AXA France IARD SA v Santander Cards UK Limited [2025] EWHC 1881 (Comm)

[2025] EWHC 1881 (Comm)

Case No CL-2021-000046

IN THE HIGH COURT OF JUSTICE

KING'S BENCH DIVISION

BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES

COMMERCIAL COURT

Venue Royal Courts of Justice, Rolls Building, Fetter Lane, London, EC4A 1NL

Date 25 th July 2025

Before

THE HONOURABLE MRS JUSTICE DIAS

Between

AXA FRANCE IARD S.A.Claimants

AXA FRANCE VIE S.A.Claimant

SANTANDER CARDS UK LIMITEDDefendants

SANTANDER INSURANCE SERVICES UK LIMITEDDefendant

And between:

Between

SANTANDER CARDS UK LIMITED

SANTANDER INSURANCE SERVICES UK LIMITED

SANTANDER UK PLC

Part 20 Claimants

Between

AXA FRANCE IARD S.A.

AXA FRANCE VIE S.A.

Part 20 Defendants

Mr Andrew Green KC, Mr Fraser Campbell KC and Mr Timothy Lau (instructed by Quinn Emmanuel Urquhart & Sullivan UK LLP) for Claimants and Part 20 Defendants Mr Adam Zellick KC, Mr David Murray and Mr Aaron Taylor (instructed by Reed Smith LLP) for the Defendants and Part 20 Claimants

Hearing Hearing dates: 11-14, 17-20, 24-26 March, 7-11 April 2025

Approved Judgment

This judgment was handed down remotely at 10.30am on 25 July 2025 by circulation to the parties or their representatives by e-mail and by release to the National Archives.

.............................

F: The Regulatory Framework 18

[1]From time to time, the court finds itself in the position of having to investigate events which took place many years or even decades ago and where, because it was never anticipated at the time that those events would later become the subject of fevered litigation, no particular steps were taken to preserve contemporaneous documentation or to crystallise recollections. Almost invariably, the key participants will have moved on, the memories of those remaining will have faded, and much of the oral evidence will come from people who either came much later to the story or were only peripherally involved (if at all) and can only speak in terms of general practice. In such circumstances, the court must perforce do the best it can on such material as is available to it.

[2]This is one of those cases. The dispute arises out of the historic mis-selling of payment protection insurance (“PPI”) over the period prior to 14 January 2005. For convenience these will be referred to as the “pre-2005”policies. (Quite what the exact starting point might have been, no-one was able to identify, although it was thought to be some time in the 1970s.) The policies in question were underwritten by various predecessors of the Claimants (“AXA”) as insurer, and sold as an adjunct to retail store cards by various predecessors of the Defendants (“Santander”) as credit lender. An important feature of this case is that for much of the relevant period, the various insurance entities and credit lending entities were all part of the GE corporate group reporting to the same profit centre within GE.

[3]As is well known, store cards are retailer-branded credit cards. They were marketed by credit lenders to customers through high-street retailers and required the customer to make a minimum monthly repayment in respect of any outstanding balance on the card. In return for a monthly premium added to the credit balance (typically 78 pence per £100 outstanding), PPI insurance could also be purchased, which typically covered some or all of the customer’s repayment obligations under the credit agreement in certain events, including, death, accident, illness, or redundancy. As premium was only charged by reference to the balance outstanding on the card, the amount could fluctuate depending on the extent to which the card was used. Furthermore, accounts could come in and out of cover as and when the balance dropped to nil. Over time, additional benefits were often added, such as price protection (which reimbursed the customer if the purchased item was subsequently reduced in price within a specified period) and purchase protection (which covered accidental damage to purchased items within a specified period.)

I: “Responsibility” for Mis-Selling as Between FICL/FACL and GECB 36

[4]The marketing of store cards and any associated PPI product mainly took place in retail stores at the till, although some PPI sales were concluded later over the telephone. The sales were conducted pursuant to arrangements made between the credit lender and the individual retailers and the insurer was not physically involved in the actual selling process. The insurance itself was a fairly standard product, albeit with minor variations in cover for different stores. No doubt reflecting this general division of input, around 95% of the policy premium remaining after payment of claims was typically paid to the credit lender while around 5% was retained by the insurer. See further paragraphs ‎50-‎52 below.

[5]During the course of the late 1990s and early 2000s, complaints relating to the mis-selling of PPI generally began to increase, leading to a veritable explosion from around 2011/2012, which was undoubtedly fuelled by the involvement of more or less scrupulous claims management companies. This ultimately resulted in redress being paid to hundreds of thousands of customers by way of premium refunds which, with interest, totalled some £38 billion across the industry for all forms of PPI. PPI was not, of course, only sold as an adjunct to store cards. It was also sold, for example, in conjunction with mortgages and personal loans. Because of the way in which the regulatory regime worked (of which more anon) and notwithstanding that the complaints overwhelmingly related to the way in which the PPI had been sold rather than the product itself, it is AXA which has been required as inheritors of the insurer’s liabilities to bear the costs of this redress in respect of pre-2005 sales amounting to almost £500 million plus more than £70 million in respect of fees payable to the Financial Ombudsman Service (the “FOS”) for complaints referred to the FOS, as well as the internal administrative costs of processing complaints.

[6]In relation to policies sold after 14 January 2005, it is Santander as inheritor of the liabilities of the credit lender which has assumed full liability for all customer redress and associated costs. This is because under the Financial Services and Markets Act 2000 (“FSMA”), the credit lenders themselves became regulated entities at that date in relation to their insurance mediation activities and, as such, were obliged by the regulators to take responsibility for any complaints arising out of policies sold thereafter.

J: The Settlement Claim 60

[7]In practice, the sales of the PPI policies in issue before me were reported by the credit lender to the insurer in bulk which meant that the insurer had no customer data which would have enabled it to handle complaints effectively. For that reason, complaints handling in relation to the pre-2005 policies was in fact undertaken on behalf of AXA by Santander, latterly pursuant to a Claims Handling Agreement (the “CHA”) concluded in December 2017 (subsequently amended in December 2018). Pursuant to the CHA, AXA has paid service charges of over £120 million to Santander.

[8]By this claim, AXA seeks to recover its losses relating to the pre-2005 policies from Santander on a number of alternative bases. No claim is made in respect of the service charges paid to Santander under the CHA which are now accepted to be irrecoverable from any Santander entity:(a) First, AXA alleges that a settlement agreement (the “Settlement Agreement”) was concluded between the parties in 2015 pursuant to which Santander agreed to take responsibility for handling all complaints arising out of the mis-selling of pre-2005 PPI policies sold by its predecessors and underwritten by AXA’s predecessors and to pay all associated customer redress, FOS fees and administrative costs;

(b) Alternatively, it alleges that Santander is obliged to indemnify AXA against any liability arising out of its predecessors’ acts or omissions pursuant to an indemnity clause contained in an agreement (the “Agency Agreement”)concluded between the relevant parties on 1 December 2000;

(c) Alternatively, it alleges that it is entitled to claim some but not all of its losses under the Civil Liability (Contribution) Act 1978 (the “Contribution Act”) on the basis that both AXA and Santander were liable to the customers whose claims were paid by AXA;

(d) Finally and alternatively, it claims damages in common law negligence for breach of duties owed directly to AXA’s predecessors by Santander’s predecessors as their agent. It is conceded that a claim on this basis for losses arising in relation to any policies sold before December 2002 is now time barred. A standstill agreement was entered into in December 2017 such that claims in relation to policies sold during 15-year period from 2002 are preserved.

K: The Indemnity Claim 66

[9]For its part, Santander pursues two counterclaims. The first is not controversial and relates to around £7 million deducted by AXA from sums otherwise payable to Santander pursuant to a profit-sharing agreement in the Agency Agreement. AXA accepts that this sum should be set off against any award in its favour. The second arises only if the Contribution Act is found to be applicable (Santander’s primary case being that it is not). In that event, Santander asserts that it is entitled to recover some £151 million that it has itself paid directly to customers in respect of pre-2005 sales on the basis that responsibility should be apportioned in whole or in part to AXA’s predecessors. A further unrelated counterclaim for rectification was conceded by AXA before trial and has now fallen away. B: THE PARTIES

[10]The corporate history on both sides is exceedingly complicated. Fortunately for both court and reader, the following can be taken as a sufficiently accurate summary for the purposes of the present dispute.

[11]The relevant policies were all underwritten by one or both of two English domiciled companies, Financial Insurance Company Limited (“FICL”) and Financial Assurance Company Limited (“FACL”). It is unnecessary to draw any distinction between them and they were referred to collectively throughout the trial as “FICL/FACL”. Historically, FICL/FACL operated jointly under the trading name of GE Financial Insurance which was part of the GE Capital corporate group until it was transferred to the Genworth corporate group (“Genworth”) in 2004. Genworth was in turn purchased by AXA on 1 December 2015 at which point, the First Claimant inherited the business, rights and obligations of FICL while the Second Claimant inherited the business, rights and obligations of FACL. Both Claimants are domiciled in France and carry on business as insurance underwriters within the AXA corporate group. Again it is unnecessary to draw any distinction between them and I shall therefore refer in this judgment simply to FICL/FACL, Genworth and AXA as appropriate.

[12]The Defendants are both members of the Santander group domiciled in England. The policies in issue in these proceedings were all sold on behalf of FICL/FACL by the First Defendant (“SCL”) which, for much of the relevant time, was called GE Capital Bank (“GECB”). GECB was originally part of the Debenhams corporate group before being acquired, first by the Burton group and subsequently, in July 1990, by the GE Capital corporate group. In 2009, GECB was acquired by the Santander group and changed its name to SCL. Pursuant to a novation agreement dated 22 January 2010, SCL novated its rights and obligations relating to UK insurance business under the Agency Agreement to the Second Defendant (“SISUK”). For the most part I shall refer in this judgment to GECB and Santander as appropriate, only drawing a distinction between SCL and SISUK where it is necessary to do so. The Third Part 20 Claimant, Santander UK plc (“SUK”) is the parent company of SCL and was joined to the proceedings for the purposes of the now-redundant counterclaim for rectification. Other than featuring in some of the contemporaneous documents it plays no substantive part in this dispute.

[13]From approximately May 1992, FICL/FACL and GECB were both part of the GE Capital group and FICL/FACL were the exclusive underwriters of store card PPI offered by GECB in the UK. It was common ground that in relation to the sale of PPI specifically (as opposed to the credit lending aspects of the store cards), GECB was acting as FICL/FACL’s agent. From 1 December 2000, the relationship between the parties was governed by the Agency Agreement referred to above, although a series of drafts makes clear that it was under negotiation from at least 1996, if not earlier. C: THE SCOPE OF THIS TRIAL

[14]It was common ground that the present hearing should determine the question of liability. The issues which arise in this respect are identified in relation to each of AXA’s separate causes of action in sections J-M below. However, quantum is altogether more complicated. AXA’s losses relate to over 650,000 individual customer complaints. With the sanction of the court, the parties sensibly agreed to proceed on the basis of a statistical sampling exercise. As a result, they have agreed that a sample of 771 complaint files (the “Complaints Sample”) can be treated as representative of the total population of complaints in issue in the proceedings, such that it can be assumed that if a particular proportion of the Complaints Sample exhibits a particular category of mis-selling, the same proportion of the overall population will have been similarly mis-sold. Various categories of mis-selling were also identified and the parties have agreed the correct categorisation of 469 files in full and a further 141 files in part. The categorisation of the remainder is still contentious.

[15]In the event, so far as concerns quantum the parties agreed a way forward which was encapsulated in a Consent Order approved by the court on 13 February 2025 to the effect that:(a) The court will determine whether each of the agreed categories of mis-selling is capable in principle of giving rise to liability on the part of Santander to AXA;

(b) For each such category, the court will also determine the type of factual circumstances in which a particular sale within that category will give rise to liability, informed by a representative sample of disclosed complaint files;

(c) The court will give such guidance in principle as it considers appropriate as to the evidence required to establish whether a particular file exhibits the relevant circumstances, for example, whether in the absence of evidence to the contrary it is appropriate to rely on a customer’s assertion as to his or her personal circumstances and/or as to how a sale took place.

[16]It is not envisaged that the court should finally settle any disputes on categorisation or determine any questions relating to extrapolation. Rather, the intention is that, armed with the guidance referred to above, the parties will exercise best endeavours to resolve these matters before returning to court for a further hearing at which the outstanding questions will either be determined or directions given for their determination.

L: The Contribution Claim 80

[17]On the opening day of trial, I voiced some concern as to how the court could or should approach the matters set out in sub-paragraphs (b) and (c) above. For example, to what extent, if at all, was the court being asked to make a determination which went further than merely confirming that specific sample cases fell into particular categories? What sort of guidance on evidence was being sought when it was not suggested that either party should be at liberty to adduce further evidence to bolster its position apart, perhaps, from referring to hitherto overlooked or unappreciated documents already in the trial bundles?

[18]By the conclusion of the hearing, the parties were agreed that there were some issues of general principle which I should determine finally as far as possible and they helpfully set them out in a schedule which I address in a separate Appendix to this judgment. They also agreed that I should be at liberty to confine myself to making particular determinations on the facts of specific sample cases if I felt that a more general statement was not possible or appropriate. That is accordingly the approach I have adopted. D: THE WITNESSES

[19]I have already alluded to the difficulty of trying to make findings about events which occurred in the distant past against the background of fading, if not already faded, memories and without a comprehensive documentary record. Given the general paucity of documentation covering the early years of the relationship between the parties, it was easy for both parties to home in on isolated specific documents in order to challenge evidence given by the witnesses. However, the fact that one or other party behaved in a particular way at one or more specific points in time clearly does not mean that it necessarily behaved consistently in that way over the entire or, indeed, any particular period of the relationship. Put more simply, the fact that assertion X might be falsified in respect of one particular date does not logically lead to the conclusion that the converse proposition, Y, is established, let alone established as an invariable practice. This is a point which obviously cuts both ways and I bear it well in mind in weighing the oral and documentary evidence.

[20]AXA called oral evidence from the following four witnesses.

[21]Mr Howard Jones: Mr Jones is now retired. He joined GE in 1994 having previously worked for most of his career in insurance, initially on the GECB side of the fence overseeing its telesales operation in Leeds. In 1995 he became responsible for the insurance side of GECB’s business and in that capacity began working together with FICL/FACL to provide PPI in conjunction with store cards. Most of his day-to-day work involved managing the insurance products sold by GECB. In about May 1996, he and GECB’s other insurance personnel moved to the FICL/FACL side of the fence where he became Client Relationship Manager with general responsibility for the GECB account, reporting to FICL/FACL’s Assistant General Manager, Mr Ken Mothersdale. In July 1998, he moved to GE Life at which point he ceased to have any involvement in store card PPI. He left the GE group altogether in 2000.

M: The Negligence Claim 89

[22]Mr Jones was an honest witness who was trying to assist the court to the best of his ability. But, as he frankly accepted, his memory of events over a four-year period more than a quarter of a century ago was very hazy. He readily admitted under cross-examination that his recollection might be mistaken and when pressed on particular details by reference to specific documents, he capitulated relatively easily. That said, he maintained that he was left with residual impressions which I am reluctant to dismiss out of hand.

[23]Mr Justin Halse: Mr Halse is also now retired. He joined FICL/FACL in 1998 after working at Lloyd’s of London. Soon thereafter, he became Marketing Finance Manager in which capacity he oversaw the finance aspects of pricing FICL/FACL’s insurance products. In 2005 he led the Deal Review Committee, which was a cross-departmental committee reviewing all new product proposals and deals with new clients or extensions to existing arrangements. In 2007 he also became head of the Risk team at FICL/FACL. His evidence mainly related to the pricing of the PPI products sold by GECB and was largelyuncontentious.

[24]Mrs Emma Smith: Mrs Smith (then Ms Brown) joined FICL/FACL after graduating in 2000, initially in the marketing team. In around 2001 she moved to the sales department where she was part of the client relationship team for Black Horse Finance Limited. In 2003, I reject Santander’s submission that this date was a mis-recollection on Mrs Smith’s part and that she only first became involved in the GECB account in September 2004. Santander argued that the earliest document which could be found to refer to her was from the later date. However, despite being pressed hard on the point, Mrs Smith’s firm recollection was that she had been involved with the account for some time (i.e., more than a few months) before 2005 and I accept her evidence. she became the Client Servicing Manager for GECB, reporting to the then Client Relationship Manager, Mrs Kate Greenaway, and it was in that capacity that she became involved with store card PPI. In 2005, she left FICL/FACL briefly to join Marsh McLennan but returned to the fold in 2006 in a completely different capacity and has remained there ever since.

[25]I found Mrs Smith to be an honest and reliable witness. She was frank and realistic in her evidence when she could not recall something but stood her ground effectively and convincingly where she could. I accept her evidence.

[26]Mr James Rember: Mr Rember joined GE Capital in 1998 as Senior Legal Counsel. In 2001, he moved to FICL/FACL as European Legal and Compliance Director supporting its consumer insurance business division. When FICL/FACL were transferred to Genworth in 2004, he moved with the business to become General Counsel of Genworth’s Credit & Lifestyle Protection business division. When that division was in turn acquired by AXA in 2015, he remained as General Counsel with an expanded role which now included the AXA group’s existing global creditor insurance business. He remained in that role until August 2023 before joining Coverys UK in December 2023 after a short period acting as consultant to the AXA group. Mr Rember’s main involvement in the matters relevant to this action related to the period from 2012 including, in particular, the events surrounding the disputed Settlement Agreement and the negotiation of the Claims Handling Agreement.

[27]It appears that Mr Rember did not part company with AXA on entirely amicable terms since he brought an employment claim against it. The claim was settled on terms which provided for him to give evidence for AXA in these proceedings in return for payment at the rate of £450 per hour with a minimum retainer of £900 per month. Nonetheless, I do not consider that the fact that he was being paid by AXA for his assistance detracts from the weight to be attached to his evidence. As he was no longer an employee and so not under any obligation to give evidence, I do not find anything objectionable in him having agreed suitable terms on which he would be prepared to devote a no doubt substantial amount of time and effort to assist. Mr Pateman and Mr Brandon-Cross were likewise remunerated for their time by Santander. The amount of Mr Rember’s remuneration is entirely appropriate for a professional person and is nowhere near a level which might be thought to cast doubt on his integrity.

[28]As it was, save in relation to one matter (addressed later), I found him to be straightforward and concise in his evidence and prepared to answer the questions put to him fully and fairly. Santander, however, sought to impugn his credibility on the basis of AXA’s conduct in relation to the CHA and the (now-conceded) claim for rectification. In simple terms, the parties had agreed that the service charges payable to Santander for handling claims under the CHA should be irrecoverable in all circumstances, the intention being to provide protection to both SCL and SISUK. But due (as we now know) to an error on the part of Santander’s external counsel, the wording of the final agreement was changed in such a way as apparently to exclude SISUK from the scope of this protection, and AXA took advantage of this mistake in order to plead a claim for recovery of the service charges. On behalf of Santander, Mr Adam Zellick KC submitted that this was discreditable in circumstances where Mr Rember had personally agreed that the charges should not be recoverable.

[29]As to this, I am not persuaded that Mr Rember acted in any way improperly, let alone dishonestly. Indeed, it is quite possible that neither side fully appreciated the implications of the drafting change when the CHA was signed. However, when proceedings came to be issued, I find that Mr Rember almost certainly suspected that Santander had made a mistake and potentially missed a trick of which AXA might be able to take advantage. Given that Santander was as big a beast in the jungle as AXA and, moreover, a litigation opponent, I do not regard Mr Rember as having been under any obligation to draw the potential mistake to Santander’s attention. He was entitled to assume that it could look after itself. It was undoubtedly unsatisfactory that the claim was not withdrawn more promptly after Santander proffered an explanation in December 2021, but the point goes no further than that and I do not find that it casts doubt on Mr Rember’s evidence more generally.

[30]I should also say that Mr Rember was meticulous about confirming oral discussions in a follow-up email. This is in stark contrast to Santander which rarely, if ever, took notes or made a record of meetings or conversations. None of Mr Rember’s emails or records was challenged as inaccurate contemporaneously and I accept them as reliable. In so far as Mr Hazell gave evidence to the effect that they were not in fact always correct, he took no steps to say so at the time or to have them corrected, and I disregard that evidence which I find somewhat unsatisfactory.

[31]The witness statement of a fifth witness of fact, Mr Mark Doherty, which related to accounting matters, was admitted by agreement without Santander requiring him to attend for cross-examination.

[32]The following witnesses gave oral evidence on behalf of Santander.

[33]Mrs Kate Greenaway: As noted in paragraph ‎24 above, Mrs Greenaway was FICL/FACL’s Client Relationship Manager for the GECB account and the line manager of Mrs Smith. Her first position within the GE corporate group in 1993 was as a Marketing Assistant. She started working with PPI directly from about 1996/1997 when she became a Client Servicing Manager for FICL/FACL’s client relationship with GE lenders, including GECB. In 1998, she moved to fill the same role in relation to certain non-GE clients and in 1999 she was promoted (with effect from 2000) to Client Relationship Manager for FICL/FACL’s PPI products. In that capacity she assumed responsibility for the management of FICL/FACL’s relationship with GECB (and others), in succession to Mr Jones. Mrs Greenaway remained with the business throughout the acquisition by Genworth but was made redundant in 2007, following which she was immediately offered a job by GECB, moving to SISUK after GECB’s acquisition by Santander. She did not have any material involvement in relevant events after 2007.

[34]It is fair to state that Mrs Greenaway’s recollection was not particularly good. However, she was a cautious and careful witness, and I found her to be honest and straightforward in her answers although I did wonder whether she was tempted at times to take refuge in her lack of clear recollection as a way of avoiding difficult questions. Nonetheless, so far as her evidence went, I accept her as a witness of truth.

[35]Mr Bradley Brandon-Cross: Mr Brandon-Cross had been involved with PPI from the mid-1980s and joined the GE group in 1994 when it acquired the company for which he was then working. From then until 2000, he was Managing Director within the GE group of, first, international business and, subsequently, UK business (which latter included FICL/FACL). He was also Chair of the PPI Committee at the Association of British Insurers (“ABI”) from 1998-1999. As such, Mr Brandon-Cross’ evidence was given at a high level, and he had very little direct knowledge of the specific matters relevant to this litigation. As a result, and despite some combative exchanges with counsel, he was not always able to support the assertions in his witness statement with concrete evidence.

[36]Mrs Philippa Owen: Mrs Owen joined the financial side of GECB in September 2003, initially as an auto pricing manager. Her first direct involvement with store card PPI was in August 2008 when she became Head of Financial Planning & Analysis, where she was responsible for managing the profit and loss account, including the financial aspects of store card PPI. From around 2009/2010, issues of consumer redress became an increasingly large part of her role, including responsibility for making adequate provision in the accounts. Despite having been due to leave Santander at the end of 2014, she was asked to stay on specifically to continue managing the financial aspects of PPI complaints and from 2015-2021 she was head of Regulatory and Remediation Finance at Santander UK.

[37]I found Mrs Owen to be a straightforward and credible witness. She was careful in her evidence, and I accept her as a witness of truth.

[38]Mr Timothy Lloyd: Mr Lloyd was formerly General Counsel and Company Secretary of Alliance & Leicester plc and moved with it when it was acquired by Santander. In October 2014, he was appointed General Counsel of Santander UK plc, a position he retained until his retirement at the end of August 2016. He was a member of Santander’s Executive Committee (“ExCo”) from 2008. As Mr Lloyd frankly accepted, he only had a walk-on part in the drama being played out between the parties, albeit an important one. He had not previously been aware of the dispute regarding PPI redress and was only made aware of it for the first time in 2015 when he was briefed to attend the critical meeting on 4 June 2015 at which AXA asserts that a settlement was agreed. His evidence was thus confined to that meeting and the comparatively brief period thereafter until he retired.

[39]Mr Lloyd was a pleasant and open witness. He gave his evidence in a fair and straightforward manner, and I have no hesitation in accepting him as a witness of truth.

[40]Mr Steve Pateman: Mr Pateman was Executive Director and Head of UK Banking at Santander UK from 2011 to September 2015 reporting directly to its CEO. He only became involved in the dispute with Genworth in 2015 and, like Mr Lloyd, his only direct participation was at the meeting of 4 June 2015.

[41]Mr Pateman was a fair and measured witness although, as he realistically accepted, his recollection of events nearly a decade ago was very limited in circumstances where he was only asked to recall them for the first time in 2023. On his own admission he could not recall precise words or details. I found it somewhat implausible that he had not been made aware of the agreements reached between Genworth and Santander in 2012 and 2013 given the size of the sums involved, although he maintained that this was indeed so. The answer may lie in the fact that, as he described it, Santander operated in a very “siloed” manner where each unit had its own clearly defined areas of responsibility and specific reporting lines, and it was only at ExCo level that cross-unit issues would be discussed.

[42]Mr David Hazell: Mr Hazell joined Santander in March 2012 as Director of Compliance for the UK group companies. He remained in that position until he moved to Santander Holdings USA Inc in the United States in November 2016. During that time, he was responsible for advising upon and monitoring compliance with regulatory rules, guidance and expectations. However, he was only marginally and occasionally involved with PPI issues as these were handled by a ring-fenced PPI Remediation team. Remediation was a separate function within the Internal Governance and Control team which was headed by Mr Jorge De La Vega and did not fall within Mr Hazell’s compliance remit. As such, he and his team were only called upon to support the Remediation team with regulatory advice when requested.

[43]Mr Hazell was nonetheless involved to some extent in the dispute with Genworth throughout 2012 until 2015. He also attended a meeting with Genworth in August 2012 as well as the meeting on 4 June 2015.

[44]As a witness, Mr Hazell was less than satisfactory. His recollection was patchy and did not always reflect the contents of his witness statement, although I do not necessarily criticise him for that as it is not unusual for dormant memories only to be triggered on being shown particular documents. However, he was unnecessarily defensive and was seemingly so determined to give away as little as possible that he repeatedly refused in cross-examination to comment on any documents to which he was not himself party – even when he was being asked to accept no more than that they said what they said. Despite being Head of Compliance, he disclaimed knowledge or responsibility for much of what was being discussed and agreed within Santander regarding PPI remediation. This was on the basis that it was all being dealt with by the Remediation team led by Mr De La Vega and that the compliance team would only get involved in response to specific requests. Towards the end of his evidence, however, he stated that he would have updated Mr De La Vega on the outcome of the 4 June 2015 meeting. In answer to a query from the Bench as to why he had taken it upon himself to do this given what he had previously said about the separation between the two teams, he claimed that he had worked closely with Mr De La Vega and he would therefore have felt it appropriate to do so. The suggestion of a close working relationship with Mr De La Vega had never previously been made in either his written or oral evidence and, indeed, was completely at odds with the thrust of his earlier testimony. It was also noteworthy that, while Mr Hazell claimed to have had very clear opinions about where liability should sit for the pre-2005 complaints which frequently differed from the views expressed by others within Santander, yet there was no evidence that he ever challenged those different views, either orally or in writing. Overall, I was left with the impression that the views he expressed in court were not necessarily views he expressed quite so trenchantly at the time. I therefore treated his evidence with some caution where it was not supported by the contemporaneous documents.

[45]A statement from a seventh witness, Mr Alan Conway, was admitted under the Civil Evidence Act 2005 as Mr Conway is sadly seriously unwell and was unable to attend court to give evidence in person. He joined Santander as an in-house solicitor in 2005 following its acquisition of his previous employer, Abbey National, and worked as a transactional lawyer with a focus on commercial contracts. At the end of October 2014, he was asked by Ms Joanna Day, his ultimate line manager, to become involved in the dispute between Genworth and Santander and he was accordingly party to the correspondence both leading up to and following the meeting on 4 June 2015, which he attended.

[46]The factual evidence conflicted in a number of respects. I have no doubt that each witness was honestly trying to assist the court to the best of his or her recollection but, given the passage of time since the events in question and the fact that many of the witnesses were only recently asked to recall what happened, it is inherently likely that to some extent their recollections were based on an element of reconstruction, whether conscious or not. On top of that, it is in the nature of things that individual participants in the same event will approach matters from their own perspective and are accordingly likely to perceive and remember things in different ways which emphasise different aspects of what occurred. For all these reasons it is entirely unsurprising that recollections may differ, and where this happens it is ultimately for the court to weigh the competing accounts as best it can in order to decide on a balance of probabilities what is most likely to have occurred.

[47]Each party called expert evidence on the marketing and regulation of credit insurance in the period c. 1990-2005: Mr Stuart Pendleton for AXA and Mr John Blundell for Santander. As one might expect, there was substantial agreement between them on most points, including in particular the standards which applied to the selling of store card PPI over the relevant period. Both experts were knowledgeable and authoritative, and I was satisfied that their evidence was a fair reflection of their honest opinion.The few matters on which they differedare discussed below in relation to the relevant issues. E: THE RELATIONSHIP BETWEEN FICL/FACL AND GECB IN PRACTICE

[48]The legal relationship between FICL/FACL and GECB is considered in more detail below. So far as the practical aspects of that relationship are concerned, however, I find the following to be established on the evidence.

[49]GECB was the dominant provider and distributor of store cards in the UK and by 2004 had captured around 50%-70% of the market, representing some 12.5 million customers. Many of these customers also purchased PPI policies which generated approximately £1 billion in premium. The precise mechanics of selling were determined between GECB and each individual retailer and FICL/FACL had no involvement in the physical sales. Indeed FICL/FACL were excluded from GECB’s relationship with the retailers (no doubt for reasons of confidentiality) and so had no direct contact either with them or with the end customers (unless and until a claim was made under the policy). Moreover, GECB was reluctant to permit FICL/FACL to have access to its telemarketing centre in Leeds (described as the “jewel in its crown”) and attempts by FICL/FACL to get visibility over customer data and, for example, penetration rates, The penetration rate refers to the proportion of insured balances and indicates the extent of PPI take-up by customers. were generally either rebuffed or ignored. Thus, FICL/FACL could not actively seek out customers for themselves and depended on policies being sold by GECB alongside the store cards. It was not in dispute that in this respect GECB was acting as FICL/FACL’s agent and that the retailer was acting as a sub-agent of GECB.

[50]As indicated in paragraph ‎4 above, around 95% of the policy premium remaining after payment of claims went to GECB as lender while 5% was retained by FICL/FACL. The allocation of premium was in theory a matter for negotiation between GECB and FICL/FACL on a scheme-by-scheme basis, although the extent to which FICL/FACL had any real bargaining power in the relationship was questionable (see paragraphs ‎55ff below). That said, the allocation of PPI premium was broadly similar across the entire industry and typically reflected the following:(a) Around 50%-70% would be payable upfront to GECB as advance commission. This commission was never repayable in any circumstances;

(b) Around 5% was retained by FICL/FACL to fund their administration and set-up costs and provide a profit margin;

(c) Around 25%-40% was retained by FICL/FACL as a claims fund. Any surplus left in the claims fund after all claims had been met (typically about 25% of the fund) was distributed as profit share. Invariably 100% of any profit share went to GECB and any shortfall in the claims fund was borne by FICL/FACL, although there was some possibility of off-setting losses against prior or subsequent periods;

(d) Around 2% was paid to the government as Insurance Premium Tax (“IPT”).

[51]After payment of claims, therefore, GECB typically retained around 61%-76% of the total premium. From this it would fund its own costs and expenses as well as any fees or costs payable under its retailer agreements albeit, on the evidence, the majority of these costs related to the store card credit lending aspects of the agreement, rather than the PPI product specifically. There were thus clear advantages to GECB in selling as many policies as possible. Not only was it profitable business in itself for GECB, but the PPI also protected GECB’s credit risk as it covered certain accidental and unforeseen events that might otherwise have affected the customer’s ability to service the loan. In addition, as explained by Mr Blundell, profits from the sale of PPI could be used by GECB to expand its position in the credit market, for example by cross-subsidising its core lending product so as to offer more competitive interest rates than its competitors.

[52]By contrast, FICL/FACL’s net margin after payment of their own costs and expenses was around 2.5%-3%. While they could theoretically earn interest on the claims fund pending disbursement, their ability to do so was necessarily constrained by the need to ensure that the fund was readily available to meet claims as and when required.

[53]Sales of PPI policies were reported by GECB to FICL/FACL in bulk by means of bordereaux without any details of individual transactions. FICL/FACL thus had no visibility over any customer data unless a claim happened to be made under a policy. For this reason, they were not in a position to understand what new types of cover might be attractive to potential customers or how existing products could be improved. I am satisfied on the evidence that the design of new products and any proposed amendments or enhancements to cover was therefore almost entirely driven by GECB and that FICL/FACL were purely reactive to its requests. This is consistent with the view expressed by the Insurance Ombudsman In a speech delivered to the IBC Creditor Insurance Conference on 10 December 1997. that it was lenders in the industry who designed PPI products.

[54]Whenever a request was made by GECB to FICL/FACL for a piece of new business (be that cover for an entirely new scheme or an upgrade to the cover offered on an existing scheme, or a completely new product), it would typically also suggest an appropriate retail price and premium split. Although sometimes the pricing negotiations would start with FICL/FACL’s wholesale price to which GECB would add a margin. FICL/FACL would then consider whether they could write the proposed cover at the proposed price. Since they did not have access to customer data, they could only rely on their experience of comparable schemes with other lenders and any change in pricing would have to be negotiated with GECB, along with the detailed allocation of premium.

[55]The evidence of AXA’s witnesses, which I accept, was that negotiations with GECB were very tough and that GECB was constantly applying pressure on FICL/FACL to increase the percentage of advance commission and reduce the claims retention, since advance commission was essentially “risk free” money from its point of view. Notwithstanding that the companies were part of the same corporate group and that FICL/FACL operated effectively as captive in-house insurers which GECB was expected to use, GECB could in theory take its business elsewhere if FICL/FACL refused to accept its proposed terms and, according to Mr Halse, this always felt like a real possibility. No doubt, as Mrs Greenaway said in her statement, the companies were working together with the shared goal of developing and maximising the business for the overall benefit of the GE group, but as Mr Brandon-Cross stated in his witness statement, FICL/FACL’s strategy was to work with GECB to sell PPI and boost income and profit for GECB, which was a separate corporate entity from FICL/FACL with separate accounts. What is clear from the evidence is that FICL/FACL’s share of the pie was towards the lower end of the ranges which were regarded as normal in the industry at the time.

[56]Further, the evidence of Mrs Smith and Mr Jones was that GECB was very demanding as regards the time in which it expected FICL/FACL to deal with requests for new cover and the level of support that it required. FICL/FACL would frequently be asked to respond to a request from GECB within a matter of days so as to meet a planned launch date for the new scheme or product. According to Mr Jones, GECB was much closer than FICL/FACL to the general GE culture of pushing to lead the market, win work and innovate. That said, FICL/FACL had the last word on the terms and conditions of any policy, for example with regard to exclusions, and some 70% of their business was with other lenders mainly operating outside the store card market.

[57]Once agreement was reached between the parties on a particular product, an addendum would be agreed which set out the basic details of the scheme, its key provisions and eligibility requirements and the premium split. The addendum also identified the party responsible for the costs of producing policy documents, marketing materials and claims forms. These addenda were intended to be appended to the Agency Agreement which was under negotiation between the parties from about 1996 and were in fact so appended when it was finally signed on 1 December 2000.

[58]Although GECB’s witnesses maintained that the relationship between the parties was amicable, having heard the witnesses, I accept that relations were difficult. Mr Mothersdale expressed it thus in 1997:

“Among the more difficult of our clients is GCF, who, because they are a sister GE company and because of the personality involved are aggressively demanding and unreasonably intolerant at times.”

Furthermore, GECB had very much the whip hand in all negotiations. Credit lending was its core business and it had a huge share of the store card market, easily outstripping its competitors. Only GECB had access to the retailers and the customers and only GECB was in a position to assess what products were likely to be attractive to end users and what would sell well. In the absence of any customer data, FICL/FACL were unable to form any view on such matters. Thus, it was GECB which effectively drove the design and pricing of products and FICL/FACL’s role was essentially relegated to that of finding a way of providing what GECB wanted. Moreover, although FICL/FACL were not dependent on GECB for the majority of their income which (as already noted) came from outside the store card market, equally it represented a significant proportion of that income which could not readily be replaced were GECB to start dealing with another insurer. I therefore reject the suggestion that GECB was a mere go-between or cipher interposed between FICL/FACL and the end customer.

[59]The actual sales process is discussed in more detail later. At this point, it need only be mentioned that the vast majority of PPI policies (around 95%) were sold at the point of sale by the retail staff of the relevant store at the same time as the customer completed a customer credit application form for a store card. This was perceived as being by far the most efficient way to achieve take-up. It almost invariably took place in public at the till while the customer was also purchasing goods from the retailer and customers were frequently offered a discount as an incentive to take out a store card. Prior to 1994, PPI was generally sold on an “opt-out” basis where it was included in the transaction unless the customer positively indicated to the contrary. However, such “negative option” selling was phased out from 1994, whereafter the credit application form for the store card included a section requiring the customer to indicate positively that he or she wanted cover, either by signing or ticking a box - so-called “opt-in” cover.

[60]It was not disputed that GECB’s primary concern was with its core lending business and that the focus of the transaction was therefore very much on the store card. On GECB’s own estimate, typically only around 20-30 seconds was devoted by sales staff to selling PPI in what was generally a fast-paced and pressured environment, particularly during end-of-season sales or other busy times when there were lengthy queues at the tills. It was also not disputed that sales staff were often part-time or temporary, for example students or teenagers doing weekend or vacation work, and that they were frequently incentivised to sell as many policies as possible. Some stores nonetheless offered customers the possibility of going to a separate part of the store to consider and/or discuss the credit agreement and ancillary PPI before signing up.

[61]A small minority of policies was sold over the telephone by GECB’s own call centre staff, often during a “welcome call” following up on the conclusion of the credit agreement. Sales by this means offered a somewhat greater opportunity to discuss the nature of the cover and explain its features. F: THE REGULATORY FRAMEWORK F.1: FICL/FACL 89. F.1.1: BIA/ABI

[62]Until FSMA was introduced in 2000, the selling of PPI was self-regulated by the industry itself. In previous incarnations FICL/FACL were both members of the British Insurance Association (“BIA”). In 1981, the BIA issued a Code of Practice covering all intermediaries other than registered insurance brokers. As a condition of membership, members undertook “to enforce the Code and to use their best endeavours to ensure that all those involvedin selling their policies observe its provisions.” It further provided that it “shall be an overriding obligation of an intermediary at all times to conduct business with the utmost good faith and integrity.” There were specific requirements for an intermediary to:

“A.1.(iii) ensure as far as possible that the policy proposed is suitable to the needs and resources of the prospective policyholder”… B(i) identify the insurance company; (ii) explain all the essential provisions of the cover afforded by the policy, or policies, which he is recommending, so as to ensure as far as possible that the prospective policyholder understands what he is buying; (iii) draw attention to any restrictions and exclusions applying to the policy.”

[63]As members of the BIA, FICL/FACL automatically became members of the ABI into which the BIA was merged in 1985. In January 1989, the ABI issued its own code of practice for intermediaries (the “ABI Code”)in materially identical terms to the prior BIA Code.

[64]In 1996, the ABI issued a Statement of Practice regarding PPI which “strongly encourages and expects its members to observe the terms of this Statement and to ensure that all those involved in selling their policies are aware of, and are asked to abide by, its provisions.” In a section headed “Selling and Marketing Principles”, the Statement provided that:

“Insurerswill comply with the [ABI Code]. The Code provides a framework of general principles within which all general insurance products should be provided… The specific requirements of the Code vary according to the type of insurance and circumstances of the prospective insured person. However, in all cases providers will ensure as far as is reasonable that the policy proposed is suitable to their needs and resources. Providers will give sufficient detail of the essential provisions of the cover afforded by the policy so as to ensure, as far as possible, that the prospective insured person understand what he/she is buying. In particular: • the suitability of a contract will be explained to those who are self-employed, those on contract or part time work, and those with pre-existing medical conditions; • details of the main features of the cover as well as important and relevant restrictions will be made available and highlighted at the time the insurance is taken out with full details being sent afterwards; • all written material will be clear and not misleading; • full details of the cover will be provided as soon as possible after completion of the contract.”

As regards the policy itself, the Statement exhorted insurers to continue to develop clearer policy wording and required policies to define the terms used. Further:

“The policy will make it clear that change in the insured person’s status, for example from permanent salaried employment to self-employment, or fixed-term contract or temporary work could significantly alter entitlement to benefit. The policy will define when the insured person will receive benefit. Any conditions relating to payment of benefits will be clearly displayed in marketing material and policy documentation. Insurers will set out clearly any variations or restrictions in cover for insured persons who are self-employed or on contract or part-time work. These terms will be clearly defined in the policy. Similarly, the effect of pre-existing medical conditions upon the extent of cover will be explained…” • the suitability of a contract will be explained to those who are self-employed, those on contract or part time work, and those with pre-existing medical conditions; • details of the main features of the cover as well as important and relevant restrictions will be made available and highlighted at the time the insurance is taken out with full details being sent afterwards; • all written material will be clear and not misleading; • full details of the cover will be provided as soon as possible after completion of the contract.”

[65]In 1999, the ABI issued Guidance in relation to the Code. This stated as follows:

“The revised ABI Code was introduced in 1989 to improve consumer protection in the selling of general insurance. It is designed to ensure that people are clear about the terms and conditions of the policy and the status of the intermediary with whom they deal. The Code is mandatory for business sold by ABI members in the UK. … What the Code Requires The Code provides a framework of general principles within which companies should sell general insurance. The interpretation of the requirements varies according to the particular type of insurance and the circumstances of the customer. ABI members, rather than intermediaries, are responsible for enforcing the Code. Members undertake to use all reasonable and practical efforts to ensure that they themselves and those selling their products comply… Complaints Handling. ABI recommends strongly that member companies join a recognised adjudication scheme such as the Insurance Ombudsman Bureau (IOB) or the Personal Insurance Arbitration Services (PIAS). Membership of these organisations can achieve the overall consumer protection which is central to the successful operation of the Code. …”

What the Code Requires

[66]At around the same time, the ABI also issued a résumé of the Code specifically for intermediaries. This contained the following:

“Interpreting the Code ABI has prepared guidance on how insurers should interpret the key requirements of the Code regarding “best endeavours”, “explain all the essential provisions” and “draw attention to any restrictions and exclusions”

. The guidance acknowledges that the interpretation of the requirements will vary according to the particular type of insurance and the circumstances of the customer. Examples of how the guidance should be applied to relevant selected classes of insurance are detailed. Dealing with the specific requirements, the following general interpretation should be applied: “Best endeavours” This aspect is relevant to ABI members who are responsible for enforcing the Code rather than intermediaries. The phrase is linked to the member “undertaking to enforce the Code” and use of the phrase “best endeavours” means that every practical effort should be made within reasonable bounds as well as in accordance with the general framework of guidance and individual insurers’ operational methods. The procedural guidance in paragraph 3 provides a more structured framework whereby this obligation on ABI members can be met more definitively. “Explain all the essential provisions” It is necessary for the intermediary (insurer, if dealing direct) to provide an overview of the policy. The detail will vary depending on the particular class of insurance. However, the proposer should have a reasonable understanding of what he is buying, whether this is explained orally or whether he is given a summary and his attention drawn to the main points. In this respect, it is important to recognise the responsibility under the ABI Statement of General Insurance Practice that insurers will work towards clearer policy wordings. The intermediary is not expected to go through all the provisions and exclusions in detail. The important feature is to identify the level of cover being provided (for example, in the case of household contents whether it is indemnity or new for old ), that the type of policy being sold suits the circumstances of the proposer and the level of protection they are seeking as far as possible. It is not good enough simply to offer, for example, an indemnity basis of cover without explaining the limitations and, indeed, that other options are available, unless, of course, the proposer wittingly asks for that type of cover. “Draw attention to any restrictions and exclusions” The same general principles outlined above apply equally here. Certain exclusions, conditions, restrictions etc under a particular policy will be common to all policyholders, for example, a condition about fraud. In those circumstances, it would not be necessary to identify these other than by reference to general exclusions applying to all policyholders of a particular type of insurance, either orally or in policyholder documentation. However, some will be more relevant and, indeed, significant to certain but not other policyholders. An example would be where benefit to self-employed people is either excluded or severely restricted for redundancy cover under a creditor insurance policy. Clearly, self employed people should be made aware of this so they can decide whether the other benefits under the policy and the premium to be paid justifies taking out such a policy. Examples of what may be required depending on the circumstances of the proposer, for some types of insurance are:- … Creditor (payment protection) Length of time payments will be made and policy duration eg monthly policies in connection with mortgage related businesses. Self-employed exclusions, exclusion of pre-existing health conditions, time limits relating to claims payments, age restrictions. … Membership of a recognised adjudicator scheme ABI strongly recommends to its members to join a recognised adjudicator scheme, eg the Insurance Ombudsman Bureau (IOB) or the Personal Insurance Arbitration Service (PIAS), to achieve the overall consumer protection central to the successful operation of the Code.” Dealing with the specific requirements, the following general interpretation should be applied: Creditor (payment protection)

Membership of a recognised adjudicator scheme

[67]From at least 1995, FICL/FACL were members of the IOB. F.1.2: GISC

[68]In 2000, the General Insurance Standards Council (“GISC”) was set up as an independent organisation to regulate the sales, advisory and service standards of members (including insurers and intermediaries). Its main purpose was to make sure that customers were treated fairly.With effect from July 2000, it promulgated its own Code for private customers which stated as follows: “The Private Customer Code This Private Customer Code sets the minimum standards of good practice which all members of GISC must follow when they deal with private customers. It gives you important protection and should help you to understand:  how insurers, intermediaries and agents, and anyone acting for them, must deal with you;  what information you should receive before you commit yourself to buying any insurance; and  how your insurance should be dealt with once it is in place. … 1 Our commitments 1.1 As members of GISC, we promise that we will:  act fairly and reasonably when we deal with you;  make sure that all our general insurance services satisfy the requirements of this Private Customer Code; …  give you enough information and help so you can make an informed decision before you make a final commitment to buy your insurance policy; … Matching your requirements 3.2 We will make sure, as far as possible, that the products and services we offer you will match your requirements.  If it is practical we will identify your needs by getting relevant information from you.  We will offer you products and services to meet your needs, and match any requirements you have. … Information about products and services 3.3 We will explain all the main features of the products and services that we offer, including:  who the insurer is;  all the important details of cover and benefits;  any significant or unusual restrictions or exclusions;  any significant conditions or obligations which you must meet; and  the period of cover. …” … Matching your requirements …

Matching your requirements

[69]FICL/FACL were members of the GISC from April 2001 at which time they also joined the GISC’s own independent adjudication scheme. F.2: GECB

[70]GECB was not an insurer and, hence, not a member of the BIA, ABI or GISC. Instead, it was from about 1992 a member of the Finance and Leasing Association (the “FLA”). The FLA adopted its own Code in about 1993 which was updated from time to time and which set out “standards of good practice for the finance and leasing industry… The Code does not, in all cases, indicate in precise terms particular action which must or must not be taken. It sets out principles which members are required to apply, with good sense and in the spirit of the Code.” Under Section 1.0, members were required to act with integrity and to act responsibly and with care in the day-to-day conduct of their business. They were also required to provide adequate training for their staff, bringing the provisions of the FLA Code to their attention and requiring them to carry out their duties accordingly.

[71]In a section headed “Credit Insurance” the FLA Code provided that:

“4.2 Details of the major items of cover and exclusions under credit protection insurance policies arranged bythe member, including eligibility criteria, shall be set out both clearly and prominently in appropriate literature. Customers shall be given this information, together with the cost of the cover, prior to making a decision to apply for insurance or at the time that an application for insurance is being made.”

An updated 2000 version of the FLA Code contained a similar provision as follows:

“3.2 Before or at the time you take out insurance, we will provide you with sufficient information to explain both clearly and prominently what the product is, the costs, and the key benefits, exclusions and eligibility criteria of your policy.”

An updated 2000 version of the FLA Code contained a similar provision as follows:

[72]While GECB was not a member of any independent adjudication scheme, complaints could nonetheless be made against it in relation to mis-selling under the FLA Code and at common law. F.3: FSMA

[73]The coming into force of FSMA on 1 December 2001 changed the regulatory landscape dramatically. From that date, FICL/FACL became regulated with regard to the underwriting of insurance and subject to the compulsory jurisdiction of the FOS. Importantly, transitional provisions introduced at the same time applied FOS jurisdiction also to historic sales that were previously subject to the jurisdiction of the IOB.

[74]From 14 January 2005, GECB also became regulated under FSMA in relation to insurance mediation activities and was thus subject from that date to compulsory FOS jurisdiction. However, because it had not previously been a member of any of the adjudication schemes listed in the transitional provisions, the FOS did not have jurisdiction over GECB in relation to complaints about mis-selling prior to 14 January 2005. Thus were sown the seeds of the present dispute.

[75]Since 1 December 2001, the regulatory redress system for customers mis-sold PPI has broadly been as follows:(a) The Dispute Resolution section of the FSA/FCA Handbook (“DISP”) contains mandatory comprehensive rules and guidance for complaints handling and redress which apply to all regulated entities from the date on which they became regulated. Since 2010, Appendix 3 of DISP has set out a detailed regime specifically for the handling of all PPI complaints, regardless of when the policies were sold. This applies much more rigorous standards than those which applied previously.

(b) A customer can make a regulatory complaint about PPI mis-selling directly against a regulated financial services company. The company may choose to offer redress in response and must decide whether or not to do so by applying the appropriate complaints handling procedures.

(c) If the complaint is rejected or the customer disputes the offer made, it can refer the complaint to the FOS which was established in 2000 and given statutory powers in relation to FSMA in 2001.

(d) If the FOS has jurisdiction over the complaint, it will assess it in accordance with broad principles of fairness, taking account of regulatory standards at the date of sale and its view of appropriate market practice, and may order the company in question to pay redress. Redress is calculated by reference to the amount necessary to put complainants in the same financial position as if the PPI had not been sold to them. This is generally taken to be the premium paid plus interest.

G: THE EMERGENCE OF PPI MIS-SELLING COMPLAINTS

[76]It was common ground that complaints about the mis-selling of PPI were virtually unheard of until the late 1990s/early 2000s. Where complaints were made, they tended to be about the rejection of claims under the cover itself, not about the way in which it had been sold.

[77]In December 1997, however, the Insurance Ombudsman gave a speech drawing attention to what he perceived as serious deficiencies in the selling of store card PPI and suggesting that this was potentially a significant problem across the industry and that what he saw was only the tip of the iceberg. Thereafter, the issue of mis-selling assumed ever greater prominence and on 30 January 2007, following an investigation commenced in 2005, the FSA issued a Final Notice to GECB requiring it to pay a penalty of £610,000 in respect of a variety of regulatory breaches in relation to PPI. These included the failure to review, amend and then operate its sales procedures so as to ensure that all customers received adequate information about the policy before making a decision to purchase, failure to monitor sales practices effectively and remedy any non-compliance, and failure to amend its training procedures to ensure consistent application of sales processes by sales staff. As noted by the FSA: “The insurance could be sold on behalf of GECB by store staff employed by high street retailers across the UK and therefore the firm operated an unusually large sales distribution model. The failures had the potential to impact upon a very large number of customers. This letter forms an important part of the narrative background although it was agreed that the findings cannot be relied upon as evidence of the truth of what is said.

[78]Three years later, the FSA sent an open letter to industry participants dated 1 August 2010 (the “Open Letter”) listing common failings across the industry as a whole in relation to PPI sales of all types, i.e., not just store card PPI. While the letter in terms only addressed failings since 2005, the Policy Statement which it accompanied expressly referred to extensive evidence of similar failings before 2005. These included:(a) Pressurising the customer into taking PPI;

(b) Automatically including PPI in a loan quotation on the assumption that the customer would want it;

(d) Selling the policy without explicit agreement from the customer;

(e) Failing to disclose fairly and clearly and in good time before the sale was concluded that PPI was optional and that it involved the purchase of an insurance policy;

(f) Failing to take reasonable steps to ensure that the customer only bought a policy under which he or she was eligible to claim;

(g) Failing to disclose clearly and fairly (where it was or ought reasonably to have been so aware) and in good time before the sale was concluded that parts of the cover did not apply to the customer in question;

(h) Failing to disclose significant exclusions and limitations fairly and clearly and in good time before the sale was concluded.

[79]Appendix 3 to DISP dealing specifically with PPI was also introduced at this time and complaints began to increase markedly thereafter. Inevitably, it was only a matter of time before the issue of PPI mis-selling was taken up by claims management companies and this led to a veritable explosion in complaints from about 2012 covering sales going back as far as the 1970s. Such complaints tended to fall into one of the following broad categories:(a) Ineligible customers: where customers were either ineligible for or would face difficulty in claiming benefits under the policy by virtue of their age, employment status, or a pre-existing medical condition;

(b) Unwilling customers: where the policy was sold without informing customers that they were in fact purchasing PPI, e.g., as an automatic add-on to the store card;

(c) Uninformed customers: where customers were not informed of relevant and significant exclusions, costs or limitations of cover;

(d) Unsuitable customers: where the policy did not represent good value to the customer because, for example, they already had sick-pay benefits or other means to cover repayments in the event of accident or sickness, or where the policy would only cover a small fraction of the outstanding balance on the store card.

[80]Given the regulatory position outlined above, Santander (on behalf of GECB) has handled and paid compensation where appropriate in respect of all customer complaints relating to policies sold after 14 January 2005 applying the procedures and approach set out in Appendix 3 of DISP as it is required to do as a regulated entity. It accepts that as between itself and AXA, it is solely responsible for these complaints and does not seek to recover any of the sums or associated costs that it has paid out in respect of them (totalling around £235 million).

[81]In relation to complaints relating to sales pre-2005, AXA as a regulated entity became liable under the FSMA transitional provisions to handle mis-selling complaints in accordance with the provisions of Appendix 3 of DISP irrespective of:(a) the date when the sales took place, i.e., whether pre- or post-2005;

(b) the fact that the policies were in fact sold by GECB on its behalf; and

[82]As such, complaints submitted after the introduction of DISP in 2010 have mandatorily had to be assessed by reference to much more stringent standards than necessarily applied when the policies in question were sold. As a result, AXA has been required to pay out the sums referred to above which it now seeks to recover, in whole or in part, from Santander. It is not in dispute that since neither GECB nor Santander was regulated prior to 14 January 2005, AXA bears sole regulatory responsibility for these complaints. However, AXA contends that as between itself and Santander, Santander is nonetheless liable for any mis-selling in relation to pre-2005 sales on one or more of the grounds asserted in paragraph ‎8 above.

H: THE FACTUAL BACKGROUND LEADING TO THE DISPUTE

[83]This section sets out a very abbreviated summary of the timeline leading up to the crystallisation of the dispute. For the most part, the issues which I have to resolve focus on specific matters or issues of principle. Nonetheless, it is helpful to place the relevant events in some sort of temporal context. As indicated above, however, it has to be borne in mind that the documentary record was by no means complete and that the gaps were not necessarily filled by oral evidence. For example, the earliest available document dates only from 1994 whereas some of the policies covered by this litigation were sold in the 1970s. There were also frequent instances of a point being raised in correspondence without any indication of how it was eventually resolved. And where, for example, there were only a few documents in which a question such as training was discussed, it was impossible to know whether that was because these were the only occasions when it was considered, or simply because they are the only records which survive.

[84]The earliest of the sales within the Complaints Sample took place in December 1982. This appears to have been an opt-out sale where the customer claimed to be unaware that PPI had been added to their account. It was not in dispute that negative option sales were phased out across the entire industry in around 1994 because they were perceived to be unfair and consumer-unfriendly. By then, the ABI and FLA Codes had both been adopted.

[85]At this time there was no written or formal agency agreement between FICL/FACL and GECB, although negotiations for an agreement seem to have commenced towards the end of 1994. An internal FICL/FACL memorandum from March 1996 notes the delay in signing the agreement and refers to this as a matter which “has been ongoing for about 18 months.” None of the witnesses was able to say why it had been considered necessary to conclude a formal agreement, given that the parties were both in the same group and had been operating the substance of an agency relationship for many years without apparent problem. None of them suggested that a written agreement was expected to do anything more than formalise the pre-existing arrangements.

[86]By 1996, a draft agreement had been negotiated with an effective date of 1 July 1996. On 2 September 1996, Mr Jones wrote a file note recording that the agency agreement was in the process of being signed and that ABI compliance was being discussed at the same time. For reasons which no-one could explain, however, the agreement was never signed until it was eventually executed on 1 December 2000 in substantially similar terms to the previous draft, albeit not identical.

[87]The emergence of PPI mis-selling as a serious issue has been described above. Prior to 2012, FICL/FACL had minimal, if any, involvement in mis-selling complaints as these tended to be directed to GECB/Santander (with whom the customer had the credit relationship) who handled them in accordance with the FLA Code to which it was subject and paid redress where considered appropriate. The only complaints that FICL/FACL received and processed directly were those relating to claims from customers under the policy itself, for example where a claim had been declined or the quantum of a claim payment was disputed. However, some mis-selling complaints could reach FICL/FACL where the complaint was declined by GECB and was then referred by the customer to the FOS who in turn referred it to FICL/FACL as the sole regulated entities for pre-2005 policies.

[88]From about 2012, the increasing number of referrals from the FOS concerning alleged mis-selling by GECB began to cause serious concern within FICL/FACL. An internal Santander report into claims handling similarly noted that the level of PPI complaints was continuing to rise in response to increasing media coverage and interest from claims management companies and the amount of its provision for potential liabilities was also commensurately increased. In a letter dated 13 March 2012 addressed to the Head of Partnership Management at SISUK, Mr Rember drew attention to the fact that Santander was declining around 90% of store card PPI complaints in contrast to its decline rate of only 10% for PPI policies overall. He stated that initial discussions had been taking place with Alison Webdale of Santander’s Compliance team about how Santander was applying Appendix 3 of DISP, pointing out that there was now a prima facie assumption that any complaint about PPI mis-selling would be upheld by the distributor and appropriate redress paid. He stated that Genworth had already received about 820 referrals from the FOS which was investigating a further 914 complaints. He invited discussion on a strategy for handling both existing and future complaints, particularly bearing in mind that Genworth did not have any of the information necessary to process complaints and calculate appropriate redress. He put forward two possible solutions, both of which assumed that Santander would be responsible for claims handling and payment of any redress.

F.1: FICL/FACL

[89]This letter led to a debate within Santander as to how complaints were being handled which involved, variously, the teams dealing with Remediation and PPI (Rich Harris and Caroline Swan), Complaints (Mr Crowley), Compliance (Ms Webdale and Mr Hazell), SCL (Mrs Owen and Mr Atkinson), SISUK (Colin Greenhill) and Legal (Rob Worthington). Internal emails identified that:(i) SCL (as the successor to GECB) was differentiating between pre- and post-2005 sales;

(ii) it was declining complaints without FOS rights in relation to the former on the basis that GECB was not regulated in respect of these sales;

(iii) it was instead referring dissatisfied customers to the FLA;

(iv) if the customer nonetheless took the complaint to the FOS, this would lead to a referral to Genworth. Such a differentiation between pre-2005 and post-2005 sales was in marked contrast to the approach adopted by the rest of the Santander group which had previously been regulated through membership of GISC. It is also clear that the different approach taken by GECB was questioned by some and that Mr Crowley and others took the view that it should be abandoned so as to bring GECB in line with the wider Santander group which drew no distinction between the position before and after 2005. Moreover, whatever the strict regulatory position, it was recognised that there might be an expectation on the part of the regulators that Appendix 3 of DISP should be applied to pre-2005 sales with customers being given FOS rights.

[90]By May 2012, Santander’s uphold rate for pre-2005 complaints was around 3% compared with at least 84% and possibly 96% for post-2005 sales. Nonetheless, on 2 May 2012 SCL sought approval from Santander’s ExCo to continue with its existing claims handling approach on the basis that this was justified by GECB’s regulatory position prior to 2005. If approved, SCL proposed that the approach would then be notified to the FSA before Genworth could take the matter further. Calculations showed that an uphold rate of 96% for complaints relating to pre-2005 sales would require an additional provision in the accounts of well in excess of £100 million – a matter which caused Mrs Owen, at least, some alarm.

[91]Approval was clearly given by ExCo since, on 4 May 2012, a meeting took place between Santander and Genworth at which Santander stated that it was not obliged to apply Appendix 3 of DISP to pre-2005 sales and had no liability to fund any redress which Genworth was obliged to pay. It seems that this position was also communicated in a call on the same day to the FSA which did not apparently object. It was subsequently confirmed by emails dated 10 May 2012 from, respectively, Mr Worthington to Genworth and Mr Hazell to the FSA. Santander did nonetheless revise its claims handling policy to ensure compliance with the ABI Code so as to “ensure robustness of approach”.

[92]On 5 July 2012, Mr Rember wrote to Mr Greenhill of SISUK fundamentally disagreeing with Santander’s stated position and indicating that Genworth proposed to start paying redress and charging Santander for both redress and any associated costs. He set out Genworth’s analysis in some detail but in essence his position was that Appendix 3 of DISP required the same approach to be taken regardless of when the sale took place, and that since Santander was solely responsible for the sales of PPI, it alone should bear any liability arising from mis-selling.

[93]By July 2012, Santander had obtained advice from KPMG as to how other market participants were handling pre-2005 complaints. In correspondence, Caroline Walters of KPMG referred to two banks in a similar position which made no distinction between pre-and post-2005sales (albeit at least one involved credit cards rather than store cards). She also drew attention to an expectation on the part of the FSA that a robust investigation should be carried out irrespective of when the sale occurred, and identified at least a risk that the FSA might regard the distinction as infringing the obligation to treat customers fairly (“TCF”). It is clear that her main concern was not so much the non-provision of FOS rights as ensuring that appropriate decisions were taken in dealing with the complaint itself.

[94]On 13 August 2012, a meeting took place between Santander and Genworth at the suggestion of the former to discuss Mr Rember’s letter of 5 July with a view to exploring a commercial solution. Mr Rember confirmed the outcome of the meeting in an email dated 14 August 2012 which was accepted to be an accurate record. It shows that the parties agreed that :(a) Santander would take over the handling of existing complaints and assume responsibility for redress payments and any associated costs incurred by Genworth. These sums would be set off against any profit share otherwise payable to Santander;

(b) In respect of future complaints, Santander was working with KPMG to review its existing claims handling process with the aim of ensuring greater consistency between pre- and post-2005 complaints. Once the review had been completed, the parties would put in place a process for reducing the number of FOS referrals and handling such referrals as were made;

(c) Pending agreement on the new process (which it was recognised might take some time), any new complaints would be passed to Santander who would handle them in the same way as existing complaints.

[95]It was thus agreed at this stage that Santander would assume responsibility for handling claims and paying redress, costs and FOS referral fees on all complaints received prior to the introduction of the envisaged new process and that any sums for which Santander was liable would be offset against profit share.

[96]At some point, it seems that a “Dear CEO” letter was sent by the FSA to Santander which caused it to adopt a more generous claims handling approach. However, not everyone within Santander appears to have been aware of the agreement reached with Genworth in August 2012. An internal email from Mrs Owen dated 9 October 2012 expressed concern that the financial implications of the agreement might not have been fully understood, and that there was a potential exposure of hundreds of millions of pounds should claims management companies get wind of the fact that Genworth was paying redress. On the other hand, a gap analysis carried out by Santander in October 2012 expressed the view that the distinction between pre- and post-2005 sales, while legally justified, did not morally give good service to customers under the “treat customers fairly” principle and recommended removing it.

[97]In November 2012, Mrs Owen carried out some further calculations of the additional provision which Santander would need to make if the uphold rate for pre-2005 sales was increased. In short, for every 1% increase in the uphold rate, a further £1.3 million would be required. Thus, an uphold rate of 92% would require further provision of £118 million. On 8 November 2012, Mr Harris (who was head of the Remediation and PPI Programme) confirmed that he was aware of the potential impact on provisioning and that changes were being made to the claims handling policy. This was clearly a reference to the new process to be put in place following conclusion of the KPMG review.

[98]On 29 November 2012, Ms Webdale drafted an email recording a meeting held with the FSA and the FOS the previous day, apparently in the context of a review by the FSA into Santander’s PPI claims handling procedures. It is not entirely clear whether this email was ever finalised or sent. Be that as it may, it referred to a concern raised by the FOS at the start of the meeting that different review standards were being applied by Santander to pre- and post-2005 complaints. Santander’s response was apparently to assure the FSA and the FOS that the same principles were being applied to both but that 2005 marked the point at which it became easier to locate original documentation or electronic copies “which they were satisfied with”. If this is an accurate description of what was said, it was more than a little economical with the truth to say the least.

[99]On 4 December 2012, Mr Rember chased Mr Hazell for an update on the introduction of the new claims handling process. On 21 December 2012, Ms Webdale circulated the proposed new policy internally within Santander and requested feedback before providing it to Genworth. By early February 2013, however, Mr Rember had still not been provided with the policy and was becoming somewhat frustrated at the delay which was starting to put Genworth in a difficult position with the FSA. By April 2013, the compensation, FOS fees and costs being disbursed by Genworth had exceeded the amount of profit share due to Santander and on 26 April 2013, Mr Rember’s colleague, Mr McDonnell, warned Santander that Genworth would thereafter look to start invoicing it instead for the amounts in question.

[100]In May 2013, Santander finally approved the new claims handling policy although it was still not shared with Genworth as had been promised. Calculations carried out by Mrs Owen’s team estimated that application of the new policy would require additional provision of £18.7 million.

[101]A further development to be noted around this time was the proposed sale by Santander of its cards business to a company called NewDay – referred to in the contemporaneous correspondence as Project Tosca. The sale to NewDay required Genworth to enter into a novation agreement and therefore depended on its co-operation.

[102]On 1 July 2013, Mr Rember wrote to Mr Hazell referring to the August 2012 agreement and recording that Mr Hazell had previously agreed and acknowledged that Santander was ultimately responsible for funding the processing and payment of redress in relation to mis-selling complaints referred to Genworth by the FOS. He noted that while Santander had indicated that the new process would be in place by the end of 2012, there had in fact been no noticeable decrease in the rate of FOS referrals and that Genworth was still receiving more than 100 per week for which it now proposed to start invoicing Santander, as there was no remaining profit share against which to effect an offset.

[103]After discussing the matter internally, Ms Webdale arranged a meeting with Mr Rember which took place on 22 July 2013. Mr Rember’s follow-up email to Ms Webdale records that they reached agreement on the following matters:(a) Santander would send Genworth the new claims handling policy for pre-2005 complaints;

(b) Santander would arrange some dates for training Genworth’s operational staff on the new policy;

(c) A process would be agreed between the parties for the handling of future FOS referrals in relation to complaints declined by Santander under the new policy. Santander would continue to fund these but wanted to have an opportunity to re-review them before Genworth responded to the referral.

[104]Ms Webdale never challenged Mr Rember’s account of the meeting directly. However, she did query internally with Mr Hazell whether it had indeed been agreed as part of the interim arrangement reached in August 2012 that Santander would be responsible for redress and costs in relation to future FOS referrals. Mr Hazell’s response was that he did not recall having discussed the question of FOS referrals after introduction of the new policy as the focus had only been on existing complaints and the need to agree a process to reduce the volume of future referrals. His position was essentially that Genworth should start handling complaints in accordance with the new policy. If it chose to pay redress outside the policy, Santander should not have to contribute. Furthermore, if the FOS overturned a decision reached under the new policy, then liability rested with Genworth alone and it was a purely commercial decision for Santander whether it wished to contribute or not. While he had previously been concerned that Santander’s previous approach to assessing complaints left it exposed to criticism, he felt this was no longer an issue under the new claims handling policy.

[105]At all events, from July 2013, Genworth began to invoice Santander directly for(i) redress

(ii) FOS fees and

(iii) its own administrative costs incurred in handling pre-2005 complaints and, notwithstanding Mr Hazell’s position, Santander paid these invoices without objection or reservation until February 2014. On 9 September 2013, Mr McDonnell followed up Mr Rember’s list of action points from the 22 July 2013 meeting, noting that the process for handling future FOS referrals was under discussion between Genworth and Ms Webdale.

[106]There matters apparently rested until February 2014 when Santander abruptly stopped paying Genworth’s invoices in so far as they related to FOS fees and Genworth’s own administrative costs. Internal correspondence within Santander suggests that this was because Santander believed there should have been no (or only minimal) further FOS referrals after the new process was introduced so that Genworth should not in fact have incurred any FOS fees or administrative costs.

[107]On 6 May 2014, Santander wrote formally to Genworth setting out its position as follows:(a) A process had been agreed with Genworth towards the end of 2013 whereby Santander would handle all pre- and post-2005 claims in accordance with its new policy. If the claim was accepted, Santander would pay the redress. If the claim was rejected, the customer would be given FLA rights;

(b) Should the complaint then be referred to the FOS and thence on to Genworth, Santander would send Genworth the necessary paperwork to enable it to deal with the claim but Genworth would be expected to endorse Santander’s original position unless new evidence or information had come to light in which case Santander would re-review it;

(d) While Santander would continue to deal with all pre- and post-2005 claims and pay redress where necessary, it would no longer pay Genworth’s administrative costs or any FOS fees.

[108]This was the first occasion on which it had ever been suggested that Santander had only been making payments on a goodwill basis and the assertion was immediately challenged by Genworth. On 22 May 2014, a conference call took place between Mr Rember and Mr McDonnell on behalf of Genworth, Mr Dewey of SISUK and Mr Schumann of Santander UK. As recorded by Mr Dewey, Santander had two objectives: first, to consider settlement of Genworth’s outstanding invoices and, second, to understand the parties’ respective positions so that they could work towards progressing Project Tosca. Santander offered to pay any remaining redress included in the outstanding invoices by offset against profit share. Genworth agreed to consider this, although its preference was to deal with the issue in its entirety and not just limited to current invoices. For its part Genworth explained that it was still receiving FOS referrals, albeit the volume was reducing. It referred to the agreement reached in August 2012 for Santander to reimburse redress, FOS fees and administrative costs which it believed still applied and which Santander had not challenged until its letter of 6 May 2014. In answer to Santander’s suggestion that the August 2012 agreement was only ever an interim arrangement pending introduction of the new process, Genworth maintained that the new process had never been intended to derogate from Santander’s continuing liability for pre-2005 complaints in principle but simply to remove Genworth (and, indirectly, Santander) from further exposure to FOS referrals. However, this had not happened.

[109]Further correspondence in which the parties debated their respective positions did not resolve the matter and on 7 August 2014, Ms Day of Santander’s Legal team wrote again expressing disappointment that Genworth was refusing to co-operate in relation to the sale to NewDay. She reiterated that Santander was under no regulatory responsibility in relation to pre-2005 claims and asserted what was said to be its definitive position, namely that it would continue to reimburse Genworth in respect of claims handled prior to 1 May 2013 when the new policy came into effect but not thereafter. It would, however, continue to pay for cases where it was determined in accordance with the new policy that eligibility criteria were not met. Mr Rember replied on 14 August 2014, challenging the underlying basis of Santander’s position but indicating that he was open to discussing a possible solution. In her formal response dated 17 November 2014, Ms Day acknowledged that different people in different departments of Santander had been taking part in discussions with Genworth at different times and that this may have created an unhelpful degree of confusion. She likewise indicated a willingness to meet on a without prejudice basis in order to find a mutually acceptable solution and draw a line under the matter.

[110]It took many months to set up a meeting. Mr Rember was particularly concerned to ensure that any attendees on behalf of Santander would be fully briefed with “knowledge of the full history” and “empowered to discuss possible solutions” with Genworth. He was assured by Mr Conway and Ms Day that the request had been forwarded to “our key stakeholders” and that key decision makers would be present. After repeated chasing by Mr Rember, Ms Day confirmed on 13 April 2015 that Santander wished to resolve the matter “effectively and holistically” without resort to litigation.

[111]Eventually, the long-awaited meeting took place on 4 June 2015. On behalf of Genworth it was led by Mr Rember and also attended by Mr McDonnell and Genworth’s Chief Risk and Compliance Officers. True to their word, Ms Day and Mr Conway had arranged for the meeting to be attended on behalf of Santander by some very senior people. It was led by Mr Pateman (the Executive Director and Head of UK Banking at Santander UK) with Mr Lloyd (Santander’s General Counsel), Mr Hazell (Director of Compliance), Ms Day and Mr Conway (both senior lawyers) and Ms Walters also in attendance. Mr Pateman and Mr Lloyd were both members of Santander UK’s most senior executive management committee, ExCo, at the time. It will be necessary to return to this meeting in the context of AXA’s primary case that a binding settlement was reached on that occasion. For present purposes, it is sufficient to note that the parties agreed that Mr Rember’s note of the meeting circulated on 5 June 2015 correctly recorded that:(a) Mr Pateman acknowledged that Santander as distributor of the policies was liable for any mis-selling and should therefore be funding redress and FOS fees and reimbursing Genworth for its administrative costs;

(b) Mr Pateman indicated that he would prefer Santander to carry out the actual claims handling without Genworth’s involvement (thereby eliminating any administrative costs to Genworth) and Genworth agreed to seek approval from the FOS for such an arrangement, failing which Genworth would appoint Santander as its agent to manage the complaints on its behalf;

(c) It was agreed that monies currently owing to Genworth should be paid by Santander insofar as they could not be set off against any profit share;

(d) Santander asked Genworth to resume discussions with NewDay and acknowledged that it would remain liable for mis-selling regardless of any agreement reached.

[112]Mr Rember’s note then set out a list of “steps to be taken to reflect the agreed outcomes of the meeting:

“1. Santander and Genworth to sign an agreement setting out their agreement on liability and the management of complaints. It was agreed that Genworth would prepare the first draft of this. 2. Santander to obtain FOS agreement to FOS contacting Santander about complaints rather than Genworth. It was agreed that SP would speak to FOS about this at his meeting with FOS later that day. DH would update JR on this. Genworth and Santander would need to agree with FOS the on going process for managing existing complaints and future ones. 3. Santander and Genworth to agree a reconciliation of the amounts owing to each in order to settle any outstanding amount. 4. Genworth to restart its discussions with NewDay with a view to commencing profit share payments to NewDay once the appropriate agreement had been signed. The obligation to pay profit share to New Day would not be retrospective and would only apply to payments after an agreed date. It was agreed that Genworth would not be responsible for any profit share payments to NewDay prior to the agreed date as these would be settled by Santander.”

[113]On 9 June 2015, Ms Day confirmed, after obtaining the “relevant inputs”,that Mr Rember’s understanding of the position as reflected in his note of 5 June 2015 was correct and that “naturally, it makes sense for this to be incorporated into an agreement.” She expressed the view that since Santander would now be assuming responsibility for the actual handling of complaints, it would no longer be reimbursing Genworth’s administrative costs as these would be eliminated. On 14 June 2015, she confirmed that Mr Pateman had discussed the claims handling position with the FOS who had raised no objections to the proposal. As it happened, Mr Pateman had a pre-arranged meeting with the FOS immediately after the meeting on 4 June 2015. Although this was to discuss an unrelated matter, he took the opportunity to raise the proposal.

[114]Genworth duly produced a first draft of the Settlement Agreement on 18 June 2015. The details were then negotiated between the parties until a final version was agreed in late October 2015. A claims handling agreement was also agreed in parallel, with finalisation subject to the signing of the Settlement Agreement. By email dated 20 October 2015, Mr Conway asked Mr Rember to arrange for five copies of the Settlement Agreement to be executed by Genworth and then sent to Santander for signature by the appropriate companies. Genworth did this on 22 October 2015. However, despite increasingly frustrated chasing emails by Mr Rember met with equally frustrated apologies from Ms Day and Mr Conway, the Settlement Agreement was never signed by Santander. It appears that Mr Pateman left Santander UK at the end of November to move to another role within the group and, according to Mr Conway’s email to Mr Rember of 19 November 2015, the directors who were authorised to sign in his place (knowing nothing about the background) had requested that the agreement be referred to ExCo for review and approval. An ExCo meeting then took place on 25 November 2015 and it is one of life’s little ironies that none of the three executives who had attended the 4 June 2015 meeting (Messrs Pateman, Lloyd and Hazell Although Mr Hazell later became a member of ExCo, he was not formally on the committee at the time. ) was present. The minutes record simply that further analysis of the agreement was ongoing due to its volume implications and that it had not yet been signed.

[115]After yet further chasing emails had received no response, Mr Rember was eventually informed by Ms Day on 14 December 2015 that the matter would be presented to ExCo before Christmas for final confirmation that it could be signed and that she had been asked to prepare yet another summary for the committee to consider.

[116]On 7 January 2016, Ms Day wrote formally to Mr Rember. After noting that Santander’s internal governance procedures required ExCo to review the proposed settlement with Genworth following Mr Pateman’s departure, she stated that new information had come to light from the FOS which suggested that some of FICL/FACL’s PPI products were flawed, in particular in relation to so-called Budget Accounts. According to Ms Day, the FOS had suggested that some of these products should never have been offered for sale at all. If so, Santander believed the question of mis-selling was irrelevant and for this reason did not intend to proceed with the Settlement Agreement. Ms Day appears to have been referring to the subject matter of a meeting between Genworth and the FOS some nine months earlier in April 2015, when the latter expressed concern that GECB’s Budget Account might not have represented good value for money as the premium was calculated as a percentage of the outstanding balance and customers buying this product typically only had very low balances.

[117]Unsurprisingly, Santander’s volte face was greeted by Genworth with dismay and, it is fair to say, outrage, not least because of GECB’s own close involvement in the design of the Budget Accounts and other PPI products offered for sale. Meanwhile, the question of how claims should be handled going forward remained unresolved. On 3 July 2017, Santander notified Genworth that it would no longer pay anything in respect of pre-2005 policies and on 31 July it stopped handling complaints in relation to such policies altogether. This effectively left Genworth over a barrel. Santander had handled all claims hitherto and Genworth did not have any of the information necessary to allow it to assume that role itself. Any interruption or delay in handling complaints would have adverse regulatory consequences and Genworth could not realistically secure claims handling services from anyone other than Santander.

[118]After exchanges of correspondence between their respective solicitors, the parties eventually concluded a stand-alone Claims Handling Agreement on 7 December 2017 whereby Santander agreed to continue to provide claims handling services in accordance with Appendix 3 of DISP in return for a service fee on the express condition that such fee would be irrecoverable by Genworth and that all redress and associated costs would be borne by Genworth/FICL/FACL. On the same date, the parties entered into a standstill agreement suspending the running of time in relation to the claims now made by AXA. On 24 December 2018, the CHA was amended and restated in respects which are irrelevant for present purposes.

[119]To complete the picture, it is necessary also to be aware that Santander’s refusal to sign the Settlement Agreement had an impact on the acquisition of FICL/FACL by AXA. The relevant Sale and Purchase Agreement was concluded on 1 December 2015 when the full scope of FICL/FACL’s potential liability for PPI mis-selling was still unknown. At that stage it was anticipated that the Settlement Agreement would be signed and that Santander would ultimately take responsibility for those liabilities. The SPA accordingly contained an express provision that 90% of such liability would be allocated to Genworth as seller but that this liability would cease on signature of the Settlement Agreement. Notwithstanding the non-signature of the agreement, Genworth refused to pay and this led to proceedings by AXA against Genworth (“the Genworth proceedings”). Following a judgment by Mr Justice Bryan in favour of AXA on liability and quantum, AXA and Genworth entered into a confidential settlement agreement in July 2020. In practical terms, therefore, it is Genworth rather than AXA which stands to benefit from success in this action.

[120]On 29 December 2020, a pre-action letter was sent on behalf of AXA to Santander and the present proceedings were commenced on 29 January 2021.

[121]Before proceeding further, it is convenient to dispose of two points.

[122]First, AXA repeatedly emphasised the gross injustice which it said would result if Santander were permitted to avoid liability to make any contribution to AXA/Genworth’s losses in this case. Mr Andrew Green KC submitted on its behalf with considerable force that, of the total redress which Genworth had paid out:(a) Some 25% (£113 million) related to a return of premium – of which GECB had enjoyed 95% of the benefit;

(b) Some 30% (£140 million) related to interest charged by GECB on those outstanding premiums, of which it had received 100% of the benefit;

(c) Some 45% (£207 million) related to statutory interest on (a) and (b) – again representing a benefit accruing almost exclusively to GECB.

[123]It is an inescapable fact that AXA/Genworth is in the unfortunate position of having been statutorily required to refund monies which it never received and to pay interest on those sums representing a benefit which it never enjoyed, while Santander, who did reap those benefits, claims that it is not obliged to contribute a penny. The outrage felt by AXA/Genworth is compounded by the fact that the Settlement Agreement under which Santander voluntarily undertook to assume the burden of making redress was simply rejected by it at the last minute when it had not only been finally agreed but had also been signed by Genworth and delivered to Santander for execution at the latter’s express request.

[124]Were it only a question of fairness and equity, AXA would be on strong ground indeed. However, it is truism that hard cases make bad law and a general feeling that AXA has been ill-used is no substitute for proper legal analysis. The court’s task is not simply to do justice; it is to do justice according to the law, and any sympathy that I might otherwise have felt for AXA’s plight must be firmly put to one side when undertaking that task.

[125]A similar point can also be made in relation to Santander’s converse complaint. This is that the DISP complaints regime pursuant to which AXA/Genworth have been required to pay the sums claimed in these proceedings (and which Santander itself has been required to apply in respect of post-2005 sales) involves the application of much more rigorous standards than those regarded as acceptable at the time the sales were made. It is also heavily weighted in favour of the consumer through evidential presumptions requiring assertions made in the complaints form to be accepted in the absence of evidence to the contrary, and by requiring complaints in certain instances (for example, negative option sales or sales to the self-employed), to be mandatorily “auto-upheld” without further investigation into the facts (such as whether the customer might in fact have known about and wanted the cover). It was submitted by Mr Zellick that if the current claim succeeds, the effect would be to hold Santander liable for sales which were properly made in accordance with the universally applied standards of the day and without proper scrutiny of the factual basis of the underlying complaint. He argued that this would be wholly unjust and unjustified.

[126]Again, however, appeals to jury points of this nature cannot avail if AXA succeeds in satisfying the court that it has a cause of action against Santander which is both factually and legally well-founded. I: “RESPONSIBILITY” FOR MIS-SELLING AS BETWEEN FICL/FACL AND GECB

[127]A recurrent theme of Santander’s submissions was that responsibility for any mis-selling was to be laid – if not entirely, then at least substantially – at FICL/FACL’s door on the basis that they had an overriding responsibility for the way in which PPI was sold. This argument was packaged in a variety of beguiling ways depending on the context in which it was deployed. Thus, in the context of AXA’s indemnity claim, it was presented as a failure by FICL/FACL to comply with a condition precedent requiring it to have complied with their duties under the Agency Agreement or as a question of causation/authorisation. In the context of the claim in negligence, it was said to be relevant to breach and causation.

[128]I put it to Mr Zellick during the course of argument that the question of responsibility might be answered differently depending on the particular relationship in which it arose. For example, the undisputed fact that FICL/FACL bore primary responsibility for the sale of PPI vis-à-vis the regulator, did not necessarily mean that they bore primary responsibility as between themselves and GECB or the end customer. Mr Zellick, however, was unperturbed by any such potential distinctions. In his submission, FICL/FACL bore primary responsibility vis-à-vis the regulator, GECB and the end customers. This is a submission which will have to be scrutinised in due course in relation to each head of claim. Nonetheless, in whatever guise the argument was presented, it raised the following issues with which it is convenient to deal with compendiously at the outset:(a) The standards to be applied to the sale of PPI policies at the relevant time;

(b) Whether, on the evidence, PPI was mis-sold by GECB in breach of those requirements;

(c) If so, whether there were any steps which FICL/FACL could reasonably have taken which would have avoided such breaches.

[129]Having done so, it will then be possible to draw the consequences of my findings in the context of each cause of action asserted by AXA. I.1: The applicable standards

[130]There was no real dispute that the appropriate standards to be applied to the sale of PPI were those set out in the applicable regulatory Codes. There was also some debate about “industry standards” more generally. However, when assessing the conduct of the parties I do not consider that they can be regarded as having acted reasonably by following lower standards than those required by the industry regulators: see Lloyd’s Bank Ltd v Savory & Co.,

[1933] AC 201

; Bolam v Friern Hospital Management Committee,

[1957] 1 WLR 582

. If the practices of the industry as a whole were uniformly poor and below those required under the ABI Code, then so much the worse for the industry.

[131]The extracts from the various codes set out above show that they did not differ over time in their essentials, at least in the period up to the introduction of FSMA. The experts primarily concentrated on the requirements of the ABI Code and I accept that this is the appropriate benchmark against which the conduct of the parties should be measured.

[132]On the basis of the agreed expert evidence, I am satisfied that someone selling PPI was obliged under the ABI Code to do the following:(a) Refer to the product and either inform the customer explicitly of the option to decline cover (opt-out sales) or obtain their express consent to purchase it (opt-in sales);

(b) Take steps to ascertain the customer’s eligibility for it;

(d) Identify the insurer;

(e) Outline the key benefits, including the nature of the cover, and its cost;

(f) Refer to the existence of exclusions and restrictions in relation to matters such as age, employment status and pre-existing medical conditions, and the fact that these were set out in the Summary of Cover (the “Summary”);

(g) Give the customer an opportunity to read the Summary before entering into a contract so that they had a reasonable understanding of what they were buying before they signed.

[133]On the basis of the expert evidence, I accept Mr Zellick’s submission that prior to 2005, the concept of “suitability” as referred to in the ABI Code was regarded as synonymous with “eligibility” (i.e., age limits and employment status) and did not include the consideration of demands and needs which would be required now. I therefore accept that it was not unreasonable at the time to take the view that if a customer was eligible to take out a policy, they would derive at least some benefit from it. I also accept that sales staff were not expected to reinvent the wheel for PPI if the relevant information as to age and employment status had already been captured as part of the credit application unless they had reason to believe (whether from the disclosed information or otherwise) that a customer was, for example, approaching the upper age limit or was a student or self-employed. In that case I accept Mr Pendleton’s evidence there was an obligation under the ABI Code to discuss and explain the eligibility requirements in more depth.

[134]As regards exclusions and restrictions, the experts agreed that – other things being equal – the salesperson was not obliged to explain individual exclusions or to discuss potentially sensitive medical conditions or financial information such as whether the customer had other resources to cover their repayments. Again, however, I consider that if the salesperson had reason to believe (whether from the disclosed information or otherwise) that a customer might, for example, have a medical condition which would affect the cover, they were obliged by the ABI Code to explain the relevant restrictions more fully;

[135]It was Mr Blundell’s view that the obligations in (e) and (f) were sufficiently met by a simple reference to the Summary, whether or not the customer took time to read it. I have serious doubts whether a bare reference without more would have been adequate, bearing in mind the emphasis placed by the ABI Code on ensuring that the essential provisions of the policy were explained in sufficient detail for the customer to understand what they were buying and the Resume published in 1999 which referred expressly to giving the customer a Summary “and [drawing attention] to the main points.” Be that as it may, both experts agreed that there would not be compliance with the ABI Code if the salesperson said nothing about the Summary but just relied on the customer reading it as part of the overall documentation given to them after signature. As Mr Blundell explained in cross-examination:

“Our view as the insurer was the summary of cover was such an important part of the sales process in terms of the information that the customer needed, so we wanted the salesperson − the salesperson would be expected to mention the product, limitations applied – Q. What do you mean by mention the product, let's just break this down? A. It covers you [for] life, [the] life so that if you die it pays off the balance. It covers accident, sickness and unemployment. So it pays a monthly instalment if you are out of work for those reasons. There are limitations both in terms of time and money. There are policy exclusions which apply. All of those are detailed in the summary of cover, please read the summary of cover, and the summary of cover to the insurer was such an important part of the process because it covers everything and without in any way wishing to undermine the sales process, we wanted the salesperson, whether it was in a branch or a retail store , to provide context within which that summary of cover would more likely be read, i.e. mentioning it covers these things, but all the detail was in the summary of cover. To merely hand the summary of cover to the customer and not say anything would not have been acceptable because you are not providing a context in which that summary of cover might be read.”

Q. What do you mean by mention the product, let's just break this down?

[136]Thus, it was common ground between the experts that if the salesperson did not mention(i) the product and the areas of cover;

(ii) the existence of exclusions/restrictions; and

(iii) the Summary prior to concluding a sale, there would be a breach of the ABI Code. They further agreed that such a breach could not be remedied by the fact that the customer was given or sent a Summary of Cover after they had signed up for the policy even within the cooling-off period. I concur. It cannot have been compliant to rely on the customer to appreciate that they had been sold a policy they may not have wanted or needed and then to cancel it.

[137]There is an obvious danger in this case of applying liberal doses of hindsight and of looking at events which took place many decades ago through a contemporary regulatory prism. However, that danger is minimised to a large degree in this case because the experts agreed that the obligations referred to above applied throughout the entirety of the period with which we are concerned. Furthermore, I find that in these respects, there was no sea-change in the standards required after the advent of FSMA. This much is made clear by the FSA Policy Statement issued in August 2010 which, in response to specific criticism that it was applying higher standards retrospectively, stated expressly that:

“Prior to January 2005, the General Insurance Standards Council (GISC) Code and, prior to the GISC Code, the Association of British Insurers (ABI) Code, constituted the industry codes of good practice in relation to the sale of insurance. In addition, the provisions in common law such as the general duty of utmost good faith in insurance contracts and misrepresentation would also have been relevant at that time. GISC members signed up to a series of commitments set out in the Code, which in our view had a similar effect to the obligations set out in the current regulatory system. For example, paragraph 1.1 of the Code set out that GISC members promise to ‘act fairly and reasonably’ when dealing with customers. This is similar to FSA Principle 6 (treating customers fairly). Applying pressure to a customer to take PPI (failing 1) is undoubtedly a failure to act fairly and reasonably and as such is contrary to paragraph 1.1 of the Code as well as a breach of FSA Principle 6. Another example is paragraph 1.3 of the Code which set out the GISC members’ promise to ‘give customers enough information and help so customers could make an informed decision before they made a final commitment to buy an insurance policy.’ This is similar in effect to FSA Principle 7, and also ICOB 2.2.3R, ICOBS 2.2.2R (which are the clear, fair and not misleading rules) and ICOBS 6.1.5R (ensuring customers can make an informed decision). Providing misleading or inaccurate information about the policy to the customer most likely would be in breach of GISC paragraph 1.3 in the same way that it would be a breach under the relevant provisions after January 2005. In addition, this could also amount to a misrepresentation under the general law, for example. There will, however, be more specific failings that will not have a corresponding commitment in the GISC Code, for example, the failure to disclose ‘price information calculated in a way to enable the customer to relate it to a regular budget’. The GISC Code has commitments regarding provision of information on costs. However, they do not descend to this level of detail. The same applies for the general law, which may not necessarily require information on price in the specific manner described in this failing.”

[138]It is therefore clear that the requirements of the ABI Code which are in issue in these proceedings did not materially change post-FSMA. The following points made by Santander are accordingly something of a red herring:(a) Mr Blundell gave evidence that the concept of mis-selling did not exist in the 1980s and 1990s. As to this, it may have been Mr Blundell’s personal experience that non-compliant sales practices were not expressly or routinely referred to as “mis-selling”, although I note that both FICL/FACL and GECB were referring explicitly to “mis-selling” in early 1998. But how they were labelled is irrelevant. Mr Blundell accepted that policies were sold to ineligible people from time to time and that is mis-selling, however it may have been described at the time. In any event, Mr Pendleton’s evidence was that what would now be regarded as “mis-selling” tended to be handled by the credit lender as a refund and processed in bulk, rather than identified specifically as a complaint. This may well have served to mask the problem from an insurer’s perspective. At all events, it is clear that complaints about “mis-selling” were beginning to emerge in 1997 in sufficient numbers to be a cause for concern as evidenced by (i) the Insurance Ombudsman’s “tip of the iceberg” speech in December 1997; (ii) Mrs Greenaway’s evidence; (iii) GECB’s own internal compliance review in November 1997 (as to which see paragraphs‎148-‎149 below); and (iv) the FSA’s Policy Statement of August 2010.

(b) Mr Blundell’s view was that there was no widespread abuse in the market before 2005. He said that he had never been aware of a problem and that as far as he was concerned, PPI was always sold in accordance with “our” interpretation of the ABI Code, by which I understood him to mean the three insurance companies for whom he worked over the period. He also said that his three employers had only experienced a very low level of complaints. He was inclined to dismiss the Insurance Ombudsman’s speech as unnecessary scaremongering and suggested that any perceived increase in claims was due to a rapid expansion of the market. However, I cannot accept that this was the sole explanation. Mr Blundell accepted that he had never himself been involved in or observed an actual sale of PPI and that all his information was anecdotal and based on what he had been told by colleagues. He further accepted that the complaints which found their way to insurers such as FICL/FACL were predominantly about the handling of claims under the policy or claims decisions rather than mis-selling. Thus, the low level of complaints to his employers could well have been explained, at least in part, by the fact that most mis-selling complaints were processed by the credit lenders as refunds and not passed back to the insurers. I therefore do not place any significant weight on this evidence. I.2: Was PPI sold by GECB in breach of the ABI Code? I.2.1: The available evidence

I.2: Was PPI sold by GECB in breach of the ABI Code?

[139]The earliest relevant document is an internal FICL/FACL memorandum dated March 1996 which records that GECB was refusing, notwithstanding FICL/FACL’s advice, to incorporate the EC Third Directive into its documentation. Moreover, it was not providing FICL/FACL on a regular basis with updated marketing material for legal approval but was simply reprinting the material on its own initiative. The documents do not show how this concern was addressed but in August that year, Mr Jones contacted Vanessa Carter (then the Marketing Manager at GECF) in relation a new client launch of which he had been informed specifically in order to discuss how the EC Directives would be communicated at or before the point of sale in PPI telephone sales.

[140]In late 1996/early 1997, FICL/FACL undertook a wholesale review of all GECB’s product literature following the ABI Statement of Practice, which Mr Jones was anxious to address. At around the same time, GECB’s status for regulatory purposes changed from that of independent broker to tied agent and Mr Jones noted in an internal email that FICL/FACL were discussing compliance in tandem with finalising the Agency Agreement. Changes to all sales and marketing documentation were also required as a result of GECB’s name change in January 1997 (referred to contemporaneously as “Project Hook”).

[141]It appears that regular quarterly meetings were held between FICL/FACL and GECB, as well as less formal meetings between Mr Jones and Ms Carter and others from time to time. At a meeting on 4 October 1996, Mr Jones chased for penetration analysis data which he had previously requested. His earlier request seems to have fallen on deaf ears, since Ms Carter had to ask again about the nature and extent of information required. Mr Jones also suggested that GECB might like to use FICL/FACL’s “wide-ranging and dynamic Training Department services”. This was left for Ms Carter to consider and advise. She never reverted.

[142]On 3 December 1996, Mr Jones wrote to Ms Carter asking her to supply all of GECB’s marketing materials in order for him to ensure that GECB was complying with the ABI Code. On a balance of probabilities, I find that this was not prompted by any request from GECB but was part of Mr Jones’s follow-up to the ABI Statement of Practice and the change in GECB’s agency status. Only some documentation was provided, however, and Mr Jones accordingly had to chase again on 4 February 1997. In his letter he expressly reminded GECB that it was a requirement of its agency agreement and its status as a FICL/FACL agent that all marketing material which mentioned insurance was submitted to FICL/FACL for approval. He also emphasised that, irrespective of Project Hook, “it is necessary to see drafts of marketing material before launch of a new scheme or when an existing scheme is changed.”

[143]On 27 February 1997, Mr Jones met Lisa Hardstaff of GECB to discuss, amongst other things, the relationship between GECB and FICL/FACL and the status of GECB as a FICL/FACL agent and its implications. Mr Jones’s handwritten annotation to the agenda records that no minutes were taken as this was “very much a training-type session”. However, the documents attached to the agenda contain notes in his handwriting which refer to the meeting and show that some discussion of training took place. They also contain two template checklists, one for creditor schemes and one for non-creditor schemes. Both checklists contained boxes for confirmation that documentation had been checked and agreed, but while the non-creditor checklist contained entries relating to client training, the former (which would have applied to GECB) did not. Mr Jones was asked about the discrepancy in cross-examination but ultimately could not recall anything to assist. In the absence of any further evidence as to what information these documents were intended to capture and the extent to which they were used by FICL/FACL – either for GECB business or business introduced by other parties – I am unable to derive much assistance from them.

[144]Mr Zellick sought to make something of the fact that a Product Information Sheet for a Frasercard scheme recorded that the premium payable to FICL/FACL did not include any training days. But that on its own does not mean that training was not provided, or offered or at least available on request.

[145]On 4 March 1997, Mr Jones raised an instance of mis-selling by GECB’s telesales unit which had been drawn to his attention by a colleague. The documents do not reveal GECB’s response, but Mr Jones’ recollection in oral evidence was that it undertook to retrain its telesales unit as a result.

[146]On 13 May 1997, Mr Jones held one of his meetings with Ms Carter. FICL/FACL’s internal minutes record that as part of a discussion about the Agency Agreement, Ms Carter asked FICL/FACL to specify the services covered by their 5.6% premium retention in order to “expand the definitions of services offered under the agency agreement with particular regard to training, research and marketing services.” Ms Carter told Mr Jones that she wished to make greater use of FICL/FACL’s training facilities and needed to know how much GECB could get for free and what it would have to pay for.

[147]In September 1997, FICL/FACL actively promoted to GECB an ABI initiative to provide laminated cards to retail staff at the point of sale which contained a guide to policy conditions and exclusions. Mr Jones’s letter of 29 September 1997 to Ms Carter stated that:

“This initiative is in response to concerns which have been raised, most recently by the Insurance Ombudsman, over sales people's over-reliance on using written, rather than oral, communication of PPI's features and benefits and, more particularly, conditions & exclusions, at the point of sale. The idea behind the laminated card is that it should be available to, & used with, the prospective customer to raise their awareness of the duties of the seller under the ABI Code of Practice to ensure that the policy being offered suits as far as possible their particular circumstances.”

He concluded by recognising that use of the laminated card was not compulsory but requested GECB to consider its use on the basis that there would be some retail sales which were outside its control.

[148]On 21 November 1997, GECB held a “workout” session as part of an internal compliance review (the “Compliance Workout”). The final record of the workout was based on a draft which contains further details of the discussion and the concerns raised by particular individuals. I have taken account primarily of the final version only. It is revealing.

[149]Numerous compliance concerns were identified across the board. “Unless otherwise specified, it should be assumed that the concern is generic to all retailers.” Of particular relevance to the present dispute were the following (which relate to the sale of the credit agreement as much as to the sale of PPI):(a) Retail staff training was self-based, end-user training with the onus on the staff to train themselves.

(b) GECB did not train all retailers and it was difficult to know how well staff were trained. The two major factors were (i) the scale of some retail operations and (ii) the amount of training that retailers would permit.

(d) Retailers did not monitor the identity of staff who had been trained and there was a particular concern about part-time and Saturday staff. Although application forms identified the salesperson, no subsequent check was made to see whether they had been trained.

(e) Generally there was only one training manual per store, which was not at the point of sale. Where sales were rung through to GECB, heavy reliance was placed on the GECB telephone operator to ensure that the agreement was signed and the ID document number recorded. For accounts opened via an online system, even this safety check was not present.

(f) While each retail chain should have had a procedure manual with an up-to-date copy in each retail outlet targeted at the staff who needed to refer to it: (i) Some chains did not and had never had a procedures manual; (ii) Procedural refinements would generally only be introduced by GECB for new retailers rather than being imposed on existing retailers; (iii) It was unclear whether the procedures manuals for older retailers had been updated. GECB had produced updates for Debenhams but did not know whether these had been incorporated into the individual manuals held in the stores; (iv) Few retailer contracts contained provision for effective charge-back in the event of failure to comply with agreed procedures; (v) It was possible that GECB was over-incentivising point of sale staff to open accounts and sell insurance;

(g) No attempt was made to make the customer read the application form although a customer would not be denied the opportunity to do so if requested. However, there had been a change of emphasis in the law which increased the obligation on GECB as a business to take an active approach to ensuring that the customer understood the agreement they were signing.

(h) Specifically in relation to the sale of PPI:

(i) There was limited training of sales staff and limited understanding on their part of the insurance product; (ii) Staff might be tempted by the incentives to operate inertia selling; (iii) Sales staff were giving the impression that a successful credit application was more likely if insurance was also taken; (iv) One month’s “free” PPI was being offered without it being made clear that the customer would automatically be charged thereafter; (v) Insurance was the subject of a separate report; Which unfortunately was not in the documentation before the court. (i) Schedule 2 to the review, which referred specifically to the account opening process for Debenhams, also recorded that: (i) The procedures manual was being revised but it was unclear whether the same version was being used in all the stores. It was thought that many managers did not update the individual manuals when updates were issued. Only one audit had been carried out on Debenhams procedures some four years previously; (ii) A significant number of complaints were being made by customers who claimed they had been charged for PPI insurance they had not requested; (iii) While GECB did not pay incentives to sales staff to open accounts or sell PPI, there was considerable management pressure from Debenhams to open accounts (this being a KPI for branch managers) and the store paid limited incentives to point of sale staff. By contrast, on the Burtons account GECB paid retail staff 50p for an uninsured account and £1.00 for an insured account with a further incentive to branch managers equivalent to 25% of branch earnings on these incentives.

[150]There is no evidence that the full scale of these problems was made known to FICL/FACL. Instead, Ms Carter contacted Mr Jones on 23 December 1997 to say that GECB had received complaints from the Trading Standards Office about inertia selling (presumably a reference to the concern identified in paragraph ‎149(i)(ii) above). She said that GECB planned to introduce a new selling message but had a major logistical problem in retraining at least 40,000 retail staff, which it wished to do by the end of January. She asked for input on the training side and stated that FICL/FACL should contribute to the cost as GECB was their tied agent.

[151]To say that this was a sanitised and greatly abbreviated version of the problems that had been uncovered by the November workout is an understatement.

[152]At all events, Mr Jones agreed to look into the training aspect and arrangements were made for him to visit GECB’s telemarketing unit in the New Year to see how it operated. A follow-up meeting then took place on 23 January 1998 which was also attended by Keith Hall of FICL/FACL and GECB’s Assistant Marketing Manager, Guy Mallon. As recorded by Mr Jones in an internal report, he was told by Ms Carter that:(a) GECB had become increasingly conscious that complaints were being escalated to a very high level with the retailers;

(b) GECB had traditionally provided one month’s free cover at the start of the policy but had always felt that the customer may have felt pressurised into purchasing PPI. The length of queues at the till (especially during sales periods) meant that sales staff were under pressure to complete the store card operating procedures as quickly as possible. It was estimated that they had 20-30 seconds to make the PPI sale.

(c) This had resulted in a sales pitch which involved minimal selling of policy benefits, with particular focus on Purchase Cover and Price Protection (where applicable), minimal discussion of cost, limited to “it’s cheap”, and reliance on the provision of one month’s free cover on the basis that the customer could always cancel if they did not want it.

(d) GECB’s proposed solution was to retrain all the sales staff so that they instead “suggested” that customers should protect their purchases and payments.

(e) Store staff training was supported by a 3-4 person unit providing technical finance and insurance support to retail branches. GECB also offered support to retailers via some 90-100 Regional Sales Managers.

(f) GECB had created a training pack containing: (i) Trainer notes; (ii) A staff training checklist; (iii) Posters and a pocket guide; (iv) Plan details outlining key product benefits.

[153]The minutes of the meeting indicate that the attendees went through the training pack and agreed that the overall concept addressed the problem of inertia selling which had been highlighted at the meeting. However, FICL/FACL indicated that the wording would also need to be specifically checked for compliance with the ABI Code and also by their legal department in relation to the product itself. GECB did not request any assistance or training from FICL/FACL beyond approval of the training pack. During the course of the hearing, Mr Zellick made much of a discussion which took place about the costs of compliance and FICL/FACL’s expressed resistance to being asked by GECB for a contribution. However, earlier and subsequent communications make clear that this related to the costs of printing and postage and had nothing to do with the provision of training as was suggested to some of the witnesses.

[154]Meanwhile, on 9 January 1998, the FLA had produced an agreed form of welcome letter to accompany every certificate sent to an insured customer following the conclusion of a sale. As stated by the FLA in its covering letter:

“Despite the requirements of the ABI General Business Code of Practice, and the more recently introduced ABI Statement of Best Practice for Payment Protection Insurance, customers in many instances are not given the marketing material, and if they are they rarely read it. This is particularly prevalent in the retail market where finance companies rely heavily on third party introduced business. … The purpose of the letter would be to thank the customer for taking out the cover in a customer-friendly way, and would then go on to reiterate the eligibility criteria and any particular pertinent aspects of the cover such as self-employed exclusions which should be brought to the customer’s attention… This is a belts and braces procedure which in no way detracts from the obligation of point-of-sale personnel to explain the essential terms and conditions of the policy, and in particular the eligibility criteria…”

[155]Ms Carter forwarded this letter to Mr Jones, indicating that GECB intended to comply with the initiative and asking for his approval.

[156]Mr Jones appears to have passed the matter to a colleague, Ms Jane Dunlop, who raised a number of queries, to which Mr Jones responded on 4 February 1998 as follows:(a) What sales aids do GECB have and have FICL/FACL signed them off? GECB used to provide store staff with the laminated card summarising the policy benefits, eligibility criteria and exclusions but these had been discontinued as they were frequently lost and it was difficult to maintain adequate and up-to-date supplies.

(b) What training materials do GECB have and have FICL/FACL signed them off? Training was nearly always provided on a “train the trainer” basis and the main training material comprised a pack of plastic cards to be retained at the point of sale together with trainer notes and a video. Mr Jones’s view was that the technical content of the material was extremely light: the selling message was brief and general and staff were asked to refer to the reverse of the credit agreement (presumably the Summary). Separate training packs were compiled for each retailer and it would be a substantial task for FICL/FACL to sign off on each of these. However, a full explanation of the policy benefits, eligibility and exclusions was included in the Summary on the back of the credit agreement and following Project Hook, FICL/FACL had seen and approved all point of sale marketing material.

(c) How many scripts are there and have FICL/FACL ever seen them? Initially a small number of scripts were used which were rigidly adhered to but this was found to impact sales adversely. There was now a huge number of scripts in use and it was not possible for them all to be checked by either FICL/FACL or GECB. The scripts were supplemented by a coaching tool which FICL/FACL would ideally have liked to see. However, the experience of both Mr Jones and FICL/FACL’s Senior Training Manager, Maureen Lloyd, suggested there was little chance of them being allowed to do so. He noted that GECB was extremely reluctant to allow access to the telesales unit which it regarded as its “jewel in the crown”.

[157]Mr Jones explained that in these circumstances it had therefore been agreed as an interim measure that GECB would refer the customer to the Summary in the credit agreement for details of the cover, eligibility and exclusions. He recorded that GECB had taken on additional resource and changed the allocation of duties within the team in order to allow a full compliance review (presumably the Compliance Workout described in paragraphs ‎148-‎149 above) which had resulted in the approach to FICL/FACL by Ms Carter on 23 December 1997. A fact-finding visit by Mr Jones and Maureen Lloyd to the telesales unit was now planned for 11 February 1998.

[158]On 6 February 1998 there was another meeting between Mr Jones, Ms Carter and Mr Mallon at which the Insurance Ombudsman’s speech in December 1997 was discussed. FICL/FACL advised GECB that GECB had a responsibility to enforce the ABI Code and that FICL/FACL had a responsibility to monitor GECB. Ms Carter requested written confirmation from FICL/FACL that GECB was acting correctly at the point of sale as well as approval of its training and selling messages and use of the training pack as a substitute for the laminated card.

[159]The visit by Mr Jones and Ms Lloyd to the telesales unit took place as planned on 11 February 1998. In his report of the visit, Mr Jones concluded that the style of the unit was now more professional and controlled but that it was still in a developmental phase. He felt that there was support for compliance but that Ms Lloyd’s observations (recorded separately) gave rise for concern. Unfortunately, these observations were not before the court.

[160]On 8 May 1998, a meeting took place between FICL/FACL and GECB specifically to discuss the introduction of GECB’s Account Cover II product and whether this was a new product or a product upgrade. During the course of this meeting, FICL/FACL advised and directed GECB specifically as to what would be required in order to ensure that the selling of Account Cover II was compliant and would not constitute negative option selling by the back door.

[161]In July 1998, it appears that GECB distributed a generic compliance pack to all its retailers in order to demonstrate to the regulatory authorities that GECB and its partner stores were taking the matter of compliance seriously. GECB asserted that it had always used the ABI Code as the benchmark for compliant selling but that, following a review, it believed that it was necessary to distribute additional reference information at store level. GECB indicated that it would support ongoing training through its sales network and task force teams. There is no evidence that FICL/FACL was ever asked by GECB to participate in or contribute to this training support.

[162]The material before the court contains version 1 (October 1998) of an Operations Manual prepared by GECB for Debenhams. In the section addressing the sale of PPI, there is a description of the product and its benefits, eligibility and exclusions. However, the manual is mainly notable in that the step-by-step guide for retail staff completing an application does not include any requirement to refer to the Summary of cover before obtaining the customer’s signature notwithstanding that GECB had apparently agreed to do this (see paragraph ‎157 above).

[163]The question of ABI compliance was again discussed between FICL/FACL and GECB at a meeting on 26 November 1998. FICL/FACL emphasised that they had a duty to ensure that GECB complied with the ABI Code. GECB’s legal department was accordingly required to respond to FICL/FACL’s questionnaire regarding ABI compliance by February 1999.

[164]This was followed by a further meeting on 17 December 1998 at which FICL/FACL spelled out to GECB that it was FICL/FACL’s agent to sell PPI and that the retailers were its sub-agents. Accordingly, GECB had to sign and return a guarantee that they and their sub-agents were ABI compliant. On behalf of GECB, Mr McPhail queried the requirement for sub-agents to explain the benefits and exclusions of cover prior to completing a sale. He said that GECB was not in a position to guarantee this and could only offer their best endeavours. He confirmed that compliance had assumed a high profile within GE, albeit seemingly only because of problems experienced recently in the United States, and asked whether it would be sufficient for GECB to make sure that staff were regularly audited, trained and updated.

[165]Several days later, Ms Dunlop had a telephone conference with GECB to discuss telephone scripts for the selling of Account Cover. This seems to have involved a customer being sent a hard copy application form which had to be returned before a sale took place. It was therefore rather different from an in-person sale in a store. At all events, while minimal details of the cover would be given over the telephone, the proposed script required customers to be expressly advised that further details could be found on the back of the application form and that they would have to sign a box to indicate that they wanted insurance cover.

[166]The documents then become rather sporadic. In April 1999, GECB prepared a Credit Operations Manual for Russell & Bromley which did not refer to PPI at all.

[167]In May 1999 there was an exchange of correspondence regarding a specific problem which had arisen in connection with the upgrade of certain covers. The exchange is relevant only because:(i) the reason the problem had arisen was because FICL/FACL’s original advice had been ignored by GECB;

(ii) it was in this context that FICL/FACL had requested an indemnity from GECB which had been resisted on the grounds that indemnities between affiliates was not normal GE policy – a point of which Mr Zellick made much in his submissions on AXA’s indemnity claim (see paragraph ‎274 below).

[168]On 21 May 1999, Ms Dunlop commented on GECB’s training pack for Account Cover II and raised a number of concerns, including the lack of reference to policy exclusions. In June 1999 further concerns were raised by FICL/FACL about GECB’s sales training video script which was assessed to fall short of point of sale requirements. Mrs Greenaway (who was then working for FICL/FACL) was asked to suggest amendments.

[169]In September 1999, GECB conducted a customer survey of its Account Cover product. This revealed that awareness of the product was generally low and that a significant proportion of customers did not realise they had the cover. Of those who cancelled cover, 40% felt it was not suitable, 29% did not ask for it, 22% were mis-sold the product and 9% felt forced to take it. The survey further indicated that a significant proportion of customers responding did not agree (even slightly) that all their queries had been answered, that staff were entirely sure about the details of cover, that they had sufficient time to consider the benefits, or that they had information about all the different levels of cover. 14% had felt pressurised into taking it. It was recognised that the attitudes of those cancelling the cover were worrying.

[170]The results of this survey do not make pretty reading, particularly after the specific focus in the previous year on compliant selling of Account Cover. However, I do not place too much weight on them, since there is no reliable information to indicate how extensive the survey was. More concerning is the concluding comment by GECB that “we must avoid passive sales because of the detrimental effect on satisfaction with the store card,” which gives some idea of GECB’s priorities, at least in the eyes of the author.

[171]In late 1999/early 2000, GECB compiled some training notes for Harrods’ forthcoming January sale. PPI was covered in a single page which did not contain any directions on how to sell the product but only a brief description of the cover.

[172]During the course of the summer of 2000, GECB was considering a proposal by Debenhams for new products to be launched on 14 August 2000. It is not clear that this directly affected the sale of PPI. However, a draft discussion document records that GECB would provide stores with a product and process training pack to contain a product overview, process flow and script. A “train the trainer” session was to take place on 25 July 2000 carried out by GECB’s Client Field Sales team. This would be followed by staff training in August 2000 on the basis of materials developed by Debenhams and GECB, also delivered by the Client Field Sales team. An Operations Manual outlining all process flows was to be drafted and included as an appendix. There is no indication that GECB sought to involve FICL/FACL in any of these steps or to seek their advice.

[173]On 1 December 2000, the Agency Agreement was signed at a meeting attended by Mrs Greenaway amongst others. There was some discussion about where GECB and FICL/FACL stood in relation to the GISC, which had been launched that summer, and also as to how best GECB could utilise its allocated 50 training days. However, GECB’s focus in this respect seems to have been on motivational training and increasing penetration rather than on compliance.

[174]Subsequent correspondence shows GECB querying with Mrs Greenaway whether there was any benefit in GECB joining the GISC, since FICL/FACL would have to become members anyway and would thereby assume responsibility for GECB’s compliance with the ABI Code in any event. GECB said that it would look to FICL/FACL for guidance. FICL/FACL’s response is unfortunately not in the papers.

[175]On 19 February 2001, some trainer notes were prepared for training in relation to Account Cover. Mrs Greenaway could not recall whether FICL/FACL had had any input into these notes but I find it more likely than not that they had, given the concerns that FICL/FACL had been raising in this regard since mid-1999. The notes expressly drew attention to the need to adhere to the ABI Code in the sales process and to ensure that the information and presentation was clear, accurate and not misleading, that the cover was sufficiently explained and that customers understood what they are buying.

[176]By October 2002, GECB was contemplating a further upgrade of the Account Cover product but was concerned about certain aspects of the proposed pricing in the light of the UCTA Regulations. In response to a query from GECB, FICL/FACL confirmed that, provided the customer was made fully aware of what they were buying and the applicable terms and conditions, the question of pricing was really an internal matter for GECB.

[177]In January 2003, correspondence was exchanged between GECB and FICL/FACL regarding the need for a signature box rather than a simple tick box to indicate a customer’s election to take PPI. GECB acknowledged that there had been significant problems in the past with a tick box when staff were incentivised simply to tick the box on behalf the customer without their explicit consent. In response, FICL/FACL insisted on a signature box absent a convincing argument to the contrary. In a follow-up email circulated within GECB on 3 April 2003, Mr Farrar of GEBC stressed that a signature box was therefore required and that push-back from the retail client was not an option. Similarly the Summary was to be included on the application form and not provided in a separate leaflet. He said:

“I need to draw your attention to the GISC rulebook, and in particular the Private Customer Code, section 3 'How you find insurance to meet your needs'. This clearly states that when selling any general insurance, the customer must be made aware of what they are buying before signing up to this… The Summary of Cover we provide at application does gives the amount of information we need to provide at the time a policy is sold. As nobody can be 100% confident about giving a leaflet out with the application form, we could be liable for breaching the rules. As members of GISC, GEIH will not allow us to put ourselves in this position. General insurance is increasingly coming under the spotlight, and I don't want to take this retrograde step now, especially as there are plans for the FSA to regulate in January 2005.”

[178]On 13 June 2003, an updated Credit Operations Manual was produced by GECB for Debenhams. As with the October 1998 version (paragraph ‎166 above), the one-page section dealing with PPI merely contained a brief outline of benefits and eligibility and contained nothing about compliant selling.

[179]By this time, GECB was well aware that it needed to prepare for statutory regulation to be introduced in January 2005 and on 24 September 2003, Mr Coorland drew up an Action Plan. This summarised the changes that would be required to ensure compliance with the new regime. Of note for present purposes are the following:(a) The FSA had accepted during the consultation period that GECB’s sales were non-advised. Accordingly, it did not need to carry out a suitability assessment. However, it still needed to conduct training and change its applications forms in line with the new requirements.

(b) The proposed new sales scripts incorporated an express reference to the Summary.

[180]Ms Greenaway could not recall whether FICL/FACL was asked to help with the training or to work on the sales scripts, although they did look at the application forms. There is nothing in the documents to suggest that GECB had insufficient resources to carry out the necessary training, or that it ever approached FICL/FACL for assistance. Mrs Greenaway’s evidence was that it would have done had it felt the need.

[181]In February 2006, Genworth confirmed that all sales literature, including scripts, policy documents and disclosure documents, had been reviewed and confirmed as compliant with current regulatory requirements. As far as Genworth was concerned, the Agency Agreement accurately reflected the parties’ respective responsibilities in this regard.

[182]Finally, in January 2007, GECB was fined £610,000 for having failed in 2005 to comply with the FSA’s regulatory requirements. An internal GECB email records that the FSA found that the sales procedure had been designed by GECB but that GECB had failed, in the light of emerging evidence, to review, amend or effectively operate that procedure. In some cases, retail staff had not followed the correct sales procedures by drawing attention to the Summary. The email recorded that GECB fully accepted responsibility for how policies were sold.

[183]I heard no evidence from anyone who was actually involved in the physical sale of PPI, whether in a retail environment or over the telephone and, as already mentioned, the experts had no such direct experience either. Nor did Santander call evidence from any store manager with responsibility for overseeing the sales process or from any GECB retail account manager or from anyone involved in providing training. There was therefore no evidence at all from Santander as to what steps were taken by sales staff to comply with the ABI Code or what measures were taken by GECB to ensure compliance, save for what I could glean from the documents. I.2.2: Findings as to breach

[184]Although I have well in mind the fact that the documentary record was incomplete, I am nonetheless satisfied that I can make the following findings as to actual selling practices on a balance of probabilities:(a) Although the papers before the court did not contain any documents earlier than 1996, I find it more likely than not that in relation to opt-out sales before they were phased out, GECB was routinely not drawing the attention of customers to the fact that they were being sold PPI which they could choose not to have. The natural inference from the findings of the Compliance Workout was that customers were not being invited to read the application form and that there was a continuing problem with inertia selling even after it had supposedly been outlawed.

(b) Even after 1994 when a positive opt-in was required, the time taken to sell PPI was very short – around 20-30 seconds according to GECB’s own estimate in 1997.

(c) There was pressure from both the stores and GECB to sell as many policies as possible and the approach adopted was fairly described as “hard sell” which, even if just within the bounds of what was acceptable, gave rise to a legitimate concern that customers were being pressurised. Even the change in selling pitch introduced in early 1998 involved positively promoting the product rather than simply presenting the customer with an option and leaving it to their choice. This may well have been in accordance with the approach adopted across the industry at the time, but that does not automatically mean that it satisfied the prevailing regulatory standards.

(d) Customers were not routinely directed to the Summary before agreeing to buy PPI. This was expressly recognised by the Compliance Workout and is consistent with the concerns expressed by the Insurance Ombudsman in his December 1997 speech and with the FLA’s “welcome letter” initiative in February 1998 which made clear that some customers were not being given any marketing material at all, let alone prior to conclusion of the sale. Even in the training notes prepared by GECB after 2000, there was no explicit instruction to refer to the Summary before concluding a sale. Most damning of all is the Action Plan proposal to change all sales script to incorporate an express reference to the Summary, from which I draw the almost inevitable inference that this had not previously been a feature of PPI sales, notwithstanding the agreement reached in February 1998 and Mr Farrar’s email of 3 April 2003. While the Summary was in many cases printed on the reverse of a triplicate form, the customer was typically only given a copy after they had already agreed to take cover and although it was open to them to ask for time to read the Summary, this put all the onus on them and was in any event irrelevant unless their attention had been drawn to the existence of the Summary in the first place.

(e) Undoubtedly a full copy of the policy wording was sent to the customer later with full documentation but, as the experts agreed, it was by then far too late. Failure to refer to the Summary before the sale was concluded could not be cured by providing documentation later.

(f) GECB’s primary concern was to maximise sales. Many sales were conducted by staff who did not fully understand the product and were incentivised to achieve as many sales as possible. While GECB ensured that its marketing materials were approved by FICL/FACL, it relied heavily on the fact that those materials were handed over on completion of the transaction and that the customer could cancel during the cooling off period. This was a very lucrative market for GECB accounting for around one-third of its net income 88% of 39% according to the initial notes of the November 1997 compliance workout session. and prior to 2005 it was largely content to pay lip-service to compliance and to absorb any mis-selling complaints which did emerge. However, GECB’s comment in the September 1999 customer survey is revealing in so far as it demonstrates that the vice of inertia selling from GECB’s point of view was the detrimental impact on customer satisfaction with the store card rather than any genuine concern about compliance. The Compliance Workout showed that there were a significant number of complaints although it is impossible to say how many, since there was no evidence on the point.

(g) The available evidence suggests that these sales practices persisted throughout the entirety of the period with which this dispute is concerned: (i) There is no suggestion in the Compliance Workout that the compliance failures disclosed were of recent origin. On the contrary, the clear impression given is that they were systemic and endemic. (ii) Mrs Greenaway’s evidence was that the practices highlighted by the FSA when they issued GECB with a Final Notice in 2007 accurately represented the sales practices followed by GECB before 2005 as well as after. She accepted that GECB applied much more rigorous techniques and practices after it became regulated, from this the natural, indeed inevitable, inference is that they were not as strict or rigorous previously. (iii) The point can also be made that the problems identified by GECB in late 1997/early 1998 were still being identified in 2007. Mrs Greenaway very fairly accepted that she could not say whether this meant that there had been failures throughout the entirety of the intervening period or just at those two points in time. However, the latter proposition would be remarkable and I cannot accept it. Indeed, it is inherently unlikely that the level of compliance on the part of intermediaries such as GECB would have got worse after regulation and GECB’s own efforts to prepare for the advent of FSMA indicates that it was trying to improve its practices from those which had previously prevailed.

[185]The ABI resume made clear that the ABI Code required the exercise of best endeavours to ensure compliance, which was explained as meaning every practical effort within reasonable bounds. I am satisfied, based on the expert evidence, that this required the steps set out in paragraph ‎132 above to be taken and that the consequence of my findings above is that there were systemic breaches of the Code in at least in the following respects:(i) A routine failure in opt-out sales to make the customer aware that they were purchasing PPI at all and that there was an option to decline cover;

(ii) A routine failure to provide even an outline explanation of the policy. I cannot say how reliable the 20-30 second estimate was, but it was GECB’s own estimate and it stuck with it. The idea that this would have been sufficient for a sales assistant to explain that the store was acting as an agent, to outline the key coverage of the product and the cost, to mention the existence of exclusions/limitations and to refer to the Summary as well as to secure an express opt-in is frankly bordering on the delusional. It was clearly inadequate and I am satisfied that it would have been both practical and reasonable to have taken longer over the sales process to ensure compliance;

(iii) A routine failure to draw attention to the Summary before concluding the sale.

[186]While these were very much systemic failings, they would inevitably have led to breaches of the ABI Code in individual cases and the application of my findings to the sample files is considered further in the Appendix to this judgment. I.3: Responsibility as between FICL/FACL and GECB: did FICL/FACL fail to exercise best endeavours?

[187]It was a constant refrain running through Mr Zellick’s submissions that FICL/FACL had the primary and overriding responsibility from a regulatory point of view for selling PPI and that they abjectly failed to discharge those responsibilities. His criticisms of FICL/FACL in this regard were aimed at three principal targets: first, a general failure by FICL/FACL to provide GECB with the necessary training and direction; second, FICL/FACL’s failure to intervene or take any corrective action when they either knew or should have known exactly how the product was being sold by GECB and its sub-agents; and third, the Budget Account policies (see paragraph ‎116 above). I.3.1: Assistance and training

[188]On the basis of the evidence set out above, I am unable to find that there was any failure by FICL/FACL to exercise best endeavours or to take reasonable steps so far as concerns the provision of assistance or training.

[189]It is abundantly clear that GECB knew about the ABI Code and what it entailed: see paragraphs ‎147, ‎154 and ‎161 above and also the evidence of Mrs Greenaway that GECB was aware at all times of the need to comply with the ABI Code. Moreover, FICL/FACL knew that GECB had this knowledge, not least because they repeatedly drew it to GECB’s attention (paragraphs ‎158, ‎163 and ‎164‎ above) and because the Agency Agreement included an express obligation on the part of GECB to comply with the ABI Code. The Code was not difficult to understand and there is nothing in the evidence which either would or should have given FICL/FACL reason to suspect that GECB itself required any training in its requirements – as opposed to the store staff.

[190]Mr Zellick nonetheless submitted that in order to exercise best endeavours, FICL/FACL should have audited the sale process by observing actual sales or watching GECB training sessions. However, given the extent to which FICL/FACL was kept firmly at arm’s length by GECB from any involvement in the retail side of the operation, the idea that FICL/FACL could have monitored the sales process in order to identify any training needs is fanciful. To the contrary, the 4 February 1998 memo referred to in paragraph ‎156 above shows that Mr Jones and Ms Lloyd very much wanted to obtain a copy of GECB’s coaching tools but regarded it as a forlorn hope.

[191]In any event, an express offer of training was made to Ms Carter in 1996 but this seems to have been simply ignored and GECB only turned to FICL/FACL for any assistance at all at the very end of 1997 following the Compliance Workout. Even then, FICL/FACL was only made aware of one very limited aspect of the findings of the workout.

[192]Thereafter there seem to have been occasional specific requests from GECB for FICL/FACL’s input into specific aspects of training/training materials and for assistance with particular issues, and this was provided promptly and readily. Examples are the provision in July 2003 of a Powerpoint presentation on the ABI’s Consistent Policy Interpretations for Creditors; the provision of training on the claims process in October 2004. Otherwise, the clear impression created was that GECB considered itself capable of designing, organising and delivering training itself. This can most clearly be seen in the fact that when GECB started preparing for FSMA, it automatically assumed responsibility for training without any reference to FICL/FACL. GECB never suggested that its resources were insufficient or inadequate to provide adequate training through a “train the trainer” approach and there is certainly no evidence that FICL/FACL were made aware that in fact store retail staff had to train themselves, that GECB had no idea who had been trained or how effective any training was, and that there was a general paucity of training documentation available in the retail stores.

[193]From 1 December 2000 onwards, the Agency Agreement expressly provided in clauses 7.1(a) and (b) that FICL/FACL should:

“(a) provide GECB with such market research, marketing and technical assistance as [FICL/FACL] think necessary to assist GECB with the marketing and sale of the Insurance; (b) provide GECBwith such specialised training in relation to the Insurance and the marketing and sale thereof as [FICL/FACL] determine in their reasonable discretion is necessary to enable GECB to comply with its obligations under this agreement, limited to 50 days annually.”

[194]Mr Zellick sought to suggest that this placed an obligation on FICL/FACL to identify any training needs on the part of GECB in relation to the sales process. However, for the reasons already set out, I regard that as a wholly impractical construction to put on the agreement in the context of the relationship between the parties, particularly when the products were largely designed by GECB itself on the basis of its own market research from which FICL/FACL was firmly excluded. In my judgment, these provisions instead had in mind assistance and training relating to the PPI product itself rather than the sales process. Thus, if a new product were to be introduced, then FICL/FACL would provide any specialised training or market research required in relation to the specific characteristics of the product. In fact, however, all PPI products tended to be variations on a common theme, most of which had been designed and proposed by GECB itself. Nonetheless, so far as the technical aspects of cover were concerned, FICL/FACL reviewed and approved all marketing materials to make sure that correct and accurate descriptions were given.

[195]I reject the suggestion put to Mr Jones that a charge for training over and above a 50-day allowance was inconsistent with the exercise of best endeavours by FICL/FACL. Given the premium split between the parties, I do not regard FICL/FACL’s obligation to exercise best endeavours as extending to the provision of services to an agent gratis. In any event, had GECB ever taken advantage of the training on offer (which it does not appear to have done), it would have been charged at a preferential rate. In fact, the discussion on 1 December 2000 about use of the allocated training days (paragraph ‎173 above) demonstrates that GECB was more interested in training for its own commercial purposes rather than in order to ensure compliance.

[196]In any event, clauses 7.1(a) and (b) only required FICL/FACL to provide such assistance or training as they considered necessary, i.e., they had a discretion. However, they could only have concluded that assistance and training were necessary if and to the extent that they were given reason to believe that it might be. In her evidence, Mrs Greenaway was unable to identify any specific example of training that could have been provided by FICL/FACL which would not have been obvious to GECB in any event. The Compliance Workout also demonstrates that GECB was perfectly aware of what compliance required and the problem was not that it had been insufficiently trained itself, but that it was either unwilling or unable to take the necessary steps to ensure compliance by the retailers.

[197]In truth, the reality is that FICL/FACL always stood ready to provide training but that GECB was uninterested. This was confirmed both by Mrs Greenaway and by Mr Brandon-Cross who accepted that there were occasions when FICL/FACL were doing their best to try to help GECB but that the latter did not want to accept that assistance. Mrs Smith’s evidence was to similar effect: the sales process was “owned by GECB” and FICL/FACL routinely offered training on product knowledge but this was rarely, if ever, taken up by GECB.

[198]The evidence does not therefore support a finding that FICL/FACL failed to exercise best endeavours in relation to either clauses 7.1(a) or (b). I.3.2: Inaction by FICL/FACL

[199]Mr Zellick’s second main criticism of FICL/FACL is that they were fully aware of how sales were being conducted and yet never made any complaint, let alone attempt to correct any deficiencies.

[200]I regard the premise of this submission as flawed. The starting point is that GECB was aware from at least November 1997 following the Compliance Workout that there were serious deficiencies in its sales processes which were leading to a significant number of complaints. However, GECB never made FICL/FACL aware of the full findings of the Compliance Workout and only a very limited and sanitised version was presented to Mr Jones in January 1998. Moreover, from FICL/FACL’s point of view, GECB gave them to understand that it recognised the issue and was addressing it. Whatever reservations FICL/FACL might or might not have had about the proposed revised selling message and the huge variety of potential scripts, it nonetheless took steps to secure GECB’s agreement to refer the customer expressly to the Summary before a sale was concluded which, if done, would at least have meant that the sale was compliant.

[201]It is not disputed that FICL/FACL reviewed and approved all marketing materials to ensure that they met the regulatory requirements. Otherwise, GECB only sought assistance in relation to specific issues and the impression created is that it was keeping the question of compliance by the retailers entirely in-house. Whether this was for reasons of confidentiality or because of resistance by the stores to the involvement of a third party matters not. What is clear is that GECB was very definitely not looking to FICL/FACL to ensure regulatory compliance at the point of sale generally. As I have found above, GECB knew for itself full well what was required and regarded the relationship with the retailers as its exclusive prerogative.

[202]This is not to say that FICL/FACL were oblivious to their responsibilities to monitor GECB’s activities as their agent. On the contrary, they were acutely aware that they bore responsibility vis-à-vis the regulator and it was for that reason that they insisted on GECB signing a guarantee of compliance on behalf of itself and its sub-agents.

[203]In the circumstances, it is unfair to criticise FICL/FACL for failing to deal with problems which they did not know about and had no reason to suspect. No doubt FICL/FACL knew in general terms the nature of the sales process that was supposed to be followed but that is very different from knowing whether it was being followed in practice. If GECB did not know what was going on at the point of sale in the stores and was taking no steps to monitor the position (as the Compliance Workout demonstrates), it is fanciful to suppose that FICL/FACL should have had any more extensive knowledge. GECB was aware of the requirements of the ABI Code to FICL/FACL’s knowledge and FICL/FACL was entitled to assume in the absence of evidence to the contrary that GECB was taking all reasonable steps to ensure compliance.

[204]I am therefore not satisfied that FICL/FACL knew that sales staff were routinely failing to comply with the requirement to draw attention to key information prior the sale, and, in particular, failing to refer to the Summary. I have accepted that a salesperson was not expected to sit down and start discussing policy exclusions, pre-existing medical conditions, financial circumstances or different levels of cover in detail. But in those circumstances, the experts agreed that an express reference to the Summary was essential and this could readily be done even in the context of a short timescale for selling.

[205]Mr Zellick argued that at least from January 1998, FICL/FACL was aware that sales were taking only 20-30 seconds to conclude and that they should have appreciated for themselves that this was inadequate (as I have held) and taken steps to direct GECB that it was not acceptable. However, it seems to me that this misses the point. The essential requirement, as FICL/FACL made clear, was to ensure that a proper explanation of the product was given to the customer before the sale was concluded. How long that would have taken is irrelevant. It would have differed from customer to customer and in any event, as already noted, FICL/FACL did take steps to secure GECB’s agreement that the Summary would be expressly mentioned before sale.

[206]Santander also complained that FICL/FACL failed to take steps proactively to monitor sales. But, as already pointed out, it is not clear what FICL/FACL could in practice have done. They approved all the marketing materials but GECB controlled all the customer data and only reported to FICL/FACL in bulk so that there was no way in which FICL/FACL could have identified potential mis-selling problems through an analysis of customer data. The bulk reports submitted by GECB did not even identify the number of new policies sold by particular retailers each month. There is no way in which FICL/FACL could have audited sales by way of “mystery shopping” on any scale which would have been meaningful, let alone representative.

[207]Mr Zellick further suggested that FICL/FACL should have provided their own sales manual to GECB. But since FICL/FACL had no access to the retailers or to the processes followed at the point of sale, any such manual would have been at such a high level of generality as to add virtually nothing to the provisions of the ABI Code and the related Guidance.

[208]I asked Mr Zellick what directions or scripts or manuals Santander say should have been provided by FICL/FACL over and above what was already set out in the ABI Code. The most he could offer in response was that it was a difficult question to answer because it depended on the prevailing standards at the time. This did not enlighten me greatly and he was unable to cast any further illumination on the point.

[209]In short, I am not satisfied that FICL/FACL ought to have known about any problems with the sales process other than those expressly drawn to their attention by GECB. Nor am I persuaded that there was anything more they could reasonably have done to ensure that proper procedures were being followed by GECB. As between FICL/FACL and GECB, the latter controlled the selling environment and FICL/FACL were effectively excluded from the relationship between GECB and the retailers. The retail agreements between GECB and the stores were kept confidential and Mrs Greenaway confirmed that FICL/FACL would not have known whether incentives were being offered to customers or staff. Nor were they routinely permitted access to the telecentre in Leeds. This may have been for legitimate reasons of confidentiality but the fact remains that FICL/FACL were only allowed in on one or two occasions.

[210]In practical terms, therefore, the most FICL/FACL could do was to put in place a system which gave reasonable reassurance that the ABI Code was being enforced by GECB. True it was that FICL/FACL were the insurance experts, whereas GECB was a credit lender. But GECB was hardly an unsophisticated ingenue. On the contrary, it was well aware of the requirements of the ABI Code and its obligation to comply with them and in my judgment FICL/FACL were entitled to assume that GECB would act responsibly in that regard and would ask if it wanted or needed input or assistance. When such requests were made, FICL/FACL provided assistance promptly and readily. In these circumstances, I am not persuaded that FICL/FACL could realistically have done more than they did. No witness could point to any occasion when they became aware of a compliance failure but were indifferent or did nothing about it.

[211]In this context, I note also that FLA’s Head of Legal & Regulatory Affairs referred in a letter dated 28 January 1998 to the fact that the Insurance Ombudsman’s Bureau had accepted that the obligation to exercise best endeavours to ensure that a policy was properly sold was satisfied if the agency agreement with the intermediary required that they provide all prospective policyholders with sufficient information to understand the key features of the cover. Clauses 6.1(a) and (g) and 6.2 (a) and (c) of the Agency Agreement achieved this by placing a primary obligation on GECB to comply with the ABI Code, including an explicit requirement not to incur any liability on FICL/FACL’s behalf otherwise than by selling PPI in accordance with the agreement.

[212]Accordingly, as between FICL/FACL and GECB compliance with the ABI Code as regards the actual sale process was in both practical and legal terms the primary responsibility of GECB. I find that there was no failure by FICL/FACL to exercise best endeavours.

[213]Even if FICL/FACL had attempted to issue more instructions or directives regarding compliance, I am not persuaded that they would have had any effect:(a) The dynamic of the relationship between the parties was that GECB did what GECB wanted and perceived was in its best commercial interests;

(b) There is evidence in relation at least to the incident described in paragraph ‎167 above, that GECB had ignored FICL/FACL’s advice;

(c) FICL/FACL could issue whatever instructions or directives they liked but they were not in any position to enforce them. Mrs Greenaway confirmed this when she gave evidence that if FICL/FACL had asked for the sales pitch to be extended, they would have been told that the retailers would not be happy to do so;

(d) The stores themselves appeared to be completely oblivious to compliance, as disclosed by the Compliance Workout, and the reality was that GECB made very little, if any attempt, to direct them in this regard. It is noteworthy that when GECB was asked by FICL/FACL in December 1998 to provide a guarantee of regulatory compliance on behalf of its sub-agents, it objected on the ground that it could only exercise best endeavours in that regard. If GECB, with its much closer relationship with the retailers, could not guarantee compliance, FICL/FACL was hardly in a position to do more;

(e) The overriding impression is that GECB only showed any interest in compliance at all when it looked as if there might be meaningful adverse consequences for it, for example, when complaints began to be received from the Trading Standards Office, when the Insurance Ombudsman’s speech started to make waves, after problems emerged in the United States (paragraph ‎164 above) and (most notably) when FSMA was on the horizon. Mr Jones’ evidence in his witness statement was to similar effect, namely that GECB normally only showed any interest in compliance when there was a statute or an EC Directive that needed to be observed. Protestations by Santander that GECB would have complied with any directives or instructions issued by FICL/FACL accordingly ring rather hollow.

[214]It is all very well to argue, as Mr Zellick did, that it was unrealistic to suppose that absolutely no useful training or instruction/direction could have been provided by FICL/FACL to GECB. However, it is Santander which is alleging that any non-compliance was caused by failures on the part of FICL/FACL and accordingly Santander bears at least an evidential burden in this respect, as he accepted. If Santander is unable to identify anything specific which it says that FICL/FACL should have done but failed to do, it is equally impossible for Santander to say how that failure would have changed anything in practice. I am not prepared on the basis of the material before me to draw the inference that some form of unidentified and unspecific training or instruction should have been given which would - in some equally unidentified manner - have prevented the compliance failures which occurred. I.3.3: Budget Accounts

[215]In so far as it is suggested that the design of Budget Accounts was inherently flawed, this is a hopeless argument.

[216]In my judgment, none of the clauses of the Agency Agreement on which Santander relies were on their true construction aimed at product design. Clause 7.1(d) was clearly aimed at the regulatory aspects of matters such as licensing, solvency and authorisation to conduct business, while clause 7.1(f) was addressing ethical business practices rather than product design. The same is true of clause 5 since I agree with AXA that the objective of the ABI Code was clearly to regulate the sale of insurance products rather than the design of the products being sold.

[217]It is arguable, I suppose, that clauses 5 and 7.1(d) could cover product design but only if it was so obviously flawed that it was unethical to sell the product at all. This is a high hurdle to meet, and while there was an allegation to this effect in Santander’s written opening it did not feature in Mr Zellick’s oral submissions or written closing. I would in any event have rejected the suggestion. Budget Accounts were in fact finance products designed originally by GECB’s predecessor for customers on limited budgets who typically maintained lower balances on their accounts. The customer would select a maximum monthly repayment level and the credit limit was then pegged by reference to that amount. While cover under the associated PPI product was limited to the chosen monthly repayment rather than the full outstanding balance, the premium tended also to be commensurately lower because of the generally lower outstanding balances on the account.

[218]The only reference to Budget Accounts as inherently flawed products appears in an internal Santander email where Ms Webdale relayed what was said to be the view of the FOS in November 2015 that Budget Accounts were flawed and should not have been sold to anyone. This was apparently on the basis that the cost of cover delivered next to no value on what the customer could get back. The email confirms that this was only a relevant concern for a limited number of brands.

[219]However, an earlier email from the FOS on 1 April 2015 explained its thinking in slightly more detail, namely that these accounts did not represent good value for customers with a higher spend or higher average balanced. While the FOS accepted that they were relatively cheaper for customers with a low spend relative to the credit limit, they did not think that such customers would actually need the policy and that “in most cases where the customer had alternative and worthwhile provision (say, 3 months’ worth of employee benefits or savings), we don’t think they’d have bought the policy with a full understanding.”

[220]Several points can be made about this. First, and perhaps most importantly, the PPI product associated with a Budget Account was a standard policy. The only difference from other types of account lay in the pricing and the credit limit, neither of which were anything to do with FICL/FACL. Secondly, the FOS seems in April 2015 to have regarded the real vice of the policies as being that they had been sold without adequate explanation of their coverage so that customers did not understand the full implications of having a Budget Account. This is more a concern about the overall package and/or the sales process, rather than the inherent design of the policy itself. Thirdly, there was a concern that the policy was unnecessary where the customer had alternative means. However, this begs the question as to whether that was necessarily the case. Fourthly, the very type of customer at which the Budget Account was targeted was those who were likely to spend less, in which case the FOS appears to have accepted that they received some value – at least where they did not have alternative means.

[221]Neither Ms Webdale nor anyone from the FOS was called as a witness and I am not inclined to place much, if any, weight on these emails. Both were hearsay and there was no proper opportunity of interrogating the allegation or establishing why the FOS had apparently hardened its view by November – if indeed it had, which in the absence of evidence cannot be regarded as certain. Without much harder evidence, I would find it difficult to conclude that the insurance product associated with Budget Accounts was so deeply flawed that it was positively unethical to sell it in any circumstances whatsoever. If the vice was that the cost/benefit relationship was not properly explained, that is something for which GECB itself was responsible.

[222]As stated in paragraph ‎252 above, the comments of the FOS in relation to Budget Accounts were used by Santander as a fig-leaf to excuse its failure to sign the Settlement Agreement. The fact that it only raised the point nine months after first becoming aware of it, does not suggest any genuine concern about the design of the product. I do not accept that there was any fault or failure to exercise best endeavours in this regard on the part of FICL/FACL.

[223]In summary, I reject Santander’s primary submission that any failings by GECB with regard to the sale of PPI was attributable to FICL/FACL’s own failure to meet their regulatory responsibility to exercise best endeavours to ensure compliance with the ABI Code or to provide appropriate training or direction. As Ms Greenaway accepted, it would be “slightly laughable” to lay the blame for the mis-selling at FICL/FACL’s door.

[224]I am therefore now in a position to turn to the specific causes of action relied upon by AXA. J: THE SETTLEMENT CLAIM

[225]AXA contends that a binding settlement was reached with Santander that it would be liable for all the consequences of any mis-selling prior to 2005, albeit not for any on-going administration costs incurred by Genworth after the new claims handling arrangements were put in place. It submits that binding agreement to this effect was reached either on 4 June 2015 at the meeting or on 9 June 2015 by an exchange of emails between Mr Rember and Ms Day. A further suggested alternative, namely that agreement was reached as a result of communications between Mr Rember and Mr Conway in October 2015 when Santander asked for execution copies of the Settlement Agreement to be distributed, is no longer pursued.

[226]I have set out the relevant sequence of events in section H above. No minutes were taken of the 4 June 2015 meeting and the closest we get to a written record is Mr Rember’s follow-up email of 5 June 2015 where (in accordance with his usual practice) he confirmed the discussions in writing so as to put something on the record. This was accepted to be an accurate record of the meeting and there is accordingly no dispute that the parties reached agreement that:(a) Santander would assume liability for redress payments, FOS fees and Genworth’s administrative costs to date, including the costs of claims handling, since it was anticipated that these would be virtually eliminated for Genworth once the new Claims Handling Agreement came into effect;

(b) Various other steps would be taken by the parties as set out in paragraph ‎112 above;

[227]It was also common ground that nothing was expressly said on either side to indicate that the agreement was subject to contract or not binding until the written agreement was signed.

[228]Mr Pateman’s evidence was that he believed the meeting was only concerned with sales which had taken place after the conclusion of the Agency Agreement on 1 December 2000 and that it was only on that assumption that he agreed Santander would continue to pay redress. However, this subjective understanding was never articulated at the meeting and there was nothing in the correspondence to suggest that any such distinction was to be drawn between pre- and post-Agency Agreement sales. Accordingly, Mr Pateman’s beliefs cannot detract from the fact that there was, as all parties accepted, an objective agreement by Santander to pay redress, FOS fees and administrative costs to date irrespective of the date of sale.

[229]Santander also argued that its consistent opinion since May 2012 had been that it was under no actual liability in respect of pre-2005 policies and that this position had never changed, notwithstanding the interim agreements reached in 2012 and 2013. That may be so, but I regard it as irrelevant. To the extent that Mr Pateman objectively agreed otherwise at the meeting without objection from anyone else, Genworth was entitled to believe that Santander had changed its mind. J.1: Agreement at the 4 June 2015 meeting

[230]Two questions arise for determination:(i) was the agreement “subject to contract”?

(ii) if not, can the agreement reached be split into discrete elements relating to liability (said to have been definitively agreed) and operational matters (admittedly left for further negotiation)?

[231]Unsurprisingly, there was no dispute between the parties as to the applicable law. The question is whether the parties intended to be bound in advance of executing a written agreement. For this purpose, the court must look at the whole course of conduct and communication between them in order to discern their objective intention. I do not accept that there is any presumption or “normal inference” to be drawn in this regard; it is a question which depends on the particular facts. Thus, in an appropriate case, the parties can be bound by an oral agreement even if they intend to memorialise it but subsequently fail to enter into a formal document: Bear Stearns Bank plc v Forum Global Equity Ltd,

[2007] EWHC 1576 (Comm)

at [171]; Newbury v Sun Microsystems Ltd,

[2013] EWHC 20180 (QB)

at [32].

[232]However, although the subjective intentions of the parties are generally inadmissible in relation to questions of construction, there is a quasi-exception where oral agreements are concerned. In this specific context, evidence of subjective understanding and subsequent conduct is admissible in so far as it tends to show, objectively, whether agreement was reached, what the terms of that agreement were and whether it was intended to be binding. In Blue v Ashley,

[2017] EWHC 1928 (Comm)

at [64], Leggatt J (as he then was) explained the rationale for the exception as follows:

“In the case of an oral agreement, unless a recording was made, the court cannot know the exact words spoken nor the tone in which they were spoken, nor the facial expressions and body language of those involved. In these circumstances, the parties’ subjective understanding may be a good guide to how, in their context, the words used would reasonably have been understood. It is for that reason that the House of Lords in Carmichael v National Power Plc [1999] 1 WLR 2042 held that evidence of the subjective understanding of the parties is admissible in deciding what obligations were established by an oral agreement.”

[233]What I understand this passage to mean is not that the court can admit evidence of what a party subjectively hoped to achieve or intended to convey by its words. Rather the court can admit evidence as to what each party subjectively understood the other party to be conveying.

[234]I accept that the context of the meeting was to find a definitive solution to a problem which had been dragging on for years and was beginning to show worrying signs of attracting regulatory attention. Genworth felt that it had been given the runaround for a very long time talking to different teams, none of which seemed to know what other teams within Santander were saying or doing. It is clear from the evidence of Mr Pateman and Mr Hazell that Santander operated in a very rigid manner where there was little co-operation or co-ordination between departments. Each division had its own reporting lines and procedures and other departments did not get involved unless asked. This was particularly the case with PPI remediation which appears to have been completely ring-fenced. Little wonder, then, that Mr Rember had become frustrated at his inability to get a consistent or coherent response from Santander and was anxious to ensure that any agreed solution would “stick”. This obviously required the involvement of senior people who could commit Santander to what was agreed.

[235]I also accept that there was no impediment from a regulatory point of view to Santander committing itself to the agreement orally. The contrary was put to Mr Rember in cross-examination but it lacked any evidential support, although that of course is not the same as saying that Mr Pateman necessarily had authority or intended to conclude a binding agreement at the meeting.

[236]The meeting itself was expressly held “without prejudice” and this is a powerful indication that the parties would not necessarily have felt under any pressure to reach a binding agreement. An equally powerful point on the other side is that “without prejudice” and “subject to contract” are paradigmatic ways of indicating that no agreement has been reached until reduced to writing: see, for example, Whitehead Mann Ltd v Cheverny Consulting Ltd,

[2006] EWCA Civ. 1303

at [42]. Yet, none of the experienced lawyers on Santander’s side thereafter used the expression in their correspondence. Nonetheless, as Whitehead makes clear, it is not essential to have an express stipulation and, in my judgment, the objective observer would have expected a regulated banking entity the size of Santander to have internal governance procedures and, at the very least, only to commit to a permanent solution of a long-standing and contentious issue in writing. It would be unusual and surprising to find it entering into a multi-million pound settlement orally on the basis of a meeting lasting less than an hour.

[237]I therefore conclude that the agreement reached at the meeting was, in effect, subject to contract. For the avoidance of doubt, this is not because the agreement was subject to any further review, or because further authority was necessarily required in order to commit to the substance of the agreement. Mr Pateman was sufficiently senior in the organisation to make the agreement. He had been briefed in the context of Genworth’s express demand that key decision-makers would be present and I find that he and Mr Lloyd had authority to bind both SCL and SISUK. No distinction was ever drawn between the companies and they were certainly held out as having authority to act on behalf of both. Moreover, none of the Santander witnesses was able to point to any formal policy requiring ExCo approval or indeed any other procedure for entering into agreements. I accept AXA’s submission that if this had in fact been the case, something would have been said either at the meeting itself or during the subsequent negotiations on the outstanding matters. If there were any doubt about this, Ms Day only confirmed the accuracy of Mr Rember’s 5 June 2015 email after obtaining the “relevant inputs” which in itself implies the existence of due authority to make the agreement.

[238]Nevertheless, none of that answers the question as to whether Santander intended to conclude a binding agreement prior to signature. All the witnesses agreed that the meeting was shorter than anticipated (estimates varied from 20-40 minutes) because Mr Pateman had to leave for his pre-arranged meeting with the FOS on another matter. This, combined with the express consensus that the agreement would need to be recorded, militates against an objective intention to reach a binding agreement there and then.

[239]Mr Green made much of the fact that Santander had previously paid out substantial sums and had made provision in its accounts on the basis that it would be assuming liability for claims handling and redress - all without any written agreement or suggestion that ExCo approval or higher authority was required. However, I do not regard this as a particularly weighty point given that the previous agreements in 2012 and 2013 were never understood to be a definitive solution for all time. On the contrary, they were clearly only interim agreements intended to hold the ring pending the introduction of the new claims handling process.

[240]AXA further relied on the fact that although Mr Hazell had insisted he had strong objections to Mr Pateman’s approach, he said and did nothing to raise this either at the time or subsequently. This is a double-edged sword. It could equally be said that Mr Hazell was reluctant to embarrass Mr Pateman in public and only kept quiet because he understood that no finally binding agreement would be made at the meeting.

[241]More importantly, even on AXA’s case, consensus still had to be reached on a number of other matters and I regard it as unrealistic to separate the acceptance of liability (Genworth’s desideratum) from those other matters (Santander’s desiderata). I therefore reject Mr Green’s submission that agreement on liability can be separated out and treated as binding on the basis that it was the price of Genworth’s co-operation in relation to the other matters.

[242]It is true that most of the matters left outstanding at the meeting were purely ancillary, for example, the reconciliation of the balances owing each way and the termination of the Agency Agreement. However, others were more substantive. For example, the respective obligations of the parties under any claims handling agreement, and the nature of any future operational costs which might be incurred by Genworth. Although it was expected that the latter would largely be eliminated by the new claims handling arrangements, this was by no means settled and it was also possible that the new arrangements might affect the extent of FOS fees to be incurred. Furthermore, what would have been the position if the FOS had not agreed to deal directly with Santander, so that costs were not reduced as anticipated? As it happens, the FOS did agree to this in principle at its meeting with Mr Pateman later on the same day, but this could not have been known at the time. I also doubt that Santander would have been quite so sanguine about concluding an agreement unless satisfactory assurances were given with regard to NewDay and this was almost certainly an important factor in Mr Pateman’s willingness to reach agreement.

[243]For these reasons, I am satisfied that the agreement reached at the meeting was a package which must be viewed as a whole and cannot be artificially dissected into separate elements.

[244]Mr Green nonetheless relied on the fact that Santander subsequently conducted itself on the basis that agreement had been reached, for example in its provisioning policy. However, this is not conclusive. As with the draft Agency Agreement, the parties may have anticipated and expected it to be concluded, but the fact that preparations for implementation were being made on the ground, or even that some employees thought it already applied, does not take the matter any further. In any event, it would have been the height of imprudence not to make appropriate financial provision immediately if it was expected that the agreement would in due course be signed.

[245]I regard it as equally irrelevant that the final version of the Settlement Agreement signed by AXA still left the CHA open for further negotiation. The position must be assessed as at the date of the meeting when it could not have been known that the parties would not in fact be able to finalise the CHA by the time that all the other matters were agreed.

[246]Decisions like RTS Flexible Systems Ltd v Molkerei Alois Müller GMBH & Co KG (UK Production),

[2010] UKSC 14

are therefore not in point. In that case, agreement was reached which was expressly subject to contract but which was subsequently performed and paid in part on an amended basis. The Supreme Court rejected the idea that there was no contract at all but equally was unable to find that only some of the terms had been conclusively agreed. In the event it held that a contract had been concluded by conduct on the terms originally agreed as subsequently varied, and that the parties had impliedly waived the “subject to contract” provision.

[247]Does evidence as to subjective understanding make any difference in this case? In my judgment it does not. On the contrary, I am satisfied that Genworth did not understand that any binding agreement had been reached at the meeting. Mr Rember clearly understood that agreement on the substance had been reached. That is not in dispute. But despite his evidence to the contrary, I cannot accept that he subjectively thought that Santander was bound to liability in advance of a full written agreement. If that were the case, it is inconceivable that he would not have raised this – forcefully – on at least one of the following occasions:(a) When Santander refused to sign the Settlement Agreement in January 2016;

(b) In the formal letter sent by AXA’s solicitors on 4 September 2017 relying on a binding agreement having been reached in 2012 as the basis of Santander’s liability. Mention of the 4 June 2015 meeting was conspicuous by its absence;

(c) When negotiating the provisions in the CHA regarding the irrecoverability of service charges. An obvious reason which could have been deployed for resisting irrecoverability was that Santander had already agreed to pay these charges;

(d) When negotiating the standstill agreement.

[248]In fact, the first suggestion that a binding agreement had been reached on 4 June 2015 was only made some years later in December 2020. The reason given for this by AXA was that it was pointless to mention it earlier because Santander had long since reneged on the agreement and AXA had no choice but to agree to its demands. I find this unconvincing. The most likely explanation to my mind is that the argument was subsequently forged in a legal crucible as an ingenious afterthought. I hasten to add that there is nothing improper in this. It is for the court to decide whether or not a binding agreement was concluded, irrespective of what the parties may or may not have thought. Litigants are entitled to present the facts in whatever legal guise seems best to them. If a particular argument only comes late to the party, that may be something which goes to its weight, but it is not objectionable to put it forward. In my judgment, however, the complete absence of any previous assertion that binding agreement had been reached in June 2015 seriously undermines any suggestion that this was in fact Genworth’s subjective understanding at the time.

[249]For their part, the Santander witnesses unsurprisingly all gave evidence that no binding contract had been concluded, but I agree with Mr Zellick that if AXA cannot get home on the basis of Mr Rember’s evidence, then Santander’s subjective understanding is irrelevant. I was, nonetheless, invited by Mr Green to draw adverse inferences from Santander’s failure to call Ms Day to give evidence. The drawing of adverse inferences from the absence of a witness is largely a matter of common sense depending on the particular context and circumstances, including:(a) whether the witness is available;

(b) whether it is reasonable to expect that they have relevant evidence to give;

(d) the significance of the points in the context of the case: Efobi v Royal Mail Group Ltd,

[2021] UKSC 33

;

[2021] 1 WLR 3863

at [41].

[250]Here, Ms Day’s fingerprints were all over the prior correspondence and the subsequent negotiations. She also attended the meeting itself and clearly had relevant evidence to give on the critical point of whether binding agreement had been reached. None of Messrs Lloyd, Hazell or Pateman could really recall exactly what was said, whereas Ms Day had been much more closely involved both beforehand and afterwards and could reasonably have been expected to have a more detailed and accurate recollection. No explanation was offered as to why she had not been called and there was no suggestion that she was either unwilling or unavailable to attend. I understand that she has now retired, but so have Mr Lloyd and Mr Conway, both of whom were proffered as witnesses. My hunch is that she was hugely embarrassed by Santander’s subsequent withdrawal from the agreement and possibly reluctant to give evidence for that reason, but that does not mean that I can infer she would have supported AXA’s case on whether a binding agreement was reached – certainly where I have found that Mr Rember himself did not have that understanding.

[251]I therefore decline to draw any adverse inference from her absence, although the failure to provide any explanation for the decision not to call her does not reflect well on Santander.

[252]In these circumstances, the precise reason why the Settlement Agreement was not signed by Santander is irrelevant. It did not necessarily have to go back to ExCo and if it had been signed in October 2015 it would have been fully binding. It was sheer bad luck from AXA’s point of view that Mr Pateman left when he did and that his replacement decided that a review was necessary. It can well be said that this was opportunistic on Santander’s part. I cannot accept that the ostensible reason given for the refusal to sign, namely concerns about the allegedly flawed design of the Budget Accounts (for which GECB itself was responsible) was genuine. Nonetheless, for as long as the agreement remained unsigned, Santander was at liberty to withdraw for good reason, bad reason or no reason at all and this no doubt presented itself as a convenient peg on which to hang its change of heart. J.2: Agreement by subsequent exchange of emails

[253]So far as concerns AXA’s alternative case based on the exchange of emails following the meeting, this cannot have led to a binding agreement if none had been reached at the meeting (which is the necessary premise of the argument). Mr Rember’s email did no more than request confirmation of the substance of what had been agreed and, as I put to Mr Campbell KC in argument, recording a non-binding agreement cannot make it binding even when coupled with confirmation that the record was accurate. In any event, Ms Day would clearly not have had authority on her own to commit Santander to an agreement of this nature if it had not already been concluded and no-one could reasonably have supposed that she did.

[254]AXA is undoubtedly entitled to feel ill-used but, as I have already made clear, abstract notions of justice and equity are no substitute for legal merit. This head of claim accordingly fails. K: THE INDEMNITY CLAIM

[255]The relevant provisions of the Agency Agreement are set out in the Annex to this judgment.

[256]Seven issues fall to be addressed:(a) The construction of clause 12.2 of the Agency Agreement, specifically whether it applies to losses arising from sales concluded before the Effective Date (i.e., 1 December 2000);

(b) Whether the losses claimed are “liabilities” within the meaning of clause 12.2;

(c) Whether the pre-condition requiring compliance by FICL/FACL with their duties under the Agency Agreement is satisfied;

(d) Causation;

(e) Whether a claim under clause 12.2 is precluded because GECB’s conduct was authorised by FICL/FACL;

(f) Limitation;

(g) Identity of the correct defendant following the novation between SCL and SISUK.

K.1: Pre-agreement sales

[257]The Agency Agreement formalised a relationship that had been ongoing for many years. This is clear from clauses 1.1 and 1.2, which I set out again for ease of reference:

“1.1 GE-CB has prior to the date of this agreement acted as the Insurers' agent in respect of the marketing and sale of the Insurance. The parties now wish to record the terms and conditions on which GE-CB shall continue to act as the Insurers' agent. 1.2 Unless otherwise stated herein, the parties agree that, notwithstanding clause 3, this agreement shall apply to and govern the marketing and sale of the Insurance under all of the Schemes (including the Schemes set out in schedule 4 in respect of which there are in existence at the date of this agreement Existing On-Risk Policies), the ongoing administration of all Existing On-Risk Policies and New Policies entered into between insured customers and the Insurers pursuant to such Schemes and the parties respective rights and obligations in respect thereof.”

[258]The agreement itself had been under negotiation since at least 1996 in substantially similar form to its eventual incarnation. None of the evidence suggested that it was intended to effect any material change in the pre-existing relationship. As such, it contained regimes for dealing with both(i) existing policies under existing retailer schemes; and

(ii) new policies under either existing or new schemes: see the definitions of “Existing On-Risk Policy” and “New Policy”.

[259]It is common ground that in relation to premium/cost allocation, administration and claims handling, the Agency Agreement drew no distinction between Existing Policies and New Policies. They were all treated the same no matter when they were sold. Nonetheless, Santander argues that for the purposes of the indemnity provisions in clause 12, a distinction between them falls to be made even though it could not point to any other clause of the agreement where such a distinction was drawn. AXA points out that, if correct, this would mean that from 1 December 2000 Santander continued to receive around 95% of any premium paid (after payment of claims) in respect of pre-agreement policies but was not to be liable for any mis-selling of the policies even if due to its own acts or omissions. Mr Green submits that such a startling asymmetry of risk and reward would require a very clear intention, particularly where one of the principal purposes of PPI in the first place was to protect Santander’s own credit risk.

[260]In this case I have reached the clear conclusion that the objective intention of the parties was that the entire Agency Agreement, including clause 12, should apply without distinction to all policies sold by GECB on behalf of FICL/FACL irrespective of the date of sale.

[261]The aim of the agreement was clearly to effect continuity in the relationship between the parties rather than to bring about any bright-line change. This is the natural inference from clause 1.2 which is underlined by the fact that clause 1.2 expressly applies “notwithstanding clause 3”, i.e., irrespective of the date on which the Agency Agreement entered into force.

[262]There was, however, a lively debate as to the correct construction of clause 1.2. On AXA’s construction, the Agency Agreement applied to:(a) The marketing and sale of insurance “under all of the Schemes” including those where there were Existing Policies;

(b) The ongoing administration of all Existing Policies and New Policies;

[263]Mr Green pointed out that “Schemes” under the agreement was defined to include existing schemes as well as new schemes. The wording was thus apt to express an intention that the provisions of the agreement should apply to the marketing and sale of any policy whenever sold. He further submitted that the agreement could have referred to “ongoing marketing and sale” or to “marketing and sale of New Policies” if it had been intended to exclude the marketing and sale of Existing Policies. Santander’s riposte was that could equally have said “including Existing Policies”. It is not uncommon for both sides to deploy arguments to the effect that the rival construction could have been expressly spelled out if it had been intended. Such arguments take the matter no further, particularly where the likelihood is that no-one addressed their minds to this particular point. The court’s task is to discern the parties’ objective intention from what they have actually said, not from what they could or might have said, which can only ever be speculative.

[264]In its written closing submissions, Santander’s case was that clause 1.2 applied to:(a) The marketing and sale of insurance after 1 December 2000 under all schemes irrespective of the date of the scheme;

(b) The ongoing administration of all Existing Policies and New Policies;

[265]In oral closings, however, Mr Zellick advanced an entirely new construction, namely that the clause applied to:(a) The marketing and sale of insurance after 1 December 2000 under all schemes irrespective of the date of the scheme;

(b) The ongoing administration of all Existing Policies;

[266]It had never occurred to anyone to read the clause in this way until virtually the dying minutes of the case. Nonetheless, Mr Zellick resolutely argued that this was the correct construction to put on the agreement and the distinction it drew between Existing Policies and New Policies indicated an overarching intention for the agreement only to apply to future conduct. The historic marketing and sale of Existing Policies was thereby excluded from its purview.

[267]I suppose this is a possible construction, but it does not seem to me at all to accord with the objective intention of the parties. The effect would be that Existing Policies and New Policies were treated together in relation to the commercial aspects of the agency (premium, commission and claims handling etc.) but not for the purposes of the indemnity provisions. This is a most unlikely intention viewed objectively. The suggested construction is also tautologous, because if the agreement applies to New Policies under head (c), it necessarily applies to the marketing and sale of those policies, making head (a) redundant to that extent.

[268]Despite Mr Zellick’s submissions to the contrary, the words “notwithstanding clause 3” clearly demonstrate to my mind that the operation of the agreement was not tied to the Effective Date, which simply set the date at which the provisions of the agreement were to start applying.

[269]Moreover, as I suggested to Mr Zellick, his construction would create a logistical nightmare in terms of reporting. According to Mrs Greenaway’s evidence, the system of bulk reporting adopted throughout the relationship did not distinguish between newly-sold policies and pre-existing policies. It simply recorded the total number of active policies under each separate scheme. There was therefore no way of identifying New Policies for the purposes of treating them differently unless GECB were to adopt a whole new reporting system and there is no evidence that the parties envisaged any such change.

[270]I therefore reject Santander’s proposed construction of clause 1.2 in favour of AXA’s interpretation.

[271]The idea of continuity was further supported by the terms of clause 4.1 where the appointment of GECB as agent was “confirmed” by FICL/FACL, and also by the factual evidence of Mrs Greenaway, Mr Halse and Mr Brandon-Cross, all of whom confirmed that there had been no intention to alter the nature of the relationship between the parties. In other words it was very much “business as usual”.

[272]The background and genesis of the Agency Agreement was to achieve a formalisation of arrangements which were already being operated. That being the case, one would expect it to be clearly stated (particularly given the long gestation period of the final agreement) if it had nevertheless been intended to draw a distinction for any specific purpose between policies sold after the Effective Date and those sold before. Further, I can see no logical or commercial reason why the parties would have intended GECB to be liable for the consequences of its acts and omissions in relation to New Policies but not Existing Policies, particularly when the Agency Agreement was concluded at a time when (to the knowledge of both parties) the impending FSMA regime did not distinguish between future and historic sales.

[273]I entirely accept that there was no express indemnity provision governing the prior relationship between the parties and that none could be implied. To that extent clause 12 was a “new” provision. But the evidence of both the factual and expert witnesses was that agency agreements between insurers and credit lenders invariably contained an indemnity agreement of this nature and that would have formed part of the factual matrix which was known to both parties. It is also true that all but one of the draft agreements before the court contained a materially identical clause. But it is trite law that reliance cannot be placed on drafts or negotiations in order to construe a contract and I leave that particular consideration out of account.

[274]Mr Zellick also prayed in aid the fact that no-one at the time anticipated the deluge of claims that would subsequently materialise and that it was therefore unlikely that anyone thought an indemnity was required. That may be so, but it is another point which cuts both ways. If no claims were anticipated, there would be no reason not to give an indemnity. He also attempted to derive support from GECB’s refusal to give an indemnity in relation to the Account Cover issue (see paragraph ‎167 above). However, that involved refusal of a specific indemnity in the context of a specific problem, and in any event it is incontrovertible that the Agency Agreement did contemplate mutual indemnities, whatever GE’s “normal” policy might have been.

[275]A more powerful argument on behalf of Santander was based on the wording of clause 12.2 itself, namely that it only applied to acts or omissions of GECB while performing its duties “under this agreement”. Mr Zellick submitted that GECB only had duties under the agreement on and after the Effective Date. By definition, therefore, acts and omissions prior to the Effective Date fell outside the scope of the indemnity.

[276]I am not persuaded by this argument. As I have concluded above, the clear intention of clauses1.1 and

1.2 was that from the Effective Date the regime set out in the Agency Agreement should be applied to the entirety of the parties’ relationship, past and future. It is not without significance that these were bespoke clauses specifically negotiated in the context of a particular longstanding relationship, and there is nothing heterodox in the idea that the duties set out in the agreement were duties which were contractually agreed to be applicable to the entirety of that relationship. The same, obviously, would be true of FICL/FACL’s duties under the agreement.

[277]There is no question of any conflict with the entire agreement provision in clause 17. On the contrary, clause 17 merely confirms that the Agency Agreement regime applies from the Effective Date.

[278]Clause 12.2 accordingly covers liabilities incurred by reason of GECB performing the duties described in the Agency Agreement as agent for FICL/FACL whenever they were performed. In this way the indemnity was tied to GECB’s activities as agent in respect of the schemes covered by the agreement. The restriction thus relates to the scope of the agency rather than being temporal. Any other conclusion would lead to a situation where there was potentially no indemnity for the mis-selling of Existing Policies sold prior to the Effective Date, but acts and omissions after that date could incur liabilities. That would be contrary to what I find to have been the objective intention of the parties to draw no distinction between any of the policies for any purpose. It would also allow an effectively standard form indemnity provision to override the clearly bespoke provisions in clauses 1.1 and 1.2.

[279]If Santander were right that the mis-selling of policies prior to the Effective Date was excluded from the scope of the Agency Agreement, I agree with AXA that it would lead to a surprising asymmetry, with the majority of the commercial benefits going to GECB and FICL/FACL picking up all of the liabilities even if attributable to GECB’s acts and omissions. In my view, much clearer wording would have been required to achieve this effect.

[280]Santander made the forensic point that if the lack of an indemnity had been thought to be a problem, then it is surprising that it took over 20 years to rectify the position. However, since it is common ground that mis-selling claims only began to take off in the late 1990s, this point has limited weight. It cuts both ways anyway. If the agreement is to be construed as Santander suggests, it is equally surprising that it was not introduced sooner.

[281]Santander’s final point was that if AXA’s construction were correct, it made no sense to limit the Agency Agreement to existing schemes while ignoring schemes which had been discontinued. However, the submission is factually flawed. Mrs Greenaway’s evidence confirmed that the only schemes which were regarded as “discontinued” for the purposes of the Agency Agreement were those under which no policies had ever been sold. Schemes under which policies had been sold which could still be reactivated or give rise to claims despite being dormant for years were not treated as “discontinued” but were regarded as existing schemes covered by the Agency Agreement. Thus Appendix 4 to the agreement included several schemes under which no premium had been collected since 1995 and under which no new policies could have been sold after the Effective Date. This also provides some support for the view that the Agency Agreement must have been intended to apply to historic marketing and sale under those schemes. K.2: Liabilities

[282]The sums claimed by AXA fall into four categories:(a) £451.9 million in redress payments made to customers either directly or indirectly via SUK under the Claims Handling Agreement;

(b) £70.6 million in respect of fees paid to the FOS in respect of complaints referred to it by disgruntled customers;

(d) £24 million paid to the Official Receiver by way of global settlement of putative claims on behalf of bankrupt estates in respect of mis-sold PPI prior to 2005 (the “OR settlement”).

[283]Santander argues that none of these payments represents a “liability” for the purposes of clause 12.2 of the Agency Agreement. It submits that “liability” in this context means a civil liability incurred to a third party which(a) could be established against FICL/FACL For simplicity, I shall continue to refer to FICL/FACL notwithstanding that with the passage of time, the relevant liability may have been that of Genworth or AXA. in the civil courts at the suit of the third party and

(b) is of a type and extent that would have been within the contemplation of the parties at the date of the agreement. K.2.1: Redress payments and FOS fees

[284]I consider these arguments first in relation to the redress payments and FOS fees.

[285]As to (a), the meaning of “liability” is a question of construction which depends on the particular facts of the case. Mr Zellick relied on certain comments of the Court of Appeal in Smit Tak Offshore Services Ltd v Youell, [1992] 1 Lloyd’s Rep. 154, 159 but those were concerned with a very different situation, namely whether a letter from the Dubai Department of Ports and Customs instructing the claimant to remove a wreck on pain of non-renewal of a licence to operate in Dubai waters created a liability for the purposes of a liability insurance policy. Unsurprisingly, the court held that it did not, although Mustill LJ stated that if the letter had been cast in the form of an instruction from the Ruler of Dubai himself – he being the source of all law – the result would have been different as the letter would then have created a legal obligation to raise the wreck.

[286]Reliance was also placed on Conister Trust Ltd v John Harman & Co.,

[2008] EWCA Civ. 841

but that concerned liability under an agreement which was altogether unenforceable under the Consumer Credit Act 1974. That again was an entirely different situation as no-one in this case suggests that FICL/FACL did not have valid obligations of some description to pay these sums. The question is “to whom”? Indeed at paragraph 87 of that case, Lewison LJ expressly accepted that “liability” does not necessarily imply enforceability, although it did in the particular context before him.

[287]Thus, neither of these authorities unequivocally supports Santander’s construction that the liability must be directly enforceable by the third party and I decline so to hold. It seems to me that the word is apt to include any liability or binding obligation which could have been enforced against FICL/FACL irrespective of by whom or on what basis, be it tort, contract, breach of statutory duty or a purely regulatory liability. In the context of an agency agreement between parties specifically relating to regulated activities, I can see no reason to exclude regulatory liabilities even if they could not necessarily have been enforced directly by the consumer against FICL/FACL in a civil action. I do, however, accept that the liability must be one which is imposed on FICL/FACL rather than one which only exists as a result of having been voluntarily assumed.

[288]In respect of the redress payments, the relevant regulatory provisions which were said by AXA to create the relevant liability were set out in DISP as follows:

“DISP 1.4 Complaints resolution rules DISP 1.4.1 R Once a complaint has been received by a respondent, it must: (1) investigate the complaint competently, diligently and impartially; (2) assess fairly, consistently and promptly: (a) the subject matter of the complaint; (b) whether the complaint should be upheld; (c) what remedial action or redress (or both) may be appropriate; (d) if appropriate, whether it has reasonable grounds to be satisfied that another respondent may be solely or jointly responsible for the matter alleged in the complaint; taking into account all relevant factors; (3) offer redress or remedial action when it decides this is appropriate; … (4) comply promptly with any offer of remedial action or redress accepted by the complainant. DISP 1.4.1 R Once a complaint has been received by a respondent, it must: (a) the subject matter of the complaint; (b) whether the complaint should be upheld; (c) what remedial action or redress (or both) may be appropriate; (d) if appropriate, whether it has reasonable grounds to be satisfied that another respondent may be solely or jointly responsible for the matter alleged in the complaint; … DISP 1.4.4 R 317. Where a complaint against a respondent is referred to the Financial Ombudsman Service, the respondent must cooperate fully with the Financial Ombudsman Service and comply promptly with any settlements or awards made by it.”

[289]These provisions had the status of delegated legislation under section 137A of FSMA and were binding on FICL/FACL. They were also rules, rather than guidance, contravention of which could be sanctioned by enforcement action. In a Policy Statement dated August 2010, the FSA expressed the clear view that breaches of the FSA Principles gave rise to liabilities and that the complaints handling rules accordingly required a firm to uphold a complaint and pay appropriate redress when the conduct in question was in breach of the Principles.

[290]FICL/FACL was thus mandatorily required under the DISP rules to offer redress where appropriate and to comply promptly when the offer was accepted. By virtue of section 138D(2) of FSMA, breach of the complaints handling rules was actionable at the suit of a consumer who suffered loss as a result of the contravention.

[291]Santander nonetheless submits that even in these circumstances, FICL/FACL was under no liability to pay redress which could have been enforced by civil action. At most, a consumer could have brought a claim on the basis that FICL/FACL had not handled the claim in accordance with the DISP rules. Admittedly they could recover damages for that breach but it would not have been a claim for redress as such.

[292]However, while Mr Zellick’s analysis of the consumer’s right of action is undoubtedly correct, I am unable to accept that FICL/FACL were under no liability to pay the redress in question. On the contrary, they were under a binding regulatory obligation, which had statutory force, to pay such redress as was appropriate under the relevant rules. The sausage which came out of the regulatory machine was accordingly a liability to pay redress and if FICL/FACL had failed to do so, not only could the regulator have taken enforcement action against them, but the customer had a civil route by which it could secure payment, even if only indirectly by way of damages.

[293]In my judgment, this is a sufficient liability for the purposes of the indemnity under clause 12.2 of the Agency Agreement, although not necessarily for other purposes.

[294]Likewise in relation to complaints referred to the FOS, FICL/FACL were under a binding regulatory obligation to comply with any FOS award. Under FSMA, FOS awards could be enforced as if they were orders of the court and, if accepted by the complainant, gave rise to a res judicata precluding further proceedings. Santander nonetheless submitted that, with one exception, none of the complaints referred to the FOS resulted in a binding award because FICL/FACL chose to pay on the basis of a provisional decision rather than a final award. Accordingly, even on the basis of the analysis above, there was no liability on the part of FICL/FACL to make payment. In my judgment, this is a distinction without a difference. The decision of the FOS may have been provisional, but it was still a decision which stood unless and until challenged by either party. I am therefore satisfied that payment pursuant to such a decision was also a liability for the purposes of the indemnity.

[295]As regards the FOS fees, the unchallenged evidence of Mr Doherty was that FICL/FACL had a legal obligation to pay fees to the FOS in respect of all complaints referred, whether ultimately upheld or not. Such fees could have been recovered by the FOS in legal proceedings and therefore fall squarely within the meaning of “liability” in clause 12.2.

[296]As regards limb (b) of Santander’s submissions, the indemnity applies to any act or omission, not just acts which amount to a breach of contract, and there is nothing to indicate that the parties intended to incorporate contractual rules of remoteness applicable to claims for damages. Indeed, it could be said that the whole point of an indemnity is to provide protection against consequences which are unforeseeable as pointed out by Simon Birt KC in Rhine Shipping DMCC v Vitol SA,

[2023] EWHC 1265 (Comm)

at [219]. The situation is very different from that in BCCI v Ali (No. 1),

[2001] UKHL 8

;

[2002] 1 AC 251

, where the court was concerned with a release of claims in a settlement agreement. In that situation, it might well be the case that the parties did not intend to release unforeseeable future claims.

[297]Accordingly, I do not see that it makes any difference in principle that at the date of the Agency Agreement, there was no regulatory route to redress such as existed later and that at that date customers could only have enforced any claims by conventional means in the ordinary courts. In my judgment, the objective aim and purpose of clause 12.2 was to provide FICL/FACL with an indemnity if it got into regulatory hot water because of something that GECB had done.

[298]The question therefore resolves into one of causation, as to which I can see no reason not to adopt the usual approach, namely that GECB’s conduct must have been the proximate or effective cause of the liability in question. On that basis there can be no doubt that the redress payments and FOS fees are liabilities which were proximately and effectively caused by the mis-selling. In any event, in the context of the relationship between the parties, they were also precisely the types of regulatory consequence which would have been in contemplation at the date of the Agency Agreement, even if the precise regulatory machinery could not itself have been foreseen. K.2.2: Administrative costs

[299]The administrative costs incurred by Genworth/AXA relate largely to the setting up of a dedicated call centre exclusively to handle PPI complaints, of which some 99.89% related to policies sold by GECB. 92% of the costs were payroll costs, while the remainder represented payments to third party suppliers for printing and postage and other services required to facilitate the operation of the call centre.

[300]Clearly Genworth/AXA had a contractual obligation to pay its staff and suppliers which could have been enforced in the civil courts. However, in my judgment this is not a liability which falls within the ambit of clause 12.2. Genworth was under no obligation to anybody to set up a separate call centre to handle claims; its decision to do so was entirely voluntary and made (no doubt for very good reason) for its own commercial convenience. As such, these costs do not in my judgment constitute a liability covered by clause 12.2.

[301]Furthermore, while the costs could be said to have been indirectly caused by mis-selling in the sense that they would not have been incurred “but for” that mis-selling, this was purely indirect consequential loss and, as such, cannot be said to have been proximately caused by the mis-selling. K.2.3: The OR Settlement

[302]None of the bankrupt estates comprised within the settlement had actually made complaints in relation to mis-selling, let alone complaints which were successful. The settlement was rather concluded by Genworth/AXA to avoid such complaints being made.

[303]By analogy with reinsurance cases such as Enterprise Oil Ltd v Strand Insurance Co. Ltd,

[2006] EWHC 58 (Comm)

; [2006] 1 Lloyd’s Rep. 186 and AstraZeneca Insurance Co. Ltd v XL Insurance (Bermuda) Ltd,

[2013] EWHC 349 (Comm)

; [2013] Lloyd’s Rep. IR 290, an indemnity against liability (of whatever nature) requires the claimant to establish that it was under an actual subsisting liability. In this case, there may or may not have been such a liability and I fail to see how the mere entering into of a voluntary settlement of claims which are feared, but which have never been advanced, can have created one. Put a different way, a settlement concluded specifically in order to avoid a regulatory liability can hardly be said to establish that liability. At best, the settlement could be said to represent costs incurred to avoid a liability, but in the absence of some doctrine akin to that of sue and labour in marine insurance, such costs are not recoverable under an indemnity: see Yorkshire Water Services Ltd v Sun Alliance & London Insurance plc, [1997] 2 Lloyd’s Rep. 21. I would in any event have held that any liability under the settlement agreement was not proximately caused by the alleged mis-selling.

[304]On that basis AXA’s claim for an indemnity cannot succeed in relation to the administrative costs or the OR Settlement. K.3: Pre-condition: compliance with FICL/FACL’s duties

[305]There was a dispute between the parties as to which of them bore the burden of proof in respect of FICL/FACL’s compliance with their duties under the Agency Agreement. Relying on Sofi v Prudential Assurance Co. Ltd, [1993] 2 Lloyd’s Rep. 559 and Paine v SJO Catlines,

[2004] EWHC 3054 (TCC)

, AXA contended that the burden was on Santander. Santander referred in turn to Braithwaite v Thomas Cook Travellers Cheques Ltd,

[1989] 1 All ER 235

which it said demonstrated the contrary. In truth, none of these cases provided unequivocal support for either proposition but, since I have not found it necessary to resort to the burden of proof, I will not lengthen this judgment further by addressing the point in detail. Suffice it to say that I would have been inclined to hold that Santander bore an evidential burden of showing that the pre-condition had not been complied with, which it would then have been for AXA to rebut as part of its legal burden of proof.

[306]The starting point is the meaning of by the words “complying with their duties under this agreement”. Clearly the clause cannot be read literally and indeed Santander accepted that not every breach by FICL/FACL of an obligation under the Agency Agreement would bar the indemnity, for example, a failure to provide monthly reports within 10 days of the month end.

[307]In the event, I have concluded that the most sensible construction of the pre-condition, which is likely to have been objectively intended, is to underline the necessity for an effective causal link between FICL/FACL’s breach of duty and GECB’s conduct. In other words, if GECB’s acts and omissions were themselves effectively or proximately caused by FICL/FACL’s failure to comply with their duties, any resulting liability would fall outside the scope of the indemnity. 308. As ultimately articulated in closing, Santander’s case was that FICL/FACL were in breach of:(a) clauses 5 and/or 7.1(d) and (f) of the Agency Agreement and their overriding responsibilities in having: (i) stood by and allowed GECB to mis-sell policies without taking any steps to ensure that they did so compliantly; and (ii) required GECB to market policies (in particular the Budget Account policies) which were considered by the FOS to have been so defective and/or to have offered such poor value for money that they should not have been offered to sale to customers at all;

(b) clause 7.1(a) in having failed to provide GECB with any or any adequate market research, marketing or technical assistance;

(c) clause 7.1(b) in having failed to provide GECB with any or any adequate specialised training in relation to the insurance and its marketing and sale.

[309]FICL/FACL’s obligation under the ABI Code both as regards their own compliance and that of GECB was to exercise best endeavours. I have found in paragraphs ‎188-‎223 above that there was no failure by FICL/FACL to exercise best endeavours and it follows that there was no non-compliance with their duties under the Agency Agreement in this respect. Even if it could be said that FICL/FACL were automatically in breach of the ABI Code by virtue of defaults by GECB, that would have been attributable to GECB, which was under its own independent obligation of compliance by virtue of clause 6.1(g).

[310]There is therefore no basis for saying that FICL/FACL failed to comply with the pre-condition in clause 12.2 and Santander’s defence to the indemnity claim on this ground fails. K.4: Causation

[311]The question here is whether the categories of loss claimed are liabilities which arise “by reason of” GECB’s acts or omissions while performing their duties.

[312]On the basis that (as I have held) the Agency Agreement applies to the marketing and sale of existing policies, it cannot sensibly be disputed that mis-selling of those policies by GECB amounted to conduct while performing its duties under the agreement, whether those duties are limited to the specific duties set out in clause 6 or regarded more broadly as the duties attendant on GECB as an agent. Indeed Santander did not argue the contrary.

[313]I have already addressed the required causal link in paragraph ‎298 above. For the reasons already given, I am satisfied that FICL/FACL’s liability for redress payments and FOS fees arose by reason of GECB’s mis-selling and are therefore covered by the indemnity. By contrast, the administrative costs and sums paid under the OR settlement cannot, even if they can properly be regarded as “liabilities” for this purpose. For the avoidance of doubt, the basis of my decision on this head of claim is that GECB is liable for all the regulatory consequences of its mis-selling and is therefore liable for all redress and FOS fees paid by AXA under the regulatory regime, irrespective of whether individual complaints would have succeeded in a court of law. K.5: Authorisation

[314]Santander argued that the manner in which the policies had been sold had been authorised by FICL/FACL and that GECB was therefore not itself in breach of the Agency Agreement.

[315]I struggle to see the relevance of whether GECB was in breach of the agreement or not. The indemnity is expressly not limited to negligence or breach of contract but is triggered by any act or omission, culpable or otherwise. More broadly, however, Santander’s submission was that FICL/FACL cannot have incurred any liability for GECB’s acts committed outside the permitted scope of the agency so that it must be assumed that the conduct in question was within the scope of the agency and had thus been authorised. In so far as this was intended to suggest that GECB’s conduct was expressly authorised, that would be tantamount to an allegation of deliberate misconduct. However, the case is not pleaded in that way and I have in any event accepted that FICL/FACL were aware only of the general process by which PPI was sold and not that systemic mis-selling was taking place. In my judgment, it would make a nonsense of the indemnity to hold that it did not apply to the carrying out of authorised activities (the sale of PPI) in an unauthorised manner (by mis-selling).

[316]Accordingly, I reject this ground of defence. K.6: Limitation

DISP 1.4.4 R

[317]The dispute in relation to limitation concerns the hoary old chestnut of whether the cause of action for an indemnity accrues on the date of the transaction under which the liability is incurred (here the date of sale) or only when the liability crystallises.

[318]In support of its argument that the correct date is the date of sale, Santander relied on the well-known case of Bosma v Larsen, [1966] 1 Lloyd’s Rep. 22. That was a shipping case where there was an express indemnity against “all consequences or liabilities arising from the Master … signing Bills of Lading or other documents or other complying with [the Charterers’] orders.” McNair J held that the obligation to indemnify against “all liabilities” was to provide indemnity against the incurring of the liability, not its discharge. He did not expressly consider the indemnity against “all consequences”. It should also be noted that this particular indemnity was given in the specific context of a charterparty where signature of the bill of lading creates a contractual nexus with anyone to whom the bill is indorsed.

[319]The preponderance of modern authority has sought to distinguish Bosma and confine it to its specific facts: see, for example, The Caroline P,

[1985] 1 WLR 553

(a case involving the implied general indemnity under clause 8 of the NYPE form of the charter) where Neill J held that in the case of an express indemnity the accrual of the cause of action was a question of construction. He thereby drew a distinction between the express indemnity against liabilities and a “general indemnity”, no doubt in an attempt to distance himself from Bosma. However, there are difficulties with that distinction which are well described in the summary of authority by Picken J in Cape Distribution Ltd v Cape Intermedia Holdings plc,

[2016] EWHC 1796 (QB)

[85]-[92] where the judge held that the critical question is not so much the subject matter of the indemnity as its proper construction in the context in which it was given: see Chitty on Contracts (35th ed.) (2023) (Sweet & Maxwell) para. 32-049. It follows that no rigid meanings can be ascribed to any particular form of words.

[320]The general modern rule, referred to by Chitty,is that time in relation to an indemnity against liability starts to run when the relevant liability is established and ascertained by judgment, award or binding settlement. In this case, as at the date of sale, any liabilities arising from mis-selling were purely contingent and not such as to cause any measurable loss or immediate impairment of FICL/FACL’s assets. They were not established or ascertained until such time as payment was made following a complaint. I see nothing in the wording of this particular indemnity to suggest that the general rule should not apply.

[321]The position might have been different if the incurring of a contingent liability had immediately diminished the value of FICL/FACL’s assets or rights – the so-called “impaired asset” line of authority reflected in cases such as Forster v Outred & Co.,

[1982] 1 WLR 86

and Co-operative Group Ltd v Birse Developments Ltd,

[2014] EWHC 530 (TCC)

. In this type of case, the indemnifier’s actions have an immediate impact on the value of the claimant’s assets or some other property, such as an equity of redemption. It can therefore properly be said that the cause of action arises at that point, because there is an actual loss which is capable of measurement. Here, however, until such time as a customer made a complaint which was either upheld or agreed, any liability was inchoate and there would have been nothing to indemnify.

[322]Appeals to the commercial consequences of one construction or another do not take the matter any further. Limitation inevitably draws an arbitrary line and it will always be possible to posit circumstances in which it works harshly against one party or the other. But that is an inherent feature of all arbitrary lines.

[323]I accordingly hold that the indemnity under clause 12.2 was triggered when FICL/FACL’s liability was established and ascertained, which for practical purposes can be taken as being the date of payment. Since it is common ground that this was all within six years of the standstill agreement, it follows that the indemnity claim – in so far as it succeeds on the facts – is not time barred. K.7: Novation and the identity of the correct defendant

[324]It is Santander’s case that the indemnity claim lies only against SCL. This depends on the construction of the novation agreement between SCL and SISUK dated 22 January 2010, whereby certain rights and obligations of SCL were novated to SISUK with effect from 1 January 2010. AXA contends that this is a matter of some importance because SISUK’s assets are greater than those of SCL and Santander has expressly declined to confirm that it will stand behind SCL should the claim against it be otherwise successful.

[325]The relevant provisions of the novation agreement were the following: “1.1 The parties to this Novation Agreement agree that as from 1 January 2010 (the "Effective Date"):1.1.1 SISUK shall have the benefit of all rights under the UK Agreement as if SISUK had executed the [Agency] Agreement instead of SCL;

1.1.2 SISUK shall perform all of the obligations of SCL under the [Agency] Agreement; and

1.1.3 Genworth accepts the assumption of liability by SISUK in place of SCL in respect of the [Agency]Agreement. 1.2 SISUK and Genworth hereby acknowledge and agree that each shall have no liability to the other in respect of any duties, obligations, causes of action, claims or liabilities whatsoever pursuant to or in connection with the [Agency] Agreement arising prior to the Effective Date or in connection with the Ireland Agreement. 1.3 SCL and Genworth hereby acknowledge and agree that each shall have no liability to the other in respect of any duties, obligations, causes of action, claims and liabilities whatsoever pursuant to or in connection with the [Agency] Agreement arising after the Effective Date. 1.4 This Novation Agreement shall be without prejudice to any accrued rights of SCL or Genworth arising prior to the Effective Date or under the Ireland Agreement.” 1.1.1 SISUK shall have the benefit of all rights under the UK Agreement as if SISUK had executed the [Agency] Agreement instead of SCL; 1.1.2 SISUK shall perform all of the obligations of SCL under the [Agency] Agreement; and 1.1.3 Genworth accepts the assumption of liability by SISUK in place of SCL in respect of the [Agency]Agreement.

[326]Had clause1.1 stood alone, it would have had the effect that SISUK not only undertook to perform all of SCL’s obligations under the Agency Agreement from 1 January 2010 but also agreed to assume any liability of SCL in respect of the agreement, irrespective of when or how that liability arose. However, clause 1.1 was qualified and limited by the provisions of clauses

1.2 and

1.3 which provided that SISUK and Genworth should have no liability to each other in respect of duties, obligations, causes of action, claims or liabilities arising before 1 January 2010, while SCL and Genworth should have no liability to each other in respect of such matters arising after that date. For this purpose, it seems tolerably clear that the word “liabilities” does not refer to the liability incurred by FICL/FACL in respect of which indemnity is sought, but rather to GECB’s liability to indemnify.

[327]There was some debate at the hearing about how these provisions should be construed given that a duty or obligation could arise before the relevant date while the cause of action to which it gave rise might only arise later. For example a breach of duty might have occurred in June 2009 but loss might not have been suffered until February 2010. This gave rise to the distinctly unattractive possibility that a particular set of circumstances might potentially be covered by both clauses or by neither.

[328]In my judgment, the straightforward approach is the most satisfactory and the objective intention to be discerned from these clauses is that where FICL/FACL could have called upon GECB to perform a duty or obligation under the Agency Agreement or if they could have issued proceedings against GECB or required it to satisfy a liability before 1 January 2010, then liability should stay with SCL. If only after 1 January 2010, then liability passed to SISUK. On this construction, 1 January 2010 itself is rather left in limbo, but it must be hoped that this does not give rise to any practical difficulty. If it does, and the parties cannot reach an amicable solutionthe court will have to deal with it.

[329]As I have found above, AXA’s cause of action for an indemnity against GECB accrued when it was called upon to pay. Since it is common ground that this was within six years of the standstill agreement, it must necessarily have been after 1 January 2010 and liability was accordingly novated to SISUK. This would also be consistent with the provision in clause 1.4 for rights accrued before 1 January 2010 to remain where they were. The fact that any mis-selling arose because of GECB’s breaches of duty prior to 2010 is irrelevant. The mis-selling had already long-since occurred and there would no longer have been any question of calling upon GECB to perform its duties in that respect at the date of the novation. K.8: Conclusion on the indemnity claim

[330]I therefore conclude that AXA has a valid claim for an indemnity against SISUK in respect of the redress payments and FOS fees but not otherwise. L: THE CONTRIBUTION CLAIM

[331]The next basis of claim asserted by AXA against Santander lies under the Contribution Act 1978, albeit this is maintained only in respect of the redress payments and the OR Settlement. AXA accepts that the FOS fees and administrative costs are not sums for which Santander was “liable” within the meaning of the Act.

[332]The Act provides as follows:

“1 Entitlement to contribution. (1) Subject to the following provisions of this section, any person liable in respect of any damage suffered by another person may recover contribution from any other person liable in respect of the same damage (whether jointly with him or otherwise). (2) A person shall be entitled to recover contribution by virtue of subsection (1) above notwithstanding that he has ceased to be liable in respect of the damage in question since the time when the damage occurred, provided that he was so liable immediately before he made or was ordered or agreed to make the payment in respect of which the contribution is sought. (3) A person shall be liable to make contribution by virtue of subsection (1) above notwithstanding that he has ceased to be liable in respect of the damage in question since the time when the damage occurred, unless he ceased to be liable by virtue of the expiry of a period of limitation or prescription which extinguished the right on which the claim against him in respect of the damage was based. (4) A person who has made or agreed to make any payment in bona fide settlement or compromise of any claim made against him in respect of any damage (including a payment into court which has been accepted) shall be entitled to recover contribution in accordance with this section without regard to whether or not he himself is or ever was liable in respect of the damage, provided, however, that he would have been liable assuming that the factual basis of the claim against him could be established. … (6) References in this section to a person’s liability in respect of any damage are references to any such liability which has been or could be established in an action brought against him in England and Wales by or on behalf of the person who suffered the damage; but it is immaterial whether any issue arising in any such action was or would be determined (in accordance with the rules of private international law) by reference to the law of a country outside England and Wales. 2 Assessment of contribution. (1) Subject to subsection (3) below, in any proceedings for contribution under section 1 above the amount of the contribution recoverable from any person shall be such as may be found by the court to be just and equitable having regard to the extent of that person’s responsibility for the damage in question. (2) Subject to subsection (3) below, the court shall have power in any such proceedings to exempt any person from liability to make contribution, or to direct that the contribution to be recovered from any person shall amount to a complete indemnity. … 6 Interpretation. (1) A person is liable in respect of any damage for the purposes of this Act if the person who suffered it (or anyone representing his estate or dependants) is entitled to recover compensation from him in respect of that damage (whatever the legal basis of his liability, whether tort, breach of contract, breach of trust or otherwise).”

[333]Five issues arise:(a) Whether AXA is “liable” for damage suffered by a third party;

(b) Whether Santander has any liability to the third party;

(d) The appropriate apportionment;

(e) Limitation.

L.1: AXA’s liability for damage

[334]Section 1(6) defines “damage” for the purposes of the Act as meaning any liability which has been or could have been established in an action against AXA by the person suffering the damage. Section 6(1) further clarifies that the basis of the liability is irrelevant. In Royal Brompton Hospital NHS Trust v Hammond,

[2002] UKHL 14

;

[2002] 1 WLR 1397

the House of Lords stated at [26] that “damage” was accordingly to be given a broad interpretation and affirmed the proposition that it meant little more than “a loss of some sort”.

[335]Nonetheless, the Act clearly requires some form of legal liability which could be established against AXA in the civil courts. I reject Santander’s submission that it must also be a “conventional” liability, which seems to me to be an impermissible gloss on the plain meaning of the statutory language.

[336]As far as the redress payments are concerned, I have already concluded that AXA was under a regulatory liability to pay redress by virtue of its obligations under DISP: see paragraphs ‎288-‎293 above. However, Santander argues that this is not a liability that could have been enforced as such in the civil courts by the complainants. The only claim that could be brought would be for default in the handling of the complaint whereas none of the claims by consumers was made on this basis.

[337]While I rejected this argument in the context of the indemnity claim, I have concluded that in this context, it is correct. Section 1(6) contemplates that the action brought by the person who suffered the damage must be one which has established or is capable of establishing the liability. However, the only civil claims that could have been asserted by consumers would have been for breach of rules 1.4.1 or 1.4.4 of DISP (set out at paragraphs ‎288 above) and, far from being in breach these rules, AXA/Genworth appear to have observed them meticulously. Likewise, it may be that AXA/Genworth were under a regulatory liability to pay redress or satisfy FOS awards, but that was because of an obligation owed to the regulator, not because they were under any liability to the consumer which the consumer could enforce directly. Moreover, the obligation under rule 1.4.1 is to pay redress when the insurer decides it is appropriate to do so. It is difficult to see how an action for breach of 1.4.1 can prescribe positively how that discretion should have been exercised.

[338]The same difficulty arises in relation to the obligation under rule 1.4.4 to satisfy FOS awards. While paragraph 16 of Schedule 17 to FSMA entitles the customer to enforce a FOS award as if it were payable under a county court judgment, this presupposes that liability has already been established by the award. Neither a claim for breach of rule 1.4.4 nor enforcement under paragraph 16 is itself capable of establishing any liability.

[339]These may seem somewhat technical distinctions because the ultimate damage suffered by the consumer and compensated by the redress payments is still the same (although I note that a similar distinction was drawn by AXA in the Genworth proceedings, albeit in the specific context of the SPA). AXA also complains that this means it is worse off as regards a contribution claim for having complied with its regulatory obligations than it would have been if it had ignored or disobeyed them. However, this is a statutory remedy and the statutory requirements must be observed. Where those requirements are clear, as they are, it is not for the courts to correct any perceived illogicality or injustice in their operation.

[340]In closing, Mr Green argued that a consumer could have claimed directly against Genworth/AXA on the basis of a liability in negligence or for breach of contract. However, AXA’s case was not pleaded or argued on that basis. The only liability asserted was a regulatory liability and in my judgment a claim for contribution on that basis cannot succeed.

[341]For the sake of completeness, if I am wrong in this conclusion, I would have rejected Santander’s submission that redress paid in accordance with a provisional decision of the FOS fell outside the scope of the Act because it was not an enforceable monetary award but only advisory. First, for the reason given in paragraph ‎294 above, namely that a provisional decision is more than merely advisory; and secondly, because I do not see why payment pursuant to a provisional decision would not have amounted to a bona fide settlement or compromise of a claim within section 1(4) of the Act. I agree with AXA that it would be perverse to adopt a construction of the Act which required the customer to pursue a complaint to the bitter end.

[342]As it is, however, I conclude that AXA is not under a relevant liability in respect of the redress payments for the purposes of the 1978 Act.

[343]Turning to the OR Settlement, AXA argued that this fell within section 1(4) of the Act. I do not agree. As discussed at paragraphs ‎302-‎304 above, the settlement did not compromise any claims because no claims had been made. It was essentially a payment made to forestall any such claims and it cannot be said that AXA was under any liability to the bankrupt estates save in so far as created by the settlement agreement itself.

[344]The claim for contribution does not therefore get off the ground in respect of either category of loss. In the circumstances, I need only deal very briefly with the remaining ingredients. L.2: Santander’s liability

[345]Assuming that my conclusion above is wrong and that AXA was under a liability to consumers in relation to the redress payments or the OR Settlement, it argues that Santander was liable for the same damage on any one of the following three bases:(a) Negligence;

(b) Negligent misstatement;

L.2.1: Negligence

[346]Mr Green submitted that in all the circumstances GECB assumed a duty of care to the customers to whom it sold PPI. I am unable to accept this submission. GECB had no direct nexus with the customers and certainly no contractual link. I am not persuaded that it was in a sufficiently proximate relationship with customers to say that it voluntarily assumed responsibility to them for the sale of the policy – at least in the absence of circumstances taking a particular case out of the norm. In these circumstances, it would not be fair, just or reasonable to impose a direct duty on GECB which ultimately only had a relationship with the customers in relation to PPI as agent for FICL/FACL and not in its own right. It is further arguable that the imposition of a generalised direct duty in this situation would cut across the contractual allocation of responsibilities as between FICL/FACL and GECB and as between GECB and the retail stores.

[347]AXA submitted that an analogy was to be drawn with insurance brokers but to my mind this is inapt. GECB was not acting as an independent broker but as a tied agent of FICL/FACL. It was therefore very firmly on the insurer’s side of the fence and I can see no justification whatsoever for any “incremental” extension of a duty of care to this situation. The fact that as between GECB and FICL/FACL, GECB was the principal economic beneficiary of the policies sold was irrelevant so far as the customer was concerned.

[348]I likewise regard it as irrelevant whether policies were sold on an advised or non-advised basis. Given the hard sell tactics employed, I accept that it is open to doubt whether these were genuinely non-advised sales. But even if they were advised, it makes no difference. Vis-à-vis the customer, GECB was still selling as agent for FICL/FACL.

[349]A more promising argument might have been that GECB was under a duty of care in its capacity as credit lender to ensure that the store card was sold in a compliant manner and that this duty extended to the sale of ancillary products. However, I did not hear argument on the point and it still does not meet the point that GECB only sold the ancillary PPI in an agency capacity. As it makes no difference to the ultimate result, I leave the point open.

[350]I accept that if a duty had existed, then on the basis of my findings above it is very likely that GECB was in breach on many occasions. However, in the context of a claim under the 1978 Act, it is not sufficient, as it is in the context of the indemnity claim, merely to point to systemic failures as having been the effective cause of the entirety of the redress payments. There is an express statutory requirement that GECB be liable for the same damage “suffered by another person” as AXA. AXA therefore needs to be able to identify actual damage suffered by a particular customer which necessarily requires analysis of each individual case to see whether it discloses a breach of duty and, if so, the extent of any damage caused as a result. This is in contrast to the claim for an indemnity or in negligence where the focus is on the loss or damage caused to AXA, where I accept that the loss can be looked at globally. L.2.2: Negligent misstatement

[351]Even if GECB was not under any general duty of care, AXA nonetheless argues that it was under a duty not to mislead customers but that it did so by impliedly representing in words or conduct that the policies had at least some benefit for the customer. AXA accepts that this argument cannot succeed in relation to customers who were not even told that they were buying PPI, but maintain it in relation to all other cases.

[352]The test for misrepresentation is what a reasonable person would have understood the words used to mean in their context. If the representation is implied, it is what a reasonable person would have inferred was being implicitly represented by the relevant conduct: IFE Fund SA v Goldman Sachs International,

[2006] EWHC 2887 (QB)

(Comm) at [50].

[353]We are not concerned for the purposes of this discussion with any express representation. That would obviously depend on the evidence in specific cases, for example where a customer was wrongly told that taking out PPI was required in order to be approved for credit. Considerations of this nature will be addressed in relation to the Sample Complaints but I am not asked in these proceedings to look at individual cases any more widely than that. Rather AXA argued that, by inviting a customer to buy PPI, GECB was necessarily impliedly representing in every case that the policy offered at least some benefit. In response, Mr Zellick submitted that there was no scope for an implied representation in a non-advised sale. I am by no means persuaded by this latter argument. Whether advised or non-advised, the test is as set out above and it is by no means impossible to envisage circumstances in which an implied representation could be made even in the context of a non-advised sale.

[354]Nonetheless, I reject the suggestion that the court can make any generalised finding of misrepresentation in particular categories of sale. The question of whether a misrepresentation was made, whether expressly or impliedly, is intensely fact-specific and cannot be determined on an indiscriminate global basis. In any event, the need to show specific damage to a specific customer applies just as much here as it does to a claim based on misrepresentation as it does to an argument that GECB was liable for straightforward negligence. L.2.3: Consumer Credit Act 1974

[355]The relevant provisions of the Consumer Credit Act are as follows:

“140A Unfair relationships between creditors and debtors (1) The court may make an order under section 140B in connection with a credit agreement if it determines that the relationship between the creditor and the debtor arising out of the agreement (or the agreement taken with any related agreement) is unfair to the debtor because of one or more of the following– (a) any of the terms of the agreement or of any related agreement; (b) the way in which the creditor has exercised or enforced any of his rights under the agreement or any related agreement; (c) any other thing done (or not done) by, or on behalf of, the creditor (either before or after the making of the agreement or any related agreement). (2) In deciding whether to make a determination under this section the court shall have regard to all matters it thinks relevant (including matters relating to the creditor and matters relating to the debtor). … (4) A determination may be made under this section in relation to a relationship notwithstanding that the relationship may have ended. … 140B Powers of court in relation to unfair relationships (1) An order under this section in connection with a credit agreement may do one or more of the following– (a) require the creditor, or any associate or former associate of his, to repay (in hole or in part) any sum paid by the debtor or by a surety by virtue of the agreement or any related agreement …; (2) An order under this section may be made in connection with a credit agreement only– (a) on an application made by the debtor or by a surety; (b) at the instance of the debtor or a surety in any proceedings in any court to which the debtor and the creditor are parties, being proceedings to enforce the agreement or any related agreement; or (c) at the instance of the debtor or a surety in any other proceedings in any court where the amount paid or payable under the agreement or any related agreement is relevant. … 140C Interpretation of ss. 140A and 140B (1) In this section and in sections 140A and 140B ‘credit agreement’ means any agreement between an individual (the ‘debtor’) and any other person (the ‘creditor’) by which the creditor provides the debtor with credit of any amount. … (4) References in sections 140A and 140B to an agreement related to a credit agreement (the ‘main agreement’) are references to– … (b) a linked transaction in relation to the main agreement or to a credit agreement within paragraph (a); …” (a) any of the terms of the agreement or of any related agreement; (b) the way in which the creditor has exercised or enforced any of his rights under the agreement or any related agreement; (c) any other thing done (or not done) by, or on behalf of, the creditor (either before or after the making of the agreement or any related agreement). (a) require the creditor, or any associate or former associate of his, to repay (in hole or in part) any sum paid by the debtor or by a surety by virtue of the agreement or any related agreement …; (a) on an application made by the debtor or by a surety; (b) at the instance of the debtor or a surety in any proceedings in any court to which the debtor and the creditor are parties, being proceedings to enforce the agreement or any related agreement; or (c) at the instance of the debtor or a surety in any other proceedings in any court where the amount paid or payable under the agreement or any related agreement is relevant.

(a) ; …”

[356]In so far as the following was not common ground, I am prepared to assume that:(a) The store card agreement was a credit agreement and the PPI policy was a related agreement within the meaning of the Act: Smith v Royal Bank of Scotland plc,

[2023] UKSC 34

;

[2024] AC 955

at [15];

(b) It does not matter for the purposes of the Act that the credit agreement and the related agreement are between different parties. A credit agreement can still be unfair in relation to matters which arise in connection with the related agreement: Plevin v Paragon Personal Finance Ltd,

[2014] UKSC 61

;

[2014] 1 WLR 4222

[357]The principles to be applied in assessing unfairness are set out in Self v Santander Cards UK Ltd,

[2024] EWCA Civ. 1106

at [55]. A broad and holistic assessment is required and there is no restriction on what can be taken into account provided the court thinks it relevant. AXA submits that the court should therefore have regard to the fact that customers considering the purchase of PPI (assuming they were aware of it at all) were likely to be concerned about their ability to repay and thus financially vulnerable and unsophisticated. It was a not a situation where they had voluntarily sought the opportunity to purchase PPI and in circumstances where they were first being offered an unsolicited store card and then PPI on top, there was a risk of them being bounced into a series of financial decisions at speed without having had a proper opportunity to think them through. Also relevant was the fact that GECB had a substantial interest in maximising sales which was not disclosed and that the retail staff were also often incentivised to achieve sales.

[358]For these reasons, AXA argued that:(a) There was unfairness on the grounds of extreme inequality of knowledge and understanding which prevented a genuine evaluation of the arrangement - a fortiori where the customer was not aware they were purchasing PPI at all;

(b) The notional opportunity to withdraw or cancel during the cooling-off period post-sale could not cure the unfairness because it would not reasonably be expected that customers would scrutinise documentation after the event and then take active steps to cancel. Fairness required a proper explanation before the sale was concluded – as, indeed, was the position under the ABI Code;

(c) The remedies set out in section 140B included an order to repay the premium. This was the same as the damage that AXA was required to compensate by way of redress payments.

[359]I agree with AXA that it does not matter that any remedy under the 1974 Act was purely hypothetical. Section 1(6) of the 1978 Act requires only a liability which has or could have been established. The authorities are clear that a notional liability is sufficient: BDW Trading Ltd v URS Corporation Ltd,

[2023] EWCA Civ. 772

;

[2024] KB 827

at [201].

[360]More difficult, however, is the fact that repayment would have been a discretionary remedy. The objective of section 1(6) is not to create or deem a liability where none can exist, but to recognise that it need not be formally established. In this case, there was a whole range of possible remedies under section 140B, some of which did not involve payment at all (for example, altering the terms of the agreement). In those circumstances, repayment might never have been ordered and it is therefore difficult to see how GECB would be said to have been liable to make such repayment.

[361]For these reasons, AXA has failed to persuade me that GECB was under any liability to consumers on the bases asserted. L.3: Same damage

[362]The 1978 Act only requires that GECB be liable for the same “damage” as FICL/FACL, not the same “damages”: Royal Brompton Hospital NHS Trust v Hammond (supra) at [27]. As recognised in that case, a practical test which can sometimes be useful is to ask whether the discharge by FICL/FACL of their liability would pro tanto discharge GECB. In this case, it undoubtedly would, and I accept that a regulatory liability to pay redress is not so fundamentally different from any liability that GECB might have incurred to customers as to take the case out of the statute altogether. Had the point been live, I would therefore not have found against AXA on this ground alone. L.4: Apportionment

[363]Ex hypothesi, the question of apportionment only arises if the contribution claim has succeeded so far, which necessarily pre-supposes that GECB was in breach of a common law duty of care or guilty of negligent misstatement or unfairness in relation to the credit agreement.

[364]However, as I have held, a claim for contribution requires a fact-specific enquiry into the facts and circumstances of each case in order to demonstrate that both parties were liable to the particular customer for the same damage. It is therefore impossible to make any a priori assessment of the appropriate just and equitable contribution. All that can be said is that, given the limited circumstances in which a claim would be likely to arise at all (i.e., where there was a specific assumption of responsibility or a specific misrepresentation), it is difficult to envisage circumstances where the apportionment would not be very heavily, if not entirely, against Santander. Likewise if Santander were held to have been in general breach of duty or to have made a generalised misrepresentation. In this context, it is not totally without significance that Santander shouldered the entire burden of making redress payments until 2015. L.5: Limitation

[365]Three potential limitation points arose in relation to the contribution claim but only one of them was the subject of any debate. This arose from the fact that section 140A of the Consumer Credit Act does not apply to “completed agreements”. Under the Act, a “completed agreement” is one under which sum is or may become payable. AXA thus accepted that credit agreements which had been cancelled prior to 6 April 2008 (when the relevant provisions came into force) were “completed” for this purpose. However, it argued that the exclusion did not apply to dormant accounts which could be resurrected at any time. It further argued that the exclusion does not bite simply because the associated PPI policy had come to an end, so long as the relevant credit agreement was still in force.

[366]I accept both propositions. So long as it is possible for a store card to be used, the credit agreement cannot be regarded as completed. Moreover, as AXA pointed out, unfairness in relation to the policy would still have tainted the credit agreement notwithstanding the subsequent expiry of the policy, and the historic premiums might still have fallen to be repaid.

[367]As for the other two limitation points, section 1(3) of the 1978 Act makes clear that a time bar of GECB’s own liability is irrelevant save in so far as the right is extinguished altogether. That was not this case, since the only time bar was under the Limitation Act which is only procedural and bars the remedy not the right, and Santander did not argue to the contrary. It was also common ground that as between AXA and Santander, a two-year time bar applies to any contribution claim such that AXA would have been limited in any event to recovery in respect of payments made on or after 7 December 2015, i.e., two years prior to the standstill agreement.

[368]As it is, however, the consequence of my findings is that the contribution claim fails altogether. Santander’s counterclaim does not therefore arise and need not be addressed further. M: THE NEGLIGENCE CLAIM

[369]As already noted, this final way of putting the claim is maintained by AXA very much as a fall-back, since it accepts that it is almost entirely time-barred by virtue of the long-stop limitation period and only survives in respect of policies sold after 7 December 2002.

[370]The issues for consideration fall under five heads:(a) Duty;

(b) Breach;

(d) Contributory negligence;

(e) Limitation under section 2 of the Limitation Act 1980.

M.1: Duty

[371]The duty asserted here is a duty owed to FICL/FACL rather than, as with the contribution claim, to the end customer. Ultimately, it was not controversial that an agent owes its principal a duty at common law to exercise reasonable skill and care in the performance of its duties Santander’s pleaded case was that there was no expectation of liability between the parties but that was not a submission which was pursued by Mr Zellick with any great vigour, if at all. It is in any event belied by the express stipulation for an indemnity in the Agency Agreement. and that at least from the date of the Agency Agreement, GECB’s duties as agent included an express obligation to comply with the ABI Code.

[372]To the extent that it was disputed, I am satisfied that this was also the position even prior to the conclusion of the Agency Agreement. As I have already found on the basis of the evidence set out at paragraphs ‎139-‎182 above, GECB knew not only(i) that FICL/FACL were members of the ABI and thus under a duty to exercise best endeavours to comply with the ABI Code themselves and to ensure that GECB did likewise, but also

(ii) that it was required to comply with the ABI Code itself. Indeed, that is what it claimed to be doing: see, for example, paragraph ‎161 above.

[373]Mr Blundell agreed that if GECB was aware that FICL/FACL had adopted the recommendations of the applicable code and were required to sell within those standards, then GECB could not itself ignore the requirements when selling. Frequently, of course, compliance would be covered by an express provision in any agency agreement, as was the case here after 1 December 2000. However, drafts of an agency agreement were under discussion from at least 1996 onwards which included comparable provisions. In these circumstances, it is pure sophistry to argue that GECB was not at all times aware that it was required itself to comply with ABI Code in its capacity as FICL/FACL’s agent.

[374]Mr Zellick’s main point under this head, however, was rather different. He submitted that it was not foreseeable prior to 2010 that a statutory regime might subsequently be put in place which would require redress to be paid for historic sales by reference to modern standards of acceptable selling and what is fair and reasonable, without regard to the rules of evidence and whether or not there was in fact any liability. He pointed to certain categories of mis-sale, such as opt-out sales or sales of Budget Accounts, which were required to be “auto-upheld” without further investigation into the facts. In these cases, compensation had to be paid notwithstanding that more detailed enquiry might have disclosed that the customer had all the relevant information and wanted the policy anyway. Relying on the recent decisions of the Supreme Court in Manchester Building Society v Grant Thornton UK LLP,

and Khan v Meadows,

;

, he argued that losses resulting from these unforeseeable matters fell outside the scope of any duty altogether because there was no sufficient nexus between the loss and the subject-matter of GECB’s duty - the so-called “duty nexus” requirement.

[375]It is fair to note that the Supreme Court’s analysis of liability for negligence in Manchester Building Society and Khan has not been universally adopted in subsequent cases. Indeed, there is some indication that its application may be somewhat limited beyond its specific context, namely the negligent provision of advice or information causing economic loss. But in any event, there is force in AXA’s submission that Santander’s argument elides the analytically separate concepts of foreseeability in ascertaining the scope of duty and remoteness.

[376]As Manchester Building Society makes clear, the scope of duty is to be determined primarily by reference to the purpose of the duty, judged objectively. In this case, the obvious purpose of GECB’s duty to comply with the ABI Code was to avoid FICL/FACL being placed in breach by virtue of their agent’s conduct. Further:(a) As held above, the relevant regulatory standards were materially the same over the entire period under consideration;

(b) It could self-evidently be expected that a breach of regulatory standards would lead to complaints being made and ensuing regulatory consequences;

(c) It is in the nature of a regulatory regime that it evolves over time. It is therefore irrelevant that the specific way in which the regulatory framework might develop could not be foreseen.

[377]In my judgment this is sufficient to bring the regulatory consequences of GECB’s conduct generically within the scope of its duty. I would observe that the nature of the compensatory scheme eventually adopted in DISP was almost certainly dictated to a large extent by the situation confronting the regulator where sales dated back decades and the likelihood of being able to carry out minute factual scrutiny of each individual complaint with any degree of rigour was vanishingly small. It is therefore unsurprising that it chose to adopt a broad and flexible approach to evidential requirements. Indeed it could be said that this is simply the consequence of the long-standing systemic failings of the industry (including GECB).

[378]Nor do I accept that the mere fact that the regulator required redress to be paid in circumstances when it might not have been recoverable in a civil claim makes it a different kind of damage or takes it outside the scope of GECB’s duty for the purposes of a claim in negligence. The redress still constitutes the measurable financial consequences of the breach.

[379]Thus, whether analysed in terms of a voluntary assumption of responsibility, or a relationship of proximity or in accordance with the Manchester Building Society framework, I am satisfied that GECB owed FICL/FACL a duty to avoid causing losses of the general kind claimed here. M.2: Breach

[380]I have already held that the appropriate standard of care was defined by the ABI Code and that GECB was in breach of the Code by virtue of its systemic failings. Since compliance with the ABI Code is itself conditioned by considerations of reasonableness (i.e., an obligation to exercise best endeavours), it is axiomatic that GECB failed to exercise all reasonable skill and care in the performance of its duties. Breach is therefore established. M.3: Causation and remoteness

[381]I have discussed above whether there was anything FICL/FACL could have done to avoid or ameliorate GECB’s failings and concluded that there was not. Santander’s argument that FICL/FACL’s losses were caused by their own acts and omissions rather than those of GECB must accordingly fail.

[382]That then leaves the question whether the specific losses claimed were caused by GECB’s breaches and/or whether they were too remote. M.3.1: Redress payments

[383]The first point to consider is whether it is necessary for AXA to prove that Santander’s breach or breaches of duty caused loss in each individual case. In my judgment, it is. Although my findings above are of systemic failings on the part of GECB which in themselves constitute breaches of duty, the focus of the claim in negligence is on whether those failings caused actual loss and damage rather than (as with the claim for an indemnity) whether they resulted in AXA incurring a regulatory liability to operate a prescribed redress scheme. I therefore hold that AXA must prove that Santander’s breach or breaches caused a loss in each individual case.

[384]So far as remoteness is concerned, however, it is Santander’s systemic failings which led to the entire problem. It would have been obvious that a systemic failure to comply with regulatory requirements would have regulatory consequences of some sort and it is beside the point that the precise nature of those subsequent consequences could not necessarily have been foreseen. On the basis that the regulatory consequences which in fact ensued required AXA to make the redress payments that it did, I am satisfied that the redress payments were not too remote a consequence of GECB’s breaches. M.3.2: FOS fees

[385]Santander’s submission that, even if it is liable for the redress payments, the FOS fees are nonetheless irrecoverable because they were unforeseeable must be rejected for the same reason.

[386]However, AXA conceded that it could not recover in negligence for fees levied in respect of complaints which were ultimately dismissed. It seems to me that this concession is correct. M.3.3: Administrative costs

[387]There is force in Santander’s submission that AXA’s internal administrative costs are irrecoverable has force. In principle, internal administrative costs that would have had to be paid anyway cannot be said to have been caused by GECB’s breach in any meaningful sense. In this case, however, almost the entirety of the costs related to additional costs incurred in setting up a dedicated complaints unit. It is possible that changes in the regulatory landscape would have required such a department to be established in any event, but I do not at present see why any genuinely additional costs of handling complaints should not be recoverable in principle.

[388]However, I did not hear full argument on the point and in any event it raises questions of quantification which cannot be determined now. This point likewise can be left for further argument if necessary. M.3.4: OR Settlement

[389]A finding that GECB is liable for the consequences of its negligence cannot on any view extend to the consequences of a settlement agreement concluded precisely in order to pre-empt such consequences. Such a settlement was not reasonably foreseeable and the claim for this head of loss thus fails in any event.

[390]It is irrelevant that the Official Receiver may have acted reasonably in negotiating a bulk settlement. That is a matter between him or her and the bankruptcy court, but it cannot affect the extent to which GECB is liable to pay compensation for losses caused by its breach of duty. In so far as AXA might say that it concluded the settlement in order to mitigate or avoid a potentially greater liability if complaints had actually been made, the short answer is that the rules of mitigation do not apply to claims in tort. M.4: Contributory negligence

[391]On the basis of my findings above there is no basis for making any deduction from AXA’s claim on grounds of contributory negligence. M.5: Limitation

[392]It is accepted by AXA that the long-stop limitation period under section 14B of the Limitation Act 1980 applies, and that section 2 of the Act additionally bars claims where more than 6 years have elapsed since AXA paid or agreed to make payment.

[393]What was controversial was Santander’s submission that actionable damage for the purposes of a claim in tort was sustained when the relevant policy was mis-sold and that AXA’s cause of action in negligence accrued at that date. If that is correct, section 2 would have the effect of barring all claims in respect of any policy sold more than six years prior to the date of the standstill agreement.

[394]There is a considerable overlap between this argument and the similar argument made in the context of the indemnity claim (see paragraphs ‎317-‎323 above) and I have no doubt that on this point AXA is correct. The present situation is almost exactly analogous with that which arose in Law Society v Sephton & Co.,

[2006] UKHL 22

;

[2006] 2 AC 543

where accountants had negligently certified the probity of a solicitor who had in fact misappropriated client funds. When this came to light much later, the Law Society compensated the defrauded clients in accordance with its statutory obligations and then claimed over against the accountants. The House of Lords held that the Law Society did not suffer damage as a result of the negligent certification until it was called upon to pay. The mere possibility of becoming obliged to pay money in the future was not damage. In reaching this decision the House approved the Australian decision in Wardley Australia Ltd v State of Western Australia (1992), 175 CLR 514 in which the High Court of Australia held that entry into a contract exposing the claimant to contingent loss or liability does not cause actual loss until the contingency is fulfilled and the loss crystallises.

[395]A distinction is therefore to be drawn between this type of case and the “damaged asset/package of rights” type of case exemplified by Forster v Outred (supra). In Forster v Outred, negligent advice from a solicitor caused Mrs Forster to enter into a mortgage to secure her son’s borrowing. It was held that she suffered an immediate measurable loss in so far as her equity of redemption was impaired as soon as she entered the mortgage. Similarly, where a solicitor negligently drafts a lease with the result that the client acquires a flawed asset. In both types of case, it can fairly be said that the asset is damaged from the outset and that the claimant suffers an actual immediate loss to that extent. This is to be contrasted with cases such as Nykredit Mortgage Bank plc v Edward Erdmann Group Ltd,

[1997] 1 WLR 1627

where there was a negligent valuation of property. The asset itself was not damaged but the negligent valuation meant that a lender who relied on it in entering into a mortgage might find that its security became insufficient as the market fluctuated. The House of Lords held that the lender only suffered loss when the value of the security in fact fell below the amount of the loan. The security rights themselves were not flawed; it was simply a question of their adequacy at a particular point in time.

[396]The present case is not remotely like the Forster v Outred line of authority. GECB’s negligence in this case did not impair the value of any other property of FICL/FACL. Mr Zellick argued rather faintly that a contingent liability to pay compensation impaired the value of FICL/FACL’s overall book of business in the sense that FICL/FACL’s rights under the policies were encumbered by the risk of having to pay compensation. However, by parity of reasoning with Sephton, this was a purely contingent liability, which was incapable of quantification unless and until a complaint was made and liability actually incurred. The contingent risk of having to pay compensation had nothing to do with FICL/FACL’s rights under the policy itself, which remained unchanged and undiminished at all times.

[397]I therefore reject Santander’s argument on this point. M.6: Novation

[398]It was conceded by AXA that liability for any negligence claims stays with SCL under the novation. In the light of my construction of that agreement, I am again not convinced that this concession was well made, but I make no finding on the point. Should it ever become a live issue (bearing in mind the very limited ambit of the negligence claim) it can always be revisited. N: CONCLUSION

[399]My decision on the matters raised for determination at this stage is therefore as follows:(a) The settlement claim fails and is dismissed;

(b) The indemnity claim succeeds against SISUK in relation to the redress payments and FOS fees but not otherwise;

(d) Subject to limitation the claim in negligence succeeds in principle, save in respect of the OR Settlement, without deduction for contributory negligence. Further argument may, however, be necessary in relation to the claim to recover FOS fees and administrative costs and as to the identity of the correct defendant.

[400]So far as concerns the Sample Complaints which I was asked to consider, the consequences of my findings are set out in a separate Appendix to this judgment.

[401]It only remains for me to express my thanks to counsel, both leading and junior, for the thoroughness and quality of their submissions. I am particularly pleased to have been addressed by junior counsel as the reduction in the number of smaller-scale commercial cases means that they all too seldom have the opportunity to hone their advocacy skills. This is a practice which is therefore much to be encouraged.