Manchikalapati & Ors, R (on the application of) v The Financial Services Compensation Scheme

This is an appeal against a decision of Mr Dexter Dias KC, sitting as a Deputy High Court Judge, to grant an application for judicial review against the Financial Services Compensation Scheme (the “FSCS”). The judge’s decision concerned a refusal by the FSCS to compensate the respondent policyholders (the “Policyholders”) for an insolvent insurer’s failure to meet its liabilities to the Policyholders for post-judgment interest and litigation costs. Compensation was declined on the basis that those amounts do not fall within the scope of the scheme set out in the Policyholder Protection Rules (the “PPR”).

The highly unfortunate factual background can be summarised as follows. The Policyholders are a group of individuals and estates of deceased persons who own long leases in a development known as New Lawrence House in Hulme, Manchester (the “Development”). The leases were acquired between 2007 and 2010, some on an “off-plan” basis. On purchasing the leases the Policyholders were each issued with an insurance policy by Zurich Insurance plc (“Zurich”) which was intended to cover structural defects (the “Policies”). The Policies provide a type of insurance referred to in the PPR as “building guarantee insurance”. Although such insurance is not strictly compulsory for a new build property, in practice lenders are likely to require it when advancing a mortgage. Further, as in this case the policyholders tend not to select the insurer, because the policies are procured by the developer and provided to policyholders as properties are sold.

The Development suffered from serious defects which led to the Policyholders making substantial claims under the Policies. Zurich declined the claims. There was also a dispute about who controlled the development, which was litigated to the Court of Appeal (Sugarman v CJS Investments LLP [2014] EWCA Civ 1239, [2015] 1 BCLC 1), but the entity that was found to be in control then went into administration.

Proceedings were issued against Zurich in March 2015. During the course of the proceedings, in March 2018, there was an insurance business transfer scheme in Ireland under which Zurich’s liabilities were transferred to another entity, East West Insurance Company Ltd (“EWIC”). The litigation culminated in a Court of Appeal decision in favour of the Policyholders: Manchikalapati v Zurich Insurance Plc and East West Insurance Company Ltd [2019] EWCA Civ 2163, [2019] 187 ConLR 62 (“Manchikalapati 2019”). The final result was a consent order in favour of the Policyholders for £9,728,420.63 (exclusive of VAT) and a costs award of 92.5% of the Policyholders’ costs. EWIC paid the £9,728,420.63 but then itself went into administration without paying the VAT, statutory interest on the judgment debt or (apart from a sum paid on account) costs.

The Policyholders sought compensation from the FSCS, which agreed that the VAT element was covered by the PPR but declined to pay compensation for EWIC’s other defaults. The FSCS’s final decision was made in a letter dated 6 May 2021 (the “Decision”). The basis for the Decision was essentially that the claims for interest and costs were not claims “under” the Policies but were claims pursuant to statute and a court order respectively (the statute being s.17 of the Judgments Act 1838, under which interest accrues on judgment debts). The Decision quantifies the interest at £784,331.88. The unpaid costs are said to be in the region of £3,283,000.

The Policyholders were granted permission for judicial review and succeeded before the judge, who quashed the Decision. The judge’s order is stayed pending the outcome of this appeal. In the meantime, the Development remains uninhabited because it is too unsafe to occupy.

I should say at the outset that I have a great deal of sympathy for the Policyholders, who have been placed in a very difficult position. They have been involved in very lengthy litigation including three trips to the Court of Appeal. Their rights under the Policies were finally vindicated in a judgment where Coulson LJ deprecated the approach of the insurers (see Manchikalapati 2019 at [82]), only for EWIC to become insolvent and for the FSCS not to cover the sums left outstanding, much of which represented expenditure which the Policyholders had to incur to establish that the insurers were in the wrong. In the meantime the Development remains unusable.

The FSCS is a body corporate that was established under Part XV of the Financial Services and Markets Act 2000 (“FSMA”) by what was then the Financial Services Authority (the “FSA”). In 2012 the FSA’s regulatory functions were split. The Bank of England through the Prudential Regulation Authority (the “PRA”) assumed responsibility for regulating financial services firms including insurers. As part of this the PRA took over responsibility for the compensation schemes for deposits and insurance. The FSA was renamed the Financial Conduct Authority (the “FCA”).

Section 2C of FSMA, inserted by the Financial Services Act 2012 (“FSA 2012”), establishes the PRA’s “insurance objective” in the following terms:

“2C Insurance objective

(1)

In discharging its general functions so far as relating to a PRA-regulated activity relating to the effecting or carrying out of contracts of insurance or PRA-authorised persons carrying on that activity, the PRA must, so far as is reasonably possible, act in a way—

(a)

which is compatible with its general objective and its insurance objective, and

(b)

which the PRA considers most appropriate for the purpose of advancing those objectives.

(2)

The PRA’s insurance objective is: contributing to the securing of an appropriate degree of protection for those who are or may become policyholders.

…”

Section 212 of FSMA provides for the establishment of the FSCS as the “scheme manager”. The compensation scheme itself was established pursuant to s.213, which required the regulators to establish it by rules made in accordance with that section. As now in force, s.213(1) provides:

“(1)

The regulators must by rules made in accordance with an order under subsection (1A) establish a scheme for compensating persons in cases where—

(a)

relevant persons are unable, or likely to be unable, to satisfy claims against them,

…

(b)

persons who have assumed responsibility for liabilities arising from acts or omissions of relevant persons … (“successors”) are unable, or likely to be unable, to satisfy claims against the successors that are based on those acts or omissions.”

Subsection (1A) requires the Treasury to specify by order the circumstances in which the FCA and the PRA respectively can make rules under subsection (1). “Relevant persons” are, essentially, authorised persons including insurers. Section 213(2) provides that the rules so made (taken together) are referred to in FSMA as “the compensation scheme”.

Section 213(3) relevantly provides:

“(3)

The compensation scheme must, in particular, provide for the scheme manager—

(a)

to assess and pay compensation, in accordance with the scheme, to claimants in respect of claims made in connection with—

(i)

a regulated activity carried on (whether or not with permission) by relevant persons...”

Section 213 also empowers the FSCS to impose levies on industry participants to fund the costs of the scheme.

Section 214 makes further provision, including to the effect that restrictions may be imposed on the kind of claims that may be entertained under the scheme and from what kinds of claimant, limits may be applied to the amounts payable and distinctions may be drawn between different kinds of claim. Section 214 relevantly provides:

“(1)

The compensation scheme may, in particular, make provision—

…

(f)

for a claim to be entertained only if it is made by a specified kind of claimant;

(g)

for a claim to be entertained only if it falls within a specified kind of claim;

…

(j)

limiting the amount payable on a claim to a specified maximum amount or a maximum amount calculated in a specified manner;

…

(2)

Different provision may be made with respect to different kinds of claim.”

Section 138G of FSMA, also inserted by FSA 2012, makes general provision in relation to the power of regulators to make rules under the Act. It provides that the power is exercisable in writing and that the instrument by which rules are made must be published.

It can be seen from the above that, although s.213(3) mandates a scheme to pay compensation in respect of claims made “in connection with” a regulated activity, ss.213 and 214 of FSMA in fact provide the PRA with a considerable degree of leeway as to the scope and extent of the compensation scheme. This is subject to the proviso that, in the context of insurance, the PRA must act in a way that is compatible with its insurance objective, that objective being “contributing to the securing of an appropriate degree of protection” for policyholders (s.2C).

The PPR is the set of rules that deals with insurance-related compensation. It forms part of the PRA Rulebook, under the title “Policyholder Protection”. The PRA has published an online version of its consolidated rulebook, but explains in a “FAQ” section that the definitive rules are contained in the rule-making instruments. It can be seen from those instruments that the rules were introduced in their current form in 2015 and have since been amended in various respects.

The most relevant provisions of the PPR as now in force are set out in the Appendix to our judgments. There are a number of defined terms. Most of the relevant definitions are in set out in rule 1.2 of the PPR, but some are contained in the PRA Rulebook Glossary, a separate list of defined terms that apply to the PRA Rulebook as a whole.

One point to note is that both the online consolidated version of the rules, and the definitive provisions in the rule-making instruments, show defined terms in italicised form, not only in the definitions section but throughout the rules. (The consolidated version also includes hyperlinks to the definitions.) While I do not consider that the scheme of italicisation is necessarily accurate in all cases, it is of some assistance in illustrating how the rules are intended to work. The italics are not replicated in the Appendix, but I have included them in citations in the body of this judgment.

In outline, the scheme of the rules is as follows. Rule 2.1 provides for the administration by the FSCS of the compensation scheme for contracts of insurance (the “policyholder protection scheme”). Rule 3 contains the main qualifying conditions for compensation. The scheme applies to “eligible claimants” and “protected contracts of insurance”, defined in rules 7 and 9.2 respectively. A “claim under a protected contract of insurance” is a “protected claim” (see rule 9.1, cross-referred to in the glossary, and the definition of “claim” in rule 1.2). It is not in dispute that the Policyholders are eligible claimants and that the Policies are protected contracts of insurance.

Rule 3 provides for compensation to be payable to an “eligible claimant” who makes a claim “in respect of” a “protected claim” against a “relevant person” or a “successor” who is in default. This is subject to conditions to the effect that the FSCS can require an assignment or subrogation in its favour and that the FSCS is satisfied that alternative steps to safeguard the position are not reasonably practicable or appropriate. In this case Zurich, as issuer of the Policies, was the “relevant person” and EWIC is its “successor”. It is not in dispute that EWIC is in default and there are no relevant alternative arrangements.

While rule 3 appears to be permissive (3.1 provides that the FSCS “may” pay compensation) it is common ground that the FSCS has no general discretion to withhold compensation if the terms of the rules otherwise provide for it. This can be seen from rule 16.1, rule 18.1 and rule 20, which all refer to “must pay”.

Rules 16 to 20 contain important provisions that determine the timing and amount of compensation. The maximum limit for certain liabilities, including those in respect of compulsory insurance and those in respect of building guarantee insurance, is 100% of the claim, whereas the limit for other claims related to general insurance contracts is 90% (rule 17). Subject to those limits, under rule 19.1 the amount payable is the amount of the “overall net claim” at the quantification date (a date selected by the FSCS for determining the liability of the relevant person or successor). This is the sum of the protected claims that the claimant has in the same category, less any liability that the relevant person or successor could have set off (rule 19.3). Rule 20 provides for calculation and payment.

The 100% limit for building guarantee policies reflects a relatively recent rule change. A PRA Policy Statement issued in October 2020 announcing the change (the “2020 Statement”) explained that the limit was being increased from 90% against the background of developments that included the Grenfell fire tragedy and its consequences, such that the PRA now considered that “some features of BGP are similar to the policies for which 100% cover is available”, including the feature that insurance is likely to be a necessary requirement of any mortgage lender. Further:

“… once known defects exist in a property covered by a BGP, the loss of BGP protection in the event of the subsequent failure of the insurer may have significant potential adverse financial and social consequences for policyholders, meaning that 100% FSCS coverage for protected policyholders represents a more appropriate degree of protection.”

The statement went on to explain that the potential adverse consequences included an inability to remortgage or sell and the “significant amount of money” that a 10% shortfall could represent. There were also potential wider market consequences. The PRA considered that the “benefits to policyholder protection” outweighed any consequential exposure to higher levies for firms.

The Policyholders challenged the Decision on three alternative grounds. First, they said that their claim fell within rule 3.1(2) because the claim for interest and costs was “in respect of” a protected claim, even if it was not itself a protected claim. Secondly, the claims for interest and costs in any event fell within the concept of “protected claim” because that definition incorporates the phrase “in respect of” via the definition of “claim”. Thirdly, the claims were owed “under” the contract of insurance and fell within the concept of “protected claim” on that basis.

The judge concluded the first ground in favour of the Policyholders but rejected the second and third. Most of the judge’s analysis relates to the first ground. In considering that ground the judge discussed the objective of the scheme and accepted that the court should not distend the scope of the scheme to cater for hardship that falls on the wrong side of the line actually drawn by the rules (paragraph 60). The judge considered that the decisive question was the meaning of “in respect of” in the definition of “claim” in rule 1.2, with the FSCS contending that it meant “for payment of” and the Policyholders saying it meant “in connection with” (paragraph 62).

The judge noted that the 2020 Statement referred to the failure of an insurer providing building guarantee policies as justifying an immediate change rather than an initial consultation phase and said that this was a reference to the failure of EWIC, a point also confirmed to us in oral submissions by Mr Handyside KC for the FSCS. The judge rejected a submission by Mr Drake KC (for the Policyholders) that it would be absurd to increase cover in this way on an urgent basis if Policyholders remained exposed to the much greater impact of unpaid costs and interest, but thought that the initiative was nonetheless relevant in casting light on the purpose of the scheme.

The judge went on to consider the case of R (Geologistics Ltd) v Financial Services Compensation Scheme both in the High Court ([2003] EWHC 629 (Admin), [2003] 1 WLR 1696) and in the Court of Appeal ([2003] EWCA Civ 1905, [2004] 1 WLR 1719) (“Geologistics”). While he accepted that that case related to a predecessor scheme with different rules, he concluded that it had persuasive force, that there was a great similarity with this case and that it gave a “valuable steer” (paragraphs 91, 103, 106 and 116). The interpretation of “in respect of” advanced by the FSCS appeared strained and would mean that the words added very little (paragraph 93). In Geologistics the Court of Appeal had concluded that a claim for costs was compensable because it was integrally connected with a liability covered by compulsory insurance. Similarly, costs and interest here were integral to a protected liability which was materially analogous to compulsory insurance. The judge suggested a taxonomy that involved asking, at stage 1, what liability was protected under the scheme and, at stage 2, asking whether any further claim is “integral to, part and parcel of or sufficiently connected to” the compensable liability (paragraph 100). In this case the costs and expenses satisfied that second stage test. The judge noted that, given the intransigent and stubborn attitude of the insurers, the sums due under the Policies would in all likelihood not have been recovered but for the litigation costs. It struck the judge as an absurdity if it was not possible to recover costs necessarily and integrally incurred to recover what was owed, and the same applied to judgment interest (paragraph 109).

The judge then tested his conclusions against the provisions of FSMA and the PPR, referring again to the increase in cover to 100% in 2020 and to other provisions of the rules. He considered that rule 2.3, which permits payment of reasonable costs of insolvency proceedings, was consistent with his conclusion. The definition of protected claim in rule 9.1 was also not inconsistent with it. Rule 9.8, which deems certain claims to be under a protected contract, did not assist. Further, none of rules 17, 19.3 and 20 affected the definition of “claim”.

In relation to the second and third grounds, the judge concluded that the costs and interest were not themselves protected claims and that they did not arise “under” a contract of insurance (paragraphs 127 and 128), so that both of those grounds failed.

The grounds of appeal can be summarised as follows:

The FSCS say that the judge failed to have sufficient regard to other provisions of the PPR and to the operation of the scheme as a whole, and was wrong to place weight on Geologistics.

The Policyholders have filed a Respondents’ Notice maintaining that the compensation sought falls within rule 3.1(2) because it is “in respect of” a protected claim and saying that the judge’s order should alternatively be upheld for reasons corresponding to the second and third grounds rejected by the judge.

For convenience, I will refer to the arguments in a similar way to the judge, namely:

The judge relied on conventional principles of statutory interpretation. While in broad terms I consider that to be the correct approach, there are qualifications. First, the rules are made by the PRA. Parliament has no direct role in the creation or amendment of the PPR, and there is no requirement to lay them before Parliament under any form of procedure (whether affirmative or negative). Beyond what is laid down in FSMA as to the overall requirements of the scheme, it therefore does not make sense to analyse the PPR in terms of Parliamentary intention. Having said that, the rules are required to be made pursuant to FSMA, which also sets out some parameters of the scheme. It is also worth noting that the definition of “subordinate legislation” in the Interpretation Act 1978 is broad enough to apply to rules such as the PPR, such that its provisions may apply: see ss.21 and 23.

Secondly, although the PPR are obviously intended to create legally enforceable rights and obligations for the FSCS, insurers and policyholders alike, it is important to bear in mind that they are drafted by a regulator for use both by policyholders and by industry participants as well as by the FSCS. They are not drafted by Parliamentary Counsel or even by a Government department in the way that some statutory instruments are. They are intended to be comprehensible by a non-lawyer and to produce a workable scheme. They should be construed with that in mind.

Subject to those qualifications, it must be right to apply broadly similar principles to those applicable in construing a statute, substituting for Parliamentary intention the intention of the body charged with making the rules, read in the light of the statutory scheme pursuant to which the rules are made. I note that this is effectively the approach put forward in Bennion, Bailey and Norbury on Statutory Interpretation, 8th ed. (“Bennion”) at 3.17 in relation to what it compendiously describes as delegated legislation. As explained in Bennion at 11.1, in conducting statutory interpretation the primary focus must be on the words used, read in their context and having regard to the purpose of the provisions. Applied to the PPR, the question is what, objectively, the rule maker must be taken to have intended by the words used in their context, having regard to the purpose of the rules and in the light of the relevant provisions of FSMA.

I would add that, in this case, it is also important to bear in mind that the scope of the rules has direct implications not only on the rights of claimants but on the position of others, in particular in relation to levies imposed on insurers to fund the scheme.

In Official Receiver v Shop Direct Finance Company Ltd [2023] EWCA Civ 367 this Court recently considered the interpretation of rules that govern the referral of complaints to the Financial Ombudsman Service. Those rules form part of the FCA Handbook and are also made under powers conferred by FSMA. Singh LJ, with whom both Carr LJ and Nugee LJ agreed on this issue (see Carr LJ’s judgment at [105] and Nugee LJ’s judgment at [113]), considered comments by Lord Neuberger MR and Arden LJ in the earlier decision of this Court in Re Lehman Brothers International (Europe) (No 2) [2010] EWCA Civ 917, [2011] 2 BCLC 184 and derived the following propositions from them at [46]:

“(1)

Ultimately it is the actual wording of a provision that must govern any decision as to its effect.

(2)

The Handbook should be read as a whole, taking an holistic and iterative approach, so that a preliminary view on one provision can be tested by reference to the rest of the relevant provisions.

(3)

The provision should be construed in the light of its overall purpose.

(4)

It should be construed on the basis that it is intended to produce a practical and commercially sensible result. The rules should be taken to be grounded in reality. The court should keep in proportion any drafting infelicities.”

Each of these points may be applied equally to the PPR.

A further point relates to the relevance of the 2020 Statement. In the same way that an explanatory note may be used “to ascertain the context of the provision and the mischief which it addresses as aids to purposive interpretation” (McDonald v Newton [2017] UKSC 52 per Lord Hodge at [30]; Bennion at 24.24), I consider that we may have recourse to the 2020 Statement as a contemporaneous document that provides assistance in determining the context of the change in the rules in respect of building guarantee insurance and the mischief at which the change was aimed. I further note that, in R (on the application of PACCAR Inc & others) v Competition Appeal Tribunal and others [2023] UKSC 28 at [42], Lord Sales also referred to the potential use of an explanatory note to resolve a specific ambiguity, as well as informing the assessment of overall purpose.

Significant reliance was placed on Geologistics by Mr Drake in his submissions to us, including that the judge was correct to rely on it but in fact underplayed its importance, so it is appropriate to consider it in some detail.

Geologistics concerned compulsory employers’ liability insurance. Under s.1 of the Employers’ Liability (Compulsory Insurance) Act 1969 (the “1969 Act”) employers were required to “insure … against liability for bodily injury or disease sustained by [their] employees”. The claimant, Geologistics, was insured by Independent Insurance Company Ltd (“Independent”). The policy also provided non-compulsory insurance, including for the costs of defending claims brought by employees. An employee, Mr Froggatt, was injured and brought a claim against Geologistics which ultimately succeeded. During the course of the proceedings Independent entered into provisional liquidation. Geologistics sought to recover its costs of defending the claim from the FSCS, as successor body to the Policyholders Protection Board, under the provisions of the Policyholder Protection Act 1975 (the “1975 Act”).

Section 6 of the 1975 Act provided:

“(4)

Subject to sections 9, 13 and 14 below and the following provisions of this section, it shall be the duty of [the FSCS] to secure that a sum equal to the full amount of any liability of a company in liquidation towards any policyholder or security holder under the terms of any policy or security to which this section applies is paid to the policyholder or security holder as soon as reasonably practicable after the beginning of the liquidation.

(5)

Subsection (4) does not apply by reference to any liability of a company in liquidation under the terms of a policy to which this section applies arising otherwise than in respect of a liability of the policyholder which is a liability subject to compulsory insurance.” (Emphasis supplied.)

Under s.6(1), the policies to which s.6 applied included those satisfying s.1 of the 1969 Act. Section 6(3) provided that “liability subject to compulsory insurance” meant any liability required under any of the enactments mentioned in s.6(1) to be covered by insurance.

The central issue was whether Independent’s liability to indemnify Geologistics for the costs it incurred in defending Mr Froggatt’s claim was “in respect of” a liability which was subject to compulsory insurance. It had been accepted that compensation was available in respect of any damages and costs that Geologistics had been found liable to pay Mr Froggatt. The dispute related to Geologistics’ own defence costs.

At first instance, Davis J accepted the submission that the costs met this test because they were “connected” with the claim brought by Mr Froggatt, and it was only by reason of that claim that the costs were incurred at all. The costs were “part and parcel” of the claim and were integrally linked to the compulsorily insured liability (see at [33] and [41]).

The Court of Appeal upheld Davis J’s decision. Waller LJ commented at [7] that the 1975 Act was concerned with the protection of policyholders. The fact that a 100% indemnity was provided and that corporate policyholders were included was consistent with the notion that “all policyholders ought to be protected completely”. Having considered some authorities and concluding at [17] that he did not gain much assistance from them because the proper construction of the words depended on their context, he returned to the wording of s.6(4) and (5), noting that the FSCS’s construction, as put forward on appeal by Sir Sydney Kentridge QC, would exclude not only defence costs but other items, potentially including costs awards made against the employer.

Waller LJ went on to say this at [23]:

“23.

In any event, it seems to me that the narrow construction placed on section 6(5) by Sir Sydney is inconsistent with there being the two subsections. If section 6(4) and (5) were intended to provide an indemnity against only that which was required to be the subject of compulsory insurance, section 6(4) could have so provided without the need for section 6(5). That alone supports the view that the words “otherwise than in respect of a liability of the policyholder which is a liability subject to compulsory insurance”, must be intended to produce the result that what the policyholder can recover under section 6(4) goes beyond the liability which must be compulsorily insured. What is contemplated is therefore that under a policy which is required to be taken out, the policyholder will be entitled to recover against the insurance company some indemnity beyond that for which statute compels insurance, but by virtue of section 6(5) that right to indemnity must still be ‘in respect of’ the ‘liability subject to compulsory insurance’.”

Waller LJ considered the policy wording, which covered both claimants’ costs “in connection with” and defence costs “in relation to” any claim. As regards claimants’ costs, he construed the words “liability subject to compulsory insurance” as descriptive of the type of liability covered by the policy as opposed to describing an established liability, and “in respect of” as “in its context intended to mean or at least include ‘in connection with’’”. He concluded that the position had to be the same as regards defence costs, construing “in respect of” as meaning “in connection with” or “in relation to” (see at [25], [26] and [28]). In reaching that conclusion he rejected an alternative argument by Sir Sydney Kentridge that attempted to distinguish claimants’ costs awards and defence costs. He explained at [29] that s.6(5) was intended to make clear that the compensation scheme did not extend to matters which were not the subject of compulsory insurance or costs incurred in respect of the same, but were nonetheless covered by a policy which also covered matters for which insurance was compulsory.

As the judge in this case recognised, Geologistics concerned different legislation. Just as Waller LJ did not find material assistance from earlier case law to which he was referred, I have not found Geologistics to be of particular assistance in resolving this case. Leaving the factual differences to one side for a moment, a significant point in Geologistics was the existence of the separate provisions in s.6(4) and (5) and the particular wording used. Waller LJ concluded that if the intention had been to confine protection to liability subject to compulsory insurance then either s.6(4) would have contained different language and s.6(5) would not have been included, or s.6(5) would have been worded differently. On its terms, s.6(4) covered all liabilities under a policy which provided compulsory insurance cover, whether or not the liability was related to the compulsory insurance or some other matter. Section 6(5) was required to reduce its scope, the question being how far it did so. Further, the analysis that permitted compensation for awards of claimants’ costs relating to liabilities that were compulsorily insured (which the FSCS had accepted were covered) had to apply in the same way to defence costs. There was no relevant distinction. Finally, unlike rule 3.1(2) of the PPR, the provisions being construed in Geologistics did not form part of a body of rules which, as I will endeavour to explain, contain other strong indicators as to the correct interpretation of “in respect of” where it appears in that rule.

The factual differences between Geologistics and this case are also important to bear in mind. In Geologistics the costs being claimed were insured risks which the failed insurer was obliged to indemnify under the terms of the policy. That is not so in this case. This was a critical feature of the Decision: see [5] above.

Mr Drake placed particular reliance on the so-called Barras principle, derived from Barras v Aberdeen Steam Trawling & Fishing Co Ltd [1933] AC 402. The principle is considered in Bennion at 24.6, which summarises it as follows:

“(1)

Where an Act uses a word or phrase that has been the subject of previous judicial interpretation in the same or a similar context it may be possible to infer that the legislature intended the word or phrase to bear the same meaning as it had in that context. This is sometimes known as the Barras principle.

(2)

This is at most a presumption the strength of which will vary according to the context: there is no rigid rule that words must be given the same meaning that they have been given in an earlier Act. The question in the end is always whether the legislature intended the term to be given the meaning it has been given previously.”

In WB v WC [2019] QB 625, Arden LJ relied on the Barras principle in a homelessness case where Parliament had successively built on the same legislation (see at [27]). She gave a number of examples of cases where the principle had been applied, and it had been considered by the Supreme Court in R (N) v Lewisham London Borough Council [2015] AC 1259 without its existence being doubted. Arden LJ noted at [28] that:

“The strength of the principle in any given case will depend on whether the context is the same and whether there are differences in phraseology and so on.”

The Policyholders maintain that the Barras principle applies in this case, such that there is a presumption that the phrase “in respect of” in rule 3.1(2) of the PPR is intended to bear the meaning accorded to it in Geologistics when interpreting the predecessor legislation. The compensation scheme was built upon without any relevant change in language on the part of the FSCS. That must be assumed to have been a deliberate decision.

I do not agree that there is a sufficient similarity between the legislation considered in Geologistics and the PPR, or between the subject matter of this case and that in Geologistics, to justify the application of any presumption. This is so even assuming (as I am prepared to do for the purposes of this case) that the Barras principle could apply to rules such as the PPR, when comparing them to the terms of earlier legislation. As already explained, Geologistics related to the interpretation of legislation which is in different terms to the provisions of the PPR that the Policyholders rely on. That legislation provided for compensation for amounts due from and unpaid by insurers in liquidation, being amounts due under the terms of a policy, but narrowed the scope of protection to amounts that related to liabilities that were compulsorily insured.

In this case the question is different. Rather than apparently broad coverage for insured liabilities which is then cut down, the question here is whether the scope of the compensation scheme, which otherwise covers claims under certain policies, extends beyond that to encompass costs awards and interest payable by the insurer under the terms of a court order and statute respectively, and not under the terms of the policy. Both the structure of the legislation considered in that case and the subject matter of the claim are materially different from the relevant provisions of the PPR and the facts of this case.

Before discussing the grounds in more detail I would make two preliminary points. First, although they are alternative grounds they are interrelated. A proper analysis of ground 1 also requires the resolution of grounds 2 and 3. With that in mind I have found it convenient to address the grounds in reverse order.

Secondly, while grounds 1 and 2 are closely related, in that both focus on the words “in respect of”, they are distinct. Ground 1 is directed at the use of those words in rule 3.1(2). Ground 2 is directed at their use in the definition of “claim” in rule 1.2 and how that fits into the concept of “protected claim”. However, the judge’s analysis in relation to ground 1 focused on the definition of “claim” in rule 1.2. It is undisputed that this was an error.

Ground 3 is whether the costs and interest were owed “under” a contract of insurance, such that they fell within the definition of “claim”. If they were so owed then compensation would be available in respect of them. This is because it is common ground that the Policies are “protected contracts of insurance”, and each Policyholder’s claim to costs and interest would then be a “claim under a protected contract of insurance”, and thus a “protected claim” within rule 9.1.

Using the italicisation as it appears in the definitive version of the rules (see [17] above), rule 1.2 defines a “claim” as:

“a valid claim made in respect of a civil liability:

(1)

owed by a relevant person; or

(2)

owed by a relevant person which has been assumed by a successor and which is based on the acts or omissions of relevant person;

under a contract of insurance.”

Mr Drake submitted that EWIC’s liability for costs and interest is owed “under” the Policies because it is owed pursuant to an insurance relationship rather than an extraneous non-insurance relationship. He referred to the distinction drawn in FSMA between regulated and other activities, with “effecting” and “carrying out” contracts of insurance being regulated activities (s.22 FSMA and article 10 of the Financial Services and Markets Act 2000 (Regulated Activities) Order 2001, SI 2001/544). The source of the liability is a regulated activity. The compensation scheme is intended to draw a distinction between liabilities owed to policyholders and other liabilities, rather than to sub-divide the former into elements that are and are not compensable. Mr Drake also relied on the use of the words in “connection with” in s.213(3) of FSMA. “Under” should be construed as “arising out of”, so as to ensure that policyholders received the full protection they were intended to have. If it was construed as confined to contractual liabilities then the reference to “civil” liability in the definition of claim would be redundant.

I have no doubt that the judge was right to conclude that the costs and interest were not owed “under” a contract of insurance. They are amounts due pursuant to a court order and pursuant to statute. The natural and ordinary meaning of “under” the contract means just that. It would cover amounts owed under the terms of the contract, but not other amounts the entitlement to which derives not from the contract but from some other source, such as a court order or statute. It cannot in my view be distorted to encompass such other amounts, and the inclusion of a reference to “civil liability” does not render that permissible. Further, the words “in connection with” in s.213(3) of FSMA must be read in the context of both the explicit power under s.214 to limit the types of claim that may be entertained, and the scope given to the PRA in s.2C to act in the way it considers “most appropriate” for advancing its objective of “contributing to the securing of an appropriate degree of protection” for policyholders. There is no reference to securing full protection, and it is clear from the scheme of the legislation that that is not what was contemplated.

If further support were needed, it can also be gained from other provisions of the rules considered under ground 1 below. For example, rule 9.8 (see [89] below) illustrates that the rule maker considered that a claim “under” a contract of insurance means what it says.

Ground 2 is whether the claims for costs and interest fall within the concept of “protected claim”. It relies on the fact that the definition of protected claim incorporates the definition of claim, and adopts an interpretation of that definition that is different to the one that is assumed for the purposes of ground 3.

The concept of protected claim is obviously central to the PPR. The PRA Rulebook glossary states that a “protected claimmeans a claim which is covered by the policyholder protection scheme as defined in Policyholder Protection 9.1”. As this indicates, the actual definition is in rule 9.1, which provides: “A protected claim is a claim under a protected contract of insurance”.

Mr Drake submitted that, reading in the definition of claim in rule 1.2 and ignoring the resulting duplicative reference to a “contract of insurance”, a protected claim is therefore “a valid claim made in respect of a civil liability owed by a relevant person … under a protected contract of insurance”. Interpreting “in respect of” as “in connection with”, in the same way as it is submitted should be done under ground 1, Mr Drake submitted that the claim for costs and interest is therefore “in respect of” a civil liability owed by EWIC under a contract of insurance.

I do not agree. First, the Policyholders’ construction requires the words “owed by a relevant person … under a contract of insurance” in the definition of claim to qualify “civil liability” rather than “claim”. In my view the more natural interpretation of the definition – and indeed the one that is assumed to be correct for the purposes of ground 3 – is that it applies to amounts owed by an insurer (by way of a civil liability) under a contract of insurance. In other words, the definition catches valid claims for a civil liability owed under the contract.

Secondly, and more fundamentally, the argument does not properly reflect rule 9.1. That is the provision that defines “protected claim”, and it expressly applies (and applies only) to claims “under” the relevant contract. The incorporation of the words “in respect of” via the definition of claim cannot overcome that clear, and obviously intended, limitation. Indeed, the clear wording in rule 9.1 lends weight to what I in any event consider to be the more natural interpretation of the definition of claim, namely that it captures claims for amounts owed under a contract of insurance.

Given that the costs and interest were not owed “under” the Policies, those amounts therefore cannot fall within the definition of “protected claim”. The judge was therefore correct to reject ground 2, albeit that he did not expressly address the Policyholders’ arguments about reading in the definition and the potential contradiction with his approach to ground 1.

There are two further points to make before leaving ground 2. First, what I consider to be the correct interpretation of the definition of “claim” requires the words “in respect of” to be read as having the narrower meaning contended for by the FSCS, namely as “for” or “for the payment of”. Although the language of rule 3.1(2) is not the same, there is a sufficient similarity in the way the words are used to provide some indication that they should not be interpreted differently. This is relevant to ground 1, and it also means that other provisions of the rules which are considered below in relation to ground 1 provide further assistance in reinforcing a narrow interpretation of the definition of claim.

Secondly, I would observe that the choice of language for the defined term “protected claim” gives a strong indication that the sorts of claims intended to be “protected” by the scheme are claims for the payment of amounts due under a contract of insurance. Other claims against insurers cannot be described as “protected” claims. The way that “protected claim” is referred to in the PRA Rulebook glossary provides some further support for this, by indicating that what is “covered” by the policyholder protection scheme is “protected claim(s)” as defined in rule 9.1, rather than anything else. This is relevant to ground 1, to which I will now turn.

Ground 1 is whether the Policyholders’ claim to be compensated for the unpaid costs and interest falls within rule 3.1(2) because it is “in respect of” a protected claim, even if it is not itself a protected claim.

Rule 3.1 provides:

“The FSCS may pay compensation to an eligible claimant, subject to 18, if it is satisfied that:

(1)

an eligible claimant has made an application for compensation…

(2)

the claim is in respect of a protected claim against a relevant person (or where applicable, a successor) who is in default…”

One of the rather confusing aspects of the PPR is that the concept of “claim” is used in more than one sense. As defined in rule 1.2 it relates, as already discussed, to a claim that a policyholder has under a contract of insurance, that is, a claim against the insurer for an amount that is owed under the terms of the policy. But the word is also used to mean a claim for compensation under the compensation scheme. This is evident both from FSMA, which refers to “claimants” and “claim” in the latter sense in ss.213(3) and 214, and from the PPR. In particular, it is common ground that the first reference to “claim” in rule 3.1(2) is to claim in that second sense. This is reflected in the lack of italicisation.

I have already referred to the use of the words “in respect of” in the definition of claim and the indication that provides in favour of a narrower interpretation of those words where they are used in a similar manner in rule 3.1(2), and to the concept of “protected claim”. I would add that “policyholder protection scheme” is also defined in the glossary as “the compensation scheme for claims under contracts of insurance” (emphasis added), thus taking a similar approach to the definition of protected claim and providing an additional indicator of the intended subject matter of the scheme.

Other provisions of the rules provide additional support for the FSCS’s interpretation that, where used in rule 3.1(2), “in respect of” means “for” or “for the payment of”, and not “integral to, part and parcel of or sufficiently connected to” as found by the judge. Of particular significance are the provisions of rules 16 to 20 dealing with the calculation and payment of compensation, especially rules 17, 19 and 20.

Rule 16 deals with timing of payment. It provides that claims must be paid by the FSCS as soon as reasonably possible after the conditions in rule 3.1 have been met and the compensation has been calculated, subject to a right to postpone in certain circumstances.

Rule 17 is headed “Limits on Compensation Payable”. It relevantly states:

“17.1

The limits on the maximum compensation sums payable by the FSCS for protected claims are set out in 17.2.

17.2

(1) For a protected contract of insurance when the contract is a relevant general insurance contract:

(a)

if the claim:

(i)

is in respect of a liability subject to compulsory insurance; or (ii) is in respect of a liability subject to professional indemnity insurance; or

(iii)

is in respect of and arises from the death or incapacity of the policyholder due to injury, sickness, or infirmity; or

(iv)

is in respect of a liability subject to building guarantee insurance;

the level of cover is 100% of the claim; and

(b)

in all other cases the level of cover is 90% of the claim; and

in each case, cover shall be determined in accordance with 19 and 20 and there is no upper limit on the amount that can be paid.”

Rule 17 therefore provides for the maximum compensation as a percentage of the claim, being (following the 2020 Statement) 100% for building guarantee insurance. It also clarifies that there is no upper limit on the amount that can be paid and provides that cover is determined in accordance with rules 19 and 20.

In this case it is less clear at first sight that the references to “claim” in rule 17.2(1) are correctly shown as being to the defined term, rather than to a claim under the compensation scheme. Read in isolation they could be referring to a claim under the scheme. However, the reference to “maximum compensation sums payable … for protected claims” in rule 17.1 indicates that “claim” in rule 17.2(1) should similarly be read as referring to the defined term.

This is further supported by rule 6, which applies where the FSCS takes measures pursuant to rule 5 to safeguard “rights of eligible claimants under protected contracts of insurance” (rule 5.1(1), emphasis supplied) where a firm is in financial difficulties (rules 5 and 6 are set out in the Appendix). Similarly to rule 17.2(1), rule 6.2(1) refers to “claims” which arise “in respect of a liability subject to compulsory insurance” or “in respect of a liability subject to building guarantee insurance”. Where used there the claims in question must be claims against the insurer, since in the context of rule 6 there is no claim for compensation under the scheme. Rather, the FSCS is taking steps to ensure that contracts of insurance are performed. Rule 6.2 requires the FSCS to ensure that policyholders receive 100% (or, where rule 6.2(2) applies, 90%) of any benefit under the contract.

This reinforces the conclusion that the compensation available is also determined by reference to the benefits to which the insured is entitled under the terms of the contract of insurance, rather than any other amount payable by the failed firm. It cannot have been intended that the amount received should differ according to whether the FSCS chooses to take, or is able to take, measures to secure performance of the contract. In other words, rule 17 should be interpreted consistently with rule 6, so that its effect is to determine the limits on compensation by reference to the claim under the contract.

If the Policyholders were correct, compensation would not only be due for amounts that are connected with but not owed under a contract of insurance, but there would be no maximum limits on compensation payable in relation to such amounts.

Before leaving rules 6 and 17, I would observe that rules 6.2(1) and 17.2(1) both use the phrase “in respect of”. They do so in a manner that has a similarity to the wording considered in Geologistics. Section 6(5) of the 1975 Act referred to a “liability … under the terms of a policy … arising otherwise than in respect of a liability of the policyholder which is a liability subject to compulsory insurance”. In rules 6.2 and 17.2, and where it is used on the second occasion in s.6(5) of the 1975 Act, it is reasonably clear that the “liability” referred to is an underlying liability in the sense of an insured risk, rather than a liability of the insurer (albeit that, with building guarantee insurance, the reference to a liability of the policyholder may well not be apt: see below). If the facts of Geologistics were repeated, it is possible that the result could be the same on the basis that the claim under the policy for defence costs would be regarded as being “in respect of a liability subject to compulsory insurance”. However, the point does not arise for decision and I will not comment further, beyond saying that I am not persuaded that the use of the words “in respect of” in rules 6.2(1) and 17.2(1), where they are used to describe the type of liability or risk covered by the policy, provides any material assistance to their correct interpretation in rule 3.1(2).

Rule 18.1, headed “Payment of Compensation”, requires the FSCS to pay any compensation that it determines is payable. This is subject to carve-outs, one of which is that arrangements have been made to secure continuity of cover under rule 5.1 (which would engage the limits in rule 6: see above). Rule 18.2 applies to claims “under” a protected contract of insurance and empowers the FSCS to secure that payments are made by a liquidator or administrator.

Rule 18 is supportive of the FSCS’s approach. If the Policyholders were correct then the proviso relating to rule 5.1 would appear to switch off any payment obligation in respect of amounts not due under the terms of the policy, provided that amounts that are due under its terms are paid as contemplated by rule 6. Rule 18.2 would also have no application. It is also worth noting that rule 3.1 is expressly subject to rule 18.

Rule 18.6 permits the FSCS to pay interest on a compensation sum, which under rule 18.7 is not taken into account in applying the limits under rule 17. Neither party relied on rule 18.6. As explained in the Decision, the FSCS’s internal policy is that it is applied only where an application for compensation has not been processed in good time. The Decision explains that the FSCS determined that there was no reason to depart from that policy in this case.

Rule 19, headed “Calculating Compensation – General”, is strongly supportive of the FSCS’s position. Rule 19.1 quantifies the amount of compensation “payable to the claimant in respect of a protected claim” as being the amount of the “overall net claim” against the relevant person or successor, subject to the limits in rule 17. It will be noted that “in respect of” is used here in a very similar fashion to rule 3.1(2). Importantly, rule 19.3 goes on to provide that the “overall claim” is the “sum of the protected claims of the same category that [the claimant] has” against the relevant person or successor in default, less any amount that can be set off against any of those claims. (I note in passing that the reference to “net” in rule 19.1 appears to be superfluous, but that is inconsequential.)

As already discussed, protected claims are claims under the contract, not other claims. Rule 19.3 leaves no scope for the concept of “overall claim” to include amounts that do not form part of a protected claim. If the Policyholders were correct then the compensation that the FSCS is required to calculate under rule 19 and pay under rules 16 and 18 would therefore not extend to the amounts for which they seek compensation. That point, and the use of “in respect of” in rule 19.1, strongly reinforce the conclusion that rule 3.1(2) is limited to claims for amounts due under the contract.

Rule 20 provides further support for the FSCS’s case. It is headed “The Compensation Calculation” and, as that indicates, deals in more detail with calculating the compensation due. In particular, rule 20.2 requires the FSCS to calculate the liability of a relevant person or successor to the claimant “under” a relevant general insurance contract “in accordance with the terms of the contract”, and (subject to rule 17.2(1)) pay “that” amount to the claimant. On its terms this can only apply to amounts due under the policy. This reinforces the point that if the Policyholders’ construction were correct then the rules would contain no provision for calculation and payment of the costs and interest claimed, bearing in mind that rule 17.2 provides that “cover shall be determined in accordance with 19 and 20”: see above.

There is a difference in wording between rule 20.1 and 20.2. Rule 20.1 requires payment of 100% of any liability of a relevant person “in respect of a liability subject to compulsory insurance”, tracking wording used in rules 6.2(1) and 17.2(1). However, far from indicating a fundamental difference in the extent of coverage between rules 20.1 and 20.2, the difference is explicable because amounts due pursuant to rule 20.2 may not reflect a liability of the insured. Indeed, that may well be the case where the policy is one of building guarantee insurance held by a homeowner. In contrast, and presumably in error, that point has not been correctly picked up in the changes made in 2020 to rules 6 and 17. The references there to a liability subject to building guarantee insurance are not strictly apt, although the overall intention is clear.

Moving on from rules 16 to 20, the Policyholders do not gain any assistance from rule 2.3 (to which the judge referred). It is understandable that the FSCS is permitted to agree to pay reasonable costs of insolvency proceedings, not least because that may allow amounts to be secured from defaulting insurers in a way that reduces the overall costs of the scheme. I also note that rule 2.3 is restricted to the costs of insolvency proceedings, not other proceedings.

Rule 9.8 is relevant and supports the FSCS’s case. It requires the FSCS to treat certain liabilities such as unexpired portions of premiums as “giving rise to claims under a protected contract of insurance”. As Mr Handyside submitted, this provision would be unnecessary if compensation were in any event available for amounts connected with claims under contracts of insurance, rather than being confined to such claims.

There was some discussion at the hearing about the scope of rules 12 and 13, which deal with assignment and subrogation. Mr Drake submitted that these conferred broad rights on the FSCS which potentially extended to the costs and interest for which compensation was sought. Mr Handyside was not able to respond in any detail because there had been no prior indication that these provisions were to be relied on. For that reason I do not propose to comment on them in any detail either. However, I note that, read literally, the assignment provisions appear to extend to rights of a claimant against the insurer that are unconnected to the policy in question. That suggests that a literal interpretation may not be the correct one. Further, the subrogation provisions, although more closely tied to the claim in question, are discretionary.

Finally, I have also considered the 2020 Statement, which is relevant in determining the context of the change in the rules in respect of building guarantee insurance and the mischief at which the change was aimed (see [38] above). However, I do not consider that it assists the Policyholders. The most relevant provisions of the rules for the purposes of this appeal predate the changes made in 2020. In any event, the text of the statement refers in more than one place to coverage for benefits “under” a contract of insurance. I can see nothing that indicates an intention that went beyond extending that coverage from 90% to 100%.

As I have already said, I have a great deal of sympathy for the Policyholders, who have been placed in a very difficult position. It is additionally unfortunate that their predicament appears to have been worsened by the timing of Zurich’s insurance business transfer to EWIC and the latter’s insolvency. If the insurance claim had not been litigated before the insolvency occurred then the FSCS may well have been involved at an earlier stage and itself determined the amount due under the Policies for the purposes of calculating compensation. Any dispute might then have been litigated against the FSCS rather than EWIC, and if the Policyholders succeeded they might have been awarded costs against the FSCS for wrongly declining compensation and the FSCS might have paid interest for any delay under rule 18.6.

However, that is not what occurred. However unfair it may appear to the Policyholders, we have no discretion to award compensation. Instead, we are obliged to apply the rules to the facts. Despite the best efforts of the Policyholders’ legal team to persuade us otherwise, in my view the PPR cannot properly be construed in a way that would extend to the costs and interest owed by EWIC. I would therefore allow the appeal and reinstate the Decision.

I have had the advantage of reading in draft the judgment of Lady Justice Falk, and I agree with her analysis, her reasoning and her conclusions. Although at first sight, and when considering the rule in isolation, the judge’s interpretation of rule 3.1(2) appeared plausible, the more one looked at the context and the language of the other rules of the scheme, the more apparent it became that it was wrong. Rule 19, in particular, is to my mind impossible to reconcile with that interpretation. Geologistics was concerned with a very different situation; the key difference was that the costs in that case were expressly recoverable under the contract of insurance which the policyholder was compelled to take out. In consequence of that, they were prima facie recoverable under section 6(4) of the predecessor scheme. The only question was whether section 6(5) was to be narrowly construed so as to deprive the policyholder of that protection because they were not a compulsory element of cover.

I also wish to associate myself expressly with my Lady’s concluding remarks and with the sympathy she has expressed for the Policyholders, who are and were entirely blameless for the predicament in which they have found themselves. I appreciate, of course, that the PRA was required to fix an appropriate level of cover under the scheme with an eye on the costs of maintaining the scheme, which is funded by a levy on insurers and therefore, indirectly, by insurance premiums. The scheme was never designed to provide a full indemnity. I am also conscious that the scheme is fundamentally aimed at the position where the insurer cannot pay, as opposed to where it appears able to afford to pay but refuses cover. Yet the upshot of confining the ambit of the scheme to what is recoverable under the policy may well be that in future, those in a similar position to these Policyholders will be loath to take the financial risk of pursuing recalcitrant insurers through the courts, as they may spend more in costs than they ultimately recover if the insurer (or its successor) subsequently runs into financial difficulty.

The option of making a much earlier claim under the scheme, leaving it to the FSCS to decide if the insurer is “in default” and if need be, judicially reviewing a negative decision, does not seem to me to provide a particularly satisfactory solution. The court is generally much better placed to determine whether the insurer is liable to pay a claim under a policy than the FSCS would be. Regardless of this, as both the judge and my Lady have acknowledged, we cannot distort the interpretation of the rules of the scheme in order to cater for hard cases.

I agree with both judgments.