11 July 2018, Mr Sutherland emailed Mr Costa asking for the average risk weights for various types of commercial loan. Ms Gillan replied the following day, saying that they were between 38% and 58%.
  166. At or around the same time, Mr Sutherland was informed by a capital specialist at the PRA that the Bank was risk weighting CLIP loans at 50% instead of 100%, with the result that the RWAs being reported to the PRA were materially lower than they should have been. It was "immediately clear" to Mr Sutherland that this would be "a serious issue". He explained:
  167. "Moving from 50% RWAs to 100% for all commercial secured loans would be a large material change, and the PRA quickly realised the magnitude of this. The material uncertainty in Metro Bank's Pillar 1 RWAs would affect how the PRA considered a number of different areas, including Metro Bank's key risks, systems and controls, the bank's AIRB application, and whether the PRA would be able to consider a Pillar 2A offset."

  168. On 13 July 2018, Mr Sutherland emailed Ms Gillan again, copying Mr Somers and Mr Costa, saying that:
  169. (1)          the "average risk weight seems low given it includes unsecured commercial loans and commercial investment loans"; and

    (2)          the Metro team at the PRA had been internally challenged on the Bank's "reporting of commercial risk weights and whether 50% risk weighting is appropriate".

  170. On 16 July 2018, Mr Somers emailed Ms Gillan, copying Mr Richardson and Mr Costa, saying:
  171. "I think we need to come clean and simplify the story. The answer is that the 50% is wrong and we know it to be. Otherwise we will be told which will be worse (s166 territory?!)."

  172. The reference to "s 166 territory" was to FSMA s 166, which gave the PRA power to be provided with information or documents "with respect to any matter", either by requiring a report, or by appointing a "skilled person" to provide such a report.
  173. Mr Richardson responded to Mr Somers and Ms Gillan within the hour, saying (emboldening in original):
  174. "A bit of boring detail to support the previous emails - As confirmed with KPMG last week, the following changes need to be made to our current reporting - The 50% risk weights for "commercial mortgages (mixed collateral)" and "commercial (other)" should be 100%, with SME support factor as appropriate."

  175. He added (again, emboldening in original):
  176. "One upside is that I think that many of the 'retail commercial mortgages' on CHL (London & Canberra) currently marked at 50% are loans to 'personal investment companies' for residential properties, so would qualify as retail under CRR Article 125 and hence 35%."

  177. On 17 July 2018, Ms Gillan responded to Mr Sutherland's first point as follows:
  178. "As you know, we are reviewing the asset classification and/or RWA assigned. The 44% is a result of applying an RWA of 50% and combining the 76% overlay for SME to the loans that qualify for the SME discount. Based on analysis undertaken as part of the asset classification review so far we now believe the 50% to be inaccurate and will be re-stating once the findings of the review are finalised."

  179. She responded to his second point by saying:
  180. "We can confirm the 50% RWA is the RWA we used in March 2018 but,  following our asset classification review, we now believe this to be inaccurate. Once the review has concluded, we will revert with further details."

    The 2018 capital raise and half year results

  181. Meanwhile, plans for the capital raise proceeded. On 13 June 2018, Mr Lane provided advice on a combined issuance of equity and AT1 [perpetual] bonds. On 12 July 2018, Linklaters provided the Bank with advice on taking "market soundings" before issuance, and on 17 July Mr Sadler advised on the key actions for the capital raise.
  182. On 24 July 2018, the Bank issued a half year trading update, which included the following passages (emboldening in original):
  183. "Capital ratios remain robust. Common Equity Tier 1 Capital ("CET1") as a percentage of risk weighted assets is 12.7%, currently exceeding our Tier 1 regulatory minimum of 9.7%. Risk weighted assets at 30 June 2018 were £6,944m. The Regulatory Leverage ratio is 4.6%...

    Further growth supported by proposed equity capital raise announced separately today. This will enhance already robust capital ratios.

    Our Pillar 2A requirement of 1.7% is currently under review with the PRA. We anticipate receiving capital relief as part of the Pillar 2A offset, in effect temporarily reducing the regulatory minimum and hence increasing management buffers, ahead of transitioning to the advanced internal ratings based approach (AIRB) on residential mortgages, expected H2 2019."

    August 2018: PBTL and CLIP

  184. On 3 August 2018, Mr Stokes emailed Mr Donaldson, saying:
  185.  "we have got to a conclusion on how we should be allocating capital to BTL/PBTL and within both what is the treatment of HMO's...Impact is that we will have to likely allocate more capital to the back book as the element  that needs to move from current 35% RWA to 100% RWA will increase as what has been classed as mortgage lending becomes commercial IPRE [income-producing real estate] lending."

  186. He went on to say that where a PBTL loan was for over five properties or was a HMO [House in Multiple Occupation], it should be risk-weighted at 100%, but that the remaining loans should remain as residential and thus risk-weighted at 35%. Mr Stokes also noted that the Bank didn't capture the number of properties on which the loan had been made, or whether it was a HMO, and asked "how do we classify this going forward", adding that the Bank needed to review "the back book" in relation to classification.
  187. Following a meeting at the Bank on 6 August 2018, Mr MacLean emailed Ms Gillan, Mr Somers and others with the actions which had resulted, which included:
  188. (1)          final sign-off of KPMG project and publication of decision trees;

    (2)          update detailed RWA numbers and analysis based upon agreed decision rules;

    (3)          understand the gap on PBTL - undertake a policy by policy review;

    (4)          design and implement new data capture process; and

    (5)          update formal reporting and discuss with PRA.

  189. On 7 August 2018, Mr MacLean emailed Ms Gillan, Mr Somers and others, suggesting that the RWAs for commercial lending depended on the loan-to-value ratio ("LTV") and could thus remain at 50% for some of the loans. Mr Somers replied the same day, saying:
  190. "No, I don't believe 50% RW is a valid choice as the PRA has used its  derogated powers to deviate from standard CRR and adopt 100% for property secured on commercial property."

  191. On the morning of 17 August 2018, Mr MacLean emailed Mr Arden, saying:
  192. "We have done further work on the RWAs. I think we are much closer to having a minimum estimate for the impact of the change in risk density...I think therefore we need to resolve this issue materially in September.

    The key points of progress are:

    · We have been through some challenge on the rules and I think we have a joined up understanding. The outcome is at the higher end of the impact ranges we have discussed as pretty much every commercial loan that is not PBTL is 100% RWA classified (or 75% if SME)

    · Magdalena and Olivier in the credit risk analytics team have been working to harmonise numbers and I don't believe we have material disagreements

    · We still have an outstanding question re £1.7b of PBTL (whether it is in fact PBTL). The credit team are testing a sample of 170 contracts to test our expected level of accuracy."

  193. On the afternoon of 17 August 2018, Mr MacLean emailed Mr Arden; the email was headed "data quality issues impacting Reg reporting" and said:
  194. "We have talked in the past about some of the issues that occur in our reg  reporting, and how they are impacted by both the historic manual nature of some of our processes, and we have talked about the investments we are  proposing that will help with that situation.

    We have also flagged data quality as an issue, but haven't really got into the detail of that in our conversations. At the moment, we are continuing to find issues with our upstream data that have caused inaccuracies in reg and external reporting...We are dependent upon people elsewhere in the bank capturing the right grain of data, capturing it accurately, and not making changes that impact our processes.

    The document attached gives a few of the material examples of this problem. I believe that we are in a situation where the risks we are running are now too high. I am keen for us to undertake a piece of work that ensures we have the right processes and controls in place to capture the right data and ensure the data quality is reliable...I think this is an end to end and complex problem and therefore finding a solution might be like eating an elephant. I think the first step is for us to highlight the need and importance of the data that we rely on, and would like your help in ensuring this kind of thing is  viewed with sufficient priority."

  195. The "document attached" included the following, under the heading "misclassification of Loan Facility Type":
  196. "An extra field 'Facility Type' (loan purpose) was added to the CRM data at the back end of last year to close an audit point. An extensive exercise was carried out by the relationship manager all across Metrobank to populate this field in the CRM system to enrich the data with the 'loan facility type'.

    Following on from this exercise, Credit Risk and Regulatory Reporting teams have performed some sampling testing on the data to check the accuracy of the data populated and we have encountered data errors. For examples:

    (1) Credit Risk have picked the top two largest exposures classified as PBTL and checked the underlying collateral secured by these loans. It appeared that these two loans classified as BTL are secured by commercial properties but have been classified as BTL.

    (2) Regulatory Reporting team have also performed some sampling testing and have observed loans secured by residential properties being classified as 'Commercial Owner Occupier'.

    Impact: This observation has resulted in us questioning the percentage of accuracy of the data populated upstream. Regulatory reporting process place a heavy reliance on data accuracy coming from upstream data. We are not able to calculate the risk weights correctly if the information we have place reliance on is not populated correctly."

  197. Under the heading "Source Data - PBTL Flag does not exist in the system to support regulatory reporting process", the text read (where "Pepper" is a reference to an external computer system used by the Bank to supplement T24, its main system):
  198. "Following on from the review of the Credit risk weights by KPMG, it has been highlighted to Metrobank that Professional BTL (PBTL) will need to be risk weighted at 100% instead of 35%. However, PBTL flag does not exist in the current Metro system and therefore it is impossible to distinguish between a BTL and a PBTL loans in order to ascertain the PBTL impact on RWA.

    Currently, we have about £1b of Pepper and T24 loans that have been flag[ged] as BTL. In order for us to be able to risk weight Portfolio BTL products correctly at 100%, we would need a flag in the system to help us to identify the PBTL loans in the T24 book.

    Impact: This will result in the understatement of the RWA numbers until such time Metrobank can start to invest resources and extensive staff time to review the loan documents to ascertain what proportion of the current BTL risk weighted at 35% that will need to be reclassified as PBTL at 100% risk weight."

  199. On 24 August 2018, Mr Somers emailed Mr Arden and Mr Donaldson, saying:
  200. "Please find attached a short note that details the RWA impacts on two key changes (CRE risk weights change and PBTL definition refinement) required to bring our RWA calculation into compliance. The impact is a circa £900m increase in RWA across Commercial and PBTL books. This represents a circa £70m increase in T1 capital.

    Mark Stokes has suggested that a meeting to discuss with you next week might be helpful for context. "

  201. Attached to that email was a PowerPoint presentation headed "RWA Calculating Review–August 2018". The first slide set out the Executive Summary, which read (our emphasis):
  202. "In September a paper will be brought to CRPAC to advise ELT [Executive Leadership Team] of inconsistencies in current RWA calculations that will result in a significant increase in RWAs.

    There are two key drivers of the increase (All RW are quoted before potential SME factors):

    1. Assets backed by commercial real-estate are currently in Metro allocated a standardised risk weight of 50%. This is based on a simplistic interpretation of the European CRR rules. Following detailed PRA statements and reviewing BIPRU confirms that the PRA have used their permitted powers of derogation to ensure that relevant assets in the UK backed by commercial Real Estate should receive a 100% RW. This interpretation has been confirmed by a full KPMG review.

    • As a result of this reclassification we estimate that RWAs increase by £640 million (June month end).

    2. PBTL assets are backed by residential properties and currently receive a 35% risk weight. Stratified random sampling however reveals that circa 37% of the balances in the book are actually secured on Multi-Family Dwellings (many leases on a single property), Houses in Multiple Occupation or Student accommodation. These should received a 100% risk weight.

    • As a result of this we estimate that RWAs will increase by a further £269 million (June month end)."

  203. Of the £640m figure, £574m related to CLIP loans. The figures had been worked out by the Credit Risk team who had built a calculator to carry out the exercise. The accuracy of the calculator had been verified by first running the figures using the current 50% risk-weighting before running them again using a 100% risk weight for commercial mortgages, and making other more minor changes.
  204. A later slide in the presentation said that "the PBTL portfolio assessment has been on a sample basis as we do not capture planning use class or the number of holders of short hold tenancies as data items" and that the sample was of 100 out of 733 loans, which had been selected on the basis of:
  205. (1)          loans which were more likely to be "high risk" of being PBTL including where they had product codes referring to "Comm_BL" or "Comm_Mtg"; where they were secured on something other than the property itself; where there was a shared title; and loans of an average property value of over £1m; and

    (2)          where the loan was for more than £1m, as this indicated that a large RWA movement could occur.

  206. Mr Arden said in his witness statement that this was the first time he was informed that (our emphasis) "the wrong risk weighting may have been applied to CLIP loans and other commercial assets". However:
  207. (1)          on 17 August Mr MacLean had emailed Mr Arden, saying "we have done further work on the RWAs...the outcome is at the higher end of the impact ranges we have discussed as pretty much every commercial loan that is not PBTL is 100% RWA classified (or 75% if SME)", see § 117 ;

    (2)          when taken to the email in cross-examination, Mr Arden reluctantly conceded that he had previously "had some passing discussion" about the CLIP loan issue; and

    (3)          he also accepted that no-one had ever said to him that "there was any doubt that the wrong risk-weighting had been applied to CLIP loans".

  208. We thus find as a fact that before 17 August 2018, Mr Arden was aware that there "was" an issue with the risk-weighting of the CLIP loans, not that there "may" have been an issue.
  209. Mr Donaldson responded to Mr Somers on 24 August 2018, agreeing to his suggestion of a meeting, and adding "we need to consider how we can minimise/offset impact as much as possible".
  210. The meeting to discuss RWAs which Mr Somers had suggested took place in the morning of 3 September 2018. Mr Arden attended, along with Mr Stokes, Mr Somers, Ms Gillan and Mr MacLean, and a new version of the PowerPoint presentation previously produced by the Credit Risk team was discussed. The two versions were essentially identical, other than that the Executive Summary now included the words: "There is limited potential mitigation in trading book assets with residential property security but this would need case by case review". In addition, an extra slide headed "decisions required" had been added. Those decisions included the following:
  211. "A decision to recognise the section of corporate loans that will be moving to 100% RWA and to release this in September returns and in the forthcoming ICAAP. We believe the PRA are expecting us to announce this and we risk significant regulatory scrutiny if we fail to act...

    A decision on the PBTL book (following from the sampling exercise that defined roughly 37% as arguably not PBTL) to either recognise that portion of the book at 100% RWA or delay (but accept that there is ultimately going to be a requirement under IRB to absorb this capital hit...

    In hope of shifting some commercial trading loans currently RWA 100% to 35% there will need to be a sampling of the commercial loans portfolio to determine what proportion may be secured on purely residential properties."

  212. The contemporaneous minutes record that
  213. (1)          KPMG had confirmed that CLIP loans should be risk weighted at 100% rather than the 50% currently allocated.

    (2)          There was "potentially a small portion of this book that was actually secured on residential property and had actually been miscoded. Analysis suggests this is likely no greater than £20m of RWA". In other words, potentially a small portion of the loans recorded as CLIP was secured on residential property and so had been incorrectly risk weighted at 50% rather than 35%.

    (3)          Mark Stokes agreed to review a detailed breakdown of exposures to confirm any potential offset effect.

    (4)          Recent PBTL sampling indicated that around 37% of the loan book was secured on a multi-family dwelling, HMO or student accommodation, but the standardised capital calculation approach did not currently require these to be risk-weighted at 100%.

  214. In the afternoon of the same day, Mr Richardson emailed Ms Gillan, saying that the Bank had "failed to spot" that the PRA required CLIP loans to be risk-weighted at 100%.
  215. Communicating with the PRA

  216. On 6 September 2018, Mr Arran Salmon, the head of the PRA's "New Banks 1&2 Teams", together with Mr Sutherland and a Ms Warren, met Mr Gunn. Both the PRA and Mr Gunn produced contemporaneous notes of that meeting. Key points were:
  217. (1)          The PRA said that the miscalculation of the RWAs made them feel "unease". Mr Gunn told the PRA that this "was clearly an error on our part and was being fixed".

    (2)          The PRA told the Bank it "would not be getting a Pillar 2 offset" but this could be revisited "once the RWA issue was fixed".

    (3)          In relation to AIRB, the PRA said that the Bank's application was "more complex/challenging than any other of [its] comparator banks" and they had "some  genuine concern about the depth of [the Bank's] use test experience and also of the depth of experience of [its] modelling team". Although the PRA had not yet carried out detailed testing, they were "uneasy" and as a result "it may take [the Bank] longer to get approval, maybe considerably so".

    (4)          The PRA were also unhappy that the Bank had put their private discussions about Pillar 2 and AIRB into the public arena, and warned that the Bank may be misleading the market by setting out clear positive expectations on both issues.

  218. On 10 September 2018, Ms Melanie Beaman, the Director of UK Deposit Takers at the PRA and a director of the Bank of England, sent Mr Hill the Bank's annual Periodic Summary Meeting ("PSM") letter. PSM letters are confidential summaries of the view taken by the PRA of a financial institution's material risks, and they delineate required mitigating actions. The PSM letter included the following passages about publishing information about confidential discussions between the Bank and the PRA:
  219. "We consider the firm should manage the expectations of external   stakeholders more effectively to avoid capital discussions leading to  reputational issues or management distraction for Metro. This risk was  highlighted by the public announcement in April when you said there would be no expectation to raise further equity in 2018, only to proceed with an equity issuance shortly thereafter.

    We were also somewhat frustrated that discussions we have had with you over IRB model recognition were recently referenced publicly, and  portrayed in a manner which may risk setting false expectations over the  timeframe for achieving IRB model approval. We have not yet reached any  formal decision on your application but we remain concerned about the depth of your relevant modelling experience.

    Similarly, in April 2018 you publically [sic] stated an expectation that Metro would benefit from the Pillar 2A offset approach before we had reviewed your ICAAP. We would not wish to see a repeat of this type of instance."

  220. The PSM letter also referred to RWAs, saying:
  221. "Finally, we understand that you are remediating the classification of commercial risk weights which will likely increase the Pillar 1 risk weighted assets that have previously been reported to us...At this stage we are
    uncertain about the materiality of any prospective adjustment to your capital position. So until this matter is satisfactorily resolved, and we have received reassurance that Metro is holding sufficient Pillar 1 capital against its commercial assets, we will not apply Policy Statement 22/17 which allows the offsetting of certain Pillar 2a variable add-ons."

  222. The reference to PS22/17 was to the PRA's approach to allowing overcapitalisation in the calculation of Pillar 1 requirements to reduce a bank's Pillar 2A requirements using the unders/overs principle, see §16(2)(c).
  223. Appendix 1 set out key actions for the Bank, which included:
  224. "CRO [Chief Risk Officer] to submit the results of the commercial risk weighting exercise and the CFO to provide an attestation on the accuracy of regulatory reporting. This should be supported by an internal audit review of the remediation exercise...

    We have chosen not to apply PS22/17 because we have been unable to confirm the adequacy of the commercial risk weights provided by you during the course of the capital assessment. Once you have completed the remediation project and we have received assurance from Internal Audit that you are correctly reporting risk weights for your commercial book, we will be willing to consider an application to apply PS22/17."

  225. On 11 September 2018, Mr MacLean emailed Mr Arden, saying:
  226. "As per previous discussions, we are now expecting to move ahead with the change of approach on commercial RWAs to include all of our commercial lending secured on commercial property as 100% RWA. Risk have spoken to Deloitte for a second opinion, and they have confirmed our   understanding."

  227. On 12 September 2018, a meeting took place between the PRA (Mr Salmon, Mr  Sutherland and others) and Mr Donaldson, Mr Arden and Ms Gillan. Contemporaneous minutes were taken both by the PRA and by Ms Gillan. Under the heading "Pillar 1 RWA reporting issue", the PRA recorded as follows:
  228. "AS [Mr Salmon] explained that the reporting error had caused agitation on the Panel as it had been perceived as the firm not being able to get the basics right.

    CD [Mr Donaldson] expressed sincere regret and apology for the error. He expects the reporting to be corrected by October/November and believes c£40m assets are in scope for re-classification [2]. He was unsure what this represented in percentage terms. Once the team has worked through the results CD will be seeking external assurance and DA [Mr Arden] will attest as to their accuracy, and the correct numbers will be contained in the new ICAAP.

    CD fully accepted the PRA's decision not to apply the Unders/Overs principle as a result, however he asked if this decision could be revisited before the next Panel. AS agreed to a discussion about this once the mistake had been corrected, noting that we would need to understand how the mistake had occurred and to be assured there is no read-across to other reporting aspects."

  229. In reliance on Ms Gillan's evidence in cross-examination we find that all those present agreed that an error had been made. She said: "no-one was hiding that. Everyone was  agreeing that that was the case."
  230. Under the heading "AIRB" the PRA minutes recorded:
  231. "CD asked what Metro could do to help move the AIRB issue along. AS explained there was no decision yet but the high level feedback so far is that the firm does not have the required level of use testing experience and that the CRR is clear on the requirements.

    AG [Ms Gillan] does not agree regarding the interpretation of the CRR, and CD added that other senior figures at the PRA had been more encouraging about Metro's prospects of getting AIRB.

    GS [Mr Sutherland] advised Metro that there would be an internal Challenge Session and there would be an opportunity for the firm to appeal the decision after that."

    September CRPAC meeting

  232. On 17 September 2018 there was a meeting of the CRPAC; Mr Donaldson, Mr Arden, Ms Gillan, Mr Stokes, Mr Somers and others attended.
  233. KPMG decision trees

  234. The first item on the agenda after the minutes of the previous meeting was "PQR Asset Classification Proposal". A paper circulated in advance was headed "PQR-Paper for Approval. Non-Retail Asset Classification". It was produced by Ms Orrell and Mr Wayne Jackson, who was also from the Credit Risk & Analytics team, and was presented at the meeting by Mr Somers. It began:
  235. "To obtain approval for the decision tree produced by KPMG to determine the correct asset classes, and hence capital treatment, for both the Standardised and IRB approaches for all Commercial assets other than PBTL, which has been reviewed separately.

    BACKGROUND
    Reporting the correct risk weight (identified via the exposure's asset classification) is a mandatory regulatory requirement and drives our 7 year plan and strategy for, amongst other things, lending and deposit growth and future levels of risk weighted assets.

    In June 2017 Internal Audit identified a gap relating to the asset  classification of non-retail exposures resulting in risk weights being wrongly allocated for some non-retail exposures. This was reported to ELT [Executive Leadership Team], Audit Committee and the PRA.

    MATERIALITY
    Non-retail assets account for 34% of our balance sheet (June 2018) and are split between our Professional BTL product and our Commercial Lending product."

  236. In reliance on the minutes of the meeting we find that "the committee accepted the recommendations for the implementation of the decision trees produced by KPMG".
  237. PBTL classification

  238. The second agenda item was "Professional Buy-To-Let Classification". A paper had similarly been circulated in advance, which included the following:
  239. (1)          Finance have been applying a 35% risk weighting to all PBTL portfolios  on T24 and Pepper, and also those acquired in the "London" and "Canberra" deals.

    (2)          The PRA's guidance states that 35% is required for owner occupied residential property and "certain Buy to Let" properties.

    (3)          There was no further explanation of what was meant by "certain BTL" but "while there is nothing to stop us including for example multifamily or HMO properties, we have good reason to believe that this is not within the spirit of what the PRA intended".

    (4)          Based on the recent sampling exercise, the best estimate was that 35% of the current PBTL book should be risk-weighted at 100%.

  240. The decisions required were summarised as:
  241. (1)          New business would be stratified to separate out residential PBTL at 35% and other PBTL at 100%.

    (2)          Resources were needed to capture additional data to enable accurate classification and modelling of eligible PBTL prior to IRB Application to PRA.

    (3)          Existing ineligible PBTL loans to retain 35% risk weighting, and to be "managed down" so that by the time the PRA approve the application for IRB to be applied to the commercial loan book, these loans would not be material.

  242. During the meeting, Mr Somers explained the background and the decisions required. He emphasised "the need to identify what assets are eligible to be retail or otherwise" and agreed to consider whether to carry out a "95% or 99% sample".
  243. Annual Review of Commercial Lending

  244. Another agenda item was the Annual Review of Commercial Lending. A pack of information was provided in advance, which included the following (all figures related to the June month end):
  245. (1)          The Bank's commercial lending portfolio totalled £2,379m, being 20% of the Bank's total lending portfolio. Of that figure, 42% were CRE loans, 38% were "trading business loans including commercial mortgages" and the balance was other loans.

    (2)          In addition, the PBTL book was £1,001m, and "commercial other" was £432m made up of Canberra commercial, London commercial and T24 commercial.

    (3)          The executive leadership team had been advised in September that there were  "inconsistencies in current RWA calculations that will result in a significant increase in risk weightings", because commercial mortgages had been weighted at 50% rather than 100%, and that KPMG had confirmed that this 100% was correct.

    (4)          As a result of the reclassification, risk weightings were estimated to increase by £640m, before allowing for any SME factor.

  246. The minutes do not refer to these parts of the information pack, and in reliance on Mr Somers' unchallenged evidence, we find that the quantum of the RWA adjustment for CLIP loans was not discussed at this meeting.
  247. September Audit Committee

  248. The Audit Committee met on the morning of 18 September 2018, the day after the CRPAC meeting. It was chaired by Mr Bernau, and three other members of the Board attended: Mr Carby, Mr Snyder and Mr Lockhart, while Mr Donaldson, Mr Arden, Ms Gillan and others attended as guests.
  249. The minutes included the following passage:
  250. "[Ms Gillan] gave the Committee a summary of the errors that were identified on the 2017 COREP (Common Reporting) returns. Inconsistencies in the reporting of commercial RWAs, specifically on commercial lending assets secured on property had been identified. Originally, the Capital Requirements Regulation had stipulated a risk weighting of 50% for these assets. The European Banking Authority Guidance had set discretion for individual national authorities to derogate from this position depending on their view of lending assets. The PRA applied a risk weight of  100%, it had not however published guidance or default data to confirm this and the required limits were therefore ambiguous. As a result the Bank was investigating revising the risk weights upwards from 50% to 100%. The error in reporting had been made based on ambiguous guidelines. There had been similar issues across the industry. The Bank had taken what it believed to be the correct approach at the time, but accepted that the error should be remediated and was working towards this."

    September NEDs meeting

  251. The NEDs met the same morning, and were briefed by Mr Gunn on his meeting with the PRA and on the PSM letter. The minutes record that the PRA had commented on "the fact that we miscalculated RWA in the commercial book" and Mr Lockhart stated that "the RWA error amounted to a £40m capital amount".
  252. September Board meeting

  253. The September Board meeting took place in the afternoon of the same day. Board members had been provided with various papers around five days in advance, including the Chief Risk Officer's report, which contained the following passage:
  254. "Standardised RWAs vs plan has exceeded appetite since March owing primarily to the reclassification of CRE assets, the review of which is still ongoing. This will remain in red until the plan is rebased to take into account the higher risk weights. This metric could increase by circa 10% (c.£40m of capital) upon completion of the asset classification project."

  255. The minutes do not record any discussion of this passage and we find on the balance of probabilities that it was not discussed.
  256. Another item was the PSM letter, also circulated in advance. The minutes record that this was "noted" and that "the classification of commercial risk weights" had been discussed, with Ms Gillan explaining the background. The minutes continued by saying (where DA is Mr Arden, CD is Mr Donaldson, and the ROC is the Risk Oversight Committee).
  257. "DA then outlined the process for remediation and confirmed that this will be brought back to ROC and Board in due course. CD explained that once fully remediated, the PRA had confirmed that they would be able to reconsider the application of Pillar 2 Offset."

  258. The Board then delegated oversight of the RWA issues to the ROC. The minutes also record that a new NED, Catherine Brown, was appointed with effect from 1 October 2018.
  259. Engagement of Deloitte

  260. At the end of September 2018, the Bank opened discussions with Deloitte for it to assist the Bank with responding to the requirement in the PSM letter that Mr Arden provide an attestation as to the accuracy of the RWAs and of the regulatory reporting. The scope of the project was set out in Deloitte's draft engagement letter dated 4 October 2018; this was finalised in the same terms on 16 October 2018:
  261. "We understand that you have initiated the Project in response to some self-identified issues around your systems and controls for the calculation of Credit Risk Weighted Assets ("Credit RWA") and COREP reporting.

    The Project objectives are to review and remediate your current policies, procedures and controls in this area. Initially, this will involve gathering and reviewing current policies and associated documents, and working with your Regulatory Reporting, Credit Risk and other relevant teams to assess your current practice.

    The Project will identify any gaps relative to industry standards and regulatory expectations, and seek to close out these gaps with 'quick win' remediation where possible.

    The Project will then set out a road map/action plan for further strategic remediation where appropriate."

  262. The detailed specification said that Deloitte would "review" both the Bank's policies and procedures relating to RWAs and COREP, and also "review known data issues"; and that "through workshops or discussions with the relevant teams" Deloitte would "document an assessment of the quality of each data item". In reliance on this letter, together with Ms Gillan's evidence, we find that Deloitte was not engaged to check the underlying data on a loan-by-loan basis or to provide corrected risk-weighted figures. The work was to be conducted in three phases spread over 9-10 weeks in total.
  263. Internal work in support

  264. Around the same time, work began within the Bank to support the Deloitte review; this was a joint project involving the risk team, the finance team and the commercial lending team.
  265. On the basis of Ms Gillan's evidence, we find that the issue which was being addressed was whether the loans had been correctly classified for regulatory reporting purposes; the teams were not checking the quantum of the loans; the identity of the borrowers, or the security on which the loans had been made. She described the work as follows:
  266. "[They] were tasked with reviewing individual loans to, firstly, understand which classification had been applied, but then applying the decision tree to ensure that the correct classification had, in fact -- so they were comparing  what was on the system versus having applied the decision tree, what
    classification should have been applied."

  267. In order to carry out that exercise, the teams therefore applied the KPMG decision trees, and, as we have already found, one of the factors in those decision trees was whether the SME supporting factor applied, see §94.
  268. Communications with the PRA

  269. On 4 October 2018, the day after the Bank received Deloitte's draft engagement letter, Mr Arden emailed Mr Sutherland as follows:
  270. "We continue to make good progress on remediating the classification of commercial risk weights in our lending portfolio. We are currently in the process of engaging an expert third party to provide external assurance over work completed to date and provide a full review of our RWA calculations
    and COREP reporting. This work will be in three distinct phases and take a number of weeks.

    As you can appreciate, and as was made clear by you in our meeting, Aileen [Gillan] and I need to be sure on the efficacy and accuracy of the results of this exercise before we submit any changes. As a consequence, COREP reporting for September will be materially unchanged, save for any business as usual movements. I trust this is in line with your expectations, though please do not hesitate to let me know if you have any questions."

  271. On 9 October 2018, Mr Sutherland responded, saying "That's fine David, thank you for confirming". His unchallenged witness evidence was that he "viewed it as a pragmatic solution to the issue of COREP submissions which required the input of specific figures when Metro Bank was in the process of conducting a remediation of its Pillar 1 commercial risk weightings and its capital requirements".
  272. Meeting with Linklaters

  273. On 5 October 2018, Mr Arden and Ms Roberts met with Linklaters, following which Ms Roberts sent an email to Mr Arden. The scope and outcomes of that meeting were disputed, and we make related findings of fact later in our decision, see §291ff.
  274. Disclosure Committee meeting

  275. The October meeting of the Disclosure Committee took place on 16 October 2018 attended by Mr Arden, Ms James and Ms Roberts. Mr Donaldson sent his apologies. The minutes record that:
  276. "A problem had been identified with the risk weight classification of some
    commercial assets; that current estimates based on sampling was c.£600m and the impact on core equity capital of c.£50m but further work was being undertaken (with the help of Deloitte) to finalise the amount; the PRA had agreed that no immediate changes necessary for the Coreg [sic] reporting; and the intention was to notify the market once finalised and resolved in line with our usual full year and Pillar 3 disclosures..

    Linklaters concurred with the view that it is neither specific nor material information at this point and was in the ordinary course of ongoing dialogue with the regulator over a complex issue and no market announcement was necessary at this point.

    The Committee understood that, once fixed, further consideration would nee dot [need to] be given to whether a market announcement was required."

  277. The minutes record the "Action" as being "Continue to monitor". Mr Jaffey noted that the committee's Terms of Reference state that part of its role was to ensure that the Bank's announcements complied with LR 1.3.3R, see §89. He invited the Tribunal to find as a fact that the committee had also considered a draft of the Q3 Update, saying:
  278. "It was put through this committee and shown to the general counsel for approval precisely to ensure that there was compliance with, amongst other
    things, the listing rules."

  279. We decline to make that finding, for the following reasons:
  280. (1)          The committee's Terms of Reference also require members to be provided, at least five days in advance, with the agenda items to be discussed, and no copy of the Q3 Update was so provided.

    (2)          Although the Terms of Reference say that meetings can also be convened "without formal papers", this was not such a meeting.

    (3)          The Bundle contained both:

    (a)           a draft of the minutes, prepared in advance, which set out each of the agenda items and a provisional view of the conclusions; and

    (b)         the final version of the minutes.

    (4)          There are differences between the two documents, and had the Q3 Update been introduced as an agenda item at the last minute, it is not credible that it would have been omitted from the final version of the minutes.

    (5)          Although Mr Arden's evidence was that "there would definitely be a draft of the Q3 in existence at this stage", he also agreed in cross-examination that "the minutes don't suggest" that it was considered at this meeting and he had no memory of it being discussed.

    Mr Somers' email

  281. On 19 October 2018, Mr Somers sent an email to Ms Gillan and others, headed "Information to Deloitte". It read (our emphasis):
  282. "Deloitte have asked Olivier [Baixas] to provide information regarding weaknesses in rwa and provision calculation process. He has a long list (ltv calculations, cross collateral, connected parties and others)..as far as I know all of which will prove non-material (the biggies I have highlighted), but to be fair I can't be sure as there are significant gaps in data recording and you can't prove what you can't measure. There is however only so much reporting change you can adopt at any one time so haven't driven these hard as believe would be more disruptive than constructive.

    There is the risk this will create more noise than light so you may need to pre warn David Arden? I have asked Olivier for a copy of anything he sends to Deloitte. If you want to review what he is sending in advance then can also do that."

  283. Mr Somers' evidence was that he was here focusing on the provisioning, because issues such as loan-to-value calculations, cross collateral and connected parties are not relevant to risk-weighting. This evidence was challenged by Mr Jaffey, but rebutted by Mr Somers on the basis that the recipients of the email would have been aware from the context that "loan value and those aspects don't affect commercial mortgages", although it could have some effect on PBTL.  
  284. Mr Pritchard subsequently asked Mr Dransfield to agree that loan-to-value "could be relevant to the RWA calculation [for] PBTL", but Mr Dransfield said that this was only the case for residential properties. In closing, Mr Jaffey said Mr Somers' evidence had therefore been shown to be "not correct", and he supported this by reference to Article 125(2)(d) of the CRR. However:
  285. (1)          Mr Dransfield did not disagree with Mr Somers: both said that loan-to-value did not affect commercial mortgages; and

    (2)          Article 125 of the CRR is headed "Exposures fully and completely secured by  mortgages on residential property". It thus relates to residential mortgages, not to  commercial mortgages, as Mr Dransfield said was the case. 

  286. Having considered all the above, we accept Mr Somers' evidence that in this email he was referring to provisioning, not risk-weighting.
  287. Meetings with Mr Hill and Mr Bernau

  288. On or around 22 October 2018, Mr Hill met with Mr Donaldson and Mr Arden to discuss what Mr Arden described in his witness evidence as "the estimated RWA correction". His unchallenged evidence was as follows:
  289. "I explained to Mr Hill that a potential £600m in RWAs was equal instead to ca. £40m in regulatory capital. Mr Hill then asked how much total regulatory capital we held, to which I told him it was around £1bn. Once I explained this, Mr Hill became noticeably less concerned and laughed. He told us not to use the RWA number when the most important number to the Bank was the potential impact on capital."

  290. Mr Donaldson's evidence (also unchallenged) about the same meeting was that:
  291. "[Mr Hill] did not think the potential impact on capital was particularly significant. He just told us to fix it. He also emphasised that the potential impact on capital (as opposed to the potential impact on RWAs) was the important number, and that this was what should be focussed on during  meetings with the Board."

  292. On 22 October 2018, Ms Gillan had her monthly one-to-one meeting with Mr Bernau, during which she briefed him on her papers for the meetings of the ROC, Audit Committee and the Board, all of which were to happen the following day, and her briefing included "discussion on the RWA issue".
  293. The October CRPAC meeting

  294. The next CRPAC meeting took place on 22 October 2018. Mr Donaldson and Mr Arden both attended, along with Ms Gillan. Mr Dransfield and Ms Orrell attended as guests, together with other members of the Bank's staff.
  295. RWA Report

  296. The first item on the agenda (after agreeing the minutes of the previous meeting) was "RWA Reclassification": the related paper from the Credit Risk and Analytics team had been circulated in advance. It was headed "Change in Risk Weighted Assets - October 2018" ("the RWA Report"). The Executive Summary began as follows:
  297. "►This paper is to advise ELT (via PQR/CRPAC) of errors in the current RWA calculations for commercial exposures and gives an indication of how the key risk parameters will drive future RWA under IRB

    ►The correction to Standardised RWAs, primarily for commercial mortgages, leads to a significant increase in RWA of £642 million.

    Commercial Mortgages - 50% to 100%

    £574m

    PBTL ineligible back book on T24

    £37m

    Business Loans ineligible as retail

    £22m

    Unsecured retail loans with collateral

    £9m

    ►Finance and Credit Risk and Analytics will continue to review RWA calculations as data and processes improve, but any further adjustments to calculations are not expected to be material."

  298. The PBTL figure of £37m was less than the £269m in the paper provided for the meeting in September, because the latter was based on both the T24 book and the Pepper book, and the former excluded the PBTL in the Pepper book. This was because the Bank understood the regulations only required the higher risk-weighting to be applied for Pillar 2A purposes: in relation to the standardised basis, the PRA had stated that 35% is required for owner occupied residential property and "certain Buy to Let" properties, and there was no guidance on what was meant by "certain" BTL loans, see the earlier discussion at §§143(2)-143(3). In consequence, as Mr Somers put it, the Bank could decide whether it "wanted to take that hit now or in the future", ie when a Pillar 2A offset was permitted. The Bank therefore decided to increase the risk-weighting for only some of the PBTL loans.
  299. That slide was followed by another giving more detail which read (emphasis and colour in original):
  300. "Following the decision tree work with KPMG, errors have been found in the current RWA calculations that will result in a significant increase in RWs.

    ►The material driver of the increase is the change in RWs applied to commercial mortgage loans.

    ►Commercial mortgages are loans to trading business companies to purchase commercial property (such as their own offices or factories) where repayment is from the trading business company's cashflow and not from a third party rental income.

    ►We are currently applying a RW of 50% whereas we should be applying a RW of 100%.

    ►Applying a RW of 100% results in additional RWAs of £574 million.

    ►Other drivers include:

    ►Some loans were assumed to be unsecured retail at a RW of 75%, but were actually secured by commercial property and hence a RW of 100%, resulting in additional RWAs of £9 million.

    ►The loans badged retail within T24 erroneously included business loans, which have now been identified and moved to the correct category, resulting in additional RWAs of £22 million.

    ►It has been agreed that back book PBTL in T24 that is not eligible for the Retail Exposure Class (HMO, Multi-family, etc) will have a RW of 100%, resulting in additional RWAs of £37 million.

    ►TOTAL IMPACT is an increase in RWAs of £642 million

    ►All impacts are as at end September 2018, and are net of the SME supporting factor where applicable."

  301. Subsequent slides included the following statements:
  302. (1)          At the end of May 2018 it came to light (via the KPMG Asset Classification project) that the Bank was not using the correct Standardised RW for commercial mortgages.

    (2)          The Bank had "wrongly assumed" that the PRA had not used the discretion given by the CRR to increase the risk weighting from 50%, and the same mistake was made by "some other challenger banks".

    (3)          Clause 4.1 of the Credit Risk section of the PRA Rulebook (see §25) only applied if the Bank could demonstrate that loss levels did not exceed 0.5%.

    (4)           KPMG had confirmed that "they were not aware of any UK regulated firm [that] has demonstrated loss levels to be less than 0.50% over the relevant time horizon".

  303. The following slide set out a detailed "impact table" analysing the numbers which made up the overall adjustment of £641m; we have called this the "Impact" slide. In relation to the RWA issue, the key figures on this slide are those relating to the changes between:
  304. (1)           the existing figures after the SME supporting factor; and

    (2)          the new figures after risk-weighting at 100% and adjusting for the SME supporting factor.

  305. The figures for the T24 book were as follows, all values in pounds.
  306. Asset

    No of accounts

    Total Loan

    Current RW

    after SME

    RW new at 100%

    RW new after SME

    Commercial (other)

    983

    1,029,173,140

    466,157,409

    1,029,173,140

    932,314,810

    Commercial mixed collateral

    144

    155,689,682

    65,765.701

    155,689,682

    131,531.381

    Commercial (OO)

    116

    91,488,458

    45,364,806

    91,488,458

    82,105,052

    Totals

     

    1,276,351,280

     

    577,287,916

    1,276,351,280

    1,145,951,243

  307. The increase as the result of the new risk-weighting of CLIP loans in the T24 book was thus £568,663,327 (1,145,951,243 less 577,287,916).
  308. The figures for the London book were as follows:
  309. Asset

    No of accounts

    Total Loan

    Current RW

    after SME

    RW new at 100%

    RW new after SME

    Corporate commercial

    2

    987,354

    654,854

     

     

    Retail commercial

    48

    10,888,777

    6,165,029

     

     

    Totals

    50

    11,876,131

    6,819,883

    11,876,131

    11,713,820

  310. The increase as the result of the new risk-weighting for CLIP loans in the London book was thus £4,893,937 (11,713, 820 less 6,819,883). When added to the figures from T24, the extra risk-weighting required for CLIP loans was £573,557,264; this was the £574m shown on the earlier summary slide.
  311. The analysis on the Impact slide also showed that there were 453 PBTL loan accounts in the T24 book, totalling £162,562,629; these were split into those which had been incorrectly weighted and those which did not need correction. The risk-weighting of a further £1,011,958,111 of PBTL loans held in "Pepper" was not amended.
  312. The minutes of the meeting record that Mr Dransfield presented this paper, explaining that the reclassification was "due to misinterpretation of the PRA rulebook" and that "the material driver" was:
  313. "reclassification of commercial mortgage loans which accounts for £572m [3] of the increase in RWAs. The rest is unsecured reclassified as commercial secured, retail in T$ [T24] which are actually business loans and back book non retail PBTL which is now classed with a RW of 100%."

  314. Ms Gillan's evidence was that no-one at the meeting dissented from that proposition or said that the figures were wrong; she agreed with Mr Stanley that the figures were understood "to be estimates, but they were best estimates [they] had at the time". Mr Dransfield's evidence was that there was no discussion of the paper, other than a question about the probability of borrowers defaulting. Mr Arden's evidence was not straightforward, see §44, but he also confirmed that no-one had disagreed, as did Mr Donaldson. We find as a fact that no-one at the CRPAC meeting dissented from the position set out in the RWA Reclassification Paper.
  315. Business and Commercial Lending

  316. The ninth item on the agenda for the CRPAC meeting was the annual review of commercial lending. Responsibility for this item sat with Mr Somers, but in his absence, Mr Dransfield took responsibility. The related slide pack consisted of 40 slides. The only reference to RWAs in the Executive Summary was in this passage:
  317. "We have experienced some issues with asset classification for regulatory reporting of RWAs. We understand that this is not unique to Metro Bank, however, signals the need for our systems, data, reporting and risk infrastructure to be robust and consistent with what is expected by Regulators and external stakeholders of a Tier 2 UK Bank.

    ►This will be resolved over a two year period when commercial IRB models (Asset Classification, Slotting and RiskCalc) are finalised, incorporated into an appropriate IT system and made mandatory for new applications and annual reviews."

  318. Nearly at the end of the slide pack (number 37 of 40) was one headed "Overview of Out of Scope Portfolios" ("the Out-of-Scope slide"), on which Mr Jaffey placed significant weight. It stated that the related portfolios comprised PBTL of £1,001m and "Commercial-other 432m", and the latter consisted of:
  319. (1)          Canberra Commercial of £176m, made up of 1,707 loans with an average loan size of £104k;

    (2)          London Commercial of £103m, made up of 763 loans with an average loan size of £135k; and

    (3)          T24 Commercial Mortgages of £153m, made up of 467 loans with an average loan size of £327k.

  320. The same slide also said that the Canberra Commercial and London Commercial were "originally badged as commercial as the borrowing entity is a limited company" but "are being treated as retail as the credit risk is reflective of a retail BTL" and that the T24  Commercial Mortgages were "being treated as retail for the same reasons as the Canberra and London Commercial portfolio".
  321. Mr Jaffey submitted in closing that the £432m of "other commercial" loans on the Out-of-Scope slide were part of the £1,029,173,140 of "other commercial" figure included in the Impact slide. In his words:
  322. "So what this slide is really saying overall is there is 432 million of commercial lending which isn't really traditional commercial lending, and it is comprised of three things, and they are all, really, retail buy-to-let conducted through companies...and, of course...they are not ordinary commercial loans. You can't risk-weight them and assess them and work out their credit risk on the usual basis as with commercial loans. The reality is that it is just retail buy-to-let."

  323. The import of his submission was that:
  324. (1)          £432m of loans identified as CLIP should have been classified as residential and so risk-weighted at only 35%;

    (2)          £432m was a significant figure, so the overall adjustment for CLIP loans would have been much less than the £574m in the RWA Report; and

    (3)          this information had been provided to the CRPAC at the time of the Q3 Update.

  325. We disagree with Mr Jaffey for the following reasons:
  326. (1)          The Impact slide in the RWA Report, and the Out-of-Scope slide in Business and Commercial Lending report were both prepared by the same Credit Risk team, led by Mr Somers. If £432m of the "Commercial - other" shown on the Impact slide were retail loans, the Credit Risk team would have known that, and would not have risk-weighted them as commercial.

    (2)          The Out-of-Scope slide says the opposite: the Canberra and London loans were "originally badged as commercial" but "are now being treated as retail".

    (3)          It is clear from the Out-of-Scope slide that the Canberra and London books are not held within T24, while the £1,029,173,140 recategorised on the Impact slide is all within T24.

    (4)          There were 983 loan accounts included in the "Commercial Other" line of the  Impact slide, but 2,973 on the Out-of-Scope slide. As Mr Stanley submitted, there is "simply not room" for those 2,973 loans to be part of the 983 "Commercial Other" recategorised loans on the Impact slide.

    (5)          None of the witnesses referred to the Out-of-Scope slide in their witness statements, and no questions were asked by Mr Jaffey about whether there was a link between it and the Impact slide, even though Mr Somers, Mr Dransfield and Mr Arden in particular could have been expected to give highly relevant evidence about that matter.

  327. We also note that Mr Jaffey did not apply to the Tribunal for any witness to be recalled, despite (a) Mr Arden and Mr Donaldson being present in the hearing room, and (b) there being time before the end of the listed hearing to accommodate the recall of a witness.
  328. In short, the supposed link between these two slides was a theory put forward by Mr Jaffey at the last minute. It had no supporting witness evidence, and was inconsistent with the information set out on the face of both slides. We find as a fact that the £432m of loans on the Out-of-Scope slide does not change the RWA adjustment required for CLIP loans of £574m shown on the Impact Slide. 
  329. The October Audit Committee meeting

  330. The Audit Committee met at 10am on 23 October 2018. A meeting pack had been circulated in advance. The meeting was chaired by Mr Bernau, and attended by Mr Carby, Mr Snyder, and Mr Lockhart, while Mr Donaldson, Mr Arden, Ms Gillan, Mr MacLean and Ms Alice Petterson, the Bank's financial controller, all attended as guests, along with other Bank employees.
  331. The Q3 Update

  332. The papers provided in advance included a draft of the Q3 Update, which contained the paragraph on which this case turns ("the Paragraph"). It read (emphasis in original):
  333. "Capital ratios remain robust. Common Equity Tier 1 Capital ("CET1") of £1,164m as at 30 September 2018 is 15.7% as a percentage of risk weighted assets, currently exceeding our Tier 1 regulatory minimum of 9.7%. This was supported by the completion of a £300m equity raise in July. Risk weighted assets at 30 September 2018 were £7,398m. The Regulatory Leverage ratio is 5.7%. Our total capital as a percentage of risk weighted assets is 19.1%."   

  334. The Paragraph thus began in the same way as the similar passage in the half-year update, see §111, by saying that capital ratios remained robust, but going on to give a higher figure for CET1 as a percentage of RWAs (15.7% compared to 12.7%).
  335. In addition to the Paragraph, the draft Q3 Update also said that "total net loans as of 30 September were £13,121m" and that loans to commercial customers were 32% of that figure, or £4,166m.
  336. The minutes of the meeting record that the Q3 Update was "noted and taken as read". Various points were discussed, but these did not include the RWAs or the Bank's capital ratios.
  337. Accounting, reporting and control report

  338. The papers for the Audit Committee meeting also included a "Q3 Update on accounting, reporting and control matters" authored by Ms Petterson. The opening paragraph said:
  339. "The purpose of this paper is to provide an update on accounting, reporting and control matters that will impact on Metro Bank's 2018 year end reporting process, and to summarise other significant matters that we would like to bring to the Committee's attention.

    Within the paper, we cover the following items:

    -  Update on new accounting standards...

    - Update on control matters: update on supplier payment practices, implementation of core asset finance system, and reporting of Risk Weighted Assets.

    -  Update on future accounting considerations:...".

    We note that our Q3 Trading Update has been prepared on a consistent basis to previous quarterly trading updates and there are no significant accounting or reporting matters to bring to the Committee's attention."  

  340. Under the heading "update on control matters" was the following passage:
  341. "Reporting on Risk Weighted Assets:

    - Following discussions at the September Audit Committee and Board meetings, we have commenced a piece of work to review our calculation of Risk Weighted Asset reporting to the regulator.

    - The work, supported by Deloitte, will review our Policies and Rule Interpretation, Data Quality, and Processes and Controls, and will create a remediation plan to resolve both short and long term issues. We expect this  plan to impact across different areas of the bank.

    - We expect to complete this work by year end, and we will update the Committee on any corrections which will be reported to the regulator and any revisions required to our reporting methodologies."

  342. The passage therefore did not give a figure for the estimated quantum of the RWA errors. The minutes record as follows (our emphasis):
  343. "Following discussions at the September Audit Committee and Board meetings, the team had commenced the work to review the calculation of Risk Weighted Asset reporting to the regulator.

    The work, supported by Deloitte, would cover policies and rule Interpretation, data quality, and processes and controls. A remediation plan to resolve both short and long term issues would be implemented. It was expected that the work would be complete by year end, the Committee would be updated on any corrections required to be reported to the regulator and any revisions required to reporting methodologies. AG reported that a separate piece of work was being carried out by the Risk function to ensure that the systems and controls for regulatory reporting were fit for purpose. The impact of incorrect reporting on the balance sheet was not insignificant and the Committee reminded Management of the importance of ensuring that this was remediated fully."

  344. Ms Gillan was asked in cross-examination about the reference to "any corrections required" and agreed that she had understood that corrections of the order of £570m were likely to be required to commercial mortgages, based on the best estimates at the time. She also confirmed that the reference to the "separate piece of work" was to that being carried out by her team and finance to support Deloitte, see §157.
  345. The October ROC meeting

  346. The October ROC meeting was held immediately after the Audit Committee meeting and there was a significant overlap in attendees. The chair, Mr Lockhart together with Mr Snyder, Mr Bernau, Mr Donaldson, Mr Arden and Ms Gillan were at both meetings. Mr Gunn and Ms Melis attended the ROC meeting, along with other Bank employees. Since ROC meetings were held quarterly, the last one had taken place in July 2018.
  347. Chief Risk Officer's Report

  348. The third item on the agenda, after the minutes of the July meeting and matters arising, was the Chief Risk Officer's report for October. This had been circulated in advance. Its summary page, under "Credit Risk" said that "standardised RWAs continue to be above threshold".
  349. The related slide headed "Credit risk–bank wide" showed standardised RWAs in amber for September and July 2018, and in red for August. This was explained by the following text, which had been updated from that in the September paper, see §151:
  350. "Standardised risk weighted assets versus plan has exceeded appetite since March owing primarily to the reclassification of CRE assets, the review of which is still on-going. This will remain in amber until the plan is rebased to take account of higher risk weights. This metric could increase by around 10 per cent (circa 40 million of capital) upon completion of the asset classification project."

  351. In reliance on Mr Arden's and Ms Gillan's evidence, we find that the first sentence relates to the previous CRE issue and final sentence relates to the correction of the risk-weighting for commercial mortgages from 50% to 100%, and that it was the latter which would cause the metric to remain in amber until that issue was rectified. As Ms Gillan said, the £40m of capital was "in the ballpark" for the reclassification of £500-£600m of CLIP mortgages as the result of their incorrect risk-weighting. We agree with Mr Stanley that two different issues had been unhelpfully conflated in this passage.
  352. Although there was a discussion at the meeting about other elements of the Chief Risk Officer's Report, there are no minuted comments about this item.
  353. The RWA Report

  354. The next topic was RWA reclassification. The RWA Report [4] which had been put before the CRPAC (see §174) had been circulated in advance. Mr Dransfield presented this paper, as he had done to the CRPAC the day before.
  355. The RWA Report thus included the slides which said that RWAs were expected to increase by an estimated £642m including £574m relating to CLIP loans, and that "any further adjustments to calculations are not expected to be material". Ms Gillan confirmed in cross-examination that both those statements reflected what she thought at the time.
  356. The minutes record the following: 
  357. "As reported to the previous meeting, the bank had identified that some commercial assets were receiving the wrong risk weighting, which had been driven by a lack of clarity and conflicting rules in the regulatory publications. The bank is now remediating this issue. The issue had affected many of the bank's peers for whom this was a wider issue. The issue had been identified in the COREP audit in 2017 and subsequently picked up by the PRA as part of the previous ICAAP submission. The management team was carrying out a separate piece of assurance work to remediate the issues and declare the quantum of the error in the next ICAAP submission. There had also been a number of minor issues. However, the commercial business was now largely correctly risk weighted. Although it would not completely mitigate against these issues, the move to AIRB would help to address this..."

  358. Two of the statements in that passage were plainly incorrect:
  359. (1)          The use of incorrect risk weightings had not been "driven by a lack of clarity and conflicting rules in the regulatory publications". It had been caused by the Bank's staff failing to identify and apply the rules which applied to UK banks.

    (2)          The commercial business was not "now largely correctly risk weighted" because, as Mr Arden agreed in cross-examination, the Bank had not yet "taken the risk-weightings through" into the CLIP numbers, and no adjustment had been made for them.

    Business and Commercial Lending Review

  360. The ROC also tabled the same report on business and commercial lending as had been circulated to the CRPAC, see §186. Although commercial lending was discussed at the meeting, there was no mention of RWAs; in particular, there was no reference to the slide which stated that as a result of the reclassification, RWAs were expected to increase by £568 million as at the September month end.
  361. The October Board meeting

  362. The Board meeting took place in the afternoon of the same day. The papers were circulated in advance, and included: Linklaters's Governance Update; the draft Q3 Update; the 2019 Budget Paper; the Chief Risk Officer's Report and the draft response to the PSM letter. No copy of Ms Roberts' email about the meeting with Linklaters was circulated (we consider this further at §§ 291ff) and neither was the exchange of emails between Mr Arden and Mr Sutherland about the COREP reporting, see §§160-161.
  363. The meeting was chaired by Mr Hill, and was attended by all the NEDs as well as by Mr Donaldson and Mr Arden. Various employees attended for particular items, as indicated where relevant below. The next following paragraphs follow the same order as the issues were discussed by the Board.
  364. Linklaters Governance Update

  365. The Governance Update was presented by Mr Sadler and Ms Beidas of Linklaters. They left the meeting when the item was concluded. There was no reference in the presentation or in the minutes to RWAs or to the earlier meeting with Mr Lane, and we find as a fact that neither RWAs nor the advice given at the meeting with Mr Lane was discussed.
  366. Audit Committee Update

  367. Mr Bernau updated the Board on the Audit Committee meeting which had taken place that morning. He told the Board that the Committee "had fully reviewed the Q3 Trading Statement and, subject to minor amendments agreed with Management, he was content to recommend them to the Board for approval".
  368. The Q3 Update

  369.  The text of the draft Q3 Update was the same as that considered by the Audit Committee other than in relation to minor amendments; in particular, the Paragraph was unchanged, other that the figure for the July capital raise had been corrected from £300m to £303m. Mr Arden confirmed in cross-examination that the Q3 Update had not previously been reviewed by Linklaters or shared with the PRA.
  370. There is no record, either in Mr Arden's contemporaneous notes or in the minutes, to the RWA issue having been discussed by the Board in the context of the Q3 update. The Applicants' evidence was as follows:
  371. (1)          Mr Donaldson said he had "pointed out [to the Board] that it was the numbers that were in COREP and the board had already been told that the COREP was not changing until we had concluded the work with Deloitte". When asked by Mr Stanley if the Board had been told that the numbers were incorrect, he said "they were the correct numbers that everybody was aware of and were agreed with the PRA" and that the Board knew "there was an issue to be worked through". He later added that "everybody knows you use the numbers from the COREP in your announcements" and "it was the best number we had".

    (2)          Mr Arden said (our emphasis) that the RWAs "would have been part of the conversation". He thus did not say that the RWAs "were" part of the conversation, but instead used the conditional voice. He also said that the lack of any documentary record of the Board being informed of the RWAs as part of the Q3 Update "doesn't mean it didn't happen".

    (3)          When Mr Arden was interviewed by the Authority, he was asked if any consideration had been given by the Board to "putting any qualifier" around the figure of £7,398m, and he said "the honest answer is no, we didn't consider a qualifier", and he confirmed this in cross-examination.

  372. We find as a fact that there was no discussion at the Board about the RWAs in the context of the Q3 Update because:
  373. (1)          there was no contemporaneous evidence that they were discussed, either in the minutes or in Mr Arden's own notes;

    (2)          Mr Arden did not give evidence that they were discussed;

    (3)          the absence of any discussion is consistent with Mr Donaldson's position that the RWA figures in the Update were correct because they were the same as the figures used in the COREP reporting, and the PRA had agreed with that approach; and

    (4)          the absence of discussion was consistent with the Board failing to consider whether to qualify the figure.

  374. Other points of note in the Q3 Update include the following:
  375. (1)          The following figures for gross loans and advances to customers as at 30 June 2018 were provided:

    Commercial loans

    £3,905m

    Residential mortgages

    £7,889m

    Other

    £259m

    (2)          Under the heading "Outlook", the Update included this sentence: "there is no update to the AIRB application for residential mortgages submitted earlier this year and we still expect an outcome in 2H19".

  376. The Board formally approved the release of the Q3 Update to the market at 7am the following morning.
  377. 2019 Budget Paper

  378. The 2019 Budget Paper was "for discussion", and included a section headed "Capital outlook", which contained this passage:
  379. "Between now and January 2019, a number of items erode the "buffer" between the (decreased) available capital resources and the (increased) regulatory requirement:

    · a further 0.5% increase in the Countercyclical buffer,

    · a 0.63% increase in the Capital Conservation buffer,

    · a review of Commercial lending risk weighted assets, and

    · IFRS16 (effective 1 January 2019).

    [These items] increase the risk that the Bank may need to raise additional capital/debt in Q3 2019 and therefore makes the outcome of the AIRB process and PRA's final response to the Bank's request for a Pillar 2A offset, of high importance to our capital planning."

  380. The paper then said that the above factors:
  381. "...increase the risk that the Bank may need to raise additional capital/debt in Q3 2019 and therefore makes the outcome of the AIRB process and PRA's final response to the Bank's request for a Pillar 2A offset, of high importance to our capital planning. In 2019 Budget we are proposing to include the benefit of Pillar 2A offset (£50m) as being recognised in our ICAAP by the end of Q1 2019, and the approval of AIRB in respect of residential mortgages being received during Q3 2019. We should have early sight of actual dates on AIRB before our full year results announcement." 

  382. The paper was "noted and taken as read by the Board".
  383. Whether the RWA issue was discussed.

  384. The minutes record that Mr Arden had highlighted an increase to capital requirements, but contain no reference to any discussion of the RWA issue. Mr Arden's contemporaneous handwritten notes also make no reference to RWAs being discussed. When asked about this in cross-examination, the exchange went like this:
  385. Mr Stanley: Are you suggesting that you did talk them through that in detail?

    Mr Arden: I gave them the Linklaters advice and the fact that the PRA had agreed that there would be no change to the COREP.

    Mr Stanley: Did you? 

    Mr Arden:  Yes. I think it would be unusual for me not to update the board on such an important element of the Q3.

    Mr Stanley: Very important, you think?        

    Mr Arden:  Essential

    Mr Stanley: Worth minuting, then, if it happened?"    

  386. Mr Arden therefore did not say "yes" initially when asked if he had talked to the Board about the RWA issue. On Mr Stanley repeating the question, Mr Arden followed his initial "yes" with another conditional sentence "it would be unusual for me not to update the Board". He subsequently answered Mr Stanley's question about the lack of any written record by saying that he was taking the minutes at the time and had overlooked this point.
  387. Mr Donaldson's witness statement said he and Mr Arden gave the Board "an update of Linklaters's advice" and said that this was "a critical driver of strategy for the Q3  Announcement".
  388. We find that the Applicants' evidence on this issue is an example of the interference caused to memory by the process of litigation, see Gestmin cited at §38. We instead place reliance on the contemporaneous evidence and find as facts that:
  389. (1)          Mr Arden did not "give" the Board Ms Roberts's email summarising Mr Lane's advice or the email exchanges with Mr Sutherland about the position in relation to the COREP reporting, because:

    (a)          those emails were not circulated in advance; and

    (b)         it is not credible that they would have been provided during the meeting and not recorded in the minutes.

    (2)          Mr Arden and Mr Donaldson did not discuss either Mr Lane's advice or the exchanges with Mr Sutherland, because it is not credible:

    (a)          that they would have done so without circulating the related emails; or

    (b)         that Mr Arden would have overlooked the need to record it in the minutes.

    (3)          Mr Arden and Mr Donaldson made no mention of the RWA issue, again because it is not credible that it would have been omitted from the minutes.

    Chief Risk Officer's Report

  390. The Chief Risk Officer's Report was the same as had previously been considered at the ROC meeting that morning. It thus contained the same brief reference on the summary page to the fact that "standardised RWAs continue to be above threshold", and the same paragraph set out at §204 which conflates the reclassification of CRE assets with the CLIP loan issue, and says that "this metric could increase by around 10 per cent (circa 40 million of capital) upon completion of the asset classification project", but without more explanation.
  391. Ms Gillan joined the meeting to present this paper. In reliance on the minutes we find that she informed the Board that:
  392. "On Credit Risk, the Bank remains within appetite overall, though standardised RWA's continue to be above threshold. Management are presently remediating the classification of commercial risk weights in the lending portfolio. Deloitte have been engaged to provide external assurance over current processes and provide a review of RWA calculations. Further updates will be provided to Board in due course, with detailed findings being taken through ROC."

  393. Mr Arden's contemporaneous hand-written notes say:
  394. "RWAs: standardised RWAs above threshold. Shared with ROC. RWAs up by c £600m."

  395. On the basis of Mr Arden's oral evidence and that given by Ms Gillan, we find that the c£600m was a reference to the adjustment for commercial mortgages (including CLIP loans) set out in the RWA Report, and we also find that the Board was told by Ms Gillan that about £600m was "the current estimate".
  396. Mr Arden agreed in cross-examination that there was no reference in the minutes to the Board being told that the underlying data was unreliable, or that the assistance of Deloitte was needed to correct major problems with the data, and we find as a fact that the Board was not so told at this meeting.
  397. Response to PSM Letter              

  398. The Board also considered a draft response to the PSM Letter. In response to the criticism of the Bank for having referred publicly to its discussions with the PRA about the AIRB model, the draft response said:
  399. "With regard to AIRB, as you know this has a material impact on our capital plans and we are keen to progress our application. Public referencing of the process merely restated the PRA's published guidelines, though again your comments in this regard are noted and accepted."

  400. The letter did not respond to the PRA's similar criticism of the Bank for having stated publicly that there was "an expectation that Metro would benefit from the Pillar 2A offset". In relation to RWAs, the draft response said:
  401. "We are making good progress on remediating the classification of   commercial risk weights in our lending portfolio. We have engaged an expert third party to provide external assurance over work completed to date and provide a full review of our RWA calculations and COREP reporting. As you can appreciate, this work will take a number of weeks before the CFO is able to attest to its efficacy and accuracy. Thereafter, the Board will review the long term capital plan in January 2019, as agreed with our supervision team."

  402. The Board approved the response, with minor amendments, none of which changed either of the passages set out above. The letter was sent on 25 October 2018.
  403. The Q3 Update and analyst calls

  404. The Q3 Update was released on 24 October 2018. It included:
  405. (1)          the figure for commercial loans of £4,166m; this was 32% of the Bank's total net loans including residential;

    (2)          the figure for the CLIP loans of £1,148m, see §208; these thus constituted over 27% of the Bank's commercial loan portfolio.

    (3)          the Paragraph, and in particular:

    (a)          the RWA figure of £7,398 million, based on the application of 50% risk weighting to CLIP Loans;

    (b)         the 15.7% figure for CET1 capital as a percentage of RWAs; and

    (c)          the 19.1% figure for total capital as a percentage of RWAs.

  406.  On the same day, Mr Hill, Mr Donaldson and Mr Arden participated in a call with analysts hosted by Bloomberg. Mr Donaldson said that "capital ratios remain robust with a total capital ratio of 19.1% and a CET1 ratio of 15.7%". In answer to a question, Mr Arden said "we expect AIRB in H2 2019 and we are very comfortable with our capital plans for 2019".
  407. On 31 October 2018, Mr Arden was asked by an analyst at Jefferies to participate in a call with investors, so as to "clarify some of the issues particularly as regards the future capital requirements". He was briefed before the call by Linklaters, as well as by the Bank's brokers, Jefferies and Bank of America Merrill Lynch ("BAML").
  408. The call took place on 2 November 2018. The first part was pre-scripted and included Mr Arden saying:
  409. "As we look forward, our application for AIRB is with the PRA, and our  expectation remains that that will be an H219 event; that provides significant upside to capital efficiency, and on that path, we see no need for an equity  raise until 2021."

  410. The second part was unscripted, and included a question from a Mr Nishil Patel, who asked:
  411. "...the risk weight on the commercial real estate portfolio, if my math is right, it is 60.4%, which just seems low, given where those standardised risk weights should be. Do you mind just helping me understand the disparity there?"

  412. Mr Arden responded:
  413. "I have not got the details to hand, so I will probably get back to you. But just rest assured, we continuously look at all the risk weightings we have, and we are constantly reviewing that. I am afraid I have not got the math to hand."

  414. When cross-examining Mr Arden about that response, Mr Stanley said:
  415. "This is the kind of problem that you get to, don't you, when you have put a statement out a few days before which has been inaccurate, and which you know is inaccurate, and someone is now asking you about it and you don't feel able to give them a straight answer."

  416. Mr Arden responded "Correct". In the course of the interview carried out by the Authority, he said he had subsequently got back to Mr Patel, and told him that "we continually review the details of our lending classifications and associated risk weights", which echoes what he had said during the call.
  417. Deloitte's reports

  418. On 1 November 2018, Deloitte issued its preliminary report, entitled "the RWA calculation and the COREP reporting process". It was based on the first two weeks of work, which involved "meetings with key people across Finance, Risk and Operations, walking through the calculation spreadsheets with the process owner, and reviewing significant aspects of rule interpretation". The "key messages" were as follows:
  419. "► The most significant mis-statement in the RWA calculation is due to the incorrect risk weighting of commercial property at 50% rather than 100%. Impact c. £600m RWA.

    ► There are multiple other data issues leading to mis-statements, but these are unlikely to be of similar size either individually or in aggregate (based on [Deloittes'] experience with other banks).

    ► Notwithstanding the above, currently it is virtually impossible to evidence the integrity of the RWA calculation or the COREP reports. This is because there are multiple gaps in the controls framework at every stage of the process, from data sourcing through to report generation.

    ► The data issues cannot be sustainably mitigated by the regulatory reporting team, and will need fixing upstream at source.

    ► This means that there are currently no firm grounds for providing attestation over COREP reporting.

    ► Inability to evidence controls has been a key concern in the PRA's thematic work around regulatory reporting, and has led to supervisory action against some firms.

    ► Lack of controls over regulatory reporting is likely to come up as an obstacle in the IRB application process."

  420.  The report also said that "there is currently no requirement [at the Bank] for upstream  creators/providers of data sourced for regulatory reporting to attest to the quality of that data" and that one of the problems encountered by the regulatory reporting team was that:
  421. "The entire commercial loans portfolio had to be manually re-classified because lack of source data/inconsistencies in the recording of loan purpose/facility type meant that the data was unfit for regulatory reporting."

  422. Under the heading "current state assessment", Deloitte said that the bank was "potentially non-compliant with the CRR" in a number of areas, including:
  423. "The bank has been incorrectly risk weighting commercial real estate lending at 50% whereas many other banks who had previously adopted this approach have already moved to a 100% risk weight. The Bank has quantified the RWA impact of applying the 100% risk weight at c. £600m."

  424. In relation to the remainder of the project work, Deloitte said that it would "focus on quick wins" but that "the data quality issues - in particular - will require longer term strategic  remediation including system and process changes to how loan data is captured up-front".
  425. On 20 December 2018, Deloitte presented their final report to Mr Arden. An accompanying slide-deck said that a total RWA adjustment of £960m was required, of which £563m related to CLIP Loans, £312m to PBTL and the balance of around £25m was other adjustments, and concluding:
  426. "it is estimated that, in aggregate, the issues identified in the RWA calculation will increase RWA by £0.9-1.0bn, and an increase in capital requirements of c.£100m, at a target capital ratio of 12.6%."

    Discussions with Linklaters

  427. On 8 January 2019, Ms Gillan emailed the PRA to organise a meeting to discuss progress on the Bank's AIRB application, and adding:
  428. "David and I would like to update Supervision on the outcome of RWA review that we have recently undertaken, supported by Deloitte. We would therefore like to set up a call or meeting as soon as diaries allow to report back on the outcome. We hope this will then allow us to revisit earlier conversations with Supervision about application of the Pillar 2A off-set approach outlined in the statement of policy on the PRA's methodologies for setting pillar 2 capital." 

  429. On 11 January 2019, Mr Arden and Ms Roberts met with Mr Lane. There were inconsistencies in the witness evidence as to what was said by Mr Arden at that meeting. Mr Stanley did not cross-examine on that point on the basis that it was not relevant to the case being made by the Authority, and for the same reason, we have decided that we did not need to make a related finding of fact. 
  430. On 12 January 2019, Mr Arden and Ms Roberts had a conference call with the Bank's brokers; this was also attended by Linklaters. Ms Roberts's contemporaneous note of the call (approved by Mr Arden) said that "the conversation focused on what is outstanding for variables and uncertainties", which were as follows:
  431. (1)          The PRA could change the £900m if they think we have been too conservative or don't agree with our new assumptions/regulatory interpretations, but the balance of probability of this is on the low side.

    (2)          The Board needs to confirm its risk appetite in relation to areas which Deloitte has identified as "grey".

    (3)          Potential mitigants are being worked through (eg reduction in RWAs by selling treasury assets).

    (4)          The potential impact on financial condition of the group is being assessed.

  432.  On 18 January 2019, Mr Arden, Mr Donaldson, Ms Gillan and Ms Roberts met with Mr Lane. In reliance on Ms Roberts's contemporaneous note we find that Mr Donaldson said that "conversation with the PRA is ongoing and a negotiation and that they have discretion to make alternative judgements/solutions in private conversations", and that:
  433. "[Mr Arden] discussed the possibility that [the PRA] would allow us to ringfence £300m of RWAs for sale (and hence potentially not require us to increase the number) and that [the PRA] could allow the remaining £600m to be phased in over a period and clarity on that is needed before we inform the market of this."

    Discussions with the PRA and the January announcement

  434. On 21 January 2019, Mr Arden and Ms Gillan met with the PRA to discuss the RWA issue. They asked the PRA whether it was possible to ring-fence £300m of PBTL loans for a future sale and whether the balance of £600m could be phased in. On the following day, the PRA wrote to the Bank refusing both requests.
  435. The PRA letter was discussed at the Board meeting which took place on the same day. The Board accepted that the "final figure of £900m was therefore confirmed" and following detailed discussions, the Board "supported management's view that a trading statement should be released to the market".  
  436. On 23 January 2019, the Bank announced a preview of its 2018 results. The announcement included the following:
  437. "Risk weighted assets at full year are expected to be approximately £8.9bn with the increase driven by both net loan growth and an adjustment in the risk weighting of certain commercial loans secured on property and certain specialist BTL loans to large portfolio landlords. Total capital ratio is expected to be approximately 15.8% as at December 31 2018."

  438. The Bank's shares fell 39%; this was the largest intra-day drop for any UK bank since 2009 and an all-time low for the Bank. Press releases were issued by firms of analysts and brokers the same day, including the following:
  439. (1)          Bloomberg published a "reaction" which said that "Metro's trading update reveals risk-weighted asset inflation is well ahead of expectation".

    (2)          Jefferies said that "the more price sensitive element of the statement was a surprise change in risk weighting on certain commercial real estate and specialist buy-to-let portfolios, leading to a 330bps decline in the total capital ratio".

    (3)           JP Morgan issued a press statement headed "an expensive mistake", which focused on the "unexpected £0.9bn RWA increase from misinterpretations of risk-weighting requirements in the CRE and BTL loan books".

    (4)          Keefe, Bruyette & Woods ("KBW") said:

    "Metro released an unexpected trading announcement earlier today, disclosing weaker profits and more importantly, an error in its risk weighted assets calculation which weighed on capital ratios materially. The shares fell -30% intra-day."

    (5)          KBW updated this later the same day, saying:

    "Since raising equity capital at £34.22 in July 2018, Metro's shares have plummeted due to a combination of factors, now down -61% since that time including falling -39% today following their negative pre-announcement for Q4 earnings."

  440. It is clear from the above that the key cause of the sudden share price fall was the increased RWAs for both CLIP and PBTL loans. Of the two, the CLIP loans adjustment accounted for almost 60% of the total (£563m out of £960m), and was the most significant element.
  441. Subsequently

  442. The Bank subsequently worked with Deloitte to carry out a programme to identify and remediate the RWA errors. This continued through 2019, and included a line-by-line review of its loan book to ensure the accuracy of the figures. Changes were made to its policies, procedures, resourcing, roles and responsibilities, data quality, culture, reporting, systems and record keeping.
  443. The PRA

  444. On 27 February 2019, the PRA appointed investigators to see whether the Bank had breached Fundamental Rules 2 and 6 of the PRA Rulebook. Fundamental Rule 2 is that a firm conducts its business with "due skill, care and diligence" and Fundamental Rule 6 is that a firm organises and controls its affairs responsibly and effectively. In the course of the investigation, the PRA asked for a "teach-in" on the Bank's governance and control framework and on the process it used to prepare regulatory reports. This was provided on 3 July 2019 by Mr Arden and Ms Gillan, together with Mr MacLean and Ms Iovino ("the Teach-In"). The Bank also shared with the PRA the scope of the work being carried out by Deloitte and its outcomes.
  445. On 21 December 2021, the PRA imposed on the Bank a penalty of £5,376,000 (after a settlement discount) for breaching Fundamental Rules 2 and 6 of the PRA Rulebook. The penalty was based on detailed findings, which included the following:
  446. (1)          The Bank had failed to ensure effective oversight and challenge of its approach to COREP reporting.

    (2)          The Bank's escalation routes to the ELT and the Board were unclear and failed to operate effectively.

    (3)          The Audit Committee provided "limited detailed challenge".

    (4)          The Bank failed to establish and implement effective controls in relation to its interpretation of relevant regulatory rules and guidance relevant to its COREP reporting.

    (5)          Its RWA calculation process remained largely manual, reflecting the Bank's limited investment in systems.

    (6)          The Bank's front-end data capture and systems did not allow it to capture all relevant information that it needed accurately to classify exposures and calculate risk weights for its lending portfolio.

  447. There was no dispute before us that points (5) and (6) above were factually correct, and we accept them as such. We consider the other points later in our decision, so far as relevant to this case.
  448. The Authority

  449. Meanwhile, during 2019 and 2020, the Authority had interviewed Mr Donaldson, Mr Arden, Mr Carby, Mr Brierley, Ms Gillan and Mr Hill under the powers given by FSMA ss 171-3, so they were "compelled interviews". Mr Donaldson was interviewed twice: during the morning of 12 June 2019 and all day on 8 December 2020; Mr Arden was also interviewed twice, during the morning of 7 June 2019 and all day on 10 December 2020.
  450. On 26 August 2021, the Authority issued a draft Warning Notice to the Bank. On 30 September 2021, the Bank said it was not disputing the breach and issued the following statement:
  451. "The Bank accepts that, in publishing RWA figures within the October Announcement that were known to include significant inaccuracies, it did not take reasonable care to ensure the October Announcement was not false and misleading or omit anything likely to affect the import of the information it contained."

  452. On 8 December 2022, the Authority issued a Final Notice to the Bank, imposing a penalty of £10,002,300 for breaching LR 1.3.3R.
  453. Meanwhile, on 17 January 2022, the Authority had issued Warning Notices to Mr Donaldson and Mr Arden; these were followed on 10 November 2022 with Decision Notices. On 7 December 2022, Mr Donaldson and Mr Arden referred the Decision Notices to the Tribunal.
  454. Mr Donaldson's and Mr Arden's careers

  455. Mr Donaldson left the Bank around a year after the announcement in January 2019, and Mr Arden left in 2022. Neither has been able to obtain a permanent role since then.
  456. The common ground

  457. By the end of the hearing, the following points were common ground:
  458. (1)          In arriving at the RWA figure used in the Paragraph, the Bank had risk-weighted the CLIP loans at 50% when it should have used 100%.

    (2)          The statement in the Paragraph that the Bank's RWAs were £7,398m as at 30 September 2018 was incorrect, and in consequence so too was the statement that total capital was 19.1% of RWAs.

    (3)          At the time of the Q3 Update:

    (a)          the Bank did not know the correct RWAs for the CLIP loans, because of issues with how the data had been categorised for RWA purposes; and

    (b)         Deloitte had been engaged to review the RWAs, working together with internal teams from the Bank's own staff.

    The Parties' cases

  459. There were two main issues in dispute: whether the Bank had breached LR 1.3.3R, and if so, whether the Applicants were knowingly concerned in that breach. We next set out each party's case in a nutshell, followed in the next parts of our decision by their detailed submissions.
  460. The Authority's case

  461. The Authority's case was that the Bank had breached LR1.3.3R of the Listing Rules, which as set out earlier, provides that an issuer, such as a bank:
  462. "must take reasonable care to ensure that any information it notifies to a RIS...is not misleading, false or deceptive and does not omit anything likely to affect the import of the information".

  463. The Authority's position was that:
  464. (1)          The Bank had breached LR 1.3.3R when it published the Q3 Update because:

    (a)          that Update contained an unqualified statement of the Bank's RWA, and the capital ratios based on it, at a time when the Bank knew that a material error had been made in relation to the CLIP loans, but it did not inform the market about that; and

    (b)         if the Bank considered that longstanding systems and controls issues meant that it could not provide accurate quarterly announcements, it was not entitled to publish announcements containing information which it knew to be unverifiable, unreliable and/or inaccurate, while it worked to establish the correct information.

    (2)          Mr Donaldson and Mr Arden, who were intimately involved in the decision to make the Q3 Update, were knowingly concerned in that breach.

    (3)          The penalties charged were appropriate and the Tribunal was asked to uphold them.

  465. It was however not the Authority's case that:
  466. (1)           the Bank was required under Article 17 of the MAR to inform the market that the RWAs previously published were incorrect (but as the Bank had decided to publish information about the RWAs, that information could not be "misleading, false or deceptive");

    (2)          the Bank had to provide a precise figure for the RWAs. If it did not know that figure; the Bank could instead have provided an estimate, or a figure with a qualification, or no figure at all;

    (3)          the Bank should not have instructed Deloitte to carry out an exercise it hoped and believed would produce more reliable figures; instead, the Authority's case was that the Bank was not entitled, while that work was being carried out, to provide figures to the market which were known to be materially incorrect; or

    (4)          that the Bank had breached the Listing Rules in relation to the RWAs for the PBTL loans, because it was a more complex issue: at the time of the Q3 Update

    (a)          there was genuine uncertainty about the regulatory requirements for PBTL loans, and

    (b)         the estimated quantum of any PBTL categorisation error was only £37m (see §174).

    The Applicants' case

  467. The Applicants' case was that the Bank had not breached LR 1.3.3R for one or more of the following reasons:
  468. (1)          The PRA had agreed that the COREP returns could continue to be made on the previous basis until the final position was clear, and it was reasonable for the Bank to follow the same approach when making the Q3 Update.

    (2)          The Bank was prevented by its obligations of confidentiality to the PRA from disclosing the RWA issue to the market at the time of the Q3 Update.

    (3)          The Bank had relied on the advice given by Mr Lane of Linklaters.

    (4)          The Listing Rules allow a bank to disclose a figure which was known to be wrong when the correct number was uncertain and being investigated.

    (5)          LR 1.3.3R was only contravened where information is materially misleading, false or deceptive or if material information is omitted, and where (as here) the correct number was unknown, there could be no material breach.

    (6)          The Bank did not know that there was a material error until (a) Deloitte had concluded its investigation; and (b) it was clear whether any RWA increase caused by the incorrect classification of the CLIP loans could be mitigated by other factors ("the Knowledge Issue").

    (7)          It would have been misleading if the Bank had disclosed only the issue with the CLIP loans but not that with the PBTL loans (and the Authority accepted that disclosure of the latter was not required).

    (8)          The alternatives suggested by the Authority, of providing an estimate, a figure with a qualification, or no figure at all, were unreasonable.

    (9)          The Bank had relied on the experienced Board and the Audit Committee.

    (10)      The Bank had relied on the advice of Ms James .

  469. If, contrary to those submissions, there was a breach by the Bank, the Applicants' case was that they were not "knowingly concerned" in that breach. If the Tribunal nevertheless agreed with the Authority that the Bank had breached LR 1.3.3R and the Applicants had been "knowingly concerned" in that breach, the Applicants submitted that the penalties charged should be replaced by a public censure or significantly reduced.
  470. ISSUE ONE: WHETHER THE BANK BREACHED LR 1.3.3R

  471. As set out at §264, when the Authority decided the Bank had breached LR 1.3.3R, the Bank accepted this was the case and did not refer the Authority's decision to the Tribunal but instead paid the penalty. However, it was common ground that the Applicants were entitled to take a different position, with Mr Stanley saying that they could not "conceivably be bound" by the Bank's decision.
  472. The parties structured their cases by reference to the reasons why the Applicants submitted that there had been no breach by the Bank. We have taken the same approach, while recognising that the burden of proof throughout remains on the Authority.
  473. The PRA and the COREP Returns

  474. As set out at §28, the COREP returns which banks are required to complete on a quarterly basis are confidential between the PRA and the bank. The first reason relied on by the Applicants as to why the Bank had not breached LR 1.3.3R was that the PRA had agreed that its COREP returns could continue to be made on the previous basis until the final position was clear, and it was reasonable for the Bank to follow the same approach in the Q3 Update. We first set out the key findings of fact, followed by the parties' submissions and our view.
  475. Findings of fact

  476. We have already made the following findings:
  477. (1)          In July 2018, the PRA identified that the Bank had been applying 50% risk-weighting to the CLIP loans rather than the correct 100%, and that this was "a serious issue", which would cause "a large material change" to the Bank's position, see §102.

    (2)          Ms Gillan responded the same month, saying that the Bank recognised that 50% was inaccurate, and would revert with further details, see §110.

    (3)          Mr Gunn told the PRA at the meeting on 6 September 2018 that "as to the miscalculations of capital required to support commercial lending, that was clearly an error on our part and was being fixed", see §131(1).

    (4)          The Bank's mistake with the RWAs for CLIP loans had "caused agitation" at the PRA because it was "perceived as the firm not being able to get the basics right", see §137.

    (5)          At the subsequent meeting with the PRA on 12 September 2018, Mr Donaldson "expressed sincere regret and apology for the error" and expected it would be corrected by  October/November.

    (6)          On 4 October 2018, Mr Arden sent Mr Sutherland the email set out at §160. This included the following:

    (a)          the Bank was making good progress in remediating the classification of the risk-weighting;

    (b)         it was in the process of engaging an external third party (Deloitte) to provide "external assurance over work completed to date and provide a full review of our RWA calculations and COREP reporting";

    (c)          that work would take a number of weeks;

    (d)         Ms Gillan and Mr Arden needed to be sure about "the efficacy and accuracy of the results of this exercise" before they submitted any changes to the PRA; and

    (e)          in consequence, COREP reporting for September would be "materially unchanged".

    (7)          Mr Sutherland viewed this as a pragmatic solution to the issue of COREP submissions which required the input of specific figures, given that Metro Bank was in the process of conducting a remediation of its Pillar 1 commercial risk weightings and its capital requirements. He responded on 9 October 2018, saying "That's fine David, thank you for confirming", see §161.

    (8)          The RWAs included in the Bank's September COREP report were therefore the same as those in the Q3 Update.

    The Applicants' position

  478. Mr Jaffey emphasised that the Bank had agreed with the PRA that it should maintain the existing RWA reporting position until the full review had been completed, and submitted that it was reasonable for the Bank to take the same approach in its Q3 Update. In oral evidence Mr Arden said that this was the normal process. He had the following exchange with Mr Stanley:
  479. "Mr Stanley: Don't you accept that a person reading a financial statistic given by a bank would assume that it was a reliable and accurate statistic?

    Mr Arden: I think a person reading a statistic on regulatory capital at a bank would assume that that number had been agreed with the PRA, or reported to the PRA, and that is exactly what had happened.

    Mr Stanley: Even if it was reported as the wrong number?...

    Mr Arden: But the PRA understood that.

    Mr Stanley: And how would I, as an analyst sitting in my office understand that, reading this number?

    Mr Arden: Until we had agreed the change in number with the PRA, the number was the number.

    Mr Stanley: So you thought it was acceptable to report an unreliable number until the PRA had agreed the reliable one?

    Mr Arden: That was the process that we had agreed with the PRA."

  480. Mr Donaldson similarly said: "everybody knows you use the numbers from the COREP in your announcements", and there was then the following exchange:
  481. "Mr Stanley: The PRA had not told you to use those numbers in anything you said to the market, had they?

    Mr Donaldson: No...We were working on the agreement that was always done, that the COREP numbers were the numbers that were published to the market...and everybody knew we would be using the COREP numbers as they were, and that is what was put in there.

    Mr Stanley: Presumably normally people would expect that the COREP numbers would be accurate.

    Mr Donaldson: Within reason, yes.

    Mr Stanley: They would be numbers that the bank believed would be right, correct?

    Mr Donaldson: Absolutely, yes.

    Mr Stanley: And on this occasion, they weren't, were they?

    Mr Donaldson: No. We had issues to resolve and we needed to resolve them..."

  482. Mr Stanley submitted that, while readers of information published by listed banks may assume that the same RWA figure would be provided to the PRA as to the market, that is because the reader would also assume that both are being provided with an accurate and reliable figure, and as Mr Donaldson had agreed, that was not the position here. The Bank and the PRA both knew that the RWA figure used in the COREP reports was wrong; as Mr Donaldson had accepted in cross-examination, the reason why the PRA had been told that the error would not be corrected in the September COREP reporting was to avoid any risk that they were misled by the figures shown in that return, but the market did not have the benefit of that explanation.
  483. The Tribunal's view

  484. We have no doubt that both Mr Donaldson and Mr Arden believed that they and the Bank were acting correctly when they used the same figure in the Q3 Update as in the COREP returns. That is clear from their oral evidence.
  485. However, we also agree with Mr Stanley that such an approach was not reasonable where the number reported to the PRA as part of its confidential COREP return was known to be wrong. Instead, the reasonable financial institution, and the reasonable CEO and CFO, would have considered separately (ie without reliance on what had been agreed with the PRA for their COREP reporting) whether the use of those numbers in the Q3 Update was a breach of LR 1.3.3R.
  486. The PRA and confidentiality

  487. The next issue was whether the Bank had acted reasonably because it had an obligation to keep confidential its discussions with the PRA about the RWA issue, and so could not disclose it to the market at the time of the Q3 Update. We next set out relevant findings of fact followed by the parties' submissions and our view.
  488. Findings of fact

  489. We have already made the following findings:
  490. (1)          On 12 July 2018, the Bank issued its half year update; this included a statement that its Pillar 2A requirement of 1.7% was "currently under review with the PRA" and the Bank "anticipated receiving capital relief from the PRA as part of the Pillar 2A Offset"; and also that the Bank also anticipated transitioning to AIRB on residential mortgages in H2 2019, see §111.

    (2)          The PSM letter sent on 10 September 2018 said that the PRA was "somewhat frustrated" that the Bank had referred publicly to the discussions on the Pillar 2A offset and to the AIRB application; that the PRA had not yet reached a conclusion on either matter and "would not wish to see a repeat" of this type of public disclosure, see §132.

    (3)          The Bank's response to the PSM letter, which was agreed at the October 2018 Board meeting, did not respond to the PRA's criticism of its public statements about the Pillar 2 offset, but said that "public referencing of the [AIRB] process merely restated the PRA's published guidelines, though again your comments in this regard are noted and accepted", see §234.

    (4)          The Q3 Update included the statement that " there is no update to the AIRB application for residential mortgages submitted earlier this year and we still expect an outcome in 2H19", see §220(2).

    (5)          On 24 October and 2 November 2018 Mr Arden told analysts "our application for AIRB is with the PRA, and our expectation remains that that will be an H219 event", see §239 and §241.

    The Applicants' position

  491. Mr Donaldson's evidence was that he had understood from the PSM letter that the Bank:
  492. "should not disclose to the public any issues that were the subject of ongoing dialogue with the PRA. This was because conversations with the PRA on such subjects are confidential and not disclosable to the market. The Bank took this seriously and the Board reiterated this to me, and in the response to the 2018 PSM Letter, confirmed that it would not make any future  disclosures to the market on any such issues under discussion with the PRA. This was another reason why disclosure to the market of the RWA Issue in the Q3 Announcement was considered inappropriate as it would, inevitably, have been in breach of that undertaking of confidentiality to the PRA."

  493. Mr Arden's evidence was that he had a similar understanding: when interviewed by the Authority before the hearing, he said that the Bank was engaged in "confidential work" and in "confidential dialogue" with the PRA about the change to the RWA, and "until that work was concluded, there was nothing to disclose".
  494. The Authority's position

  495. Mr Stanley submitted that ensuring the accuracy of the information in the Q3 Update did not involve the publication of any matters that were confidential to the PRA for two reasons:
  496. (1)          The PRA had asked the Bank not to disclose confidential discussions about discretionary matters, in particular the Pillar 2A offset and the AIRB application. The Bank's failure to follow the risk-weighting rules was of an entirely different nature.

    (2)          The Bank did not in any event respect the confidentiality of its discussions with the PRA; that is evident from its statement in the Q3 Update that " we still expect an outcome [to the AIRB application] in 2H19" and Mr Arden's similar responses to analysts. Mr Stanley said the Bank had "simply ignored the PRA's request".

    The Tribunal's view

  497. We agree with Mr Stanley that it is not credible that the Bank considered it was bound by a requirement of confidentiality not to disclose accurate information about the RWAs in the Q3 Update, and it is also not credible that Mr Donaldson or Mr Arden took that view at the time. That is for the reasons given by Mr Stanley, to which we add the following:
  498. (1)          The discussions with the PRA about Pillar 2A and the AIRB application concerned the exercise by the PRA of a discretion which they were still considering. Whether, and if so when, that future discretion would be exercised was unknown, and that decision was a matter for the PRA.

    (2)          There is a fundamental difference between the Bank's discussions with the PRA about those two matters, and informing the PRA that the Bank had made an historic error in the categorisation of its CLIP loans which it was remediating.

    (3)          The decision not to amend the COREP reporting was made by the Bank, and Mr Arden informed the PRA of that decision. Mr Sutherland simply accepted the approach the Bank had decided to take, see §§160-161.

    (4)          In its reply to the PSM letter, the Bank:

    (a)           did not even respond to the PRA's criticism of its public statements about the Pillar 2 offset, see §234; and

    (b)         its statement that the Bank's public referencing of the AIRB discussions "merely restated the PRA's published guidelines" was incorrect: the Bank told the market in the half-year update that:

    "We anticipate receiving capital relief as part of the Pillar 2A offset...ahead of transitioning to the advanced internal ratings based approach (AIRB) on residential mortgages, expected H2 2019."

    (5)          The statements to analysts on 24 October and 2 November 2018 that "our application for AIRB is with the PRA, and our expectation remains that that will be an H219 event" was exactly the sort of public disclosure that the PRA was asking the Bank not to make.

  499. In any event, if the Bank and the Applicants had considered they were bound by confidentiality, such a position would not have been reasonable. The reasonable financial institution would have distinguished between informing the PRA that it had made an historic error which it was remediating, and asking the PRA to exercise various discretions in the future.
  500. Mr Lane's advice

  501. The next issue was whether the Bank had relied on advice given by Linklaters at the meeting on 5 October 2018. There was significant disagreement about this meeting, but we begin by making findings of fact which were not in dispute; we have included under that heading certain findings based on Mr Arden's evidence.
  502. Findings of fact not in dispute

  503. At the September 2018 board meeting, a new NED, Catherine Brown, was appointed with effect from 1 October 2018, see §154. On 5 October 2018, Mr Lane, Mr Sadler and Mr Cheyne came to the Bank's offices to provide Ms Brown with a training session covering the Listing Rules Corporate Governance, the Bank's Share Dealing Policy and its Disclosures Procedures Manual.
  504. Mr Arden and Ms Roberts were aware that the training session was taking place, and decided to ask whether Mr Lane would be available for a meeting subsequently. In the course of, or shortly before, that training, Ms Roberts popped into the training room to say that she and Mr Arden would like to speak to Mr Lane afterwards.
  505. That meeting took place in Mr Arden's office. It had not been pre-arranged; no papers, written instructions or briefing was provided to Linklaters either in advance or at the meeting. No note of the meeting was taken, and Linklaters was not asked to confirm in writing the advice which was given.
  506. After the meeting, at 14.22 the same day, Ms Roberts sent Mr Arden an email which was headed "For DisCo" and read as follows:
  507. "David, please review and add to minutes below as needed:

    David Arden and I met with Linklaters (John Lane, Jonty Sadler and Rupert Cheyne) at 11.45am today to discuss the ongoing work and dialogue with the PRA on the classification of our RWAs. We explained that we had identified a problem with the risk weight classification of some commercial assets; that current estimates based on sampling was c.£600m and the impact on core  equity capital of c.£50m but further work was being undertaken (with the help of Deloitte) to finalise the amount; that the PRA had agreed that there were no immediate changes necessary for our CoReg [COREP] reporting; and the intention was to notify the market once finalised and resolved in line with our usual full year and Pilar 3 disclosures. Linklaters concurred with our view that it is neither specific or material information at this point and was in the ordinary course of ongoing dialogue with the regulator over a complex issue and no market announcement was necessary at this point."

  508. Mr Arden read the email, and responded around 30 minutes later saying "Perfect - thanks Jo". He did not think anything needed to be added. He copied his reply to Ms James, Ms Roberts' email was appended. Neither he nor Ms Roberts sent a copy of the email to Mr Lane, or to anyone else at Linklaters. However, Mr Arden discussed the meeting with Mr Donaldson the same day, and during October he also told Ms Gillan.
  509. Mr Arden agreed in cross-examination that the following information in the email was incorrect:
  510. (1)          The statement that the "current estimates" of c.£600m were "based on sampling". As we have already found, see §123, the CLIP loan figures had been arrived at on the basis of recalculating risk-weighting of those loans in the Bank's books by 100% instead of 50%, along with other minor changes.

    (2)          The statement that the PRA had "agreed that there were no immediate changes necessary" for the Bank's COREP reporting. Instead, at the date of the meeting with Linklaters, Mr Arden had emailed Mr Sutherland setting out what the Bank planned to do, but had not received a response.

  511. Ms Roberts's email is the only contemporaneous evidence of what happened at this meeting, which Mr Arden read at the time and agreed was "perfect". However, from the witness box, he initially suggested that the email had incorrectly recorded what had been said in relation to the two points in the previous paragraph. However, in reliance on the contemporaneous evidence, and on Mr Arden's subsequent responses in cross-examination, we find that the email is not wrong, and that Mr Arden told Mr Lane that the £600m was "based on sampling"; and had also told him that the PRA had agreed no immediate change was required to the COREP reporting.
  512. In addition to the points recorded by Ms Roberts in the email, it is clear from Mr Arden's own evidence that he also gave Mr Lane the following information, which he accepted was incorrect:
  513. (1)          He told Mr Lane that the PSM letter had raised the risk-weighting "as a potential issue", but agreed under cross-examination that the PSM letter had not said there was a "potential issue"; instead it had referred to it as an issue which the Bank had accepted needed remediation.

    (2)          He told Mr Lane there were "regulatory interpretation issues" about the RWAs, when he knew at the time that the position had been confirmed by both KPMG and Deloitte, and that on 12 September 2018, he and Mr Donaldson had told the PRA that the Bank had made an error, see §§137-138.  

  514. Mr Arden also agreed in cross-examination that the Q3 Update was not mentioned during the meeting, and in particular:
  515. (1)          he did not inform Linklaters that that the Bank was going to report the RWAs as part of the Q3 Update; and

    (2)          he did not ask for advice on whether the Bank could report the RWAs on a basis it knew to be incorrect, pending completion of the Deloitte work.

  516. On 16 October 2018, the text of Ms Roberts's email (other than the first two sentences) was replicated in the minutes of the Disclosure Committee meeting, as can be seen from the text set out at §163. That is consistent with the header on the email, which was "for Disco" and with the opening sentence "David please review and add to minutes below as needed".
  517. After the meeting on 5 October 2018, there was no further discussion or  communication between Mr Lane and the Bank for the rest of that calendar year on any disclosure or market-related issue.
  518. Who was at the meeting

  519. There was a dispute as to who attended the meeting; Mr Lane's evidence was that Mr Sadler and Mr Cheyne left at the end of the training session, but Mr Arden said that all three attended.
  520. Mr Sadler and Mr Cheyne were asked by the Authority to give evidence about the meeting, but neither was able "definitively" to recall it; Ms Roberts was asked about it when she gave oral evidence but could not add to her witness statement. 
  521. In reliance on Ms Roberts's contemporaneous email, and taking into account that Mr Lane did not attend the Tribunal to be cross-examined, we find that all three did attend. However, we also find that the advice was given only by Mr Lane, because:
  522. (1)          Mr Arden's oral evidence was that he and Mr Lane did most of the talking;

    (2)          Ms Roberts's evidence was that (our emphasis) Mr Arden did "most of the talking and the Linklaters partner asked questions";

    (3)          Mr Arden's witness statement refers only to statements he said were made by Mr Lane (and does not refer to anything said by the other two Linklaters attendees) and he repeated this from the witness box, saying "Mr Lane provided advice"; and

    (4)          Linklaters's contemporaneous time record states that Mr Lane spent two hours at the Bank, which included both the training session and the meeting; Mr Sadler spent two hours on preparing for and attending the training session, but with no reference to the meeting, and Mr Cheyne spent 1.5 hours preparing for and attending the training session, but again without reference to the meeting. Although we have found they did attend, we also find that they did not give advice; had they done so, it is unlikely that they would have omitted any reference to the meeting when they completed their timesheets.

    How long was the meeting

  523. Mr Lane's evidence was that the meeting lasted around 15 minutes. Mr Arden's evidence was inconsistent: in his interview with the Authority on 10 December 2020, he said it was "half an hour" or was "maybe a little shorter"; in his witness statement, which was deposed on 28 February 2024, he says it lasted for "approximately 45 minutes", and in oral evidence he said that it was "longer" than the 15 minutes given by Mr Lane. Ms Roberts' evidence was that it lasted "around 30-40 minutes".
  524. Taking into account all of the above, including the fact that Mr Arden's original evidence was given close to the time of the events in question, and recognising that Mr Lane did not attend to be cross-examined, we find that the meeting lasted around half an hour and was thus a short meeting.
  525. Linklaters' practice when giving advice

  526. Mr Brierley, the Bank's previous CFO, gave unchallenged evidence that it was "not unusual to take advice from Linklaters on the duty of disclosure arising in relation to any issues surfacing from the Bank's finances" and he gave the example of asking for advice on the application of the Class Tests in LR 10, Annex 1. He said:
  527. "Such an issue required discussion of financial information (provided by the Finance function) for the purposes of advice and guidance by our lawyers and, if appropriate, the brokers, the result of which would be a concluded strategy on a disclosure issue by Management."   

  528. Mr Lane's evidence was that the frequency of Linklaters's interactions with the Bank depended on whether they were advising on a transaction (when there would be frequent interactions) or other times, when interactions were less frequent. We accept that evidence, which is reflected in Linklaters's involvement with the capital raises, see §75 and §§111-112.
  529. Mr Lane also said that interactions with the Bank "could involve formal instructions or might occur via email or verbally", but Linklaters did not have an ongoing role pro-actively to monitor the Bank's market announcements. Mr Jaffey asked us to reject this evidence, on the basis that Linklaters were "asked to sign off the half-yearly announcement" in July 2018. However, it is clear from the context that this sign off was required because Linklaters were advising on the capital raise which happened on the same day, see §111. Mr Lane's evidence is also consistent with that given by Ms Roberts: she said that the Bank would rely on Linklaters when they "had a transaction coming up" and that the firm would "sometimes" be involved in reviewing interim trading statements "particularly if, for example, we might be issuing a prospectus". We therefore accept Mr Lane's evidence that Linklaters were not asked to sign off each of the Bank's trading statements.
  530. Knowledge of the impending Q3 Update

  531. In reliance on Mr Brierley's evidence, we find that Linklaters as a firm were aware that the Bank had adopted a pattern of quarterly reporting in March, June, September and December. Mr Lane accepted that he knew that the Bank announced its results on a quarterly basis, but said he did not know the specific dates when they were published. However, on 13 June 2018, Mr Lane had sent the Bank a presentation about a possible combined equity and perpetual bond issuance, and one of the slides in that presentation stated that the Q3 results would be issued on 24 October 2018. We therefore find that Mr Lane had been informed of the date on which the Q3 Update would take place.
  532. Mr Lane's evidence was also that he was unaware that the Bank's practice was to include its RWA figure in its interim results announcements. Mr Jaffey asked us to reject that evidence by reference to a passage contained in the written representations drafted by Linklaters on behalf of the Bank, which were submitted to the Authority's Regulatory Decisions Committee ("RDC") on 8 March 2022 (our emphasis):
  533. "Notwithstanding the fact that the Bank did not specifically seek advice on the content of the October Announcement, Linklaters would have been aware that the Bank's Q3 trading update...would have to include RWA figures."

  534. No contemporaneous or other evidence has been provided to support the italicised phrase, which is in the conditional voice. It does not provide a basis for us to reject Mr Lane's statement that he did not have that knowledge, and we accept his evidence. 
  535. Mr Lane also said that he had never been provided with the Q3 Update and did not discuss the Q3 Update with the Bank. Mr Jaffey asked us to reject that evidence by reference to the analyst call which took place on 31 October 2018. Mr Arden was briefed before that call by Linklaters, as well as by Jefferies and BAML. Mr Jaffey invited us to find as facts that (a) either Mr Lane carried out the briefing, or it was carried out by one of the other two Linklaters attendees at the meeting on 5 October 2018, and (b) the person who carried out the briefing was also aware of and had seen the Q3 Update.
  536. We decline to make those findings, because:
  537. (1)          Mr Arden could not remember who from Linklaters had briefed him, see §239.

    (2)          Although someone at Linklaters did brief Mr Arden, there is no evidence as to who that person was. Those at Linklaters who were giving advice to the Bank also included teams led by Mr Manketo and Mr Kumar, and Linklaters also had "a strong undercast" of qualified lawyers, see §66.

    (3)           It was Mr Lane's evidence that he had never seen the Q3 Update, and had he briefed Mr Arden some three weeks after the meeting, it is not credible that neither he nor Mr Arden would have remembered that he had done so.

    (4)          As noted above, Mr Sadler and Mr Cheyne were asked by the Authority to give evidence about the meeting, but they were not called by the Authority because neither were able "definitively" to recall it. However, there is no property in a witness, and had the Applicants wanted to call either or both of those individuals in order to ask whether they had carried out this briefing, they could have done so.

  538. In summary we therefore find as facts that:
  539. (1)          Mr Lane was aware that the Bank would issue its Q3 Update on 24 October 2018;

    (2)          he was not provided with a copy of that Update before, during or after the meeting with Mr Arden and Ms Roberts; and

    (3)          he did not know that the Bank's practice was to include its RWA figure in its interim results announcements.

    What was said by Mr Arden at the meeting

  540. As set out at §§297-300, we have already found that Mr Lane was told by Mr Arden:
  541. (1)          there were "regulatory interpretation issues" about the classification;

    (2)          this had been raised as a "potential issue" by the PRA;

    (3)          the estimate of £600m was based on sampling; and

    (4)          the PRA had agreed that no immediate changes were necessary.

  542. However, none of those statements was true at the time they were made. In addition:
  543. (1)          Mr Arden told Mr Lane that the Bank was "potentially having to re-weight the CRE loan portfolio", when the Bank had told the PRA that it was remediating the issue so as to use the correct 100% risk weighting for the CLIP loans; and

    (2)          Ms Roberts's email twice refers to the existence of an "ongoing dialogue" with the PRA about the issue, when there was no dialogue: instead, the Bank had simply told the PRA that it was remediating the issue. We make the reasonable inference from the duplicate mention in Ms Roberts's email that Mr Arden also told Mr Lane that there was an ongoing dialogue.

  544. Ms Roberts also recorded that Mr Arden had told Mr Lane that the Bank had "a problem with the risk weight classification of some commercial assets", and Mr Lane recalls that Mr Arden "explained the general nature of the issues". When Mr Arden was asked in cross-examination whether he had also told Mr Lane that it was certain that an adjustment would be required, he repeatedly said "I don't recall"; he would only confirm that he had talked about "the regulation interpretation issues and the data quality issues". We make the reasonable inference that Mr Arden did not tell Mr Lane that the Bank had misapplied the rules requiring that the CLIP loans be risk weighted at 100% and that in consequence an adjustment would be required.
  545. Mr Lane's evidence was that:
  546. "The impression I received from David Arden was that there was a high level of uncertainty regarding whether an adjustment would be required. Also, if there was to be an adjustment, the amount could be highly variable due to underlying data and regulatory interpretation issues. Further work was needed to be done with Deloitte along with further interactions with the PRA."

  547. He also said that:
  548. "David Arden indicated that he didn't feel that the issue would necessarily be material in terms of impacting Metro Bank's growth ambitions and future financial performance. He indicated he thought that even in a worst-case outcome Metro Bank would remain well capitalised, in excess of its minimum regulatory capital requirements."

  549. Mr Arden responded to this evidence in his second witness statement, saying:
  550. "I stated that I did not believe the RWA issue would be material in the context of the Bank's future strategy...However, I did not think the RWA issue was an irrelevant issue for the Bank...If I thought it was immaterial I would not have sought the advice of Linklaters."

  551. Mr Arden thus confirmed Mr Lane's recollection that he had said the RWA adjustment would not be material. He did not deny that he had told Mr Lane that "further work was needed to be done with Deloitte along with further interactions with the PRA" before the Bank could know the position.
  552. We find as a fact that as the result of the information provided by Mr Arden in the course of the meeting, Mr Lane had the incorrect impression that:
  553. (1)          there was a high level of uncertainty as to whether an adjustment would be required;

    (2)          if there was to be an adjustment, the amount could be highly variable;

    (3)          this was due to underlying data and regulatory interpretation issues;

    (4)          even in the worst case the outcome was not necessarily material; and

    (5)          the position would not be clear until Deloitte had done further work and there had been further interactions with the PRA.

    Confidential matter?

  554. Ms Roberts's contemporaneous note ended by saying:
  555. "Linklaters concurred with our view that it is neither specific or material information at this point and was in the ordinary course of ongoing dialogue with the regulator over a complex issue and no market announcement was necessary at this point."

  556. Thus the contemporaneous email refers to Mr Lane having agreed that the matter was "complex", but Mr Arden's evidence to the RDC on 8 March 2022, which was repeated in his witness statement, was that Mr Lane had said (our emphasis):
  557. "It seems to me that you are dealing with a complex and confidential matter with your prudential regulator and, until that work is completed, there is nothing to disclose."

  558. Ms Roberts said in her witness statement that the legal advice received was "that it was a confidential issue that was being worked out with the regulator so no further action needed at that time". However, under cross-examination she was unable to answer questions about the advice, and relied only on the email she had sent previously. We thus place no reliance on this part of her witness statement.
  559. Mr Lane's evidence was as follows:
  560. "I do not recollect using these words. I do not believe I gave advice that no announcement obligation could arise before either completion of work or agreement of the PRA. I do not recall any discussion of confidentiality  obligations to the PRA."

    The Tribunal's finding

  561. The first reference to Mr Lane having said that the matter was "confidential" appears over three years after the meeting in Mr Arden's evidence to the RDC. From the witness box, Ms Roberts was unable to do more than rely on the text of her email. We therefore rely on the contemporaneous evidence in that email, which agrees with that given by Mr Lane.
  562. We find that there was no reference by Mr Lane to the discussions with the PRA as being confidential. That finding is also consistent with those we have already made about which of the Bank's discussions with the PRA were confidential, see §285-289.
  563. The purpose of the meeting

  564. Mr Lane's evidence was that he understood the purpose of the meeting was to seek advice on whether a pro-active announcement was required by the MAR, which as we said at §31ff, requires the publication "as soon as possible" of:
  565. "information of a precise nature, which has not been made public, relating, directly or indirectly, to one or more issuers or to one or more financial  instruments, and which, if it were made public, would be likely to have a  significant effect on the prices of those financial instruments..."

  566. As we have already found, see §294 and §300, no copy of the Q3 Update was provided to Linklaters before the meeting, and Mr Arden accepted in cross-examination that it was not mentioned in the meeting; he did not tell Mr Lane the Bank was going to report the RWAs as part of that Update; and he did not request advice on whether the Bank could report the RWAs it knew to be incorrect, pending completion of the Deloitte work.
  567. Despite that evidence, Mr Arden also stated that:
  568. (1)          the purpose of the meeting was to obtain "general advice" on disclosure;

    (2)          Mr Lane had advised that "until Deloitte's work was complete and until the position was confirmed with the PRA, there was nothing to disclose to the market", and

    (3)          this advice covered both a proactive disclosure to comply with the MAR, and "any disclosure of the issue", including by way of the Q3 Update.

  569. We agree with the Authority that the purpose of the meeting was to establish whether the Bank had to make a pro-active disclosure under the MAR, and it was not to ask for advice on the Q3 Update. We make that finding of fact for the following reasons:
  570. (1)          Ms Roberts's contemporaneous email said (our emphasis) "Linklaters concurred with our view that it is neither specific or material information at this point...and no market announcement was necessary at this point". The MAR requires a disclosure if information is "sufficiently precise" and would be likely to have "a significant effect" on the share price. Although the word "specific" is not identical to "precise" and "material" is not identical to "significant effect", the meanings are similar.

    (2)          Article 17 of the MAR requires that "an issuer shall inform the public as soon as possible of inside information which directly concerns that issuer". Ms Roberts's email concludes by saying (our emphasis) that Linklaters had advised that "no market announcement was necessary at this point". That is consistent with Mr Arden having asked Mr Lane to advise on whether an immediate announcement was required under the MAR.

    (3)          Ms Gillan was told about the meeting by Mr Arden during October 2018, and she too understood that Linklaters' advice was directed to the question of whether there was inside information that needed to be disclosed.

    (4)          It is not credible that Mr Arden would have sought advice on the Q3 Update without referring to it during the meeting or providing a copy of the draft announcement either before or during the meeting.

  571. That finding is consistent with the fact that Ms Roberts's email was not circulated to the Board before its October meeting, when the Q3 Update was approved, see §228.
  572. Reasonable to rely?

  573. Mr Jaffey submitted that when issuing the Q3 Update it was reasonable for the Bank to rely on the advice given by Linklaters in the meeting, even though that Update was not mentioned by either Mr Arden or Mr Lane for the following reasons:
  574. (1)          It is the duty of lawyers to identify the full legal framework and ensure
    any advice covers the steps its client should take, for example to comply with its
    regulatory obligations generally.

    (2)          If a lawyer's intention is to limit the advice to anything short of a comprehensive advice based on the full legal framework, it is for the lawyer to ensure any limitations or caveats are explicit and recognised (whether at the time of the advice or in subsequent communications).

    (3)          The Bank relied on Linklaters for proactive advice generally; in the context of this issue, Linklaters knew that the Q3 Update was imminent but did not limit or caveat the advice.  

    (4)          Even if Mr Lane's advice was limited to the MAR, it would have been reasonable for the Bank to understand that disclosure was also not required under LR 1.3.3R.

  575. In relation to the first three of those points, Mr Stanley submitted that the Bank had failed properly to brief Mr Lane, because Mr Arden did not mention the Update in the meeting; no copy of the Update (or any other papers) was provided in advance, at the meeting, or subsequently; and Mr Arden gave partial and incorrect information to Mr Lane.
  576. In relation to the final point, Mr Stanley reiterated that the issue was not whether the Bank had a duty to disclose the RWA issue at the time of the Q3 Update, but whether it was a breach of LR 1.3.3R to disclose figures which it knew to be incorrect. The two issues were separate and distinct. The Bank had not sought or received advice on that second issue. The Applicants' approach failed to distinguish between:
  577. (1)           an obligation to inform the market about the RWA issue; and

    (2)           how the Bank should proceed where it does choose to publish an RWA figure to the market.

  578. We have no hesitation in agreeing with Mr Stanley for the reasons he gave. We add that even if the Bank had intended to obtain Linklaters' advice about the Q3 Update, it would have failed to instruct Linklaters properly or with reasonable care and could not, as a result, reasonably have relied upon their advice.
  579. Overall conclusion on legal advice

  580. For the reasons explained above, we find as a fact that Mr Lane did not give the Bank advice about the Q3 Update, but instead about whether immediate disclosure was required under the MAR, and it was not reasonable for the Bank to treat his advice as if it did relate to the Q3 Update.
  581. No breach if uncertain and under investigation?

  582. The next issue was whether the Listing Rules allow a bank to disclose a figure which was known to be wrong, when the correct number was uncertain and being investigated. We first set out Mr Jaffey's submissions, followed by those of Mr Stanley and our conclusions.
  583. Mr Jaffey's submissions

  584. Mr Jaffey submitted that when a business announces its management information "the market understands...that that information is subject to review and revision and emerging information", and thus an announcement is not "misleading" so as to engage LR 1.3.3R if the figure in question "is subject to revision following quantification of an emerging issue".
  585. He referred to the version of the Listing Rules which existed in 2002; this contained LR 9, headed "General obligation of disclosure for companies", and included the following:
  586. "9.1  A company must notify the Company Announcements Office without delay of any major new act developments in its sphere of activity which are not public knowledge which may,

    (a) by virtue of the effect of those developments on its assets and liabilities or financial position or on the general course of its business,  lead to a substantial movement in the price of its listed securities;...

    9.2 A company must notify the Company Announcements Office without delay of all relevant information which is not public knowledge concerning a change:

    (a) in the company's financial condition;

    (b) in the performance of its business; or

    (c) in the company's expectation as to its performance

    which, if made public,  would be likely to lead to substantial movement in the price of listed securities

    9.3  ...

    9.3A A company must take all reasonable care to ensure that any statement or forecast or any other information it notifies to, or makes available through, the Company Announcements Office is not misleading, false or deceptive and does not omit anything likely to affect the import of such statement, forecast or other information.

    9.4 The company need not notify the Company Announcements Office of information about impending developments or matters in the course of negotiation and could give such information in confidence to recipients within the categories described in paragraph 9.5...

    9.5. The categories of recipient referred to in paragraph 9.4 are:

    (a) the company's advisers and advisers of any other persons involved or who may be involved in the development or matter in question;

    (b) persons with whom the company is negotiating, or intends to negotiate, any commercial, financial or investment transaction...

    (c) representatives of its employees or trades unions acting on their behalf; and

    (d) any government department, the Bank of England, the Monopolies and Mergers Commission or any other statutory or regulatory body or authority."

  587. Mr Jaffey emphasised LR 9.4, which had specifically provided that companies need not notify the market about "information about impending developments or matters in the course of negotiation". He said that the same exception continued today, referring to Section 2.5.2G of the Authority's Disclosure, Guidance and Transparency Rules sourcebook, which reads:
  588. "(1) Delaying disclosure of inside information will not always mislead the public, although a developing situation should be monitored so that if circumstances change an immediate disclosure can be made.

    (2) Investors understand that some information must be kept confidential until developments are at a stage when an announcement can be made without prejudicing the legitimate interests of the issuer."

  589. That passage was, said Mr Jaffey, consistent with Regulation 17(4) of the MAR, set out at §34 above. He summarised the position as follows:
  590. "Participants in the London Stock Exchange understand when they see an RIS announcement that it will always have a metaphorical asterisk next to it that information in here may well be affected by these type of matters [ie impending developments or matters in the course of negotiations] and you have to read this RIS announcement subject to caveats, which everyone in the market understands."

    Mr Stanley's submissions

  591. Mr Stanley's main point was that both sets of provisions relied on by Mr Jaffey concerned the withholding of information about uncertain future events, such as a take-over. They do not allow an issuer to publish incorrect information about other matters, such as its assets or liabilities, on the basis that it is investigating the quantum of the error. He said in closing:
  592. "There are many circumstances in which you are not under an obligation to say anything, but if you do say something you are under an obligation to take care that it should be accurate."

  593. He also made the following subsidiary points.
  594. (1)          The 2002 Listing Rules were replaced in 2005 as part of the implementation of the MAR, and the Q3 Update was issued in 2018, so the earlier provisions ceased to apply some 13 years before the issues with which this case is concerned.

    (2)          In any event, the 2002 Listing Rules not only included LR 9.4 but also LR 9.3A, which was in materially similar terms to the current LR 1.3.3R: it required that a company which makes an announcement must take "reasonable care to ensure that... [it] is not misleading, false or deceptive and does not omit anything likely to affect [its] import".

    (3)          Although Article 17(4) of the MAR does allow an issuer to delay disclosure, that provision is subject to stringent conditions, one of which is that delay was "not likely to mislead the public" and another is that "immediate disclosure is likely to prejudice the legitimate interests of the issuer". Here, neither of those conditions was met.

    (4)          This case was in any event not about delaying disclosure, but about publishing incorrect information. In Mr Stanley's words:

    "it would be one thing not to mention negotiations, it would be quite another thing to put out an announcement saying 'there are no negotiations taking place' when, in fact, there were. That would not be permissible."

    The Tribunal's view

  595. We agree with Mr Stanley that there is a difference between (a) deciding not to disclose something (such as negotiations) because they come within one of the exceptions permitted by the Authority and (b) disclosing a figure which is known to be wrong. None of the provisions on which Mr Jaffey relied allows the latter.   
  596. To borrow his language, there is no "metaphorical asterisk" or "caveat" attached to a company's published results. In other words, the market does not understand that (a) those results may contain figures known by the company to be materially incorrect, and/or (b) that incorrect position will remain hidden until further investigations have been completed.
  597. No material breach if unknown

  598. The next issue was whether LR 1.3.3R was only contravened where information is materially misleading, false or deceptive or if material information was omitted. Mr Jaffey submitted that as the quantum of the adjustment required was unknown until the Deloitte review was completed, the Bank did not know it was material, and thus the Q3 Update contained "no material inaccuracy".
  599. Mr Stanley accepted that LR 1.3.3R was only contravened where information was materially misleading, false or deceptive or if material information is omitted, saying:
  600. "there is always going to be some sort of threshold. Information is not false if it is substantially true. It is not misleading if it is substantially correct. It is not lacking something that would affect the import of the information if the effect of...saying something more would be trivial or inconsequential."

  601. However, in his submission, it was plainly wrong to say that a bank could issue a statement it knew to be wrong, simply on the basis that there was uncertainty. By way of example, if there was real uncertainty as to whether a figure was wrong by (a) a material £600m or (b) only by an immaterial £2m, it was not permissible for a bank to proceed on the basis that the error was immaterial.
  602. We again agree with the Authority. LR 1.3.3R requires that an issuer such as a bank "must take reasonable care to ensure that any information it notifies to a RIS is not misleading, false or deceptive and does not omit anything likely to affect the import of the information". If a bank knows there is an error, and knows it might be material, it cannot just assume it is not material, because that would be to omit something which is "likely to affect the import of the information" which is being disclosed.
  603. The knowledge issue

  604. We then considered whether the Bank did not know there was a material error until (a) Deloitte had concluded its investigation, and (b) it was clear whether any RWA increase caused by the incorrect classification of the CLIP loans could be mitigated by other factors ("the mitigants"). We first summarise some of the relevant findings of fact already made, followed by the parties' position overall and then by their detailed submissions on particular points.
  605. Key findings already made

  606. In September 2018 (see §146(3)), the executive leadership team had been advised that there were "inconsistencies in current RWA calculations that will result (our emphasis) in a significant increase in risk weightings", because commercial mortgages had been weighted at 50% rather than 100%, and that KPMG had confirmed that this was correct.
  607. The RWA Report considered in October 2018 by the CRPAC and the ROC stated that (see §§174-185):
  608. (1)          CLIP loans had been wrongly risk-weighted in the past;

    (2)          when correctly risk-weighted, the related RWA increased by £574m;

    (3)          further adjustments to the figures were not expected to be material.

  609. No-one at the CRPAC meeting (including Mr Donaldson and Mr Arden) dissented from the position set out in the RWA Report, see §185).
  610. Ms Gillan told the Board that around £600m was the current estimate for the RWA adjustment; that figure was based on the analysis in the RWA Report, see §232.
  611. At the time of the Q3 Update (see §237 and §208):
  612. (1)          the Bank's commercial loans were £4,166m out of total net loans of £13,121m;

    (2)          the CLIP loans were £1,148m, and so constituted over 27% of the Bank's commercial loan portfolio;

    (3)          the estimated RWA adjustment for CLIP loans was £574m.

  613. The RWA adjustment figure for the CLIP loans reported to the PRA in January 2019 following Deloitte's review was £563m, see §249.
  614. The Authority's overall position on the knowledge issue

  615. Mr Stanley said it was plain that the Bank and the Applicants knew that risk-weighting the CLIP loans at 50% instead of 100% was incorrect, and they also knew at the time of the Q3 Update that correcting that error would materially change the RWA figure.
  616. He relied on the paper provided to the CRPAC in September 2018; on the passages from the RWA Report set out above; on Ms Gillan's confirmation that she had understood that corrections of the order of £570m were likely to be required for CLIP loans, and that any further adjustments were not expected to be material, see §202 and §208.
  617. Although Mr Stanley accepted that there was some uncertainty at the margin, he said that at the time of the Q3 Update, the Bank knew that the RWA figure of £7,398m had been materially misstated, because it included a figure for CLIP loans which was too low by over £500m. He also disagreed with all the reasons put forward by the Applicants as to why they did not have that knowledge. We set out his specific submissions under the relevant headings below.
  618. The Applicants' overall position on the knowledge issue

  619. The Applicants' position was that the Bank did not know there was a material error at the time of the Q3 Update, because of one or more of the following, each of which we consider in turn below.
  620. (1)          Uncertainty as to the rules on classification.

    (2)          Data issues, including the application of the SME supporting factor and the possibility that some of the CLIP loans were residential mortgages and so had been wrongly classified.

    (3)          Uncertainty as to whether the increased RWAs could be significantly reduced or eliminated by one or more "mitigants".

    Rules on classification

  621. The Applicants' position was that at the time of the Q3 Update, there was uncertainty about whether the Bank had to risk-weight CLIP loans at 50%. We agree with Mr Stanley that this is unsustainable. We have already made the following findings:
  622. (1)          In May 2018, KPMG had informed the Bank that it was wrong to use 50% risk weighting, see §100.

    (2)          On 17 July 2018, Ms Gillan informed the PRA that the 50% was inaccurate, see §109 and §110.

    (3)          There were then numerous internal communications to the same effect, culminating in the PowerPoint presentation headed "RWA Calculating Review - August 2018" together with the meeting with the PRA and the PSM letter.

    (4)          At some point before 11 September 2018, Deloitte confirmed to the Bank that the 50% risk weighting was wrong, see §136.

    (5)          There is no contemporaneous evidence to the effect there was any doubt as to the regulatory requirement, see our findings at §§184ff.

  623. As set out at §25, the PRA did allow a 50% risk-weighting where the condition in Rule 4.1 of the Credit Risk section of its Rulebook was met. That condition was that a bank's "annual average losses stemming from lending secured by mortgages on commercial  property in the UK did not exceed 0.5% of risk-weighted exposure amounts over a representative period". However, we have also found as a fact that the RWA Report included a slide saying that the Credit Risk and Analytics team had considered this exception, but that KPMG had confirmed that "they were not aware of any UK regulated firm that has demonstrated loss levels to be less than 0.50% over the relevant time horizon", see §§177(3)-177(4).
  624. There was no subsequent reference in the contemporaneous evidence to any consideration being given to this Rule, and Mr Arden's oral evidence was that Deloitte had told him it "would be very difficult to achieve" and that no time was spent considering it as a  possibility.
  625. On the basis of the foregoing, we find as facts that at the time of the Q3 Update:
  626. (1)          it was known that the exception to Rule 4.1 did not apply to the Bank; and

    (2)          there was no uncertainty as to the rules on classification which applied to the Bank's CLIP loans.

    Data issues

  627. A key part of the Applicants' case was that the Bank did not know there was a material error in the RWAs for the CLIP loans, because of the data issues within the Bank. We first consider the nature and extent of those issues, and then their materiality.
  628. Nature of the data issues

  629. We have already made the following findings of fact: 
  630. (1)          On 17 August 2018, Mr MacLean emailed Mr Arden about errors in "upstream data that have caused inaccuracies in reg and external reporting", see §§118-119.

    (2)          In November 2018, Deloitte reported that there were "multiple gaps in the controls framework at every stage of the process, from data sourcing through to report generation", see §245.

    (3)          The PRA subsequently confirmed that the Bank's "front-end data capture and systems did not allow it to capture all relevant information that it needed to accurately classify exposures and calculate risk weights for its lending portfolio, see §261.

  631. We thus make the further finding that at the time of the Q3 Update, there were data issues within the Bank's systems.
  632. We also find that those errors related to the categorisation of loans rather than to their existence or quantum or to the security held over them. That is clear from the following earlier findings:
  633. (1)          Mr MacLean referred to problems with regulatory and external reporting, see §118;

    (2)          Deloitte was engaged to review the classification issues, see §155;

    (3)          the Bank's internal team were looking at the individual loans "to understand which classification had been applied" and to compare that with the correct classification, see §158;

    (4)          Deloitte concluded that the data issues meant that it was "virtually impossible to evidence the integrity of the RWA calculation or the COREP reports", see §245; and

    (5)          the PRA similarly said that the Bank was "unable to accurately classify exposures and calculate risk weights for its lending portfolio", see §261(6).

    Extent of the data issues

  634. In oral evidence, Mr Donaldson and Mr Arden went further than the findings we have made above. Mr Donaldson said he "didn't believe the data" and "didn't trust the data"; he asked "how could I have any trust of the systems and controls within the organisation", and declared that "we had lost faith in the [Bank's internal] team". Mr Arden similarly said "I'm not sure at this stage that I believed anything that was being presented to me"; and "I didn't quite believe anything that the internal team were telling me".
  635. We agree with Mr Stanley that this evidence was not reflective of the position at the relevant time. If Mr Donaldson and/or Mr Arden had no confidence whatsoever in the Bank's financial data, this would have been reflected in the contemporaneous evidence. Instead, that evidence shows that the concerns were about the classification and risk-weighting of loans. 
  636. As Mr Stanley said, were we to be wrong in that finding, and instead were to accept Mr Donaldson's and Mr Arden's evidence that at the time of the Q3 Update they had a total lack of confidence in the Bank's systems and data, that would not help their case, because the Bank would then have been unable to publish any financial information to the market.
  637. Effect on materiality

  638. The key question in this context is whether the uncertainties over data categorisation were such that the materiality of the RWA adjustment required for the CLIP loans was unknown until the Deloitte review was completed. The Applicants' case was that the following key uncertainties existed, each of which we consider in turn below:
  639. (1)          the SME supporting factor; and

    (2)          the possible miscategorisation of loans as CLIP when they were secured on residential property.

    SME supporting factor

  640. We have already found as facts that in August 2018, the RWA adjustment for CLIP loans was quoted before any SME supporting factor, but that the figures provided in the subsequent RWA Report in October 2018 were explicitly stated to be net of SME adjustments, see §118 and §174.
  641. The figures in the RWA Report were arrived at after the Bank's staff had applied the KPMG decision trees to the loan books, and one of the factors in those decision trees was whether the SME supporting factor applied, see §159.
  642. Mr Donaldson said that by the time of the Q3 Update he did not know whether the SME adjustment was right or wrong, because he had been provided with some figures which included it, and some which did not, so he had "lost faith" in the data and that was why Deloitte had been instructed. We do not accept that evidence, because:
  643. (1)          the Bank's staff had applied the KPMG decision trees to calculate the figures for the October 2018 paper;

    (2)          it was no part of Deloitte's work to carry out a loan-level review to identify whether loans which benefitted from the SME factor had been correctly identified; and

    (3)          there is no contemporaneous evidence to the effect that anyone considered the SME factors used in the RWA Report were incorrect.

  644. Even if some loans which were entitled to the SME factor had been overlooked or wrongly classified, others might have been wrongly identified as benefitting from that factor, as Mr Donaldson agreed under cross-examination.  
  645. We therefore find as a fact that there was no material uncertainty at the time of the Q3 Update about the application of the SME factor, and we reject Mr Donaldson's evidence to the contrary.
  646. Residential property

  647. The other reason put forward as to why the CLIP loan estimate of £574m could not be relied upon was because some of the loans classified under that heading might have been residential loans, for which the correct risk weighting would have been 35%.
  648. However, the only contemporaneous evidence about this possibility was as follows:
  649. (1)          On 16 July 2018, Mr Richardson emailed Ms Gillan and Mr Somers about the need to change the risk-weighting for CLIP loans (see §108), and added this final paragraph:

    "One upside is that I think that many of the 'retail commercial mortgages' on CHL (London & Canberra) currently marked at 50% are loans to 'personal investment companies' for residential properties, so would qualify as retail under CRR Article 125 and hence 35%."

    (2)          On 17 August 2018, Mr MacLean emailed Mr Arden with a list of data issues, one of which was that "Regulatory Reporting team have also performed some sampling testing and have observed loans secured by residential properties being classified as 'Commercial Owner Occupier'", see §119.

    (3)          The presentation pack provided for the meeting between Mr Arden, Mr Somers, Ms Gillan and others on 3 September 2018 to discuss the RWA issue stated that "there is limited potential mitigation in trading book assets with residential property security but this would need case by case review", see §128.

    (4)          The minutes of that meeting said (see §129) that "potentially a small portion of this book that was actually secured on residential property and had actually been miscoded. Analysis suggests this is likely no greater than £20m of RWA."

  650. Mr Arden agreed in cross-examination that unless virtually all of the CLIP loans had been wrongly classified, and should instead have been residential loans, the result would still be a "substantial uplift" in the RWA number.
  651. We find as a fact on the basis of the contemporaneous evidence that the number of residential loans which were miscategorised as CLIP was small, in the region of £20m, and that this was immaterial in the context of the overall risk-weighting correction required for CLIP.
  652. Conclusion on data issues

  653. We find as a fact that at the time of the Q3 Update, Mr Donaldson, Mr Arden and through them the Bank knew that although there were data issues, they also knew that the effect of changing the risk-weighting of the CLIP loans was material.
  654. The mitigants overall

  655. It was also the Applicants' case that the RWA figure published as part of the Q3 Update was not known to be materially incorrect, because the Bank reasonably believed that the increased RWA required as a result of the CLIP loans would be offset by one or more of the following mitigants:
  656. (1)          The PRA allowing the Bank to move to an AIRB basis.

    (2)          The PRA exercising its discretion to allow a Pillar 2A offset.

    (3)          The PRA allowing the Bank to "phase in" the 100% risk-weighting for the CLIP loans.

  657. The Authority's position was that at the time of the Q3 Update, the Applicants did not expect that any of these mitigants would be permitted by the PRA, and that in any event, they would not have affected the RWAs for the CLIP loans.
  658. We agree with the Authority: even if the Applicants were correct about one or more of the mitigants, they would only have offset the extra capital required as a result of the increased RWAs for the CLIP loans. They would not have changed the RWAs themselves, so the RWA figures published at the time of the Q3 Update would still have been materially wrong. Nevertheless, as the mitigants formed a major part of the Applicants' case, we have considered each of them below.
  659. The AIRB application

  660. We have already found as facts that:
  661. (1)          the Bank wanted to move from standardised basis to IRB; it had developed a plan to do this over five years, beginning in February 2018 when it applied for residential mortgages to move onto that basis. The Bank's plan was to follow this in turn by buy-to-let mortgages, revolving unsecured mortgages, and finally by commercial loans, see §91. Thus, the application which had been made at the time of the Q3 Update related only to residential mortgages, not to CLIP loans.

    (2)          The Bank's own expectation, set out in its 2018 half year update, was that it would make the transition to IRB for residential mortgages in the second half of 2019, see §112.

    (3)          At the meeting with Mr Gunn on 6 September 2018, the PRA said that the Bank's application was "more complex/challenging than any other of our comparator banks" and they had "some genuine concern about the depth of [the Bank's] use test experience and also of the depth of experience of [its] modelling team". Although the PRA had not yet carried out detailed testing, they were "uneasy" and as a result "it may take [the Bank] longer to get approval, maybe considerably so", see §131(3).

    (4)          The PSM letter dated 10 September 2018 warned the Bank about speaking publicly about the application, saying that this "may risk setting false expectations over the timeframe for achieving IRB model approval" and adding that the PRA had "not yet reached any formal decision on [the Bank's] application but we remain concerned about the depth of your relevant modelling experience", see §132.

    (5)          At the meeting on 12 September 2018, the PRA said there "was no decision yet but the high level feedback so far is that the firm does not have the required level of use testing experience". When Ms Gillan and Mr Donaldson challenged that response, Mr Sutherland advised the Bank that "there would be an internal Challenge Session and there would be an opportunity for the firm to appeal the decision after that".

    (6)          The 2019 Budget paper which was discussed at the October Board meeting, said that the Bank was hoping that AIRB approval for residential mortgages would be being "received during Q3 2019", adding "we should have early sight of actual dates on AIRB before our full year results announcement".

  662. Mr Donaldson agreed in cross-examination that, given the PRA's response at the meeting on 12 September, the application for AIRB was not "looking particularly promising" and was in any event "some way in[to] the future".
  663. We find that the reasonable issuer in the Bank's position would not have considered that obtaining IRB for the residential mortgages would mitigate against the increased RWAs for CLIP loans, given:
  664. (1)          its own expectation, in the same Board meeting as the Q3 Update was approved, was that this would not be achieved before the third quarter of 2019, a full year ahead; and

    (2)           the warnings from the PRA about the Bank's unrealistic expectations and the reference in the meeting on 12 September to appealing the outcome of the decision-making process.

    Pillar 2A Offset

  665. As we explained at §16(2)(b), the PRA can agree that a bank carry out an "overs and unders" calculation so that any overcapitalisation under Pillar 1 can be used to reduce its Pillar 2A requirements, in accordance with the approach set out in PS22/17.
  666. Submissions

  667. Mr Jaffey submitted that the Bank considered there was "real uncertainty" about whether the Bank's capital would need to increase as a result of the CLIP loans issue, because that increase would have been more than cancelled out by a Pillar 2A offset.
  668. Mr Stanley accepted that Pillar 2A "would have been a genuine mitigant", but would not have reduced the RWA adjustment itself. In addition, it was also a possible future adjustment, which at best would have made the Bank's position "retrospectively correct".
  669. Findings of fact

  670. We have already found the following facts:
  671. (1)          On 12 July 2018, the Bank issued its half year update, which included a statement that its Pillar 2A requirement of 1.7% was "currently under review with the PRA" and the Bank "anticipated receiving capital relief from the PRA as part of the Pillar 2A Offset".

    (2)          On 10 September 2018, the Bank received the PSM letter, which stated that the PRA was "somewhat frustrated" that the Bank had referred publicly to the discussions on the Pillar 2A offset and said "we would not wish to see a repeat" of this type of public disclosure, see §132.

    (3)          The PSM letter also said that, as the Bank was remediating the classification of commercial risk weights, and this " will likely increase the Pillar 1 risk weighted assets", the PRA was "uncertain about the materiality of any prospective adjustment to [the Bank's] capital position". It continued:

    "Until this matter is satisfactorily resolved, and we have received  reassurance that Metro is holding sufficient Pillar 1 capital against its  commercial assets, we will not apply Policy Statement 22/17 which allows the offsetting of certain Pillar 2a variable add-ons.".

    (4)          At the meeting with the PRA on 12 September 2018, Mr Salmon agreed to a discussion about the offset once the mistake had been corrected, noting that the PRA would need to understand how the mistake had occurred and to be assured there was no read-across to other reporting aspects, see §137.

    (5)          In the October 2018 Board meeting, the Bank noted and took as read the 2019 budget estimate, which had included the benefit of the Pillar 2A offset of £50m as being actioned "by the end of Q1 2019", see §223.

    (6)          On 8 January 2019, Ms Gillan emailed the PRA to arrange a meeting to discuss the outcomes of the Deloitte work on the RWA issue, and added "we hope this will then allow us to revisit earlier conversations with Supervision about application of the Pillar 2A off-set approach", see §250.

  672. Mr Arden accepted in cross-examination that he "didn't believe at any stage that there was a promise" by the PRA that once the RWA issue had been remediated, the Bank would allow a Pillar 2A offset, and he also said that it "was not going to be on the table until [the Bank] had got the RWAs right". Mr Donaldson similarly accepted that discussions about the Pillar 2A offset could not begin until after the remediation of the RWAs.
  673. On the basis of all the above, we find as further facts that:
  674. (1)          the PRA did not promise that the Bank would obtain the Pillar 2A offset; and

    (2)          the PRA would not consider whether or not to allow the offset until after the RWA issue had been remediated.

  675. In relation to quantum, Mr Donaldson said in his witness statement that if the Bank had obtained the offset, it "would have reduced its overall capital by over £100 million; a reduction of an equivalent of well over £1 billion of RWAs". However, under cross- examination he was unable to explain the basis for that figure. In re-examination, his attention was drawn to the £50m in the 2019 Budget paper, and asked whether this helped him with the value of the proposed Pillar 2A offset. He responded: 
  676. "I think, therefore , it is the 100 plus million that we had in the work that we had done, and it must be offsetting the overs and unders with the worst case that we considered we would get post the Deloitte work"

  677. We understand him to mean that he thought the £50m in the 2019 Budget paper was a net figure after the RWA capital adjustment of some £40-50m had been deducted from his £100m estimate. However, that is not what the paper says. We prefer to rely on the contemporaneous evidence, and find as a fact that at the relevant time, the Bank's estimate of the effect of a Pillar 2A adjustment was £50m.
  678. Conclusion on Pillar 2A offset

  679. We find that the reasonable issuer in the position of the Bank would not have published an RWA figure which it knew to be incorrect, on the basis that the effect of the CLIP loan adjustment would have been cancelled out if the PRA had granted a Pillar 2A offset. We come to that finding because the PRA had made remediating the RWAs a pre-condition before the offset could be considered, and in any event had not promised it would grant the offset after that remediation had taken place.
  680. Phasing in

  681. We have already found as facts that on 18 January 2019, Mr Arden discussed with Mr Donaldson, Mr Lane and others "the possibility" that the PRA would allow the Bank "to ringfence £300m of RWAs for sale...and allow the remaining £600m to be phased in over a period", and that Mr Arden had noted that this "potentially" would mean that the Bank did not have to change its RWA figures. On 22 January 2019, the PRA rejected that possibility.
  682. Mr Donaldson's evidence was that the idea of phasing in the extra RWAs was put forward by Deloitte; when asked in cross-examination when it was considered, he initially avoided giving a straightforward answer, but finally agreed that it was not discussed until after the Bank had received the results of the Deloitte review, and we so find.
  683. Mr Jaffey invited the Tribunal to find that, had the Bank made this request at an earlier time, it would have been refused, and that this finding should be made as an "adverse inference", because the Authority could have called witnesses from the PRA but did not do so. We do not need to make an adverse inference to find on the balance of probabilities that had the Bank suggested phasing-in at an earlier stage, it would have been rejected. That is because the PSM letter said the PRA was " uncertain about the materiality of any prospective adjustment to your capital position " and required Mr Arden to provide an attestation.
  684. However, we add that:
  685. (1)          the question is hypothetical: the possibility was not raised sooner because it was first suggested by Deloitte after the conclusion of their review; and

    (2)          had the possibility been considered earlier, it could have been raised with the PRA even though it could not have been implemented until after attested figures were provided.

  686. Mr Jaffey also asked the Tribunal to make adverse inferences that (a) the PRA had received the request at "the appropriate time" and (b) the PRA had the power to grant the request but it was refused. However, neither point is relevant to the issue we have to decide, which is whether it would be reasonable for an issuer to provide RWA figures to the market which it knew to be incorrect, on the basis that the PRA might in the future agree to phase in the RWA adjustment. We have no hesitation in finding that the answer to that question is plainly no. A reasonable issuer would not have decided in October 2018 that it could publish figures it knew to be wrong, on the basis that the PRA might allow the correction to be phased in, when that possibility had never been raised with the PRA.
  687. PRA discretion

  688. On 18 January 2019, during a meeting with Mr Arden, Ms Gillan and Mr Lane, Mr Donaldson said that the PRA had "discretion to make alternative judgements/solutions in private conversations", see §253, Mr Arden's oral evidence was that the PRA might have disapplied the 100% risk-weighting requirement for the CLIP mortgages so as to allow the Bank to revert to the 50% in the CRR. In closing, Mr Jaffey referred to a published list of waivers permitted by the PRA and the Authority (but without providing a related document) as support for the submission that until January 2019, the Bank did not know the PRA would not exercise that wide discretion in the Bank's favour.
  689. However, under cross-examination, Mr Arden had conceded that no-one in the Bank, no-one at the PRA and no external adviser, had ever suggested at the time of the Q3 Update that a waiver might be possible, and that this was first raised "later in the year". We therefore find as a fact that the possibility that the PRA might have issued a waiver so the Bank did not have to follow its published policy on RWAs for CLIP loans was not considered before or at the time of the Q3 Update. 
  690. A reasonable issuer would not have decided in October 2018 that it could publish figures it knew to be wrong, on the basis that the PRA might have agreed to disapply its published requirements for the risk-weighting of CLIP loans, when that possibility too had never been raised with the PRA.
  691. Taking all the above into account

  692. It was part of the Applicants' case that the above factors taken together meant that the Bank did not know, at the time of the Q3 Update, that the error in the RWA figure provided to the market was materially incorrect. We have considered each of the factors individually, and find that none provides a basis for a reasonable issuer to come to that conclusion; the position is the same if the factors are taken together.
  693. Overall conclusion on the Knowledge Issue

  694. For the reasons set out above, we find that a reasonable issuer in the position of the Bank would have known that the error in the RWA amount caused by the incorrect weighting of the CLIP loans was material.
  695. The PBTL Loans

  696. It was not part of the Authority's case that the Bank had breached LR 1.3.3R by publishing an RWA figure which included the PBTL loans on the previous basis, ie without any reference to, or adjustment for, a change to the risk-weighting of those loans. The Applicants submitted that it would have been misleading for the Bank to disclose the issue with the CLIP loans but not that with the PBTL loans, and thus that the Bank had been right to take the same approach to both types of loan.
  697. We begin our consideration of this point by summarising the findings we have already made about the PBTL loans; we have also made additional findings: these begin at §424 below.
  698. Findings of fact

  699. On 18 May 2018, Ms Iovino copied Mr Arden on an email which said that "some potential gaps in the RWA calculation" had been identified following a meeting with KPMG. She attached a paper written by the Credit Risk and Analytics team which said that loans relating to Housing Association Properties; Houses in Multiple Occupation and Student Accommodation should all be reclassified from residential to commercial, and so given a risk weighting of 100% rather than 35%.
  700. On 3 August 2018, Mr Stokes emailed Mr Donaldson to say that where a PBTL loan was for over five properties or was a HMO, it should be risk-weighted at 100%, but that the Bank didn't capture the number of properties on which the loan had been made, or whether it was a HMO.
  701. On 6 August 2018, it was agreed that the Bank would undertake a policy by policy review in order to "understand the gap on PBTL".
  702. On 17 August 2018, Mr MacLean emailed Mr Arden, saying that there was "an outstanding question re £1.7b of PBTL (whether it is in fact PBTL). The credit team are testing a sample of 170 contracts to test our expected level of accuracy".
  703. On 17 August 2018, Mr MacLean informed Mr Arden that the Credit Risk and Regulatory Reporting teams had performed some sampling testing of PBTL, but that as there was no PBTL flag in the Bank's system it was "impossible to distinguish between a BTL and a PBTL loans in order to ascertain the PBTL impact on RWA" and this would remain the case until the Bank was able to "invest resources and extensive staff time to review the loan documents to ascertain what proportion of the current BTL risk weighted at 35% that will need to be reclassified as PBTL".
  704. On 24 August 2018, Mr Somers sent Mr Arden a PowerPoint presentation, which said stratified random sampling "reveals that circa 37% of the PBTL loans required recategorisation" and that the sampling had taken place as the Bank does not "capture planning use class or the number of holders of short hold tenancies as data items"; and that the sample of 100 out of 733 loans had been selected on the basis of:
  705. (1)          loans which were more likely to be "high risk" of being PBTL including where they had product codes referring to "Comm_BL" or "Comm_Mtg", where they were secured on something other than the property itself; where there was a shared title; and loans of an average property value of over £1m; and

    (2)          where the loan was for more than £1m, as this indicated that a large RWA movement could occur.

  706. On 17 September 2018, a paper about PBTL classification was considered at the CRPAC meeting. It said that, based on the recent sampling exercise, the best estimate was that 35% of the current PBTL book should instead be risk-weighted at 100%, and this would increase the RWAs by £269 million.
  707. The paper also said that the PRA's guidance was that 35% is required for owner occupied residential property and "certain Buy to Let" properties; that the phrase "certain BTL" was not further defined, so there was nothing to stop the Bank including for example multifamily or HMO properties, the Credit Risk team had "good reason to believe that this is not within the spirit of what the PRA intended".
  708. The Bank did not inform the PRA that there were any errors in the risk-weighting of the PBTL loans, either when emails were exchanged with Mr Sutherland or at the meeting which took place in September 2018, or at any other point before the communications and meeting in January 2019.
  709. The RWA Report considered at the October 2018 CRPAC and Audit Committees said that risk-weighting for the miscategorised PBTL in T24 was estimated to result in additional RWAs of £37 million.
  710. The difference between the £269m estimate in September and the £37m in October was because the former included both the T24 loan book and the Pepper loan book, and the latter excluded Pepper. This was because the Bank had taken the view that the regulations only required the higher risk-weighting to be applied for Pillar 2A purposes, and as the Bank was still operating on the standardised basis, it could decide for itself whether to apply the higher risk-weighting, and it limited the change to the PBTL loans in the T24 book.
  711. On 20 December 2018, Deloitte's final report said that correctly classifying the PBTL loans was estimated to result in an additional £312m of RWAs. Deloitte therefore took the view that all the PBTL loans needed to be risk-weighted at 100%, and the Bank did not have the option to use the lower risk-weighting, even while it continued to use the standardised basis.
  712. As is clear from the foregoing, there were different views on whether 100% risk weighting was required for PBTL loans under the Bank's current (standardised) approach, and this was echoed in the witness statements and oral evidence.
  713. (1)          Mr Somers' position was that 100% risk-weighting was not required by the regulations unless or until the Bank moved from the standardised approach, and that it wasn't therefore necessary for the Bank to make such a large correction in January 2019. 

    (2)          Mr Sutherland's witness statement repeated the PRA rule that the Bank should have applied the 100% risk weighting to "certain PBTL Loans", but without further details. In his oral evidence, he said that when the Bank made the January announcement "people were talking about it in the [PRA's] office because there was a debate about the classification of professional buy to let versus normal buy to let."

    (3)          The Authority accepted that (in contrast to its position on the CLIP loans) there was genuine uncertainty about the regulatory requirements for PBTL loans.

    (4)          Neither party led any expert evidence on the issue.

  714. We find as a fact on the basis of Mr Somers' evidence, which was not inconsistent with that given by Mr Sutherland, that at least at the time of the Q3 Update, there was genuine uncertainty as to which of the PBTL loans had to be reweighted at 100% and when the reweighting had to take place.
  715. As is clear from the above findings:
  716. (1)          The PRA's rules were not definitive, referring to "certain PBTL loans".

    (2)          As a result, there was genuine uncertainty as to whether 100% risk-weighting was required for the Bank's PBTL loans.

    (3)          The Bank's system did not include a flag to identify PBTL loans.

    (4)          The Bank's estimate was based on sampling of individual loans.

    (5)          At the time of the Q3 Update, the best estimate was that an extra £37m of RWAs was required.

    (6)          The total RWAs in that Q3 Update was £7,398 million, so adding £37m would have increased that figure by 5%.

    (7)          Significant work would be required to obtain a more exact figure.

    (8)          After the Bank's teams had worked with Deloitte between October 2018 and January 2019, and based on Deloitte's view as to the regulatory requirement, the estimate increased to £312m, over eight times greater.

    Submissions and the Tribunal's view

  717. Mr Jaffey relied on the fact that the RWA number included in the Q3 Update included an incorrect figure for both types of loan, and submitted that the Authority's different approach as between the CLIP loans and the PBTL loans could not be justified. In other words, if the Bank acted reasonably in respect of the PBTL loans, the same was true for CLIP loans.
  718. However, we agree with the Authority that the position was different. In relation to the PBTL loans there was genuine regulatory uncertainty as to whether any adjustment was required, together with significant doubt as to the quantum. The best estimate of the adjustment, if one was required, was £37m, only 6% of the £574m for the CLIP loans.
  719. We thus do not agree with Mr Jaffey that the Bank acted reasonably because it took the same approach to both types of loan.
  720. Whether the alternatives were unreasonable

  721. The Authority had decided that the Bank breached LR 1.3.3R by publishing an RWA number which it knew was materially incorrect, and that it should instead have provided no figure at all, an estimate or a qualified figure.
  722. The Applicants' position

  723. The Applicants' position was that those three alternatives were unreasonable, and that the Bank's decision to use the existing basis was therefore the only reasonable option. In relation to the three possibilities, Mr Jaffey made the following submissions.
  724. (1)          The Bank could not omit the RWA figure from its Q3 Update, because it had always been included, and leaving it out would have caused significant market disruption.

    (2)          Had the Bank provided an estimate, this would have had to be adjusted again once Deloitte had done its work, in particular, to include the adjustment for the PBTL loans, and that too would have caused market disruption.

    (3)          He described the third option as follows:

    "explaining that the Bank had discovered an issue with its calculation of RWA and therefore the RWA figure in the Q3 Announcement was wrong (or might be wrong), even though the Bank was unable to quantify the extent of the error, had not yet completed work to ascertain the correct position and could therefore not confirm its capital position to the market. Such an unprecedented, unquantified and unclear announcement would likely have been catastrophic, as well as being unnecessary in law [and] would have had serious prudential consequences..."

  725. The Applicants also relied on Mr Arden's evidence that:
  726. "On 26 January 2023, I informally met with Mr Charles Woods of the PRA. Mr Woods was at the time of the Q3 Announcement (and remains) a Director of the PRA, holding the position of Head of UK Banks Supervision, and I understand is the person with overall responsibility for the prudential regulation of the Bank. At our meeting, I raised the FCA's criticism that the Bank should have qualified its capital position in the Q3 Announcement. In response, Mr Woods audibly laughed and said that the PRA 'would have had something to say about that'."

  727. Mr Jaffey submitted that the Tribunal should make an adverse inference from the fact that the Authority had not called Mr Woods, and that "Mr Arden's account is therefore undisputed. Mr Woods described the PRA's position to Mr Arden accurately".  
  728. The Authority's submissions

  729. The Authority's position was that LR 1.3.3R does not require issuers to publish accurate and complete information only if it is reasonable to do so, but instead requires them to take reasonable care to "ensure" that published information is accurate and complete, and this is always the position. Mr Stanley said that the Rule confers no discretion on issuers to decide that it would be in their best interests to mislead the market, or that it would be in the public interest to do so. Moreover, the Bank:
  730. "...did not simply provide a RWA figure that was known to be incorrect, and did not merely omit the important information that the figure was under review and realistically bound to be revised upwards: it positively invited the reader to draw favourable conclusions about Metro Bank's capital ratios from that figure ('remain robust'), and did so where Metro Bank's capital ratios were regarded as commercially important."

  731. Mr Stanley asked the Tribunal to reject Mr Jaffey's submission that making a qualified announcement was "unprecedented"; his skeleton includes the example of the Q3 Update issued 7 November 2024 by Wood Group Plc, which said that the Group had commissioned an independent review to focus on accounting, governance, and controls, including the  question of whether any prior year restatement may be required, and was unable to give an answer at that stage. The announcement also said that "The results presented in this trading update, and our full year outlook, are before any potential impacts from the independent review". In addition, the Bundle contained extracts from statements made by other listed companies which had been qualified, such as one by Tesco Group plc in 2014 which read (emphasis added):
  732. "On the basis of preliminary investigations into the UK food business, the Board believes that the guidance issued on 29 August 2014 for the Group profits for the six months to 23 August 2014 was overstated by an estimated £250m. Some of this impact includes in-year timing differences. Work is ongoing to establish the extent of these issues and what impact they will have on the full year."

    The Tribunal's view

  733. We agree with the Authority that if a listed company decides it will report figures quarterly, it must report accurate and reliable figures quarterly. If, for any reason, including internal control or data failings, it is unable to report accurate figures, as Mr Stanley said, it has to "come clean about that". It does not have the choice to report a figure that is known to be inaccurate or not believed to be reliable, without qualification. It is irrelevant that qualification may be embarrassing or commercially inconvenient.
  734. Although the example provided in Mr Stanley's skeleton argument post-dated the relevant period, and the other examples in the Bundle related to disclosures under the MAR, we nevertheless agree with the Authority that listed companies did sometimes qualify their announcements. Moreover, the Bank never assessed whether it should issue a qualification to the RWA figure, see §218(2); in other words, the Bank did not reject this option on the basis that it might mislead investors, the Bank simply did not consider it at all.
  735. In relation to Mr Arden's hearsay evidence about his conversation with Mr Woods, we agree with Mr Jaffey that both the fact of that conversation and Mr Arden's recollection of what Mr Woods said was unchallenged.
  736. However, it does not follow that Mr Woods "described the PRA's position to Mr Arden accurately". We also take into account that:
  737. (1)          the conversation took place in January 2023, over four years after the Q3 Update;

    (2)          as Mr Sutherland said, "the market disclosure regime was not within the remit of the PRA"; and

    (3)          the PRA were not asked to give a view at the time of the Q3 Update as to how the Bank should comply with the Listing Rules.

  738. We decline to make a finding of fact as to the PRA's view in October 2018 as to the approach the Bank should take to complying with LR 1.3.3R.
  739. Reliance on the board and the Committees

  740. The Applicants' case was that the Bank had also acted reasonably because they had followed proper governance processes involving the Board, the Audit Committee, the Disclosure Committee and the CRPAC. Mr Jaffey submitted that the Tribunal should draw an adverse inference from the failure of the Authority to call Mr Hill or any of the NEDs to give evidence, and asked us to find that "the Board were given a full account in relation to the RWA issue".
  741. In Efobi cited at §62 above, Lord Leggatt said that in deciding whether or not to make such an inference, it was relevant "what other relevant evidence there was bearing on the point(s) on which the witness could potentially have given relevant evidence". In this case, there is extensive contemporaneous evidence as to what the Board were told, and we are able to decide this question on the basis of the facts we have already found, as summarised below.
  742. Findings of fact

  743. Our findings of fact about the membership of the Board and the main Committees are at §85ff. In particular:
  744. (1)          Mr Arden and Mr Donaldson were the only executive directors of the Bank;

    (2)          Most of the NEDs were experienced in financial services and many had worked for other listed entities.

    (3)          The Audit Committee was chaired by Mr Bernau; Mr Carby, Mr Lockhart and Mr Snyder were members. Mr Donaldson, Mr Arden and Ms Gillan attended as guests.

    (4)          The CRPAC was not attended by any Board members.

    (5)          The members of the Disclosure Committee were Mr Donaldson, Mr Arden and Ms James.

    September

  745. The September Audit Committee was informed by Ms Gillan that "inconsistencies in the reporting of commercial RWAs, specifically on commercial lending assets secured on  property had been identified". The Committee was also informed, wrongly, that the regulatory position was "ambiguous", although it was also told that it had been "accepted that the error should be remediated", see §148.
  746. At the September NEDs meeting, Mr Gunn reported on the meeting with the PRA and on the PSM letter, and Mr Lockhart stated that "the RWA error amounted to a £40m capital amount", see §150.
  747. The September Board meeting took place the same afternoon (see §§151-153). The Board was provided with:
  748. (1)          Ms Gillan's report, which said (in a passage not discussed during the meeting) that the RWA reclassification "could" increase regulatory capital by around £40m, but did not include the estimated increase to the RWAs; and

    (2)          the PSM letter, which was "noted"; the minutes record that the Board was told that the classification of commercial risk weights had been discussed with the PRA; that remediation was taking place and that "once fully remediated, the PRA had confirmed that they would be able to reconsider the application of Pillar 2 Offset".

    October Audit Committee

  749. The October Audit Committee was provided with the text of Ms Roberts's email about the meeting with Mr Lane, see §163. However, we have already found that this text suffered from the following errors and omissions (see §§316-318):
  750. (1)          there were "regulatory interpretation issues" about the classification of the loans in question, when there was no doubt about the regulatory interpretation;

    (2)          reweighting had been raised as a "potential issue" by the PRA, when the PRA had been told by the Bank that it had made an error;

    (3)          the estimate of £600m was based on sampling, when it was not;

    (4)          there was an "ongoing dialogue" with the PRA about the issue, when there was no dialogue: instead, the Bank had informed the PRA that it was remediating the issue.

  751. The Audit Committee was also provided with a "Q3 Update on accounting, reporting and control matters" (see §§199-201), which stated in the opening section (our emphasis):
  752. "our Q3 Trading Update has been prepared on a consistent basis to previous quarterly trading updates and there are no significant accounting or reporting matters to bring to the Committee's attention."  

  753. That report did include a section headed "Reporting on Risk Weighted Assets", but this stated that the Bank had "commenced a piece of work to review our calculation of Risk Weighted Asset reporting to the regulator"; that the work was expected to be completed by the end of the year, and that the Committee would be advised as to "any corrections which will be reported to the regulator and any revisions required to our reporting methodologies". It did not give a figure for the estimated quantum of the RWA errors. The related minutes instead record (our emphasis): 
  754. "It was expected that the work would be complete by year end, the Committee would be updated on any corrections required to be reported to the regulator and any revisions required to reporting methodologies...The  impact of incorrect reporting on the balance sheet was not insignificant."

    October ROC meeting

  755. The ROC meeting was the first since July. The committee was provided with:
  756. (1)           the Chief Risk Officer's Report, the summary page of which contained a brief reference to the fact that "standardised RWAs continue to be above threshold", while the more detailed slide unhelpfully conflated the reference to the historic issue of reweighting CRE assets with the new CLIP loan problem. The same slide also stated (again, our emphasis) that "this metric could increase by around 10 per cent (circa 40 million of capital) upon completion of the asset classification project", but without more explanation. Although there was a discussion at the meeting about other elements of the report, there are no minuted comments about this item, see §§204-206;  

    (2)          the RWA Report, which included slides which said that RWAs were expected to increase by an estimated £642m including £574m relating to CLIP loans, and that "any further adjustments to calculations are not expected to be material". However, the minutes record that the Committee was also told that "the commercial business was now largely correctly risk weighted" and that the wrong risk weighting "had been driven by a lack of clarity and conflicting rules in the regulatory publications". Neither of those statements was correct, see §§210-211; and

    (3)          the report on business and commercial lending as previously circulated to the CRPAC, see §186. The Executive Summary said (our emphasis) that the Bank had "experienced some issues with asset classification for regulatory reporting of RWAs"; that this was "not unique to Metro Bank" and would be "resolved over a two-year period". Although the minutes show that commercial lending was discussed, there was no mention of RWAs.

    October Board meeting

  757. The first substantive item on the agenda for the October Board meeting was a governance update from Linklaters, but neither RWAs nor Mr Lane's advice formed part of the discussion, see §215. The second item was Mr Bernau's summary of the Audit Committee meeting which had taken place that morning: Mr Bernau said that the Committee recommended the Q3 Update to the Board. However, as already noted above, the Audit Committee:
  758. (1)          had been provided with incorrect and incomplete information about the meeting with Mr Lane;

    (2)          had been told that there were "no significant accounting or reporting matters to bring to the Committee's attention";

    (3)          references to the RWA issue were limited to stating that work was in process and that the Committee "would be updated on any corrections required"; and

    (4)          no quantum was given, the only relevant reference was the statement that "the  impact of incorrect reporting on the balance sheet was not insignificant".

  759. The Q3 Update was approved by the Board without reference to the RWA issue, see §219.
  760. The next relevant item was the 2019 Budget Paper. This listed four items which eroded the capital buffer, one of which was "a review of Commercial lending risk weighted assets". However, the discussion at the meeting did not expand on that statement; the Board was not provided with the quantum of the error, and no reference was made to Linklaters's advice, see §228.
  761. That was followed by the Chief Risk Officer's Report which had also been provided to the ROC. As noted above, it said (our emphasis) that "this metric could increase by around 10 per cent (circa 40 million of capital) upon completion of the asset classification project". In the meeting, the Board were told that "management are presently remediating the classification of commercial risk weights in the lending portfolio", with Deloitte providing "a review of RWA calculations"; they were also told by Ms Gillan that the "current estimate" for the RWAs for commercial mortgages was around £600m, see §232.
  762. The position of the parties

  763. When the Authority issued the Warning Notices to Mr Donaldson and Mr Arden on 17 January 2022, these stated that neither the Audit Committee nor the Board had been informed of the estimated quantum of the RWA error relating to the CLIP loans. However, the Authority later accepted Ms Gillan's evidence that the Board was told during the October 2018 Board meeting that the current estimate was around £600m.
  764. The Applicants' position was thus that "the Authority now accepts that the Q3 announcement was approved by the Bank in accordance with its usual governance processes", and that this showed that the Bank had acted reasonably.
  765. However, the Authority's amended position was that "the information provided to the Board and Audit Committee does not support any inference that the Bank complied with its obligations under LR 1.3.3R", because that information "downplayed the CLIP Loan error and the adjustment to the Bank's RWA that would be necessary to correct it", and because the Board was only told of the £600m figure after it had already approved the Q3 Update.
  766. The Tribunal's view

  767. We agree with Mr Stanley that the information provided to the Audit Committee and the Board "downplayed the CLIP Loan error and the adjustment to the Bank's RWA that would be necessary to correct it". From the findings of fact summarised at §445ff above, we highlight in particular the following (although other relevant points are also contained in those findings). We consider that the position is the same in relation to the ROC.
  768. The Audit Committee

  769. The Audit Committee was told in September that there were "inconsistencies" in the reporting of RWAs, and also wrongly informed that the regulatory position was "ambiguous". In October, the Committee was provided with:
  770. (1)          the text of Ms Roberts's email about the meeting with Mr Lane, this incorrectly stated that there were "regulatory interpretation issues" about the classification; that the £600m was based on sampling and that there was "ongoing dialogue" with the PRA about this "potential issue"; and

    (2)          the "Q3 Update on accounting, reporting and control matters" which said there were "no significant accounting or reporting matters to bring to the Committee's attention"; that work to review the calculations of RWAs would take the rest of the year, and the Committee would be "updated on any corrections required" albeit that "the impact of incorrect reporting on the balance sheet was not insignificant".

  771. Thus, although the Committee was told that the estimate was £600m, they were also told that this was based on sampling; that there were regulatory uncertainties and that the quantum would not be known until the end of the year. The first two of those statements was incorrect, and although it is true that further work was continuing on the quantum, there was no doubt at the time of the October meeting that it was of the order of £574m because further adjustments to calculations were "not expected to be material".
  772. The Board

  773. The NEDs were told in September that "the RWA error amounted to a £40m capital amount", but at the subsequent Board meeting they were provided with Ms Gillan's report, which said that the RWA reclassification "could" increase regulatory capital by around £40m, and did not include the change to the risk-weighting itself.
  774. The October meeting of the ROC was attended by five of the ten NEDs. The Committee had been provided with the RWA Report, which included the slides which said that RWAs were expected to increase by an estimated £642m including £574m relating to CLIP loans, and that "any further adjustments to calculations are not expected to be material". However, the ROC was also told, incorrectly, that "the commercial business was now largely correctly risk weighted". It is thus not surprising that these NEDs went to the subsequent Board meeting thinking that there was no continuing issue.
  775. At the October Board meeting, in reliance on the information provided to the Audit Committee, Mr Bernau recommended the Q3 Update to the Board. However, that information was incomplete and in important respects also incorrect. The Board was provided with the 2019 Budget Paper and the Chief Risk Officer's report, but neither gave a clear explanation of the CLIP loan error or of its materiality. Although the £600m figure was provided orally by Ms Gillan, this was after the Q3 Update had been approved, and given in the context of the incorrect and incomplete information already provided to Board members.
  776. We therefore reject the Applicants' submission that the Bank acted reasonably because proper internal governance procedures had been followed. Some of the information provided at the relevant meetings was partial, some was wrong, and the overall impression was misleading. A reasonable issuer in the position of the Bank would have ensured, before the Q3 Update was considered by the Board, that the position as set out in the RWA Report was explained and understood.
  777. Reliance on Ms James

  778. It was also the Applicants' position that the Bank had acted reasonably because it had a "highly experienced General Counsel...who can be expected to have advised the Committee if the existing legal advice was inadequate or insufficient". Mr Jaffey submitted that as the Authority did not call Ms James as a witness, the Tribunal should make an adverse inference and:
  779. "...proceed on the basis that Ms James had no concerns regarding the scope of Linklaters' advice regarding disclosure, the manner in which they had been instructed or the application of the advice (as it had been reported to them) to the decisions that the Bank was taking and that Mr Arden  proceeded on that basis."

  780. In correspondence before the hearing, Baker & McKenzie, on behalf of the Applicants, asked the Authority to call various people as witnesses on the basis that the firm considered it "highly unlikely that [they] would be willing voluntarily to co-operate with the Applicants". None of the related letters in the Bundle contain a request for the Authority to call Ms James. The Applicants have not anywhere suggested that she would also be "highly unlikely" to attend voluntarily as a witness: we note that others who formerly worked at the Bank - Mr Brierley, Ms Gillan and Ms Roberts - all gave evidence on behalf of the Applicants.
  781. As we said at §315(4), there is no property in a witness, and had the Applicants wanted to call Ms James, they could have done so. If she had been unwilling, and if the Applicants considered her evidence to be important, they could have asked the Tribunal for a witness summons. It is not enough to rely on the burden of proof, and say that if the party with that burden does not call a witness, the Tribunal should infer that her evidence would have helped the other party. In any event, we have other relevant evidence which in our judgment is sufficient to make findings of fact on this issue, see the citation from Efobi at §62. 
  782. Findings of fact

  783. We have already made the following findings about Ms James's role in the Bank in the relevant period.
  784. (1)          She was a member of the Disclosure Committee, see §89.

    (2)          She was copied on the email from Ms Roberts which summarised the meeting with Mr Lane, see §162.

    (3)          She attended the meeting of the Disclosure Committee which took place on 16 October 2018, the other attendees being Mr Arden and Ms Roberts. The Committee considered the text of the email from Ms Roberts, and "understood that, once fixed, further consideration would be given to whether a market announcement was required"; the action recorded was "continue to monitor", see §§163-164.

    (4)          She was copied on Ms Roberts's note of the subsequent meeting with Mr Lane on 11 January 2019, and replied saying "Noted", see §251.

    (5)          She was not a member of CRPAC, the Audit Committee, or the Risk Oversight Committee, and did not attend any of the meetings of those Committees during the relevant period as a guest; she was not a member of the Board and she did not attend either the September or October Board meetings as a guest.

  785. In their witness evidence, both Applicants refer to Ms James' attendance at the Disclosure Committee meeting, and Mr Arden also says that she "relied heavily on Linklaters for legal support" but neither Applicant makes any other substantive references to her involvement or role. Ms Gillan's evidence was that the regular communications between Ms Roberts and Linklaters was "frustrating for Sally-Ann James (General Counsel) as she had little control over the fees that built up as a result".
  786. Submissions

  787. Mr Jaffey submitted that "where, as here, a listed firm's General Counsel is fully involved in identifying issues, taking advice and ensuring that it is acted on, reasonable care will have been taken". He also relied on Ms James having seen Ms Roberts's email summarising the meeting with Mr Lane, saying:
  788. "She did not suggest that Linklaters had been inadequately or incompletely briefed. She did not suggest that the external legal advice that had been  obtained was insufficient, and did not suggest that more advice was needed."

  789. Mr Stanley's position was that Ms James' role did not change the Authority's position.
  790. Discussion

  791. As is clear from the findings of fact set out above, Ms James saw the email from Ms Roberts summarising the meeting with Mr Lane. But she did not attend any of the Committees at which various information about the RWAs was provided, and she did not attend the Board meetings. She was not copied on, or in contact with, Mr Sutherland at the PRA. There was no evidence, either written or oral, as to Ms James having been given any  information about the RWA issue after she received a copy of Ms Roberts' email.
  792. It is therefore unsurprising that Mr Jaffey's submissions placed significant weight on that email. However, as we have already found, it contained the following incorrect statements, see §299ff and the summary at §317:
  793. (1)          That there were "regulatory interpretation issues" about the RWAs, when the Bank knew and accepted that the incorrect risk-weighting had been used.

    (2)          The PRA had raised the risk-weighting "as a potential issue", when the PRA had been told by the Bank that it had made an error.

    (3)          That current estimates of £600m were "based on sampling", when they were not.

    (4)          There was "ongoing dialogue" with the PRA, when this was not the position.

  794. The email did not mention the Q3 Update, or the Listing Rules, or that the Bank had misapplied the requirement that CLIP loans be risk weighted at 100%: it instead referred only to "a problem with the risk weight classification of some commercial assets".
  795. The final part of the email said
  796. "Linklaters concurred with our view that it is neither specific or material information at this point and was in the ordinary course of ongoing dialogue with the regulator over a complex issue and no market announcement was necessary at this point." 

  797. This reflected the requirements in the MAR that disclosure must be made if information is "sufficiently precise" and would be likely to have "a significant effect" on the price of the shares, and that "an issuer shall inform the public as soon as possible of inside information which directly concerns that issuer". We have found as a fact (see §340) that the purpose of the meeting was not to give the Bank advice about the Q3 Update, but was instead whether immediate disclosure was required under the MAR. Although Ms Gillan did not see the email, she too understood that Linklaters' advice was directed to the question of whether there was inside information that needed to be disclosed.
  798. Given the above, we disagree with Mr Jaffey that this was a case where Ms James was "fully involved in identifying issues, taking advice and ensuring that it is acted on". Instead, Ms James was only provided with Ms Roberts's email, which contained significant inaccuracies, and the reasonable reader of that email would have understood Linklaters to have advised that there was no need for an immediate market announcement. We thus reject the Applicants' submission that the Bank acted reasonably in relation to the Q3 Update because it relied on Ms James.
  799. Overall conclusion on Issue one

  800. For the reasons set out above, we find that the Bank breached LR 1.3.3R because it failed to take reasonable care to ensure that the Q3 Update was not false or misleading and did not omit anything likely to affect the import of the information contained within it. Instead, the Q3 Update contained an unqualified statement of the Bank's RWA, and the capital ratios based on it, at a time when the Bank knew that a material error had been made in relation to the CLIP loans, but it did not inform the market about that error.
  801. ISSUE TWO: KNOWINGLY CONCERNED

  802. The second main issue was whether Mr Donaldson and/or Mr Arden were "knowingly concerned" in the Bank's breach.
  803. The legal principles

  804. There was a fundamental disagreement between the parties as to what was required for a person to be "knowingly concerned" in a breach. We first set out the law, followed by the parties' submissions and our view.
  805. The statutory provisions

  806. FSMA s 73A is headed "Part 6 Rules", and begins as follows
  807. "(1) The  FCA  may make rules ('Part 6 rules') for the purposes of this Part.

     (2) Provisions of Part 6 rules expressed to relate to the official list are referred to in this Part as 'listing rules'."

  808.  FSMA s 91 is headed "penalties for breach of Part 6 rules". Part 6 sets out the provisions relating to listing, and begins:
  809. "(1) If the FCA considers that–

    (a) an issuer of listed securities...

    has contravened any provision of listing rules, it may impose on him a penalty of such amount as it considers appropriate.

    ...

    (2) If, in the case of a contravention by a person referred to in subsection (1) ('P')...the FCA considers that another person who was at the material time a director of P was knowingly concerned in the contravention, it may impose upon him a penalty of such amount as it considers appropriate."

  810. The Authority decided that the Applicants came within FSMA s 91 because they were "knowingly concerned" in the Bank's breach of LR 1.3.3R.
  811. We were not taken to any case in which courts or tribunals have considered the meaning of the term "knowingly concerned" in FSMA s 91, but the same phrase occurs in other parts of that Act. Those sections include s 382, which is headed "restitution orders" and so far as relevant provides as follows:
  812. "(1) The court may, on the application of the appropriate regulator...make an order under subsection (2) if it is satisfied that a person has contravened a relevant requirement, or been knowingly concerned in the contravention of such a requirement, and

    (a) that profits have accrued to him as a result of the contravention; or

    (b) that one or more persons have suffered loss or been otherwise  adversely affected as a result of the contravention.

    (2) The court may order the person concerned to pay to the regulator  concerned such sum as appears to the court to be just having regard–

    (a) in a case within paragraph (a) of subsection (1), to the profits appearing to the court to have accrued;

    (b) in a case within paragraph (b) of that subsection, to the extent of the loss or other adverse effect;

    (c) in a case within both of those paragraphs, to the profits appearing to the court to have accrued and to the extent of the loss or other adverse effect. "

  813. Plainly, the Authority is not here seeking a restitution order against the Applicants, but the section is relevant to the parties' dispute about the meaning of "knowingly concerned", as we explain below.
  814. Burton v Bevan

  815. In Burton v Bevan [1908] 2 Ch 240 at 246-247, Neville J considered whether the defendant had "knowingly contravened" a particular statutory provision, and said at p 720:
  816. "I think that 'knowingly' means with knowledge of the facts upon which the contravention depends. I think it is immaterial whether the director had knowledge of the law or not. I think he is bound to know what the law is, and the only question is, did he know the facts which made the act complained of a contravention of the statute?"

    Scandex

  817. In Securities and Investment Board v Scandex [1998] 1 WLR 712 ("Scandex"), the Court of Appeal considered a claim brought by the Securities and Investment Board ("SIB"), a predecessor to the Authority, against a director of Scandex, a company incorporated in Denmark. Scandex had contravened s 3 of the Financial Services Act 1986 ("FSA 1986"), which provided that:
  818. "No person shall carry on, or purport to carry on, investment business in the United Kingdom unless he is an authorised person under Chapter III or an exempted person under Chapter IV of this Part of this Act."

  819. The SIB sought a compensation order against the director under s 6(2) FSA 1986, for having been "knowingly concerned in the contravention" by Scandex of s 3. The director's defence was that he honestly believed Scandex was authorised to carry on investment business in Denmark and thus was exempt from the requirement to obtain similar authorisation in the United Kingdom.
  820. Millett LJ stated at p 717 that to be "knowingly concerned" so as to be liable under s 6(2), the director had to be knowingly concerned in the contravention. He went on to say that the contravention in this case had three elements: (i) the carrying on of an investment business; (ii) in the United Kingdom; (iii) by a person who is not authorised under FSA 1986. He ruled that to be "knowingly concerned", the director must have knowledge of all three elements. At p 720, he also endorsed the dictum set out above from Burton v Bevan, saying that the director "is to be judged on the facts as he believed them to be, but on the law as it is".
  821. Capital Alternatives

  822. In FCA v Capital Alternatives [2018] 3 WLUK 623 ("Capital Alternatives"), McCahill J reviewed the case law and gave guidance, which so far as relevant is as follows:
  823. "797. S.382 FSMA gives the Court jurisdiction to grant restitution orders against those 'knowingly concerned' in a contravention of a 'relevant requirement', in addition to the primary contraveners.

    798. In SIB v Pantell (No.2) [1993] Ch 256 [Pantell] at 264D-E, at first instance, Browne-Wilkinson VC stated of 'knowingly concerned':

    'The most obvious example of a person 'knowingly concerned' in a contravention will be a person who is the moving light behind a company which is carrying on investment business in an unlawful manner. Professor Gower in his report, which was the basis on which the Act was introduced, specifically pointed out the mischief of directors hiding behind the corporate veil of companies...If, as is often the case, the company is not worth powder and shot, it is obviously just to enable the Court, as part of the statutory remedy of quasi-rescission, to order the individual who is running that company in an unlawful manner to recoup those who have paid money to the company under an unlawful transaction.'

    799. The learned Judge there identified the most obvious example of a person who is 'knowingly concerned' in a contravention, namely the 'moving light' behind a company which has contravened a relevant requirement, but, in my judgment, the matter is not limited to those who are the moving lights behind the contravening entity. Each case must be considered on its own unique facts.

    800. In the Court of Appeal in the same case and the same report (at 283G), Steyn LJ held that proof of actual knowledge is essential but not enough. Mere passive knowledge is not sufficient and actual 'involvement in the  contravention must be established'.

    801. The concept of 'involvement' is a broad one, covering those who pull the strings at a directorial and/or managerial level (this would include the 'moving lights' in the contravening entity) and could, in an appropriate case, include those who are involved at a lower level, depending on their knowledge and participation in the contravention.

    802. In SIB v Scandex Capital Management [1998] 1 WLR 712, the Court
    of Appeal, at 720F-H, confirmed that the relevant knowledge is knowledge
    of the facts on which the contravention depends, and that it is immaterial as to whether or not the individual knows that such facts constitute a relevant contravention. This is because the individual is presumed to know what the law is, and ignorance of the law is no defence."

    Avacade

  824. In FCA v Avacade Ltd and others [2020] EWHC 1673 (Ch), ("Avacade") at [454], Adam Johnson J adopted the summary of the law set out in Capital Management, which he found to be "entirely correct".
  825. The appellants in Avacade had taken legal advice (see [449(ii) of the judgment]), but Johnson J said at [471] that this was irrelevant, because "knowingly concerned" requires knowledge of the facts "and it is immaterial whether or not the individual knows that such facts constitute a relevant contravention".
  826. Ferreira

  827. In the case of FCA v Ferreira [2022] EWCA Civ 397 ("Ferreira"), Snowden LJ gave the only judgment with which Newey and Lewison LJJ both concurred. Ms Ferreira had been the director of a company which had breached FSMA s 21; at the relevant time that section read as follows:
  828. "(1) A person (' A ') must not, in the course of business, communicate an invitation or inducement to engage in investment activity.

    (2) But subsection (1) does not apply if—

    (a) A is an authorised person; or

    (b) the content of the communication is approved for the purposes of this section by an authorised person."

  829. It was common ground that Ms Ferreira was "actually involved in the contravention": the issue was whether she had been knowingly concerned in the company's contravention. Ms Ferreira had lost at the High Court before Bacon J ("the Judge"), who began her decision by finding as follows, consistently with Scandex and Burton v Bevan:
  830. "the concept of being 'knowingly concerned in a contravention' for the purposes of section 382 required satisfaction of two discrete elements, namely (i) that the person must have been actually involved in the contravention, and (ii) that the person must have had knowledge of the facts on which the contravention depends."

  831. Snowden LJ said at [20] that "it was not disputed before us that these initial steps in the Judge's reasoning were correct". However, the Judge went on to find that Ms Ferreira also had "knowledge of the facts on which the contravention depends", because, in the Judge's view to satisfy that part of the test:
  832. "all that is required is knowledge that a communication has been made which invites or induces investment activity or claims management activity, and knowledge that this is in the course of business."

  833. The Judge recognised that this finding was inconsistent with the approach taken by Millett LJ in Scandex, but said that this was because the term "knowingly concerned" in FSA 1986 was to be read differently from its usage in FSMA s 21. In coming to that conclusion, the Judge took into account that one of the purposes of s 382 was "to prevent directors from hiding behind the corporate veil of the infringing company", see the citation from Pantell set out in Capital Alternatives. The Judge acknowledged at [117] of her decision that, in consequence:
  834. "In almost every case where a person is 'concerned' in a breach of section 21 FSMA they are likely to have the requisite degree of knowledge, since all that is required is knowledge that a communication has been made which invites or induces investment activity or claims management activity, and knowledge that this is in the course of business."

  835. Ms Ferreira appealed on the ground that the Judge's interpretation of s 382 was wrong, and that to have been "knowingly concerned" in a contravention of section 21(1), a defendant must have had knowledge (whether actual, or imputed on the basis of wilful blindness) that the relevant communication was not approved.
  836. Snowden LJ noted at [27] that the Judge had accepted that as a consequence of her analysis, the word "knowingly" in s 382 would "serve little or no purpose", commenting that this was "not a promising starting point for an argument on statutory interpretation". He went on to find that the Judge had been wrong to distinguish the usage of the same term in FSA 1986 from that in FSMA s 21, and endorsed the approach taken by Millett LJ in Scandex, saying that:
  837. "...knowledge of the facts which make the act complained of a contravention of the statute must include knowledge of the factual circumstance that prevents a potentially relevant disapplication from operating."

  838. At [32] he gave this example:
  839. "Suppose a statute were to prohibit any communication inviting or encouraging the making of an investment, but also provided that such prohibition is not to apply at weekends. It would not be sufficient to establish liability under section 382 if a defendant director knew that an advertisement inviting an investment had been placed in a newspaper by his company. Those facts alone would not indicate whether a contravention of the prohibition had occurred. The missing fact which the director would also have to know is that the advertisement was not in a newspaper published at a weekend."

  840. Finally, he considered the reasons given by the Judge for her conclusions, which included this one:
  841. "...one of the purposes of introducing powers to make a restitution order against someone who was 'knowingly concerned' in unlawful investment activity was to prevent directors from 'hiding behind the corporate veil' of an insolvent infringing company."

  842. Snowden LJ then said at [47]:
  843. "...the Judge interpreted section 382 in a way that imputed to the legislature
    an intention to impose personal liability on directors (or others) simply on the basis that they knew of the actions that the company was taking in the course of its business. That would be a far-reaching step indeed. Business is normally conducted, and investment opportunities are routinely offered, by companies with limited liability. The interpretation adopted by the Judge would result in limited liability being disregarded irrespective of whether the company was in fact rendered insolvent by the contravention of FSMA, and in a much wider set of circumstances than those in ...which the courts have conventionally thought it appropriate to pierce the corporate veil. Such grounds conventionally require some finding that the directors or corporators have established the company as a sham or façade for the purposes of some fraud. The corporate veil has never been disregarded simply because the directors were aware of the actions that their company was taking in the course of its business. In my judgment, the intention to introduce such a radical departure from the principles of limited liability in the financial services field should not be attributed to the legislature in the absence of some very clear indication - of which there is none."

    Submissions on Ferreira

  844. Submissions on this issue were made on behalf of the Applicants by Mr Pritchard. He initially said, in reliance on the above passage from Ferreira, that a director could only be "knowingly concerned" if he had acted fraudulently, but then amended his position, and said there had to be "personal wrongdoing", which would include fraud or recklessness, or
  845. "something that takes it out of the ordinary course of business of being a director...It could be a lack of integrity, but it is not just doing your usual business as a director of the company."

  846. Mr Stanley said that this was plainly wrong. There was no reference in the statute to a person only being "knowingly concerned" if he was guilty of fraud, recklessness, lack of integrity or "personal wrongdoing". In the main parts of Ferreira, Snowden LJ had clearly explained what was required for a person to be "knowingly concerned", and he made no reference to a requirement for there to be "personal wrongdoing".
  847. The Tribunal's view of Ferreira

  848. We agree with Mr Stanley, for the reasons set out below.
  849. The words of the provision

  850. What is meant by "knowingly concerned" is, as Snowden LJ said, a matter of statutory interpretation. In Roberts v Secretary of State for the Environment, Transport and the Regions, Ex p Spath Holme Ltd [2001] AC 349, Lord Nichols said at p 396:
  851. "Statutory interpretation is an exercise which requires the court to identify the meaning borne by the words in question in the particular context. The task of the court is often said to be to ascertain the intention of Parliament expressed in the language under consideration. This is correct and may be helpful, so long as it is remembered that the 'intention of Parliament' is an objective concept, not subjective. The phrase is a shorthand reference to the intention which the court reasonably imputes to Parliament in respect of the language used."

  852. He added that "an appropriate starting point is that language is to be taken to bear its ordinary meaning in the general context of the statute". 
  853. Here, the statutory phrase is "knowingly concerned". There is nothing in the words themselves, or in their context, which justifies or requires them to be read as applying only where the person in question has committed some "personal wrongdoing", such as fraud, recklessness or acting without integrity.
  854. The ratio of Ferreira

  855. In Ferreira at [20], Snowden LJ confirmed that the phrase "knowingly concerned" required that the person must have:
  856. (1)          been actually involved in the contravention, and

    (2)          had knowledge of the facts on which the contravention depends.

  857. The issue before the Court of Appeal was what was meant by "knowledge of the facts".  
  858. In its judgment, the Court overruled the Judge's finding that this requirement was met if the person had "knowledge that a communication has been made which invites or induces investment activity or claims management activity, and knowledge that this is in the course of business". Instead, the person had also to have knowledge of all the elements which make up the contravention, including knowledge of the factual circumstance that prevents a potentially relevant disapplication from operating. In Ms Ferreira's case, she did not know the Company's communications had not been validly approved, and so she was not "knowingly concerned".
  859. The ratio of the judgment thus endorses the decision in Scandex as to what is required for a person to be "knowingly concerned", namely that the person has to have knowledge of all the elements which make up the contravention. There is no suggestion that any sort of "personal wrongdoing" is additionally required.
  860. The corporate veil

  861. The issue discussed at [47] is that the Judge had found support for her (incorrect) view by reference to one of the purposes for which s 382 had been introduced. Taking it in stages:
  862. (1)          Where a company is carrying on investment business in an unlawful manner, a customer's right of action lies against the company: that is the effect of the corporate veil.

    (2)          Where a company has no assets, that right of action is worthless.

    (3)          In that situation, s 382 allows regulators to issue a restitution order so as to require the directors personally to pay compensation; in other words, it allows the corporate veil to be pierced.

    (4)          Such an order can only be made if the director is "knowingly concerned" in the company's breach. In other words, being "knowingly concerned" is a precondition for the making of a restitution order.

  863. Snowden LJ said that if the Judge were to be correct, restitution orders could be made against those who were simply "aware of the actions that their company was taking in the course of its business", and the Judge was plainly wrong to have come to that conclusion. Instead, breaching the corporate veil by issuing a restitution order requires that the company is being used "as a sham or façade for the purposes of some fraud".
  864. The passage on which Mr Pritchard relied therefore does not change the meaning of "knowingly concerned" from that which had already been set out earlier in Ferreira. Instead, the passage states that a restitution order can only be made if a person is both knowingly concerned and the company is being used as a sham or façade for wrongdoing. In the case of the Applicants, the Authority have not sought a restitution order, so the passage is of no relevance.
  865. Forster: meaning of "knowingly concerned"

  866. In FCA v Forster [2023] EWHC 1973 (Ch) at [230]-[231], Gleeson J considered the meaning of "knowingly concerned", and repeated the two limbs of the test approved in Ferreira. Having cited [47] of that judgment, he went on to say:
  867. "237. I think that Snowden LJ's meaning here is clear - a director of a firm who has the ordinary knowledge of the firm's activities which a director is expected to have should not automatically be liable under s.382 on a breach by their firm of a regulatory requirement. Directors of financial firms are entitled to the same company law protections as directors of any other company...the proposition as articulated by Snowden LJ seems clearly correct. The ordinary rules of director's liability are not dispensed with simply because the firm concerned engages in unauthorised financial business. The FSMA does provide the FCA with the ability to pursue directors personally, but an essential threshold condition for such pursuit is, as Snowden LJ put it, 'some fraud'.

    238. The conclusion from this is that in order for s.382 to apply, there must be some involvement by a director in the contravention of a regulatory rule by his company which goes some way beyond the normal involvement of a director in the affairs of the company."

  868. That judgment in Forster, including these paragraphs, thus reflects our analysis of Ferreira. The test for "knowingly concerned" is that set out above at §496, and the passage at [47] of Ferreira instead explains what is required before a regulator can issue a restitution order under s 382 to a person who was "knowingly concerned".
  869. Mr Pritchard submitted that "you can replace s 382 with s 91" in the paragraphs from Forster cited above, so that the Authority was only able to impose a penalty for being "knowingly concerned" if the director's involvement in the contravention "goes some way beyond the normal involvement of a director in the affairs of the company".
  870. We disagree. The cited passages are concerned with the power given by s 382 for a court, on the application of a regulator, to order a director to make restitution for wrongs carried out by the company: as explained above, such a restitutionary order pierces the corporate veil because the individuals are being required to compensate third parties for the actions of the company. The penalties imposed on the Applicants are not restitutionary. They are a punishment imposed by the Authority for being knowingly concerned in the breach carried out by the Bank. The passage at [47] of Ferreira and those at [237]-[238] of Forster are thus irrelevant to the position of the Applicants.
  871. Forster: reliance on legal advice

  872. The other passage from Forster relied on by the Applicants concerned reliance on legal advice. In Forster, the Authority had decided, inter alia, that:
  873. (1)          certain investments which had been marketed and sold by a number of investment companies of which Mr Forster was a director, were each a "collective investment scheme" ("CIS"), and

    (2)          Mr Forster was "knowingly concerned" in breaches by the investment companies and/or by his co-director, of the regulations relating to the marketing and promotion of CIS investments.  

  874. Mr Forster's case relied in part on the fact that the investment companies had obtained legal advice in the form of two Opinions from "experienced independent Counsel" who had confirmed that the CIS provisions did not apply, and it was submitted that Mr Forster was therefore not "knowingly concerned" because he had relied on those Opinions.
  875. Gleeson J first considered the legal point. He noted that in Burton v Bevan, Scandex, and Avacade, it had been held that an incorrect view of the law was irrelevant, because the director was assumed to know the law. At [246] he summarised the submissions made on Mr Forster's behalf:
  876. "where a lay client seeks and obtains legal advice from an appropriately qualified professional, he cannot reasonably be expected to form a view on the correctness or otherwise of the legal advice which he has received".

  877. Gleeson J then said:
  878. "This is, in principle, a good argument. To describe a person as 'knowingly concerned' in a contravention of the law in circumstances where he has obtained independent advice that the activity concerned is not in contravention of the law is to strain the meaning of the word 'knowingly' beyond any reasonable compass."

  879. However, he went on to say at [249] that a lay client "both can and must interrogate the factual assumptions on which the advice which he has received is based". He then considered the factual assumptions provided for the first Opinion, and said "Mr Forster must have known that this was not the way in which the scheme actually operated". He continued at [253]:
  880. "If he had read the McGee opinion dispassionately, it should immediately have been clear to him that what it was in fact saying was that, in different
    circumstances, and against a different factual matrix, the contracts which had been put in place would not necessarily constitute participation in a scheme. This opinion is entirely correct. However, it does not help Mr Forster in this case."

  881. The second Opinion was, said Gleeson J, also based on incorrect facts. He summarised the position at [257]:
  882. "The key points here seem to me to be twofold. First, it is absurd to suggest that a lay client should not rely on the advice which he has received as regards the legal analysis which it contains. Provided that he has sought the advice of an appropriate professional, he cannot be criticised for relying on the advice which he has received. Second, however, is that all legal advice is necessarily based on assumed facts. Legal advice cannot take any other form than that "if the facts are X, the conclusion is Y". The lay client cannot be expected to hold any view as to the legal content of such an opinion. However, what he can be expected to do is to consider the statement of facts on which the opinion he has received is based. If that statement of facts does not correspond to the truth as he knows it to be, he cannot rely for any purpose on the advice which he has received because he knows it to be based on false premises. Where he knows (or should know) that the factual matrix on which the advice given to him is based is incorrect, it is simply not open to him to say that he relied upon that advice."

    The Applicants' submissions

  883. The Applicants' position was that, by reference to the principles in Forster, they were entitled to rely on Mr Lane's advice: Mr Jaffey's skeleton said that it was "not for a lay client such as the Applicants" to "assess the scope and adequacy of a lawyer's advice before deciding whether or not the advice can be relied on".
  884. The Authority's submissions

  885. The Authority's position was that it was clear from Burton v Bevan, Scandex and Avacade, that ignorance of the law was not a defence: as Neville J said in Burton v Bevan "the only question is, did [the person] know the facts which made the act complained of a contravention of the statute?".
  886.  However, the Authority also accepted that if an issuer was relying on legal advice about an announcement, and that advice: 
  887. "reasonably led the issuer to believe that the announcement is accurate and complete, [t]he receipt of such advice might arguably demonstrate that the issuer had taken reasonable advice to ensure accuracy and completeness."

  888. Mr Stanley said, however, that this was not the Applicants' position, because Linklaters were not asked to advise on whether the Bank could include RWA figures which it knew to be materially wrong in the Q3 Update. Instead, Mr Lane was asked to advise about whether the MAR required the Bank to make a proactive market announcement about the CLIP loan error.
  889. The Tribunal's view

  890. As set out above, Gleeson J said in Forster that if a lay client provided a lawyer with a statement of facts, and "that statement of facts does not correspond to the truth as he knows it to be, he cannot rely for any purpose on the advice which he has received because he knows it to be based on false premises".
  891. We have already found as facts that:
  892. (1)          Mr Arden made a number of incorrect statements to Mr Lane, see §§317-318.

    (2)          The facts provided to Mr Lane gave him a false impression of the situation, see §324.

    (3)          Mr Lane was instructed to advise on whether the Bank had to make a pro-active disclosure under the MAR, and was not asked for advice on the Q3 Update, see §319.

    (4)          Mr Lane was not provided with a copy of the draft Q3 Update either before or after the meeting with Mr Arden and Ms Roberts, see §316.

  893. Since Mr Arden gave Mr Lane an incomplete and incorrect picture of the facts, and did not ask for advice on the Q3 Update, then, to borrow the phraseology from Forster, it is not open to him to say he was not "knowingly concerned" on the basis that he had relied on that advice.
  894. We considered whether the same was true for Mr Donaldson. He had not been at the meeting with Mr Lane, and was not copied on the email from Ms Roberts; he also did not attend the Disclosure Committee on 16 October 2018, where the meeting was discussed. However, Mr Arden told him about the meeting the day it had happened. Mr Donaldson said in his witness statement that:
  895. "My clear understanding from the conversation with Mr Arden was that until Deloitte's work had been concluded and discussions to agree the final capital position with the PRA had taken place no market announcement was necessary. In other words, it was my understanding that the Bank did not need to make any specific disclosure about the RWA Issue in the Q3  Announcement and I relied on that advice."

  896. Mr Donaldson agreed in cross-examination that he didn't know whether or not Mr Lane was shown the draft Q3 Update, but said he trusted Mr Arden and Ms Roberts. He also agreed that his understanding of Linklaters' advice was that "there was no need to say anything until [the Bank] had corrected matters with the PRA".
  897. We find as a fact that Mr Donaldson understood from his conversation with Mr Arden that Linklaters had advised that "no market announcement was necessary" about the RWAs until Deloitte had finished its work and the Bank had discussed the outcomes with the PRA. However, he did not understand that Linklaters had advised that the Bank could report a RWA figure in its Q3 which it knew to be materially incorrect. The two are different. This is thus not a case where, as Gleeson J put it, Mr Donaldson had "obtained independent advice that the activity concerned is not in contravention of the law". The advice he knew the Bank had received related to whether or not the Bank had to make an announcement (so as to comply with the MAR), not about whether the Bank could continue to provide the market with a figure it knew to be wrong.
  898. The principles summarised and the issues remaining

  899. On the basis of the case law set out above, we summarise the position as follows. To be "knowingly concerned" in a breach:
  900. (1)          a person must have been actually involved in the contravention; merely passive knowledge is not sufficient; and

    (2)          must have had knowledge of the facts on which the contravention depends; and

    (3)          it is immaterial whether he had knowledge of the law, unless:

    (a)           he had received and was relying on independent legal advice that the activity concerned was not in contravention of the law; and

    (b)         that advice was based on a correct and complete factual matrix.

  901. We have already found under Issue One that the Bank had issued the Q3 Update which contained an unqualified statement of its RWA, and the capital ratios based on it, at a time when the Bank knew that a material error had been made in relation to the CLIP loans, but it did not inform the market about that. The Bank had therefore breached LR 1.3.3R because it failed to take reasonable care to ensure that the Q3 Update was not false or misleading and did not omit anything likely to affect the import of the information contained within it.
  902. It was common ground that the Applicants were "involved in the contravention" and did not have merely "passive knowledge". We have also found that the Linklaters' advice did not satisfy the requirements at §537(3) above. The only remaining issue was therefore whether the Applicants had the necessary "knowledge of the facts on which the contravention depends".
  903. We have considered this issue in the following stages:
  904. (1)          whether the Applicants knew that CLIP loans had been wrongly risk weighted at 50% and should have been risk-weighted at 100%;

    (2)          whether the Applicants knew that the best estimate of the CLIP loan error was material in the context of the figures reported by the Bank in the Q3 Update;

    (3)          whether the Applicants knew, based on their reasonable belief, that at the time of the Q3 Update the best estimate of the CLIP loan error was significantly less than the estimate, such that the figure included in the Q3 Update was not materially incorrect; and/or

    (4)          whether the Applicants knew, based on their reasonable belief, that at the time of the Q3 Update the effect of one or more mitigants meant that the total RWA figure in the Q3 Update was not materially incorrect.

    (1) Whether the Applicants knew the CLIP loans had been wrongly risk-weighted

  905. We have considered this issue separately for Mr Arden and Mr Donaldson.
  906. Mr Arden

  907. We find as a fact that Mr Arden knew that the CLIP loans had been wrongly risk weighted at 50% and should have been risk-weighted at 100%, on the basis of our earlier findings summarised below.
  908. (1)          He had received the email of 24 August 2018 from Mr Somers and the attached PowerPoint presentation, see §118.

    (2)          He attended the meeting with Mr Stokes and others on 3 September 2018, at which the PowerPoint presentation was discussed, and where the minutes recorded that KPMG had confirmed that CLIP loans should be risk weighted at 100% rather than 50%, see §§128-129.

    (3)          He received the email from Mr MacLean on 11 September 2018, which said that Deloitte had given a second opinion that the CLIP loans should be risk-weighted at 100%.

    (4)          He had participated in the meeting with the PRA on 12 September 2018 at which the PRA were told about the incorrect risk-weighting, see §137.

    (5)          He attended the September CRPAC meeting which approved the KPMG decision trees; these included the higher risk-weighting for CLIP loans, see §§141-142, and at which an information pack on commercial lending set out the consequences, see §146.

    (6)          He attended the September Audit Committee at which Ms Gillan explained the errors, see §148.

    (7)          He attended the September Board Meeting, at which Ms Gillan explained the issues with the classification of commercial mortgages, and he outlined the process for remediating the errors, see §§151-153.

    (8)          He received the RWA Report before the October CRPAC meeting, see §174; this stated that correcting the CLIP loan issue would increase the risk-weighting by £574m and that "any further adjustments to calculations are not expected to be material". He also attended the meeting at which the Report was discussed, and expressed no doubts about the information in that Report, see §185.

    (9)          He attended the October ROC meeting, at which the RWA Report was also tabled and discussed.

    Mr Donaldson

  909. We make the same finding about Mr Donaldson. It is clear from the findings of fact summarised below that he knew the CLIP loans had been wrongly risk-weighted at 50% and should have been risk-weighted at 100%:
  910. (1)          He had received the email of 24 August 2018 from Mr Somers and the attached PowerPoint presentation.

    (2)          He had participated in the meeting with the PRA on 12 September 2018 at which the PRA were told of the errors and that they would be remediated.

    (3)          He attended the September Audit Committee meeting at which Ms Gillan explained the errors; see §118, §137 and §148.

    (4)          He received the RWA Report before the October CRPAC meeting, see §174.

    (5)          He attended the CRPAC meeting at which that Report was discussed and did not express any disagreement with the position it set out, see §185.

    (6)          Under cross-examination, he agreed that he knew that risk-weighting the CLIP loans at 50% was wrong.

    (2) Whether the Applicants knew the estimated CLIP loan error was material

  911. We also find as a fact that both Applicants knew the estimated CLIP loan error was material to the RWA figure being reported as part of the Q3 Update, for the following reasons:
  912. (1)          It was self-evident from the size of the estimated CLIP loan adjustment of £574m in the context of the Bank's loans, see §237 and §208.

    (2)          Both Applicants attended the September 2018 CRPAC meeting, which considered the Annual Review of Commercial Lending pack, see §146(3). This included the following (our emphasis):

    "The executive leadership team had been advised in September that there were 'inconsistencies in current RWA calculations that will result in a significant increase in risk weightings', because commercial mortgages had been weighted at 50% rather than 100%, and that KPMG had confirmed that this was correct."

    (3)          The RWA Report similarly said that the errors "will result in a significant increase in RWs", see §175.

    (3) Whether the Applicants knew the error was significantly less

  913. We next considered whether the Applicants knew, based on their reasonable belief, that at the time of the Q3 Update the true amount of the CLIP loan error was significantly less than the estimate, such that the figure included in the Q3 Update was not materially incorrect.
  914. We have already made the following findings of fact:
  915. (1)          The RWA Report said that the error was £574m, and that "any further adjustments to calculations are not expected to be material"; at the CRPAC meeting at which that Report was discussed, neither of the Applicants expressed any disagreement with the position it set out, see §185.

    (2)          There was no uncertainty about the regulatory interpretation of the PRA's requirements, instead, both Applicants knew at the time of the Q3 Update that the rules required the CLIP loans to be risk-weighted at 100%, see §366.

    (3)          There was no uncertainty as at the time of the Q3 Update about the application of the SME factor, see §380.

    (4)          The number of residential loans which were miscategorised as CLIP was small, in the region of £20m, and this was an immaterial difference to the overall risk-weighting correction that was required, see §385.

  916. We have also found as a fact that the CLIP loans estimate was not based on sampling, see §123; instead, it was the PBTL figure which was calculated in that way. However, Mr Donaldson said in his witness statement that one reason he did not accept the estimate was because he understood the CLIP loan figure to have been based on "high-level sampling", and that as a result it was "not clear that any adjustment would be necessary". However, under cross-examination he accepted that this may have been because at the time he wasn't distinguishing between CLIP loans and PBTL loans.
  917. Given the clear information provided to Mr Donaldson about the basis for the CLIP loan estimate, we find that there was no reasonable basis for him to hold the view, at the time of the Q3 Update, that the estimate was based on "high level sampling".
  918. Mr Arden gave similar evidence, saying that when he was given the estimated figures in September 2018, he "was aware that the methodology...was based on high-level sampling", but he accepted under cross-examination that this was not the position. As with Mr Donaldson, we find that there was no reasonable basis for him to hold that view. 
  919. In the light of the above, we find that the Applicants had no reasonable basis on which they could have thought that the estimate of £574m was much too high and thus that the RWA figure reported in the Q3 Update was materially correct.
  920. (4) The mitigants

  921. The final issue under this heading was whether the Applicants reasonably believed that, as the result of one or more mitigants, the total RWA figure in the Q3 Update was not materially incorrect.
  922. The position of the parties

  923. The Applicants' position was that they knew, based on their reasonable belief, that at the time of the Q3 Update the best estimate was incorrect because the increased RWA required as a result of the CLIP loans would be offset by one or more of the following mitigants:
  924. (1)          The PRA allowing the Bank to move to an AIRB basis.

    (2)          The PRA exercising its discretion to allow the Bank a Pillar 2A offset.

    (3)          The PRA allowing the Bank to "phase in" the 100% risk-weighting for the CLIP loans.

    (4)          The PRA agreeing to waive the regulatory requirement for CLIP loans.

  925. The Authority's position was that at the time of the Q3 Update, the Applicants did not expect that any of these mitigants would be permitted by the PRA, and that in any event, they would not have changed the Bank's RWAs at the time of the Q3 Update.
  926. The Tribunal's view

  927. We have already considered each of the above mitigants in the context of the Bank, see §§387-408, and decided as follows:
  928. (1)          The reasonable issuer in the Bank's position would not have considered that the possibility of obtaining AIRB for residential mortgages would mitigate against the increased RWAs for CLIP loans.

    (2)          The position was the same for the Pillar 2A offset.

    (3)          Phasing-in was not considered until the end of 2018, and even if it had been, a reasonable issuer would not have decided that it could publish incorrect figures on the basis of this possibility, which had never been raised with the PRA.

    (4)          The position was the same in relation to the PRA's discretion over waivers of rules.

  929. It is also clear that, at best, the mitigants would only have reduced or eliminated the extra capital required as a result of the increased RWAs for the CLIP loans. The mitigants would not have changed the RWAs themselves: the figures published at the time of the Q3 Update would still have been materially wrong, and the Applicants knew that was the case.
  930. Conclusion on issue two

  931. For the reasons set out above, we find that the Applicants were knowingly concerned in the Bank's breach of LR 1.3.3R.
  932. ISSUE THREE: PENALTIES

  933. As set out at §484, FSMA s 91 provides that if the Authority considers that a director "was knowingly concerned in the contravention it may impose upon him a penalty of such amount as it considers appropriate". The Authority imposed a penalty of £223,600 on Mr Donaldson and a penalty of £134,600 on Mr Arden, by reference to the approach set in their Decision Procedure and Penalties Manual ("DEPP"). The Applicants submitted that the Authority should instead have issued a public censure, and that in any event the quantum of the penalties were "grossly disproportionate".
  934. The Tribunal's approach

  935. In Carrimjee v FCA [2015] UKUT 79 (TCC) at [15], the Tribunal (Judge Herrington, Mr Bottriell and Ms O'Neill) held that the Tribunal is not bound by the Authority's policy when making an assessment of a financial penalty, but that the Tribunal will pay the policy due regard when carrying out its overriding objective of doing justice between the parties, and in so doing will look at all the circumstances of the case.
  936. In FCA v Da Vinci Invest Ltd and others [2015] EWHC 2401 at [201], Snowden J (as he then was) said in the context of the imposition of a penalty for market abuse:
  937. "It was the FCA's submission, and I accept, that in determining any penalty under section 129, the starting point for the court should be to consider the relevant DEPP penalty framework that was in existence at the time of commission of the market abuse in question. To do otherwise would risk introducing an inequality of treatment of defendants depending upon whether the proceedings were taken against them under the regulatory route or the court route and depending upon how long the proceedings had taken to come to a conclusion. By the same token, however, in common with the Upper Tribunal, the court is not bound by that framework, or by the FCA's view of how it should be applied. But if the court intends to depart from the framework in a particular case, it should explain why it considers it appropriate to do so."

  938. This case concerns not only the parts of the DEPP which relate to penalties but also those which relate to public censure, but we consider that our approach should be the same as that set out in Carrimjee and Da Vinci, with which we respectfully agree.
  939. The Applicants did not invite us to depart from the principles set out in the DEPP and although we recognised that we are not bound by it, we have not identified any reason to depart from the approach it sets out.
  940. Public censure or penalty

  941. We first set out the parts of the DEPP which are relevant to deciding whether to impose a censure or a penalty, followed by the parties' submissions and our view.
  942. The DEPP

  943. DEPP 6.4 is headed "Financial penalty or public censure" and DEPP 6.4.1 reads:
  944. "The FCA will consider all the relevant circumstances of the case when deciding whether to impose a penalty or issue a public censure. As such, the factors set out in DEPP 6.4.2 are not exhaustive. Not all of the factors may be relevant in a particular case and there may be other factors, not listed, that are relevant."

  945. The above passage thus cross-refers to DEPP 6.4.2, which sets out the following "particular considerations" that may be relevant when the Authority decides whether to issue a public censure rather than a penalty:
  946. (1)          Whether deterrence may be effectively achieved by issuing a public censure.

    (2)          Factors in favour of a financial penalty include:

    (a)          The person has made a profit or avoided a loss as the result of the breach.

    (b)         The breach is "more serious in nature or degree" and "other things being equal, the more serious the breach, the more likely the FCA is to impose a financial penalty".

    (c)          The person has a poor disciplinary record or compliance history.

    (3)          Factors in favour of public censure include:

    (a)          If the person has admitted the breach; has provided "full and immediate co-operation" with the Authority, and has taken steps to compensate those who suffered losses as a result of the breach.

    (b)         Exceptional circumstances, for instance serious financial hardship.

  947. Since one of the "particular considerations" to consider in deciding whether to issue a censure or a penalty is "the seriousness of the breach", the guidance at DEPP 6.5B is also relevant, because it sets out how seriousness is assessed when considering penalties in non-market abuse cases.
  948. Factors considered in assessing seriousness include those relating to the impact of the breach and its nature. In relation to the former, DEPP 6.5B(8) reads as follows:
  949. "(a) the level of benefit gained or loss avoided, or intended to be gained or avoided, by the individual from the breach, either directly or indirectly;

    (b) the loss or risk of loss, as a whole, caused to consumers, investors or other market users in general;

    (c) the loss or risk of loss caused to individual consumers, investors or other market users;

    (d) whether the breach had an effect on particularly vulnerable people, whether intentionally or otherwise;

    (e) the inconvenience or distress caused to consumers; and

    (f) whether the breach had an adverse effect on markets and, if so, how serious that effect was. This may include having regard to whether the orderliness of, or confidence in, the markets in question has been damaged or put at risk."

  950. Factors relating to the nature of a breach include:
  951. "(a) the nature of the rules, requirements or provisions breached;

    (b) the frequency of the breach;

    (c) the nature and extent of any financial crime facilitated, occasioned or otherwise attributable to the breach;

    (d) the scope for any potential financial crime to be facilitated, occasioned or otherwise occur as a result of the breach;

    (e) whether the individual failed to act with integrity;

    (f) whether the individual abused a position of trust;

    (g) whether the individual committed a breach of any professional code of conduct;

    (h) whether the individual caused or encouraged other individuals to commit breaches;

    (i) whether the individual held a prominent position within the industry;

    (j) whether the individual is an experienced industry professional;

    (k) whether the individual held a senior position with the firm;

    (l) the extent of the responsibility of the individual for the product or business areas affected by the breach, and for the particular matter that was the subject of the breach;

    (m) whether the individual acted under duress;

    (n) whether the individual took any steps to comply with FCA rules, and the adequacy of those steps;

    (o) in the context of contraventions of Part VI of the Act, [which includes the vires for the Listing Rules] the extent to which the behaviour which constitutes the contravention departs from current market practice;..."

    The Authority's position

  952. The Authority's position was that the Applicants had acted negligently and their misconduct was serious, and it was therefore appropriate to impose a financial penalty rather than issue a public censure. In relation to "the impact of the breach", the Authority found as follows in relation to both Applicants:
  953. "Metro Bank is a premium listed issuer that was listed on the FTSE 250 at the time of the breach. The Bank's inclusion of a total RWA figure calculated using the incorrect CLIP Loans risk weight, without any qualification, had the potential to mislead its investors and affect the import of the information contained in the October Announcement. The fact that, following the January 2019 announcement, there was a drop in Metro Bank's share price of 39% on the day of the announcement and adverse market commentary indicates that Metro Bank's breach of LR 1.3.3R, in respect of which [each Applicant] was knowingly concerned, had a serious  adverse effect on financial markets and risked damaging confidence in the financial markets (6.5B.2G(8)(f)).

    The existence of the CLIP Loans error only became known to investors three months after the breach, when the January 2019 announcement was released. This delay caused a risk of loss to new and existing individual shareholders who traded between the October Announcement and the January 2019 announcement (DEPP  6.5B.2G(8)(c))."

  954. The Authority accepted that the 39% drop in the Bank's share price was not solely attributable to the incorrect RWAs for the CLIP loans, but said it "played a substantial part in that fall", and that failure to inform the market of the true position meant that investors proceeded under the positive misapprehension that the published RWA figure was reliable, when it was not.
  955. In relation to the nature of the breach, the Authority took into account that as CEO and CFO, both Applicants held senior positions within the Bank (DEPP 6.5B.2G(9)(k)), although in relation to Mr Arden, the Authority noted that he did not "have significant industry experience, with this being his first senior role at a listed firm".
  956. The Authority also found that the Applicants were both:
  957. "...aware that Metro Bank calculated the total RWA figure in its October Announcement by, amongst other things, incorrectly applying a risk weighting of 50% to its CLIP Loans and failed to take adequate steps to ensure that Metro Bank complied with LR 1.3.3R. [They] failed to ensure that Metro Bank considered whether the October Announcement ought to have included a qualification or sought legal advice or input from its  professional advisers on this question. [They] also failed to ensure that the  Audit Committee and the Board considered whether the inclusion of the  inaccurate RWA figure in the October Announcement without any qualification was appropriate (DEPP 6.5B.2G(9)(n))."

    The Applicants' position

  958. The Applicants' case was that a censure was the appropriate sanction, because, by reference to the "particular considerations" set out above:
  959. (1)          This would effectively achieve deterrence.

    (2)          The Applicants did not make a profit or avoid a loss.

    (3)          There was little or no culpability on the part of the Applicants, who believed they were following legal advice and that the Bank was following proper governance procedures.

    (4)          The Applicants took steps to inform the PRA and liaised with them to remedy the issue.

    (5)          Neither has a disciplinary record.

    The Tribunal's view

  960. In relation to the "particular considerations" set out at DEPP 6.4.2 which are identified as being relevant to whether public censure or a penalty is appropriate, it is clear that some of those circumstances apply to the Applicants: they received no financial gain as the result of the breach, and they have no prior disciplinary record; on the other hand, the Applicants did not admit that there was a breach, and there are no exceptional circumstances.
  961. However, as is clear from DEPP 6.4.1 "not all of the factors may be relevant in a particular case". The key relevant factor in the Applicants' case is whether the breach was so "serious in nature or degree" that public censure is inappropriate.
  962. We agree with the Authority that publishing an RWA number which was known to be materially incorrect had serious consequences: in particular, investors proceeded under the positive misapprehension that the published RWA figure was reliable when it was not. If a qualified RWA figure had been published instead, investors would have faced a known risk (albeit with an uncertain final outcome) and would have made their investment decisions accordingly.
  963. We do not agree with Mr Jaffey that the Applicants had "little or no culpability" for those serious consequences. They were the only two executive directors of the Bank; the only legal advice obtained related to compliance with the MAR and not to the Q3 Update (see §340), and the information provided to the Audit Committee and the Board downplayed both the CLIP Loan error and the adjustment to the Bank's RWA that would be necessary to correct it (see §§460-466). We therefore agree with the Authority that this is a case where public censure is insufficient, and we move on to consider penalties.
  964. The penalty framework

  965.  DEPP 6.1.2 states that the principal purpose of imposing a financial penalty is to promote high standards of regulatory and/or market conduct by deterring persons who have committed breaches from committing further breaches, and helping to deter other persons from committing similar breaches, as well as demonstrating generally the benefits of compliant business.
  966. DEPP 6.5.2 states that the FCA's penalty-setting regime is based on the following principles:
  967. (1)          disgorgement - a firm or individual should not benefit from any breach;

    (2)          discipline - a firm or individual should be penalised for wrongdoing; and

    (3)          deterrence - any penalty imposed should deter the firm or individual who committed the breach, and others, from committing further or similar breaches.

  968. DEPP 6.5.3 sets out a five-step framework to determine the appropriate level of financial penalty, with further details provided in subsequent chapters, and it states that these steps "will apply in all cases".
  969. In relation to non-market abuse cases such as this one, the relevant framework is at DEPP 6.5B, and begins as follows.
  970. Step 1: Disgorgement

    The FCA will seek to deprive a firm of the financial benefit derived directly from the breach...

    Step 2: The seriousness of the breach

    (1) The FCA will determine a figure which will be based on a percentage of an individual's "relevant income". "Relevant income" will be the gross amount of all benefits received by the individual from the employment in connection with which the breach occurred (the "relevant employment"), and for the period of the breach. In determining an individual's relevant income, "benefits" includes, but is not limited to, salary, bonus, pension contributions, share options and share schemes; and "employment" includes, but is not limited to, employment as an adviser, director, partner or contractor.

    (2) Where the breach lasted less than 12 months, or was a one-off event, the relevant income will be that earned by the individual in the 12 months preceding the end of the breach. Where the individual was in the relevant employment for less than 12 months, his relevant income will be calculated on a pro rata basis to the equivalent of 12 months' relevant income.

    (3) This approach reflects the FCA's view that an individual receives remuneration commensurate with his responsibilities, and so it is reasonable to base the amount of penalty for failure to discharge his duties properly on his remuneration. The FCA also believes that the extent of the financial benefit earned by an individual is relevant in terms of the size of the  financial penalty necessary to act as a credible deterrent. The FCA recognises that in some cases an individual may be approved for only a small part of the work he carries out on a day-to-day basis. However, in these circumstances the FCA still considers it appropriate to base the relevant income figure on all of the benefit that an individual gains from the relevant employment, even if their employment is not totally related to a controlled function.

    (4) Having determined the relevant income the FCA will then decide on the percentage of that income which will form the basis of the penalty. In making this determination the FCA will consider the seriousness of the breach and choose a percentage between 0% and 40%.

    (5) This range is divided into five fixed levels which reflect, on a sliding scale, the seriousness of the breach. The more serious the breach, the higher the level. For penalties imposed on individuals there are the following five levels:

    (a)  level 1 - 0%;

    (b)  level 2 - 10%;

    (c)  level 3 - 20%;

    (d)  level 4 - 30%; and

    (e)  level 5 - 40%."

  971. Pausing there, as Mr Jaffey said in closing, these percentages are based on a person's gross income, but any penalty will be paid out of net income, ie after tax and National Insurance Contributions.
  972. The DEPP 6.5B framework continues as follows:
  973. "(6) The FCA will assess the seriousness of a breach to determine which level is most appropriate to the case.

    (7) In deciding which level is most appropriate to a case against an individual, the FCA will take into account various factors which will usually fall into the following four categories:

    (a)  factors relating to the impact of the breach;

    (b)  factors relating to the nature of the breach;

    (c)  factors tending to show whether the breach was deliberate; and

    (d)  factors tending to show whether the breach was reckless."

  974. The factors relating to the impact and nature of the breach have already been set out at §§565-567. The DEPP continues by reference to factors which tend to show that a breach was deliberate or reckless (neither of which are relevant here), followed by factors which are likely to be considered "level 1 factors", "level 2 factors" or "level 3 factors". These include:
  975. "(a) little, or no, profits were made or losses avoided as a result of the breach, either directly or indirectly;

    (b) there was no or little loss or risk of loss to consumers, investors or other market users individually and in general;

    (c) there was no, or limited, actual or potential effect on the orderliness of, or confidence in, markets as a result of the breach;

    (d) the breach was committed negligently or inadvertently;..."

  976. DEPP 6.5.3(3) records that "a penalty must be proportionate to the breach" and says that the Authority "may decrease the level of the penalty arrived at after applying Step 2 of the framework if it considers that the penalty is disproportionately high for the breach concerned".
  977. Step 3 is "mitigating and aggravating factors" and provides that the Authority may increase or decrease the figure arrived at following Step 2, to take into account factors which aggravate or mitigate the breach. We say more about Step 3 at §604 below.
  978. Step 4 is "adjustment for deterrence". DEPP 6.5B.4 begins by saying:
  979. "If the FCA considers the figure arrived at after Step 3 is insufficient to deter the individual who committed the breach, or others, from committing further or similar breaches then the FCA may increase the penalty." 

  980. Step 5 is "settlement discount" and is relevant where the person on whom the penalty is imposed comes to an agreement with the Authority.  
  981. Applying the Steps

  982. The Authority applied the framework set out above to both Mr Donaldson and Mr Arden. In relation to Step 1, disgorgement, the Authority accepted that neither had obtained any financial benefit from the breach.
  983. In relation to Step 2, the breach lasted less than twelve months. Mr Donaldson had been employed by the Bank throughout the previous twelve months, so his "relevant income" was his earnings for that period of £1,115,904. Mr Arden had been employed by the Bank since March 2018, so his "relevant income" was £673,191, a pro-rata share of his annual earnings.
  984. In relation to seriousness, the Authority had accepted that the breach was neither deliberate nor reckless; we have already set out its view in relation to both "the impact of the breach", and the nature of the breach,
  985. In relation to the "level 1 factors", "level 2 factors" and "level 3 factors", the Authority held that the following factors were relevant:
  986. (1)          no profits were made or losses avoided by the Bank or the Applicants as a result of the breach (DEPP 6.5B.2G(13)(a)); and

    (2)          the Applicants were knowingly concerned in Metro Bank's breach of LR 1.3.3R; this was negligent, having been caused by a lack of competence (DEPP  6.5B.2G(13)(d)).

  987. Taking all the above into account, the Authority decided that the seriousness of the breach was Level 3. The penalties before considering mitigating or aggravating factors were thus £223,180 for Mr Donaldson (£1,115,904 x 20%) and £134,638 (£673,191 x 20%) for Mr Arden.
  988. The Authority went on to decide that there were no mitigating or aggravating factors (Step 3); that the figure at Step 3 was sufficient to act as a deterrent to the Applicants and others (Step 4) and there was no settlement, so Step 5 did not apply.
  989. Step 2(1)-(3): Earnings

  990. The Applicants disagreed with the earnings used by the Authority as the basis for calculating the penalty charged on Mr Arden under Step 2(1)-(3). It was not in dispute that the Authority had included a "sign-on bonus" of £160,000 in Mr Arden's "relevant earnings". Mr Jaffey said:
  991. "That payment was made to compensate him for the loss of a bonus payment that his former employer (Sainsbury's Bank) would otherwise have paid. It was therefore not paid by the Bank as a reward for his services in the relevant period. Instead, it compensated him for services that he had performed in the past for a different employer, which are unrelated to the alleged wrongdoing, and which he forfeited by leaving."

  992. He went on to say that the £160,000 was therefore not "relevant income". Mr Stanley disagreed, saying that relevant income is "the gross amount of all benefits that a person received from the employment". As Mr Arden had received the sign-on bonus as part of his remuneration after he joined the Bank in March 2018, there was no basis to exclude it from the definition of "relevant income".
  993. The Tribunal's view

  994. We started with the meaning of "relevant income". This is "the gross amount of all benefits received by the individual from the employment" in question. The term "benefits" is then widely defined (our emphases) to include "salary, bonus, pension contributions,  share options and share schemes" and is not limited to those items. There is thus no exclusion for signing on bonuses.
  995. In addition, some "benefits" arise from work carried out during more than one year: for example share schemes and share options are typically earned over a period, at the end of which entitlement crystallises. The term "benefits" thus includes all sums earned in the year, irrespective of when the work was done. Consistently with that approach, there is no reason to exclude earnings which compensate for the loss of income from a previous employer.
  996. We also took into account the reasons given in the DEPP for using a person's earnings as the basis for the penalty, which are as follows:
  997. (1)          The first reason is that individuals receive remuneration commensurate with their responsibilities, and so it is reasonable to base the amount of penalty for failure to discharge their duties properly on the remuneration received. Here, the Bank had decided to pay Mr Arden earnings which included the signing-on bonus, and must therefore have decided that the total package was commensurate with his responsibilities, otherwise they would not have agreed to make the payment. In other words, it was part and parcel of the package to which he signed up when he took the new role.

    (2)          The second reason is that "the extent of the financial benefit earned by an individual is relevant in terms of the size of the financial penalty necessary to act as a credible deterrent". The Authority clearly took this into account, because it decided that a penalty calculated based on Mr Arden's earnings was sufficient to act as a deterrent to him and others and no additional amount was required under Step 4, see §593.

    (3)          The third reason is that even where a person's work is only partly related to a "controlled function", it is still appropriate to base the relevant income figure on all of the benefit an individual gains from the relevant employment. This, like the definition of "benefits" shows that a broad-brush approach is taken: there is no scope for apportioning a person's earnings to take into account different job roles.

  998. Taking into account all the above, we therefore agree with the Authority that Mr Arden's earnings at Step 2(1)-(3) of the calculation includes his signing-on bonus.
  999. Step 2(4)-(7): Seriousness

  1000. For the reasons already set out above, see §§568-571 and §591, the Authority's position was that the seriousness was at level 3. The Applicants submitted that the seriousness did not reach that level, for the following reasons, all of which we have already considered and rejected:
  1001. (1)          Reliance on legal advice, see our findings at §340.

    (2)          The announcement in January 2019 related to matters other than just the CLIP loans, and thus "there is therefore no evidence as to the extent of the share price movement that was due to the disclosure of the CLIP Loans issue", see §§257-258.

    (3)          It would have been reckless (and misleading) for the Bank to make an
    announcement of an unspecified and unquantified problem with its regulatory capital position, see §§432-440.

  1002. Mr Jaffey also pointed out that at the time the Authority decided that seriousness was at level 3, it had also decided that the Applicants had "failed to ensure" that "the estimated quantum of the correction was brought to the attention of the Audit Committee and the Board". However, on the basis of Ms Gillan's evidence, the Authority subsequently accepted that the Board had been told during the October 2018 meeting that the current estimate was around £600m. Mr Jaffey said that, in consequence, the level of the penalty should be reduced.
  1003. However, we have found that some of the information provided to the Audit Committee and the Board was partial, some was wrong, and that the overall impression was misleading, see §§460ff. We also take into account that one of the relevant factors in deciding on the level of seriousness is whether there was "no or little loss or risk of loss to... investors or other market users individually and in general". The breach in this case did expose investors to loss, see §575. 
  1004. Taking into account all other relevant matters, including the seriousness of the breach; the fact that the Applicants were negligent rather than reckless and did not act deliberately, and that penalties are paid out of post-tax income, we agree with the Authority that level 3 penalties are appropriate.
  1005. Step 3: Mitigation

  1006. Step 3 at DEPP 6.5B.3 is "mitigating and aggravating factors". Neither party considered that there were any aggravating factors, and the Authority's position was that there were also no mitigating factors. We first set out the related text from the DEPP; then consider the parties' submissions together with our view.
  1007. DEPP

  1008. DEPP 6.5B.3 reads as follows, so far as relevant to this case:
  1009. "(1) The  FCA may increase or decrease the amount of the financial penalty arrived at after Step 2...to take into account factors which aggravate or mitigate the  breach. Any such adjustments will be made by way of a percentage adjustment to the figure determined at Step 2.

    (2) The following list of factors may have the effect of aggravating or mitigating the  breach:

    (a) the conduct of the individual in bringing (or failing to bring) quickly, effectively and completely the  breach to the  FCA's attention (or the attention of other regulatory authorities, where relevant);

    (b) the degree of cooperation the individual showed during the investigation of the  breach by the  FCA, or any other regulatory authority allowed to share information with the  FCA;

    (c) whether the individual took any steps to stop the breach, and when these steps were taken;

    (d) any remedial steps taken since the breach was identified, including whether these were taken on the individual's own initiative or that of the  FCA or another regulatory authority;

    (e) whether the individual has arranged his resources in such a way as to allow or avoid disgorgement and/or payment of a financial penalty;

    (f) whether the individual had previously been told about the  FCA's concerns in relation to the issue, either by means of a private warning or in supervisory correspondence;

    (g) whether the individual had previously undertaken not to perform a particular act or engage in particular behaviour;

    (h) whether the individual has complied with any requirements or rulings of another regulatory authority relating to the  breach;

    (i) the previous disciplinary record and general compliance history of the individual;

    (j) ...

    (k)  whether  FCA guidance or other published materials had already raised relevant concerns, and the nature and accessibility of such materials;..."

    Submissions and discussion

  1010. The Authority decided that no mitigation should be applied to the penalty amount arrived at after Step 2. Mr Jaffey submitted that there were mitigating factors, and we agree. So far as relevant, we have incorporated the parties' submissions into our findings below.
  1011. Co-operation

  1012. Factor (b) of the factors listed in DEPP 6.5B.3 is the degree of co-operation during the investigation of the breach. The Applicants co-operated fully with the Authority and with the PRA; both Applicants attended interviews in June 2019 and December 2020. Mr Arden also led the Teach-In provided to the PRA on 3 July 2019, to assist them with their investigation, see §260. Mr Jaffey submitted that all of those points were relevant to co-operation.
  1013. The Authority's position in relation to Mr Donaldson was that he "had a legal obligation to attend his interview with the Authority and to give answers" and so the interviews were not relevant to mitigation. In relation to Mr Arden, the Authority took the same position, and added that:
  1014. "The teach-in presentation was given to the PRA only, and the admissions contained in it concerned historical governance, controls and the process for preparing COREP reports. The presentation and its contents were therefore not directly relevant to the Authority's investigation and so the Authority does not consider that Mr Arden's co-operation should be considered a mitigating factor."

  1015. We disagree. The co-operation referred to in factor (b) is that shown by individuals "during the investigation" carried out by the Authority. Interviews are a significant part of the approach taken to investigations, and while it is true that in a compelled interview, a person must answer the questions asked , there can nevertheless be varying degrees of co-operation and there is no basis for excluding interviews from factor (b) as a matter of principle. In relation to the Applicants, as Mr Jaffey said, "there has never been any attempt by either individual to present anything other than a full and honest account of the events to the Authority".
  1016. The Authority also dismissed as irrelevant Mr Arden's co-operation with the PRA, but factor (b) expressly includes co-operation with "any other regulatory authority allowed to share information with the  FCA", and that description plainly includes the PRA.
  1017. Remediation

  1018. Factor (d) relates to remediation. Mr Jaffey said that both Applicants worked with Deloitte and the regulatory authorities to correct the breach, and to improve the systems and controls which had been part of the underlying reasons for the incorrect RWAs having been applied in the first place.
  1019. The Authority's position is that factor (d) does not apply, because:
  1020. "Metro Bank disclosed the corrected RWA figure to the market in the January 2019 announcement to comply with its obligations under Listing Principle 1 and Article 17 of MAR rather than in order to remedy directly any harm arising from its breach of LR 1.3.3R."

  1021. We again disagree with the Authority: the factor reads "any remedial steps taken since the breach was identified", so it is wrong to exclude the action taken by Mr Donaldson and Mr Arden to remedy the CLIP loan issue. The Authority's submission that the factor applies only where a person has taken steps "to remedy directly any harm arising" unreasonably narrows the wording of this part of the penalty framework.
  1022. Compliance with the PRA's requirements

  1023. Factor (h) is whether the person complied with the requirements of another regulatory authority. The Bank kept Mr Sutherland informed of the RWA issue, and both Mr Donaldson and Mr Arden met him and others from the PRA on 12 September 2018; they explained what had happened and what the Bank was doing; on 9 October 2018, Mr Sutherland agreed with Mr Arden that the COREP reporting would remain unchanged while the Bank continued its work. Although this co-operation occurred before the breach, it is nevertheless relevant. On 21 January 2019, Mr Arden and Mr Donaldson met with the PRA and asked whether the Bank could phase-in the adjustment but this was refused the following day. The January announcement took place on 23 January. The Bank, via Mr Arden and Mr Donaldson, clearly co-operated with the PRA throughout.
  1024. However, the Authority says that the Applicants:
  1025. "...did not bring this to the attention of the Authority, and so the Authority does not consider the fact that the PRA was notified of the RWA error to be a mitigating factor."

  1026. That is, however, to ignore the wording of factor (h), which refers to complying with "the requirements of another regulatory authority". We again agree with Mr Jaffey that this factor is applicable.
  1027. Communications with the Authority

  1028. Factor (f) is whether the Authority had told either of the Applicants that it had concerns about the issue, either in a private warning or supervisory correspondence. The answer to that question is plainly "no", but Mr Jaffey emphasised that in contrast to the PRA, the Authority had not given responsibility for supervising the Bank to a dedicated team, although the Bank requested this on numerous occasions and despite it being the seventh-largest bank in the UK. When the Bank needed to contact the Authority, it had to do so via the Contact Centre (see our findings at §72). Mr Jaffey submitted that the absence of supervisory support and the lack of a conduit through which to contact the Authority meant that there was no equivalent to the regulatory discussions which took place with the PRA.
  1029. The Authority asked us to find that the lack of a dedicated supervisory relationship was irrelevant to this factor. We agree. There is no evidence that the Bank would have taken a different approach to the CLIP loan issue, had the Authority made itself more accessible. The Bank were able to ask for advice on the Listing Rules, not only from the Authority's Contact Centre, but also from Linklaters, but failed to do so. Instead, Mr Arden only asked for advice on the MAR, and even then he provided partial and incorrect information.
  1030. No negative factors

  1031. Factors (e), (g) and (i) did not apply: the Applicants did not rearrange their resources so as to avoid payment of the penalties, they did not breach any previous undertaking and they had no disciplinary record.
  1032. Other consequences

  1033. It was not in dispute that the Applicants have been punished in other ways: in particular, neither has been able to obtain a permanent post, see §267. Mr Donaldson said that this was the result of the Authority's enforcement action and the subsequent issuance of the Decision Notices, adding that the Authority's actions had also caused wider financial and reputational repercussions and had a "continued mental and physical impact".
  1034. We accept that the principal purpose of imposing a financial penalty is to deter the individuals and others from committing further breaches, as well as demonstrating generally the benefits of compliant business, see DEPP 6.1.2 cited at §577. The Applicants' position was that these wider consequences were themselves a sufficient deterrent and no penalty was required in addition.
  1035. We nevertheless decided that the wider consequences are not a relevant factor, for the following reasons:
  1036. (1)          We thought it unlikely that the reluctance of potential employers to offer the Applicants a permanent role was entirely due to the actions taken by the Authority in investigating and penalising the Applicants. On the balance of probabilities, the actions taken or not taken by the Applicants during the relevant period also play a part, and it is not realistically possible for us to disentangle the two elements.

    (2)          The fact that employers and others may take a negative view of a person who has been issued with a penalty by the Authority is also not relevant to mitigation; it is instead a reasonable and likely consequence of the Authority deciding to issue the penalty.

    (3)          The Authority has to take a view on mitigation when it sets the penalty. It is plainly not possible for it to take into account the future reactions of employers and others to its own issuance of the penalty.

    (4)          Although an application to the Tribunal is a complete rehearing of the issues which gave rise to the Authority's decision, that does not mean that we should take into account the consequences of the Decision Notices themselves.

    Difference between the Applicants?

  1037. We were not invited to distinguish the position of the Applicants, other than that in the context of Step 2, the Authority noted that Mr Arden did not "have significant industry experience, with this being his first senior role at a listed firm".
  1038. However, Mr Arden was much closer to the events in question than Mr Donaldson: he was the CFO; he had the meeting with Mr Lane, and it was Mr Arden who was required by the PRA to attest to the RWA numbers. Both Applicants attended the CRPAC meeting on 22 October 2018, at which the RWA Report was considered. Having considered all relevant matters, including the position taken in submissions, we have decided not to distinguish between Mr Donaldson and Mr Arden.
  1039. Conclusion on mitigation

  1040. We determine that the penalties are to be mitigated by 25%. Mr Donaldson's penalty is thus reduced from £223,100 to £167,325 and that of Mr Arden from £134,600 to £100,950. Setting the penalties at that level is, in our judgement, proportionate to the seriousness of the breach, as mitigated by the factors which we accepted should be taken into account.
  1041. DISPOSITION

  1042. For the reasons set out above, we agree with the Authority that the Bank breached LR 1.3.3R and that Mr Donaldson and Mr Arden were knowingly concerned in that breach.  
  1043. However, we determine that the appropriate action for the Authority to take in relation to the References is to mitigate the penalty charged on Mr Donaldson from £223,100 to £167,325 and to mitigate the penalty charged on Mr Arden from £134,600 to £100,950. Our decision was unanimous. 
  1044. The Decision Notices are remitted back to the Authority for it to give effect to our determination.
  1045. ANNE REDSTON

    UPPER TRIBUNAL JUDGE

     

    Release Date 16 June 2025



[1] The sets of questions linked with one another, so the output is sometimes described as "a decision tree" and at other times as "decision trees" but nothing turns on that.

[2] It was common ground that Mr Donaldson's reference to his belief that "c£40m assets are in scope for re-classification" should have been recorded as a statement that the recategorisation would affect £40m of Pillar 1 capital.

[3] This appears to be an error for £574m but nothing turns on that.

[4] The two versions were identical, other than in relation to one additional sentence which is not relevant to this decision.

About BAILII - FAQ - Copyright Policy - Disclaimers - Privacy Policy amended on 25/11/2010