Prudential Assurance Company Ltd and Rothesay Life Plc, Re

This case raises for the first time before the Court of Appeal the approach that the court should adopt in dealing with applications to sanction transfers of insurance business under Part VII (“Part VII”) of the Financial Services and Markets Act 2000 (“FSMA”).

On 16 August 2019, Snowden J, a judge with considerable experience in this field, exercised his discretion under section 111(3) of FSMA to refuse an application by The Prudential Assurance Company Limited (“PAC”) and Rothesay Life Plc

(“Rothesay”) (together the “appellants”) for the court to sanction a scheme (the

“Scheme”) providing for the transfer from PAC to Rothesay of some 370,000 annuity policies written by PAC.

The judge refused sanction for two main reasons. First, despite the fact that PAC and Rothesay had equivalent Solvency Capital Requirement (“SCR”) metrics, Rothesay did not have the same capital management policies or the backing of a large wellresourced group with a reputational imperative to support it over the lifetime of the annuity policies. Secondly, it had been reasonable, in the light of PAC’s sales materials, age and reputation, for policyholders to have chosen PAC on the basis of an assumption that it would not seek to transfer their policies to a third party provider.

PAC and Rothesay now appeal that decision with the judge’s permission. In the broadest of outline, PAC and Rothesay make the following three points.

First, PAC and Rothesay contended that the judge failed to accord adequate recognition or weight to (a) the commercial judgment of PAC’s board, (b) the conclusions of the independent expert that the risk of PAC or Rothesay needing external support in the future was remote, (c) the regulators’ lack of objection to the Scheme and the continuing future regulation of Rothesay, and (d) the prejudice that a refusal to sanction would cause to PAC and Rothesay.

Secondly, PAC and Rothesay submitted that the judge accorded too much weight to the objecting policyholders’ contentions that (a) they chose PAC on the basis of its age and established reputation, (b) they had reasonably assumed that their annuity would be provided throughout its term by the same provider, and (c) there were distinguishing features of an annuity.

Thirdly, PAC and Rothesay argued that the judge ought not to have concluded on the evidence that there was a material disparity between the external financial support potentially available for each of them.

A number of policyholders made submissions to support the judge’s judgment. They submitted that the SCR metrics employed by the Prudential Regulation Authority (the “PRA”) were entirely based upon analyses of the current year and made no attempt to predict what risks the future might hold. The judge had been justified in taking his own view of those risks. The individual policyholders who made submissions explained how they had chosen PAC for its age and reputation. They made powerful submissions about how those features allowed them peace of mind. They said that they did not want to be the subject of repeated Part VII transfers in the future.

The PRA and the Financial Conduct Authority (the “FCA”) (together the

“Regulators”), as interested parties, have neither supported nor opposed the appeal. Instead, they have sought to assist the court in their capacities as the appellants’ prudential regulator and conduct regulator respectively.

The Association of British Insurers (“ABI”) intervened to express the concerns of the industry, submitting that the principles applicable to Part VII schemes had to be clear and predictable, given the heavy time and cost commitment required to prepare them. Moreover, the requirements of the regulatory regime were “calculated to ensure that the insurer could still pay out to policyholders after the occurrence of a 1-in-200 year stress event”, (Footnote: ¹) which falsified the judge’s view that he could speculate about Rothesay requiring capital support over the life of the annuities.

Applications for the sanction of Part VII schemes, including insurance company transfer schemes, are a significant part of the work of the Insolvency and Companies List in the Business and Property Courts of England and Wales. Despite that fact, as we have said, there has never before been a substantive appeal of a sanction decision to this court. Moreover, it is rare for such applications to be refused. For that reason, we have paid special attention to the arguments in this case, and have tried to explain in this judgment how the jurisdiction should generally be exercised.

On 22 June 2020, Patten LJ ordered that, in the event of the appeal being allowed and the judge’s decision being set aside, the Court of Appeal would not deal with the sanction of the Scheme but would remit the application for sanction to the High Court.

Our judgment deals with some essential background, the relevant legislation, the distinctions to be drawn between different types of insurance business, the essential authorities, the judge’s judgment, and finally with the grounds of appeal raised by PAC and Rothesay.

The following essential background is broadly taken from [1] to [85] of Snowden J’s judgment.

PAC is an English company wholly owned by M&G plc, which was in turn owned by Prudential plc until completion in October 2019 of the Demerger to which we refer below. PAC has been in existence since 1848 and has a reputation as a venerable and substantial business. In 2018, the Prudential group, of which PAC then formed part, had gross assets of over £508 billion. The gross assets of the demerged M&G Prudential group, of which PAC now forms part, are smaller but still very substantial.

The reputation of PAC and the Prudential group, built up organically over many years, is encapsulated in the following description on the website of Prudential plc: (Footnote: ²)

“Providing financial security since 1848

Successive generations have looked to Prudential to safeguard their financial security - from industrial workers and their families in Victorian Britain to over 26 million customers worldwide today. Our financial strength, heritage, prudence and focus on our customers’ long-term needs ensure that people continue to turn to our trusted brands to help them plan for today and tomorrow.” (emphasis added)

Rothesay was established in 2007 by The Goldman Sachs Group, Inc to conduct business as a specialist provider of annuities. Goldman Sachs sold its interests in Rothesay between 2013 and 2017. At the time of the first instance hearing, Rothesay was almost entirely owned by Blackstone Group LP (“Blackstone”), GIC Private

The proposed Scheme would transfer approximately 370,000 non-profit, in-payment annuity contracts (the “Transferring Policies”). These contracts provide for regular payments of fixed sums, in some cases increasing in line with objective criteria such as indices that measure inflation, over the lives of the annuitants and others. For convenience, we refer to the contracts as policies and to the annuitants as policyholders. The Scheme was designed to reduce PAC’s regulatory capital requirements in connection with a planned demerger (the “Demerger”) of the Prudential group. The Demerger was announced on 14 March 2018 and was to separate the Prudential group’s business into two segments: the UK and Europe on the one hand; and Asia, the US and Africa on the other. M&G Prudential would separate from the Prudential group and focus on the UK and European market. Prudential plc was to focus on the Asian, US and African market. The need for PAC to reduce its regulatory capital requirements arose from the transfer of two Hong Kong subsidiaries from PAC to a Prudential plc subsidiary. One of those subsidiaries made a substantial contribution of excess (surplus) capital to PAC’s capital position under the recast EU

PAC entered into two agreements with Rothesay on 14 March 2018: a business transfer agreement (the “Business Transfer Agreement”) and a collateralised reinsurance agreement (the “Reinsurance Agreement”). The Transferring Policies represent around 90% of the business reinsured by Rothesay under the Reinsurance Agreement, as the agreements also covered additional retail and bulk annuity policies, bringing the total number to be transferred to around 400,000, with a gross best estimate of liabilities (“BEL”) for PAC of about £12.9 billion.

In simple terms, the purpose and effect of the Reinsurance Agreement was to transfer the majority of the economic risk and reward of the annuity business covered by the agreement from PAC to Rothesay. When the agreements were signed, the assets backing the annuity policies were transferred by PAC to Rothesay as part of the premium for the reinsurance, and were held by Rothesay in custody accounts. However, the contractual obligations under the policies remained with PAC. The Business Transfer Agreement expressly contemplated that the parties would cooperate to achieve the actual transfer of that business through the Scheme. If the Scheme were implemented, the Reinsurance Agreement would be modified to exclude the Transferring Policies, and the assets transferred to Rothesay under the Reinsurance Agreement would be released from the custody accounts to Rothesay.

Neither of the PRA and the FCA objected to the Scheme. The PRA certified that the relevant certificates had been obtained as required by section 111 of FSMA, referred to at [36] and [37] below.

The independent expert was Mr Nick Dumbreck FIA, a consulting actuary of Milliman LLP and a member of the Institute and Faculty of Actuaries. Mr Dumbreck produced a number of reports and letters considering the effects of the Scheme on three groups of policyholders: the Transferring Policyholders, the existing nontransferring Rothesay policyholders, and the existing non-transferring PAC policyholders. The interests of the Transferring Policyholders were the most significant. Mr Dumbreck’s overall conclusions were that the Scheme would not have a material adverse effect on the security of benefits or reasonable expectations of policyholders with PAC or Rothesay.

The requirements of Solvency II formed a key part of the independent expert’s analysis. The judge summarised the Solvency II metrics at [45] to [48]:

On 31 December 2018, PAC’s shareholder-backed business had assets of around £60.7 billion, Technical Provisions of £51.9 billion, Own Funds of around £8.8 billion and an SCR of around £5.1 billion. PAC’s SCR coverage ratio was 140%. On the same date, Rothesay had assets of around £36 billion, Technical Provisions of about £32 billion, Own Funds of £3.89 billion and an SCR of £2.16 billion. Its SCR coverage ratio was 180%. These figures include the effect of the Reinsurance Agreement. Both companies provide a very high level of security for policyholders, notwithstanding Rothesay’s more favourable SCR coverage ratio.

The companies had different capital management policies. PAC did not disclose its confidential capital management policy, but its details satisfied Mr Dumbreck. PAC used a “solvency intervention ladder” whereby different management actions would be undertaken if the SCR coverage ratio fell below a series of set levels. Rothesay targeted an SCR coverage of between 130% and 150%. Actions would be taken to restore the coverage if it fell below 130%. While these differences make direct comparison difficult, Mr Dumbreck concluded that, based on current risk levels, PAC’s approach provided slightly higher levels of security for policyholders.

With respect to the possibility of parental support, Mr Dumbreck noted the significant resources available to Prudential plc and its reputational incentives for providing support to its subsidiaries. However, he also noted that solvencythreatening scenarios might similarly constrain the financial resources of Prudential plc. In addition, PAC would no longer be a subsidiary of Prudential plc after the planned Demerger and would no longer benefit from its parental support. By contrast, Rothesay’s parent was unlisted and lacked comparable financial resources. Mr Dumbreck cited the corrective actions in Rothesay’s capital management policy and the possibility of Rothesay’s parent raising capital from shareholders and the debt markets in support of his conclusion that both companies’ capital management policies provided “a very high level of security for policyholders”. Critically, he also concluded that “scenarios which could lead to the entire own funds of either company being dissipated are so extreme that any comparison of probabilities is subject to a very high degree of uncertainty”.

As regards the size and risk profile of each company, Mr Dumbreck noted that PAC, as a multi-line insurer, had a wider variety of business and was exposed to a broader range of risks. That also meant that it had a greater diversification of risk and its SCR coverage ratio took account of that fact. Rothesay was smaller and accordingly faced a smaller variety of risks. Consequently, Mr Dumbreck concluded that the companies’ respective SCR coverage ratios provided “broadly equivalent” levels of protection against risks.

Mr Dumbreck examined the effect of the Scheme on policyholders’ reasonable expectations, namely receipt of benefits, administration, management and governance of the policies in line with the policies’ contractual terms, and standards of service. In each instance, Mr Dumbreck concluded that the policyholders’ reasonable expectations would continue to be met after the implementation of the Scheme. Mr Dumbreck was also satisfied that the Scheme would not have a material effect on the security of benefits or reasonable expectations of holders of nontransferring policies of either PAC or Rothesay.

Some 1,000 policyholders lodged objections to the Scheme and a number of them made submissions to Snowden J and, as we have said, to this court. The policyholders said that they specifically chose to buy an annuity from PAC on account of its reputation in the market and did not feel comfortable with Rothesay taking over responsibility for their policies.

Part VII of FSMA is headed “Control of Business Transfers” and comprises sections 103A to 117. (Footnote: ³) It covers a wide range of financial business transfers: general and longterm insurance business (sections 104-105), banking business (section 106), reclaim fund business (section 106A) and ring-fencing transfer schemes (section 106B). By way of explanation, reclaim funds are companies that manage monies credited to dormant bank and building society accounts, while a ring-fencing transfer scheme enables a bank to transfer those parts of its business which are required to be separated and ring-fenced pursuant to Part 9B of FSMA, introduced in response to the financial crisis of 2008-2009.

Although Part VII contains provisions particular to each type of transfer scheme, the exercise by the court of its discretion to sanction any type of scheme is ultimately subject to the same broadly-expressed statutory criterion in section 111(3) that “in all the circumstances of the case, it is appropriate to sanction the scheme”.

It is common ground that the Scheme is an “insurance business transfer scheme” within the provisions of sections 105(1) and (2)(a). (Footnote: ⁴)

Section 104 provides that “[n]o insurance business transfer scheme is to have effect unless an order has been made in relation to it under section 111(1)”. Section 107 makes provision for an application to be made to the court for an order sanctioning an insurance business transfer scheme. Section 108 allows the Treasury to make regulations imposing requirements on applicants under section 107, and provides that the court may not determine an application if the applicant has failed to comply with a prescribed requirement. (Footnote: ⁵)

Section 109 makes provision for an application under section 107 in respect of an insurance business transfer scheme to be accompanied by “a report on the terms of the scheme”, which may only be made by a person nominated or approved by the appropriate regulator (Footnote: ⁶) and appearing to it to “have the skills necessary to enable him to make a proper report”. The scheme report must be in a form approved by the appropriate regulator. (Footnote: ⁷)

Section 110 explains who has the right to be heard on an application under section 107. In this case, those parties include the PRA, the FCA and “any person … who alleges that he would be adversely affected by the carrying out of the scheme”. (Footnote: ⁸)

Section 111(1) of FSMA “sets out the conditions which must be satisfied before the court may make an order under this section sanctioning an insurance business transfer scheme”. Section 111(2) requires the court to be satisfied that the appropriate certificates and authorisations to conduct the transferring business have been obtained. As the judge said, all such certificates and authorisations had been provided in this case, so compliance with section 111(2) was not an issue.

Most importantly for our purposes, however, section 111(3) provides, as earlier mentioned, that “[t]he court must consider that, in all the circumstances of the case, it is appropriate to sanction the scheme”.

Section 112 provides in some detail for the ancillary orders that the court may make if it makes an order sanctioning a scheme under section 111(1).

Given the wide range of businesses that may be transferred under Part VII and the even wider range of circumstances in which such transfers may be proposed, it is immediately apparent that application of the deliberately broad terms of section 111(3) will require consideration of different factors, depending on the business and the circumstances. There can be no single test nor a single list of factors that can be applied in all cases. This may be contrasted with the court’s discretion to sanction a scheme of arrangement under Part 26 of the Companies Act 2006, where agreement to the scheme must first be obtained from meetings of the members or creditors concerned. Irrespective of the circumstances of the case or whether the scheme is proposed with members or creditors, the requirement for class meetings enables the court, having satisfied itself that the class was fairly represented by those at the meeting and that the majority was acting in good faith and not to promote interests adverse to those of the class, to apply the single test of whether the scheme is one that an honest and intelligent person, a member of the class concerned and acting in respect of their interest, might reasonably approve: see Re Telewest Communications plc (No 2) [2004] EWHC 1466 (Ch), [2005] 1 BCLC 722 at [20]-[22].

There has been a plethora of first instance decisions approving insurance business transfer schemes under Part VII. When considering those decisions, it is important to understand the nature of the business with which the court was concerned. There is an infinite variation in the books of business held by insurance companies. The type of business varies over time as economic trends change. It is undesirable for us to try to produce a definitive categorisation. In broad terms, however, there are two important distinctions. The first is between general insurance business (e.g. motor and household insurance) and long-term business (e.g. life assurance and annuities). The second is between policies that vest a discretion in the insurer, most obviously in those having a with-profits element, and those that do not. A book of annuities such as that sought to be transferred in this case, of course, has no such discretionary element. It is simply a long-term debt owed to each policyholder, payable on a regular basis during their life (or lives).

When considering insurance business transfer schemes, these distinctions need to be kept firmly in mind, because different interests are engaged. In the case of the transfer of with-profits business, for example, the interests of the insurers and their shareholders have to be balanced against those of the policyholders, and the interests of different classes of with-profits policyholders may need to be balanced against each other. In the case of a book of in-payment annuities, on the other hand, the interests that require consideration are, at least primarily, those of the creditor policyholders. Moreover, the factors to be considered in transferring general business are different again.

The authorities that deal with the court’s sanction of insurance business transfer schemes must be viewed in their specific context.

In this section, we refer to only 5 of the most important sanction decisions.

Re London Life Association Ltd (21 February 1989, unreported) (“London Life”) concerned the transfer of mostly (75%) with-profits business from a mutual life insurance company to a larger Australian mutual life insurance society. It was a merger of the two societies and, because London Life was a mutual society, the transfer engaged the with-profits policyholders’ interests as members as well as their interests as policyholders. Hoffmann J explained in his judgment the origins of the need for court approval of transfers of life insurance business in the Life Assurance Companies Act 1870, which gave regulatory powers for the first time to the Board of

Hoffmann J then dealt with the nature of the court’s discretion saying that it had to be “exercised according to principles which [gave] due recognition to the commercial judgment entrusted by the company’s constitution to its board”. The court was, he said, first concerned with whether a policyholder, employee or other person would be adversely affected by the scheme in the sense that it appeared likely to leave him worse off. It did not, however, follow that any scheme which did so had to be rejected. The question was whether the scheme as a whole was “fair as between the interests of the different classes of persons affected”. The court did not have to be satisfied that no better scheme could have been devised. The choice as between different schemes and the details of them were matters for the board and for negotiation between transferor and transferee. The court could either confirm or reject the scheme proposed. Finally, Hoffmann J said that the question of whether policyholders would be adversely affected by the scheme was largely actuarial and involved “a comparison of their security and reasonable expectations without the scheme with what [they] would be if the scheme were implemented”. Whilst he did not say these were the only considerations, they were obviously very important.

The proposed transfer of with-profits business considered by Evans-Lombe J in Re Axa Equity & Law Life Assurance Society plc and Axa Sun Life plc [2001] 1 All ER (Comm) 1010 (“Axa”) was complex and a far cry from the straightforward transfer of in-payment annuities proposed in the present case. Its purpose was to deal with the very substantial “inherited estate” that had built up in the with-profits fund. Under the scheme, eligible policyholders were entitled to elect to receive cash payments or additional bonuses on their policies (incentive payments) and thereby waive their rights to receive distributions from the inherited estate. Evans-Lombe J described the primary question at [7] as being whether the scheme was unfair because Axa’s offer to policyholders of the incentive payments did not represent a reasonable price to compensate them for their interest in receiving distributions in the future. EvansLombe J summarised 8 principles that were to be derived from London Life at [6] as follows:

“(1)

The 1982 Act confers an absolute discretion on the Court whether or not to sanction a scheme but this is a discretion which must be exercised by giving due recognition to the commercial judgment entrusted by the Company’s constitution to its directors.

(2)

The Court is concerned whether a policyholder, employee or other interested person or any group of them will be adversely affected by the scheme.

(3)

This is primarily a matter of actuarial judgment involving a comparison of the security and reasonable expectations of policyholders without the scheme with what would be the result if the scheme were implemented. For the purpose of this comparison the 1982 Act assigns an important role to the Independent Actuary to whose report the Court will give close attention.

(4)

The FSA by reason of its regulatory powers can also be expected to have the necessary material and expertise to express an informed opinion on whether policyholders are likely to be adversely affected. Again the Court will pay close attention to any views expressed by the FSA.

(5)

That individual policyholders or groups of policyholders may be adversely affected does not mean that the scheme has to be rejected by the Court. The fundamental question is whether the scheme as a whole is fair as between the interests of the different classes of persons affected.

(6)

It is not the function of the Court to produce what, in its view, is the best possible scheme. As between different schemes, all of which the Court may deem fair, it is the Company’s directors’ choice which to pursue.

(7)

Under the same principle the details of the scheme are not a matter for the Court provided that the scheme as a whole is found to be fair. Thus the Court will not amend the scheme because it thinks that individual provisions could be improved upon.

(8)

It seems to me to follow from the above and in particular paragraphs (2) (3) and (5) that the Court, in arriving at its conclusion, should first determine what the contractual rights and reasonable expectations of policyholders were before the scheme was promulgated and then compare those with the likely result on the rights and expectations of policyholders if the scheme is put into effect”.

In Re Royal Sun Alliance Insurance plc [2008] EWHC 3436 (Ch) (“Royal Sun Alliance”), David Richards J was dealing with a transfer of general business

David Richards J explained the principles enunciated in Axa at [10] by saying that only sub-paragraphs [6(1)-(4)] were really in point in relation to general business. The reference to the “commercial judgment entrusted to the company’s directors” was probably more in point in relation to the transfer of long-term business than general business, although it was plainly “a matter for the board of the company to decide whether it is going to put forward any proposal for the transfer of business”. The reference to “reasonable expectations” related to with-profits policyholders, though it had relevance to general business most obviously “as to the levels of service provided by the insurer to its policyholders”.

David Richards J concluded at [11] with this most often cited passage:

“Accordingly, in approaching this application I shall be concerned to see whether there is any material adverse effect on the position of policyholders in any of the three groups to which I have referred. The word “material” is important. The court is not concerned to address theoretical risks. It might be said that a transfer of business from a very large company to a large company involved a reduction in the cover available to the transferring policyholders, but assuming that the transferee is in a financially strong position it matters not that the level of cover in the transferee is less than that in the transferor. What the court is concerned to address is the prospect of real, as opposed to fanciful, risks to the position of policyholders”.

In Re Scottish Equitable plc and Rothesay Life plc [2017] EWHC 1439 (Ch)

Warren J dealt with the question of fairness mentioned by Evans-Lombe J in [6(6)] as follows at [63]:

“Care must be taken over the use of the word “fair”. This is not the subjective view of a policyholder or even of the judge. An objective view must be formed, a view reached against the objective standards and the factors appropriate to take into account. To take an extreme example, a scheme would not be unfair because it transferred business from a Scottish company to an English company even though a particular policyholder selected the company in the first place precisely because it was Scottish rather than English”.

Later at [114]-[116], Warren J returned to the question of policyholder choice:

“114.

… Miss Hutchins emphasises the unfairness, as she sees it, of compelling her elderly father to transfer to a new company from the venerable [Scottish Equitable] which he deliberately chose. He wants to be given a choice, in particular to transfer to LGAS rather than to [Rothesay].

115.

There are two points to make. Firstly, the venerable position of SE is not, I am afraid, of itself a relevant factor. Even venerable institutions can fail as those who work in this area of the law are well aware. In any case, SE is part of a group, about the age and venerability of which I have no idea. So the point, if it had any force, is not made good.

116.

Secondly, a newish body, that is to say, RL, is not to be regarded as an unsuitable provider simply because it is new otherwise we could never have new entrants into the market for transfers. The question is not its age but its financial strength, record and expectations. As to this, the independent expert and the regulators are clearly satisfied about its financial strength, there is no criticism made of its record, and I have no reason to think that it will not be properly and prudently managed into the future”.

In Re Barclays Bank plc [2018] EWHC 472 (Ch) (“Barclays”), Sir Geoffrey Vos C dealt with the sanction under Part VII of a ring-fencing transfer scheme. Such a scheme must be supported by the report of a skilled person who is required to answer the statutory question in section 109A(4), namely (1) whether persons other than the transferor are likely to be adversely affected by the scheme, and (2) if so, whether the adverse effect is likely to be greater than is reasonably necessary in order to achieve the purpose of the scheme. The court summarised the appropriate approach to sanction in such cases at [100]:

“in exercising its discretion, the court must keep in mind, in addition to the contextual and other matters I have already mentioned, the following main factors:-

(i)

The court’s discretion is unfettered and genuine and is not to be exercised by way of a rubber stamp.

(ii)

The design of a ring-fencing transfer scheme is a matter for the board of the bank concerned. There may be many possible approaches to the design of a statutorily-compliant ring-fencing transfer scheme that will affect stakeholders differently. The choice is for the directors of the bank concerned, acting properly in accordance with their duty under section 172(1) of the Companies Act 2006 (which is to act in the way they consider, in good faith, would be most likely to promote the success of the company having regard to matters including those specified in that subsection).

(iii)

The adverse effects of a ring-fencing transfer scheme must be viewed through the lens of the statutory question, so that the court must consider, with the aid of the Skilled Person, first whether persons other than the transferor are likely to be adversely affected by the scheme, and, if so, whether the adverse effect is likely to be greater than is reasonably necessary in order to achieve the statutory purposes. In considering whether persons are likely to be adversely affected by the scheme, regard need only be had to those adverse effects that are (i) possibilities that cannot sensibly be ignored having regard to the nature and gravity of the feared harm in the particular case, (ii) a consequence of the scheme, and (iii) material in the sense that there is the prospect of real or significant, as opposed to fanciful or insignificant, risk to the position of the stakeholder concerned.

(iv)

Even if the statutory question is answered negatively, it will not automatically follow that a proposed scheme will be rejected. The court’s approach will depend on all the circumstances, including the balance between the chosen design of the scheme, the benefits that will be achieved by the scheme, and the nature of the adverse effects identified, all viewed through the lens of the approach inherent in the statutory question itself.

(v)

The court will give weight to the views expressed to it by the Skilled Person and by the Regulators, and will fairly evaluate the weight to be given to views expressed to it in statements of representations made by stakeholders”.

At [114]-[119], the judge dealt with his general approach to the exercise of discretion under section 111(3). He said that the approach indicated by Axa and Royal Sun Alliance meant that, in the overwhelming majority of cases, “the court’s inquiry would focus on the questions of whether policyholders’ security of benefits and reasonable expectations of service standards would be adversely affected by the proposed scheme”, and that the court would “regard the first of those issues as “primarily” a matter of actuarial judgment, and in that respect [would] give close attention to the views of the independent expert and the regulators”. At [115], however, the judge said that it was clear that this was not the full extent of the factors that the court might take into account in the exercise of its discretion under section 111(3). The question for the court was whether it was appropriate to sanction the scheme, and the court was required to take into account all the circumstances of the case. London Life and Axa acknowledged that this conferred a very broad discretion. The judge did not consider that the court was “constrained by the same actuarial analysis or regulatory criteria derived from Solvency II that necessarily [shaped] the approach of the independent expert and the regulators”. As Briggs J had said in Re Pearl Assurance (Unit Linked Pensions) Limited [2006] EWHC 2291 (Ch) at [6], the court had a “discretion of very real importance, which [was] not in any way intended simply as a “rubber-stamp” for the opinion of the independent expert or the views of the regulators”.

The judge cited the general approach suggested by Warren J in Scottish Equitable at [56]. (Footnote: ⁹) He said this about Warren J’s approach:

“117.

I agree with Warren J that the fact that Part VII exists means that policyholders are not given a veto over what insurers wish to do for commercial reasons. But neither, in my judgment, is there any presumption in favour of a transfer for such reasons. In each case a balance has to be struck, and it must be for the commercial parties to the proposed transfer to satisfy the court that “in all the circumstances of the case, it is appropriate” to sanction a change to the contractual status of the policyholders.

118.

I would also accept that in striking the balance of interests to which Warren J referred, any purely subjective likes or dislikes of policyholders carry little or no weight. That was, I believe, what Warren J had in mind in paragraph [63] of Scottish Equitable when he gave, as an extreme example, a policyholder who objected to a scheme for the transfer of business from a Scottish company to an English company. It was also the reason why (at paragraphs [95]-[98]) Warren J rejected the objection from a policyholder that he simply “did not like” Rothesay, and had made a number of unsubstantiated allegations about Goldman Sachs which, at the time, was a shareholder of Rothesay. But as I have indicated, I do not consider (and I do not believe that Warren J decided) that the only factors that the court can take into account are those that can be reduced to part of an actuarial or risk-based analysis”.

Snowden J gave, in effect, 6 reasons for ultimately exercising his discretion to refuse to sanction the scheme: (i) annuitants cannot change their provider, (ii) policyholders would reasonably have assumed that PAC would not transfer its obligations, (iii) the SCR metrics did not provide a complete answer to the question of security of benefits for policyholders - the court had to consider the respective capital management policies of the transferor and transferee, (iv) the disparity between the external support potentially available for PAC and Rothesay, (v) the age and reputation of the transferee, and (vi) PAC was not prejudiced by the refusal to sanction the Scheme as it had already achieved its main business purpose (to reduce the SCR of its shareholder-backed business so as to facilitate the Demerger) by entering into the Reinsurance Agreement.

In relation to his first reason that annuitants cannot change their provider, he said at [120] that the particular nature of an annuity policy was an important factor in the exercise of his discretion. The purchase of an annuity was, for many people, one of the most important financial decisions that they will ever make. An annuity policy was often purchased on retirement with a large single premium payment derived from a person’s pension scheme. In relation to these policies, once it was purchased, the policyholders could neither encash the policies and take their lump sums out again, nor change annuity provider.

For that reason, the judge said at [122] that policyholders would be particularly concerned to select “a company with a good reputation and financial standing who they trust, because once selected, they will have no choice but to rely upon that company to provide them with essential income over a potentially very long period which could run into several decades”. These considerations were apparent throughout the representations that the judge received from opposing policyholders. At [126] Snowden J concluded by saying that it had to be appreciated that the Scheme’s impact on policyholders was very different to the effect of a scheme for the transfer of general insurance policies. (Footnote: ¹⁰)

In relation to his second reason, the judge said at [127]-[128] that, whilst it did not assist in the context of Part VII to ask whether policyholders would have a reasonable expectation that their insurer would not seek to transfer their policy, (Footnote: ¹¹) he considered that there was considerable force in the submission made by the opposing policyholders that “they reasonably assumed that PAC would not transfer its obligations under the annuity policies to any other company”. It was unrealistic to assume that the average prospective annuitant would have any independent knowledge of the possibility of transfer of the policy under Part VII. None of the documents in this case made any reference to Part VII or to the possibility of transfer. It was entirely reasonable for policyholders reading PAC’s statements to make the assumption that it would only be PAC that would be providing the annuity for the rest of their lives.

In relation to his third reason, the judge saw no reason at [133] to “doubt Mr. Dumbreck’s analysis based upon the current SCR coverage ratios of PAC and Rothesay”, and recognised that it was to be given considerable weight in the exercise of his discretion. Nonetheless, the judge thought it was “also necessary to consider the respective capital management policies of the two companies and to understand how they might each react to an unexpected deterioration of their financial position”.

In relation to the judge’s fourth reason, he summarised at [135]-[143] the evidence concerning the prospects that external support would be available for each of PAC and Rothesay in the event of deterioration in their financial position. He repeated Mr. Dumbreck’s view, which he shared, that “in the event of a deterioration in PAC’s financial position it [was] likely, for obvious reputational reasons, that Prudential plc would provide the necessary financial support to PAC”. (Footnote: ¹²) In contrast, the judge recorded at [141] Mr Dumbreck’s statement that “Rothesay’s parent company does not have any substantial resources that could be made available in the event that

Notwithstanding Mr Dumbreck’s view, the judge considered at [145] that the disparity between the external support potentially available for PAC and Rothesay was prima facie a material factor affecting the interests of policyholders to be taken into account in the exercise of his discretion. Moreover, he was not persuaded that the risk of failure of either PAC or Rothesay was fanciful. (Footnote: ¹⁴) Questions of the relative lack of availability of parental support for Rothesay were, therefore, material. Even though (a) the current SCR and SCR coverage ratios of both companies meant that there was a very high degree of confidence (in excess of 99.5%) that neither company would fail to maintain capital assets at least equal to its Technical Provisions in the next year, and (b) the judge had no reason to doubt the PRA’s statement that there was nothing to suggest any imminent deterioration in Rothesay’s current balance sheet, he did not think that there could be anything like the same level of confidence that a material deterioration of the balance sheets of either company might not occur at some time over the decades during which the annuities would require payment. It was necessary to have regard to the fact that, if the remote risk of insolvency were to eventuate, the result for policyholders would be catastrophic. The likelihood of support being available for PAC, and the relative uncertainty of whether it would be available for Rothesay, could not be dismissed as an immaterial factor. The judge did not agree with Mr Dumbreck that a comparison could be made between the comfort to be obtained from the current capital resources, regulatory capital requirements and capital management policies of the two companies on the one hand, and the potential for parental support on the other. The two were not alternatives: the availability of parental support was capable of providing additional comfort in the event that the comfort to be derived from the current financial strength of the two companies proved misplaced.

Finally at [153], the judge said that David Richards J’s comments in Royal Sun Alliance about the court being concerned with “the prospect of real, as opposed to fanciful, risks to the position of policyholders” were made in the context of the transfer of short term business. The duration of the annuity policies are many times greater. The impact of a default upon annuitants would be catastrophic in comparison, and if transferred from PAC, the annuitants would not have the opportunity to renew with another insurer.

The judge concluded on his fourth reason at [154]-[155] by saying that he did not consider that he could disregard as fanciful the possibility that PAC or Rothesay might require external financial support over the lifetime of the annuitants in this case. There was a material difference in the potential availability of assistance for the two companies: “if the need arose, PAC would be likely to be supported from the very substantial resources of the Prudential group; but no equivalent measure of comfort is available in relation to Rothesay”. That conclusion was unaffected by the possibility of the Financial Services Compensation Scheme (“FSCS”) paying full compensation. The relevant question was whether the Scheme made a material change in the security of benefits for policyholders as a result of the change from PAC to Rothesay. If FSCS compensation were relevant, it would apply in all annuity cases, and it would mean that there would be no purpose in any detailed analysis of the respective financial strengths of the transferor and transferee companies. That was not, the judge noted, an approach which had been adopted in any previous case.

In relation to the judge’s fifth reason, the judge did not agree with Warren J in Scottish Equitable at [115] that the respective ages and reputations of transferor and transferee were not relevant factors. Where consumers had made a choice of annuity provider based on such factors, that was something that the court could take into account. No-one could ever guarantee that a financial institution could not fail. The fact that the PRA or an actuary could not quantify a firm’s venerability or reputation in capital terms did not mean that those matters had to be disregarded by the court. The judge said at [160] that the court’s role under section 111(3) was not simply intended to replicate the Regulators’ risk-based approach or the independent expert’s actuarial approach. Consumers were not generally given access to the detailed financial information and Solvency II metrics. Instead, they were expected to select an insurer based upon other factors. The purchaser of an annuity policy who decided to place reliance on the longevity and established reputation of an insurer could not be said to be acting irrationally or unreasonably. The court was entitled to have some regard to the reasons for their choice. At [164], the judge noted that the facts of this case were distinguishable from those which Warren J considered in Scottish Equitable. He accepted that he should give some weight to the policyholders’ contractual choice when exercising his discretion.

In relation to the judge’s sixth reason, the judge thought that, because the Reinsurance Agreement had already passed the economic risk to Rothesay, he could take into account that PAC had already achieved its substantial business purpose, so that if the Scheme were not sanctioned, PAC’s capital benefits would not be lost. The judge said at [170] that he could take that into account in striking the balance between the interests of policyholders and commercial parties. The judge did not consider that the other prejudices relied upon were of major significance. A continuing regulatory capital requirement of £100 million represented a very small proportion of PAC’s capital benefits from the Reinsurance Agreement. Rothesay entered into the Reinsurance Agreement knowing that the Scheme was subject to the sanction of the court: “[t]here is a distinct air of bootstraps in Rothesay suggesting that it would be prejudiced if the Scheme was not sanctioned, by being deprived of the opportunities which it would have for saving administration costs, managing risks, or for using different or more innovative investment techniques for the assets that back the transferring annuities, which it would be able to do if the Scheme was sanctioned”.

The judge said at [175] that he saw no reason to conclude that policyholders would be adversely affected by the Scheme in relation to the standards of service.

The judge concluded at [177]-[184] that the independent expert’s opinion, with which the Regulators did not disagree, that the implementation of the Scheme would cause no material adverse effect upon the security of benefits and reasonable expectations of Transferring Policyholders as regards service standards and governance was entitled to considerable weight. But it was not determinative. The court could take into account a wider set of factors in striking a balance between the interests of policyholders and commercial parties. The 6 factors weighed heavily against the exercise of the court’s discretion to sanction this Scheme. In terms of the criteria that the opposing policyholders relied upon to select their annuity provider, Rothesay was very different from PAC. It was a relatively new entrant without an established reputation in the business. In the event of insolvency during the long life of the policies, the reliance which policyholders would have to place upon an uncertain capital-raising exercise from the investors in Rothesay or the markets more generally was a material disadvantage of the Scheme to Transferring Policyholders.

Finally, the judge concluded at [183] that he did not accept that his refusal to sanction the Scheme would make it more difficult for PAC to utilise Part VII in future or for Rothesay to acquire further annuity policies.

We have already indicated at [5]-[7] the issues raised by the appellants’ grounds of appeal, and at [3] the two main reasons that Snowden J gave for refusing to sanction the Scheme, namely (i) the disparity between PAC and Rothesay in the security of policyholders’ benefits over the lifetime of the annuity policies, and (ii) the reasonableness of the policyholders’ choice of PAC on the basis of an assumption that it would not transfer their policies.

As we see it, however, the three central issues raised by the appeal are as follows:

Two subsidiary issues were also argued as follows:

If the appellants succeed on one or more of these issues, the court will need to consider the consequences for the appeal, and for the further disposal of the sanction application (the “disposal issue”).

Before dealing with each of these 6 issues, we think it helpful to set out what we derive from the legislation and the previous authorities. Whilst we have been referred to literally dozens of first instance sanction decisions, we have already cited the ones that we find most helpful.

The judge hearing an application for the sanction of an insurance business transfer scheme under Part VII should first, we think, identify the nature of the business being transferred and the underlying circumstances giving rise to the scheme.

As we have already indicated, different considerations affect different types of business. For example, the court considering the transfer of a book of annuities in payment will be primarily concerned with the interests of the transferring policyholders, whereas a transfer of with-profits business may raise directly the question of fairness between the policyholders remaining with the transferor, the transferring policyholders, and the companies themselves and their shareholders. Transfers of some types of business may engage the interests of employees or other stakeholders in the transferor or transferee companies.

The circumstances giving rise to the scheme proposed will also affect the approach of the court. For example, many schemes will reflect commercial transactions between transferor and transferee companies for the benefit of those companies. Other schemes will be occasioned by external events (such as the departure of the UK from the European Union) or the financial or other commercial circumstances of the transferor. Some may take the form of a rescue of the business retained or transferred.

The discretion of the court has frequently been said to be unfettered and genuine and not to be exercised by way of a rubber stamp. (Footnote: ¹⁵) That is true but, as in the exercise of all discretions, the court must take into account and give proper weight to matters that ought to be considered, and ignore matters that ought not properly to be taken into account. The correct identification of which matters fall on which side of the line in particular transfer situations has caused some confusion in this, and perhaps other, cases.

From our reading of the decided cases, we have detected a tendency on the part of those presenting these applications, in many cases accepted by the judges hearing them, to treat the judgments of Hoffmann J in London Life and Evans-Lombe J in Axa

In a case such as the present, the paramount concern of the court will be to assess whether the transfer will have any material adverse effect on the receipt by the annuitants of their annuities, and on whether the transfer may have any such effect on payments that are or may become due to the other annuitants, policyholders and creditors of the transferor and transferee companies. The court will also be concerned to assess whether there may be any material adverse effect on the service standards provided to the transferring annuitants or policyholders. Whether any other factors require consideration will depend on the circumstances of the case.

The first duty of the court is carefully to scrutinise the reports of the independent expert and the Regulators, and the evidence of any person required to be heard under section 110 including those that allege that they would be adversely affected by the carrying out of the scheme. The court must understand the opinions presented and is entitled to ask questions about them as necessary. It will do so, in particular, with a view to identifying any errors, omissions, or instances of inadequate or defective reasoning.

In the absence of such defects, however, the court will always, in exercising its discretion, accord full weight to the opinions of the independent expert and the Regulators. That does not mean that the court can never depart from the recommendations of the expert or the non-objections of the Regulators, but it does mean that full weight must be accorded to them, so that a court would not depart from such recommendations and non-objections without significant and appropriate reasons for doing so. This is particularly so in relation to the financial and actuarial assessments required as regards the security of financial benefits. Whilst the judges hearing Part VII applications have considerable experience of the actuarial and specialist issues reported on by both the expert and the Regulators, the court is not itself an expert and should not substitute its own expertise for that of the entities required or entitled by statute to proffer those opinions.

This approach to the exercise of the court’s discretion applies to the crucial question of whether the proposed scheme will have any material adverse effect on policyholders, employees or other stakeholders. An adverse effect will only be material to the court’s consideration if it is: (i) a possibility that cannot sensibly be ignored having regard to the nature and gravity of the feared harm in the particular case, (ii) a consequence of the scheme, and (iii) material in the sense that there is the prospect of real or significant, as opposed to fanciful or insignificant, risk to the position of the stakeholder concerned. In some cases, it may also be relevant for the court to consider whether there would be such material adverse effects in the event that the scheme was not sanctioned.

Even if the court finds that the proposed scheme will have a material adverse effect on some group or groups of policyholders, it may still sanction the scheme in the exercise of its discretion. For example, this might occur if the scheme is in the nature of a rescue of the business. If there are differential effects on the interests of different classes of person affected, the court will need to consider whether the proposed scheme as a whole is fair as between those interests.

The court should adopt the same approach to the exercise of its discretion (described at [82] above) when making the more general comparison between the positions that would exist with or without the proposed scheme in respect of (a) the security of the policyholders’ benefits, and (b) the standards of service and corporate governance that the policyholders can expect. In many cases, this comparison will entail the court’s consideration of the contractual rights and reasonable expectations of policyholders, including the standards of service and governance that can be expected if the scheme is implemented.

Once the court has undertaken the evaluations we have mentioned, the court will decide whether or not to sanction the proposed scheme, if, under section 111(3) it is, in all the circumstances of the case, appropriate to do so. It cannot require the applicants to vary or alter the scheme, even though that may sometimes be the effect of the court expressing its concerns. The choices of both the scheme itself and its detailed terms are for the directors of the transferor and transferee concerned. The primary duty of those directors is, of course, to promote the success of their companies.

We think that these two issues are most conveniently dealt with together. The way in which the PRA now regulates insurers under FSMA and Solvency II has a significant bearing on the prospect of those insurers requiring external financial support in the future.

The judge dealt with “capital management, corrective actions and likely support from other sources” at [135]-[155] as we have summarised above at [57]-[60]. There was argument before us about whether the judge had really said that Mr Dumbreck thought that the risk of Rothesay needing external financial support was remote. We do not think it much matters. The judge undoubtedly did decide that less external financial support was potentially available for Rothesay than for PAC, and that that was a material factor to be considered. The essential questions are, therefore, whether the judge was right to conclude that there was a material disparity between the external support potentially available for each of PAC and Rothesay, and, even if there were such a disparity, whether that was a material factor.

In relation to Issue 1, the appellants submit that the judge wrongly dismissed as “essentially supposition” both Mr Dumbreck’s view, and the evidence of Mr Stoker, that, even in adverse conditions, Rothesay’s ultimate owners would be motivated to protect their existing investment. A future projection as to the external support likely to be available to each of PAC and Rothesay in the event of their financial deterioration was equally speculative. The judge should, the appellants submit, have accepted Mr Dumbreck’s opinion that the likelihood of either company being required to call upon external support in the first place was remote.

On this point, we do not accept that the judge was justified in dismissing Mr Dumbreck’s view that there was, in essence, no disparity between the financial resilience of PAC and Rothesay either now or in the future. We think the judge misunderstood the nature of the continuing regulation of authorised insurers and its significance.

Mr Tom Weitzman QC, leading counsel for the PRA, explained to us how the PRA applies the mandatory Solvency II regime by reference to the following diagram reproduced at [35] in Snowden J’s judgment in Re Rothesay Life plc and Monument Life Insurance DAC [2020] EWHC 2185 (Ch):

An insurer’s financial resources are primarily articulated in terms of Minimum Capital Requirement (“MCR”), (Footnote: ¹⁶) SCR and SCR coverage at a particular point in time. That point in time relates to a one-year period going forward. Solvency II is nonetheless a risk-based regime which also has regard to forward looking matters such as risk management and governance. (Footnote: ¹⁷) Solvency II, for example, requires each insurer regularly to complete an “own risk and solvency assessment” (“ORSA”). (Footnote: ¹⁸) The ORSA requires insurers to undertake a forward-looking assessment of risks, solvency needs and adequacy of capital resources, (Footnote: ¹⁹) and involves a range of stress and scenario tests.

The BEL shown in the diagram above is calculated on the basis of the insurer’s expected future obligations over the lifetime of the policies. It uses the most up-todate financial information and best estimate actuarial assumptions, and represents the present value of the projected cash-flows. (Footnote: ²⁰)

As the diagram shows, the BEL plus the risk margin, which is what an insurer would charge on top of the BEL to take over the relevant liabilities, (Footnote: ²¹) makes up the Technical Provisions. (Footnote: ²²) There are then other liabilities and the insurer’s own funds on top of the Technical Provisions. The SCR is the amount held by the insurer above its Technical Provisions which covers the 1-in-200 year stress event within the next 12 months. (Footnote: ²³) One year is selected as being a sensible period by reference to which to carry out such an assessment.

As paragraph 76 of the PRA’s approach to insurance supervision published in October 2018 says: “[t]he Solvency II regime compares the level and quality of capital held by an insurer (including the firm’s ability to raise more capital if needed) with the capital requirements applicable to that firm. These requirements are calculated to ensure that the insurer could still pay out to policyholders after the occurrence of a 1-in-200 year stress event, where the stress event used in the calculation reflects the risk profile of the particular insurer”.

In determining what constitutes a 1-in-200 year stress event, insurers are required to identify quantifiable risks to which they are exposed. (Footnote: ²⁴) Such risks will include falls in equity markets, default of debtors, adverse interest rates or exchange rate movements, changes in longevity, and a large number of other quantifiable risks. Mr Weitzman gave one example of an unquantifiable risk. He referred to an adjustment in the business operating environment such as the change in the personal injury discount rate which was announced by the government in 2017 and caused a number of insurers to increase substantially their estimate of liabilities.

Insurers are also required to have in place plans, including group support, to restore capital in the event of adverse events, so that they meet their SCR requirements within a specified period. (Footnote: ²⁵)

The judge accepted at [132] that, measured by their SCR coverage ratios, the relative financial strengths of PAC and Rothesay were comparable, and that the fact that PAC was currently larger than Rothesay did not, of itself, mean that Rothesay policyholders would have less security of benefits. Moreover, he accepted, correctly we think, Mr Dumbreck’s opinion that Rothesay’s lower resilience due to less diversification of risk was taken into account in Rothesay having a proportionately higher SCR. The judge, concluded, therefore, again correctly, that Mr Dumbreck’s analysis based upon the current SCR coverage ratios of PAC and Rothesay was to be given considerable weight in the exercise of his discretion.

It was, however, at [148] of his judgment that the judge, we think, took a wrong turn. He pointed to the independent expert and the PRA having relied on the SCRs, the SCR coverage ratios, and Rothesay’s balance sheet as at 31 December 2018. He seems to have thought that the expert and the PRA were not justified in looking at these metrics at a specific date to support their opinion that there was only a remote chance of parental support being needed in the future. As we have explained, the date at which these metrics are judged is not the critical feature. The expert and the PRA were judging the solvency metrics of the companies as at that date on the basis of an analysis of their likely resilience to a 1-in-200 year stress event within the coming year. But the fact that Rothesay would continue to be regulated under the same rules from year to year into the foreseeable future meant that the present conclusion of the expert and the PRA were valid parameters for Rothesay’s future security. The judge was not justified in adding his own speculative conclusion at [149] that he could not be confident that the companies’ balance sheets would not deteriorate materially over the life of the annuities.

In our judgment, it is important to understand, as the PRA submitted, that its prudential assessment of a scheme involves consideration of the future. The PRA takes account of the transferor’s and transferee’s respective abilities to measure, monitor and manage risk and to conduct their business prudently. That includes their ability to take corrective action in the event that there is a material deterioration of their balance sheets. The PRA also takes account of the fact that the transferor and transferee will continue to be supervised by the PRA on an ongoing basis, and its

It is not correct to submit, as Mr Barry Isaacs QC did on behalf of the policyholders he represented, that the independent expert and the PRA are looking only at the company’s financial security over a one-year period.

Accordingly, we do not think that the judge was justified in concluding that there was a material disparity between the potential need for external support for each of PAC and Rothesay. He disregarded the opinion of the expert and the PRA as to the appellants’ future financial resilience on the false basis that those opinions were themselves founded upon only a snapshot of the current year. They were not for the reasons we have given.

The judge was making a comparison between the likelihood of each of PAC and Rothesay receiving external (mainly parental) financial support in the event of their financial deterioration during the life of the annuities. The financial support he was considering was, of course, of a non-binding and non-contractual nature, since it was described as “reputational” or for “comfort”. (Footnote: ²⁶) We do not think that the likelihood of non-contractual parental support being available in the future was a relevant factor for the judge to take into account. None of the expert, the PRA or the court can assume that any non-contractual parental support will be available in the future, let alone long into the future. Any insurer, who complies with the requirements of Solvency II, is at liberty to dispose of its excess own funds above its Technical Provisions and SCR at any time. Parents can never be required to support their subsidiaries’ capital whether for reputational reasons or on any other basis. Moreover, parents of insurers are always at liberty to sell their regulated subsidiaries to others with lesser resources. Indeed, the Demerger which was in contemplation at the time of the hearing before the judge, and has since been implemented, involved PAC ceasing to be a subsidiary of Prudential plc and becoming part of a smaller group. Thus, even if the judge were justified in commenting that he could not disregard as fanciful the possibility that PAC or Rothesay might require external financial support over the lifetime of the annuitants, (Footnote: ²⁷) it was not a factor that could be taken to override the conclusions reached by the independent expert and the PRA as to the financial stability and resilience of Rothesay on the basis of current Solvency II metrics and its ongoing regulation.

For these reasons, we think the judge was wrong to consider whether there was a material disparity between the external non-contractual support potentially available as at the date of the sanction hearing for each of PAC and Rothesay. It was not a relevant factor once the Solvency II metrics were satisfied. Insurers can be sold, as we have said, and can reduce their surplus own funds at any stage. That is why the evaluation of an insurer’s Solvency II metrics, taken together with the prospect of its continuing regulation, is both necessary and normally sufficient to measure its resilience to future events.

It may be commented that the judge was doing what he criticised the expert and the PRA for having done. He was considering what non-contractual financial support might be available in the future to each of PAC and Rothesay by reference to their respective economic positions as at the date of the sanction hearing. The relevant issue was the appellants’ financial resilience throughout the life of the annuities. That latter question could not be answered by speculation as to what future parents might offer by way of non-contractual financial support, but only by looking at the companies’ compliance with the current and future regulatory regime.

We should emphasise that we are not saying that the judge was wrong to say at [178] that the court can, in exercising its discretion under section 111(3), take into account a wider set of factors than the actuarial ones that guide the analysis of the independent expert and the statute-based mandate of the Regulators. As regards security of policyholders’ benefits, however, those factors must go to the question of whether the Scheme will have a material adverse effect on those policyholders. Such matters may certainly include matters which fall outside the PRA’s assessment of the proposed scheme, but do not include speculation as to what future owners of an insurer may or may not wish non-contractually to do in the future to support their regulated subsidiary.

For the same reasons, the judge did not accord adequate weight to the conclusions of the independent expert that the risk of PAC or Rothesay needing external support in the future was remote. The judge should not have tempered his reliance on that opinion with his own speculation that Rothesay might require external financial support during the life of the annuities (as he did at [149], [154] and [181]). It is true, of course, that Rothesay, like any insurer, might at some stage during the life of annuities suffer a material deterioration in its balance sheet and require external financial support. But the whole thrust of the Solvency II regime is to evaluate those risks going forward and to ensure that steps are taken to mitigate them as and when they arise. The judge’s speculation about what additional parental funds might or might not be available in the future added nothing to the analysis undertaken by the PRA and the independent expert.

Under this issue, some argument was addressed to the question of whether the judge had been right at [155] to say that the availability to policyholders of the FSCS was irrelevant when the court was asking whether the Scheme would make a material change in the security of benefits for policyholders. This was not a ground of appeal, so the court is not strictly obliged to deal with the point. Nonetheless, we cannot see how the FSCS could have been relevant to what the judge had to decide in this case. The FSCS is a scheme of last resort, and is applicable equally to the holders of Transferring Policies before and after the transfer. As the judge suggested, if the existence of 100% protection for policyholders were a crucial factor, it would be hard to see why the expert needed to undertake a detailed analysis of the respective financial strengths of the transferor and transferee companies.

The reasoning at [90]-[106] above also resolves Issue 2. For the reasons already given, the judge did not give adequate weight to the Regulators’ lack of objection to the Scheme, and the continuing future regulation of Rothesay. As we have said, the judge was right at [178] and at [115] to point out that the court’s discretion was not constrained by the same actuarial analysis or regulatory criteria that shapes the approach of the expert and the Regulators. Full weight should, nonetheless, have been given to the fact that the PRA and the FCA had expressly considered the Scheme in the light of their statutory objectives, which include (a) assessments on a continuing basis of the suitability of Rothesay’s controllers, and (b) monitoring the safety and soundness of Rothesay going forward. (Footnote: ²⁸)

Accordingly, we conclude as follows on Issues 1 and 2:

This and the following issues are no longer of such significance now we have decided that the judge was wrong on the question of future parental financial support.

The appellants submitted that the judge wrongly attached weight to the fact that consumers had chosen Prudential for its age and reputation. The consumers may have been justified in doing so, but the court could not take that into account in the Part VII context. The court had access to the insurers’ detailed financial information and Solvency II metrics, and the opinions of experts and Regulators, which all provided a far more reliable guide to the security of policyholders’ benefits than any subjective factors which a policyholder may have considered prior to inception. The appellants

The objecting policyholders placed considerable reliance on the judge’s reasoning and on five main factors identified by Mr Isaacs as follows: (i) there were no Solvency II metrics for the decades during which annuities are paid, so that reputation and venerability cannot be a proxy for a metric which does not exist; (ii) since all insurers comply with Solvency II, policyholders must select their provider on the basis of other factors; (iii) whilst not all such factors are relevant to discretion (e.g. subjective dislikes), it did not follow that only actuarial factors are relevant, (iv) reputation and venerability was in this case a key aspect of how PAC sold its policies; its own materials recognised that the value of an annuity was not just the payments, but also the provision of peace of mind, and (v) even the expert recognised that the security of benefits only depended primarily on factors other than the level of prominence and age of the company, so that concepts of reputation and venerability fell to be measured subjectively by the judge.

As we have noted at [65], the judge disagreed with Warren J in Scottish Equitable at [115] that age and reputation were irrelevant. The court could, he said, take into account the reasonable choices made by policyholders, even though neither the PRA nor an actuary could place a value on a firm’s venerability or reputation.

This issue is, in our judgment, resolved against the objecting policyholders for similar reasons to those given under Issues 1 and 2.

As we have said, the court will not depart from the view of the expert or the nonobjections of the Regulators without a proper and relevant reason. In relation to security of policyholders’ benefits, the question is whether the Scheme would have a material adverse effect on the policyholders. If the policyholders’ prospects of being paid are essentially the same with and without the Scheme, it is hard to see how there can be any material adverse effect on the security of benefits caused by the Scheme.

It is for this reason, and the reasons given above, that we agree with what Warren J said in Scottish Equitable as follows:

“63.

[fairness] is not the subjective view of a policyholder or even of the judge. An objective view must be formed, a view reached against the objective standards and the factors appropriate to take into account. To take an extreme example, a scheme would not be unfair because it transferred business from a Scottish company to an English company even though a particular policyholder selected the company in the first place precisely because it was Scottish rather than English. …

114.

… Miss Hutchins emphasises the unfairness, as she sees it, of compelling her elderly father to transfer to a new company from the venerable SE which he deliberately chose. He wants to be given a choice, in particular to transfer to LGAS rather than to RL.

115.

There are two points to make. Firstly, the venerable position of SE is not, I am afraid, of itself a relevant factor. Even venerable institutions can fail as those who work in this area of the law are well aware. …

116.

Secondly, a newish body, that is to say, RL, is not to be regarded as an unsuitable provider simply because it is new otherwise we could never have new entrants into the market for transfers. The question is not its age but its financial strength, record and expectations”.

We have already explained why the judge was not justified in making an adverse comparison between the financial strength, record and expectations of PAC and Rothesay.

The subjective factors relied upon by the objecting policyholders are not, as we have said, relevant to be taken into account in the exercise of the court’s discretion.

So far as Mr Isaacs’ five main factors are concerned: (i) the Solvency II metrics taken together with ongoing regulation do indeed take account of the future prospects of an insurer, (ii) whilst it is true that all insurers must comply with Solvency II, the subjective choice of policyholders is not a relevant factor when a judge exercises the discretion under Part VII, and (iii) we do not say that only actuarial factors are relevant to the discretion the court exercises, but for any factor to be relevant, it must create some material adverse effect on policyholders in relation to their security of benefits. As to (iv) and (v), we accept that PAC sold its policies on the basis of its reputation, venerability and the purchaser’s putative peace of mind and we accept that at least some, perhaps many, policyholders reasonably chose to take their annuities from PAC by reference to these, among other, factors. We accept, as did the expert, that non-actuarial factors may be relevant in some cases, but subjective factors of this sort are not, given the extensive financial and actuarial evidence available to the court on this application.

For these reasons, we conclude on this issue that the judge ought not to have accorded any weight to the facts that the objecting policyholders (a) chose PAC on the basis of its age, venerability and established reputation, and (b) reasonably assumed that PAC would provide their annuity throughout its lengthy term.

On this issue, PAC and Rothesay submit first that the judge at [40] inappropriately limited the necessity to respect the commercial judgment of the transferor’s board to issues concerning the design of the scheme. In evaluating the Scheme, he had not adequately recognised: (i) the fact that PAC’s board had exercised its commercial judgment in concluding that the transfer was beneficial and appropriate, (ii) that the board had complied with its duties under section 172(1) to act in the way it considered “in good faith, would be most likely to promote the success of the company” having regard to a number of matters including the interests of the company’s employees, the need to foster the company’s business relationships with suppliers, customers and others, (iii) that judgment was exercised against the background that PAC was required to adhere to the FCA’s principles for businesses, which include (under principle 6) the requirement to pay due regard to the interests of its customers and to treat them fairly.

Put briefly, we do not think that the judge made the error contended for by PAC and Rothesay. The judge plainly understood the effect of Hoffmann J’s dictum in London Life and Evans-Lombe J’s first point at [6(i)] in Axa that the discretion had to be exercised according to principles which gave due recognition to the commercial judgment entrusted by the company’s constitution to its board. He realised that he was looking for a material adverse effect on the policyholders’ security of benefits.

It is once again important to have regard to the context in which Hoffmann J and Evans-Lombe J referred to the commercial judgment of the directors. In London Life, the entire business of London Life was to be transferred to an Australian mutual society, with the transferring with-profits policyholders necessarily losing their membership of London Life. It was not to any significant extent simply a transfer of actual or contingent liabilities of the sort typically found in ordinary policies of general insurance or life assurance or in annuities. The terms of transfer were a matter for detailed negotiation between the boards of the two societies, with each advancing the interests of their own with-profits policyholders, who were also the members of each society. There were no doubt many possible outcomes to those negotiations and it was the duty of London Life’s board to agree what they considered to be the best and fairest terms obtainable in the circumstances. Fairness was a vital element because the interests of different groups of London Life policyholders had to be balanced against each other. It is for these reasons that Hoffmann J said, correctly in the circumstances of that case, that the question was whether the scheme as a whole was “fair as between the interests of the different classes of persons affected” and that the choice as between different schemes and the detail of them were matters for the board and for negotiation between transferor and transferee.

Similarly, the commercial judgment of the directors was an important consideration in those cases where (a) with-profits policies were transferred as part of a process of demutualising a mutual society, and (b) a bank was deciding on an appropriate ringfencing transfer scheme. (Footnote: ²⁹)

In Axa, there was a different balance of interests to be struck. Longstanding withprofits funds had in many cases accumulated large “inherited estates”, representing surpluses in the fund not required for the expected level of reversionary and terminal bonuses. These estates were often referred to as 90/10 funds, because if a distribution were made from it, policyholders’ reasonable expectations would usually require that 90% was apportioned to policyholders and 10% to the insurer’s shareholder funds. The assets representing the with-profits fund, including the inherited estate, were beneficially the property of the insurer, not the with-profits policyholders. The entitlement of policyholders to 90% of any distribution was dependent on the insurer’s decision to make a distribution. The scope for a continued inexorable increase in the inherited estate, with only limited use being made of it, was obvious. The scheme proposed by Axa, and sanctioned by the court, was designed to reduce the inherited estate, which stood at £1.68 billion, in a mutually beneficial way. The issue for the court was whether the incentive payments proposed for electing policyholders represented a fair price for the contingent rights that they would give up. Axa’s proposal was the subject of intense scrutiny and discussion between Axa and the regulators and the independent actuary (as he was then known). The commercial judgment of the board of Axa was that the deal was fair, but that was, in truth, of limited value since their duty was to act in the best interests of Axa. It is to be noted that, although Evans-Lombe J referred in [6(i)] to the board’s commercial judgment as one of the factors to emerge from Hoffmann J’s judgment, he did not rely on it, or even again refer to it, in his reasons for sanctioning the scheme.

The court is entitled to assume that, in proposing a scheme, the directors of the transferor and transferee companies are acting in accordance with their statutory and other duties. If it became apparent that they were not doing so, it is very difficult to see that the court could contemplate sanctioning the scheme. In most circumstances, the commercial judgment of the directors has little, if any, further role to play. The facts of London Life illustrate a case where it did have a substantial role, but such cases have been comparatively rare and we see no real part for their commercial judgment in the court’s consideration of whether to sanction a scheme such as the present.

Had the judge been right about the relevance and existence of a disparity between the availability of non-contractual external financial support for transferor and transferee, even reasonable commercial decisions would have little, if any, part to play. The same applies had the judge been right to say that relevant factors included (a) the policyholders’ choice of PAC on the basis of its age, venerability and established reputation, and (b) the policyholders’ reasonable assumption that PAC would always provide their annuities. As we have held, however, the judge was not right on these two points. We do not think, however, that he made any error of law in his approach to the commercial judgment of the PAC’s board.

The appellants submitted that the judge was wrong to conclude at [182] that the detriments to PAC and Rothesay caused by the continuation of the Reinsurance Agreement did not constitute significant prejudice when set against the fundamental change in status and material disadvantage to transferring policyholders. The judge

Again, we can deal with this issue briefly. In our judgment, the judge was, quite correctly, looking for a material adverse effect on the policyholders’ security of benefits. Had he been right that the two main factors he identified did provide such a material adverse effect, we do not consider that in the circumstances of the present case any prejudice to the commercial interests of PAC or Rothesay would have been in point. Those commercial interests were unrelated to the interests of the transferring policyholders and, if the scheme had involved a material adverse effect to policyholders, we are unable to see why their interests should give way to the commercial interests of the companies and their shareholders. Since the Scheme was promulgated by the commercial parties knowing that it needed the sanction of the court, it does not seem to us that the kind of prejudice they pointed to would carry great weight if there had been shown to be a material adverse effect on policyholders.

In our judgment, the judge did not make any error of law in his approach to the question of the alleged prejudice that a refusal to sanction would cause to PAC and Rothesay.

As we have said, the judge made errors in his approach to the exercise of his discretion as to the sanction of the Scheme under section 111(3). He ought not to have concluded that there was a material disparity between the non-contractual external support potentially available for each of PAC and Rothesay. In any event, such a disparity was not a material factor. Moreover, he failed to accord adequate weight to the expert’s conclusion that the risk of PAC or Rothesay needing external support in the future was remote, to the Regulators’ lack of objection to the Scheme, and to the continuing future regulation of Rothesay. Finally, he ought not to have accorded any weight to the fact that the objecting policyholders chose PAC on the basis of its age, venerability and established reputation, and reasonably assumed that PAC would always provide their annuities.

These errors mean that the judge’s exercise of his discretion cannot stand. As Patten LJ ordered on 22 June 2020, in the event of the judge’s decision being set aside, the question of whether the Scheme should be sanctioned would be remitted to the High Court. The judge hearing the renewed application for sanction will take into account the determination of this court as to how the discretion under section 111(3) is to be exercised.

For the reasons we have given, this appeal will be allowed, and the matter will be remitted to a judge sitting in the Insolvency and Companies List of the Business and Property Courts of England and Wales. Without any disrespect to Snowden J, we think it would be preferable for the renewed sanction hearing to be heard by another judge of the Chancery Division.