The Financial Conduct Authority v BlueCrest Capital Management

This appeal raises two separate issues, one in relation to the powers of the Financial Conduct Authority (‘the FCA’), and the other as to the jurisdiction of the Upper Tribunal when an FCA decision or notice is referred to it. The FCA has statutory powers and responsibilities under the Financial Services and Markets Act 2000 (‘FSMA’), which prior to 2012 fell upon its statutory predecessor, the Financial Services Authority (‘the FSA’). I shall refer to the relevant authority over the whole period as the FCA for convenience, at the expense of strict accuracy.

One of the relevant powers and responsibilities is authorising firms to carry out regulated activities by granting, varying or revoking ‘permissions’ under Part 4A of FSMA. Those powers include a power to impose requirements on such firms if it appears desirable to the FCA in order to advance its operational objectives. The operational objectives include the consumer protection objective which by s. 1C is “securing an appropriate degree of protection for consumers”. The FCA also has disciplinary powers to impose a penalty on authorised persons who breach requirements in Rules made by it pursuant to s. 137A FSMA, which are set out in its Handbook. Those Rules include:

The Principles of relevance to the appeal are Principles 7 and 8 which provide:

“7.

Communications with clients

A firm must pay due regard to the information needs of its clients, and communicate information to them in a way which is clear, fair and not misleading.

8.

Conflicts of Interest

A firm must manage conflicts of interest fairly both between itself and its customers and between a customer and another client.”

A client is a defined term whose meaning is set out in COBS 3.2. It includes any person to whom the firm provides or intends to provide a service in the course of a regulated activity, including a potential client. Customer is also a defined term which in the present case means the same as client.

COBS 4.2 is a Rule giving effect to Principle 7. It provides that “A firm must ensure that a communication or financial promotion is fair, clear and not misleading.”

The Respondent (‘BCMUK’) is an English Limited Liability Partnership and part of the BlueCrest private fund management group founded in 2000. During the relevant period (1 October 2011 to 31 December 2015) BlueCrest ran two funds: an ‘Internal Fund’ only open to certain partners and employees of BlueCrest, whose stated purpose was to attract and retain senior partners and other key members of staff; and an ‘External Fund’ which was open to external investors outside BlueCrest. Each had a typical master fund/feeder fund structure. Each master and feeder fund had an investment manager which was an entity in the BlueCrest group. The investment managers of the External and Internal Funds appointed BCMUK to act as one of the sub-investment managers for the Funds pursuant to a series of sub-investment management agreements.

The proceedings concern two notices given by the FCA to BCMUK:

Both the FSN and DN were based on the same alleged wrongdoing by BCMUK. In short, the FCA alleged that BCMUK failed properly to manage a conflict of interest which is alleged to have arisen by reason of BCMUK acting as sub-investment manager in respect of both the External Fund and the Internal Fund; and, specifically to have favoured the Internal Fund, and thereby BlueCrest senior partners and key staff including portfolio managers, at the expense of the External Fund by reason of BCMUK’s allocation of traders between the Funds; and by use of a proprietary quantitative trading programme called ‘Rates Management Trading’ by the Funds, which was said to have exacerbated the conflict. The allegations, which are serious, are denied by BCMUK, but for the purposes of the applications under appeal were assumed to be true.

BCMUK exercised its right to refer the notices to the Upper Tribunal (Tax & Chancery Chamber); it referred the FSN on 27 October 2021 pursuant to s. 55Z3 FSMA and the DN on 22 November 2021 pursuant to s. 208(4)(b) FSMA. Each of those sections provides that what is referred to the Upper Tribunal is “the matter”. The references were consolidated and certain preliminary issues were heard together by Upper Tribunal Judges Timothy Herrington and Rupert Jones. The appeal and cross appeal to this court are from their closely reasoned decision of 21 June 2023: [2023] UKUT 00140 (TCC) (‘the UT Decision’).

The FCA’s case as set out in the FSN and DN was based solely on an alleged breach by BCMUK of Principle 8. So too was its case in support of the FSN and the DN as pleaded in its original Statement of Case in the references (‘the S/C’).

The UT Decision was concerned with three applications. One was an application dated 13 September 2022 by BCMUK to strike out the parts of the S/C which supported the FCA’s case on the FSN imposing the redress requirement (‘the strike out application’). The application was brought on the basis that the relevant statutory power to impose a redress requirement upon which the FCA relied, namely s.55L FSMA, when read with s.404F(7) FSMA, only permitted the FCA to impose such a scheme on an individual firm where it would have such power under s. 404 to impose a market wide redress scheme, and that is only where four conditions are met, being loss, causation, breach of duty and actionability; and that the FSN and S/C did not disclose any real prospect of meeting any of those conditions. The Upper Tribunal (a) accepted BCMUK’s case that s. 55L required fulfilment of the four conditions; (b) rejected its case that the matters alleged in the S/C were not capable of fulfilling the first three conditions of loss, causation and duty; but (c) accepted its case that the S/C was not capable of fulfilling the actionability requirement, it being common ground that a breach of Principle 8 by itself is not actionable. The effect of the decision to allow the application on this basis was that the FCA had no power to impose the redress requirement in the FSN and it could not be maintained.

The FCA appeals against the decision on the strike out application, contending that (a) the power to impose a redress requirement in s. 55L is not circumscribed by s. 404F(7) and is not subject to any of the four conditions; (b) alternatively if it is circumscribed, there is no actionability condition.

By a Respondent’s Notice BCMUK contends that the strike out decision should be upheld on the following additional grounds:

The second application considered by the Upper Tribunal was an application dated 29 July 2022 pursuant to rule 5(3)(c) of the Tribunal Procedure (Upper Tribunal) Rules 2008 (‘the UT Rules’) by which the FCA sought to amend its S/C (‘the amendment application’). While certain amendments were agreed by BCMUK, others were contested. The UT Decision put the contested amendments into four categories. The first contested amendment was refused and there is no appeal from that aspect of the decision. The second contested amendment raised allegations in support of the FCA’s case that there had been a breach of Principle 8 (‘Amendment 2’). The Upper Tribunal allowed that amendment. The third and fourth contested amendments (‘Amendments 3 & 4’) relied on a breach of Principle 7 and Rule 4.2.1 of COBS respectively. The Upper Tribunal refused those amendments on the grounds that there was no jurisdiction to allow them because they did not fall within the matter referred.

The FCA appeals against the decision to disallow Amendments 3 & 4. BCMUK cross appeals against the decision to allow Amendment 2. By a Respondent’s Notice BCMUK also seeks to uphold the refusal of permission for Amendments 3 & 4 on the grounds that:

The outcome of the strike out application is potentially affected by the appeal against the decision on the amendment application in relation to Amendment 4, because if Amendment 4 is allowed, it will introduce a case of breach of COBS Rule 4.2.1, which, as is common ground, is actionable.

The FCA’s application notice dated 29 July 2022 also included the third application determined by the Upper Tribunal, which was an application to serve a rejoinder which further addressed its case on redress in support of the FSN (‘the rejoinder application’). BCMUK opposed the application on the same grounds upon which it relied for the strike out application, and the Upper Tribunal refused the application for the reasons it gave in striking out the redress case. The FCA appeals against this decision on the same grounds as it relies upon in relation to the strike out decision, and the issues which arise are the same as those which arise on the appeal and Respondent’s Notice on the strike out application.

Ground 1 of the FCA’s appeal is that the UT Decision was wrong to hold that it may only exercise its statutory power to impose a single-firm redress requirement subject to the conditions that: (1) persons have suffered financial loss or damage (‘the Loss Condition’); (2) the loss or damage has been caused by a wrong on the part of the firm (‘the Causation Condition’); (3) the loss or damage was suffered by a person or persons to whom the relevant regulatory duty was owed (‘the Duty Condition’); and (4) the wrong which caused the loss or damage was an actionable wrong in respect of which a remedy or relief would be available in legal proceedings (‘the Actionability Condition’), together ‘the ‘Four Conditions’, as formulated in the UT Decision; alternatively that it was wrong to hold that the power is subject to the Actionability Condition.

Ground 1 involves the Respondent’s Notice points that:

Ground 2 of the FCA’s appeal is that the UT Decision was wrong to treat its jurisdiction as a narrow one which precluded an ability to allow Amendments 3 & 4. This ground raises the cross-appeal and Respondent’s Notice grounds that:

The Upper Tribunal dealt with the amendment application issues before dealing with the strike out issues. Before us the issues were dealt with in the reverse order, reflecting their order in the grounds of appeal. I will address them in the order adopted by the parties and the grounds of appeal.

Section 55L appears in Part 4A of FSMA which is concerned with permission for persons to carry out regulated activities. It provides, in relevant part:

“Imposition of requirements by FCA

…

(2)

The FCA may exercise its power under subsection (3) in relation to an authorised person with a Part 4A permission (whether given by it or by the PRA) (“A”) if it appears to the FCA that—

…

(c)

it is desirable to exercise the power in order to advance one or more of the FCA's operational objectives.

(3)The FCA's power under this subsection is a power—

(a)

to impose a new requirement,

(b)

to vary a requirement imposed by the FCA under this section, or

(c)

to cancel such a requirement.

The operational objectives referred to in s. 55L(2)(c) are identified at s. 1B(3) as the consumer protection objective, the integrity objective and the competition objective, each of which is defined in ss. 1C, 1D and 1E respectively. Section 1C(1) identifies the consumer protection objective as being “securing an appropriate degree of protection for consumers”; and s. 1C(2) identifies 7 matters to which the FCA must have regard in deciding what degree of protection for consumers may be appropriate. “Consumers” are defined in s. 1G in very wide terms. The exercise of such a power is, by s. 55L(4), referred to as the FCA’s “own-initiative requirement power”.

Section 55N identifies requirements which may be imposed under s. 55L. They may require the person to take, or refrain from taking, specified action (s. 55N(1)); and may extend to unregulated activities (s. 55N(2)). Section 55N(5) provides:

“A requirement may refer to the past conduct of the person concerned (for example, by requiring the person concerned to review or take remedial action in respect of past conduct).”

Section 404 provides:

“404 Consumer redress schemes

(1)

This section applies if—

(a)

it appears to the [FCA] that there may have been a widespread or regular failure by relevant firms to comply with requirements applicable to the carrying on by them of any activity;

(b)

it appears to it that, as a result, consumers have suffered (or may suffer) loss or damage in respect of which, if they brought legal proceedings, a remedy or relief would be available in the proceedings; and

(c)

it considers that it is desirable to make rules for the purpose of securing that redress is made to the consumers in respect of the failure (having regard to other ways in which consumers may obtain redress).

(2)

“Relevant firms” means—

(a)

authorised persons;

(b)

payment service providers; or

(c)

electronic money issuers.

(3)

The FCA may make rules requiring each relevant firm (or each relevant firm of a specified description) which has carried on the activity on or after the specified date to establish and operate a consumer redress scheme.

(4)

A “consumer redress scheme” is a scheme under which the firm is required to take one or more of the following steps in relation to the activity.

(5)

The firm must first investigate whether, on or after the specified date, it has failed to comply with the requirements mentioned in subsection (1)(a) that are applicable to the carrying on by it of the activity.

(6)

The next step is for the firm to determine whether the failure has caused (or may cause) loss or damage to consumers.

(7)

If the firm determines that the failure has caused (or may cause) loss or damage to consumers, it must then—

(a)

determine what the redress should be in respect of the failure; and

(b)

make the redress to the consumers.

(8)

A relevant firm is required to take the above steps in relation to any particular consumer even if, after the rules are made, a defence of limitation becomes available to the firm in respect of the loss or damage in question.

(9)

Before making rules under this section, the FCA must consult the scheme operator of the ombudsman scheme.”

Section 404(9) requires consultation with the ombudsman scheme operator in relation to rules establishing a redress scheme because by s. 404B, where a complaint to the ombudsman includes subject matter which falls to be dealt with or has been dealt with under a consumer redress scheme, the ombudsman is obliged to determine the complaint by reference to what in its opinion the determination under the redress scheme should be or should have been.

Section 404A provides:

“404A Rules under s.404: supplementary

(1)

Rules under section 404 may make provision—

(a)

specifying the activities and requirements in relation to which relevant firms are to carry out investigations under consumer redress schemes;

(b)

setting out, in relation to any specified description of case, examples of things done, or omitted to be done, that are to be regarded as constituting a failure to comply with a requirement;

(c)

setting out, in relation to any specified description of case, matters to be taken into account, or steps to be taken, by relevant firms for the purpose of—

(i)

assessing evidence as to a failure to comply with a requirement; or

(ii)

determining whether such a failure has caused (or may cause) loss or damage to consumers;

(d)

as to the kinds of redress that are, or are not, to be made to consumers in specified descriptions of case and the way in which redress is to be determined in specified descriptions of case;

(e)

as to the things that relevant firms are, or are not, to do in establishing and operating consumer redress schemes;

(f)

securing that relevant firms are not required to investigate anything occurring after a specified date;

(g)

specifying the times by which anything required to be done under any consumer redress scheme is to be done;

(h)

requiring relevant firms to provide information to the FCA;

(i)

authorising one or more competent persons to do anything for the purposes of, or in connection with, the establishment or operation of any consumer redress scheme;

(j)

for the nomination or approval by the FCA of persons authorised under paragraph (i);

(k)

as to the circumstances in which, instead of a relevant firm, the FCA (or one or more competent persons acting on the FCA’s behalf) may carry out the investigation and take the other relevant steps under any consumer redress scheme;

(l)

as to the powers to be available to those carrying out an investigation by virtue of paragraph (k);

(m)

as to the enforcement of any redress (for example, in the case of a money award, as a debt owed by a relevant firm).

Section 404F provides:

“404F Other definitions etc

(1)

For the purposes of sections 404 to 404B—

“redress” includes—

(a)

interest; and

(b)

a remedy or relief which could not be awarded in legal proceedings;

“specified” means specified in rules made under section 404.

(2)

In determining for the purposes of those sections whether an authorised person has failed to comply with a requirement, anything which an appointed representative has done or omitted as respects business for which the authorised person has accepted responsibility is to be treated as having been done or omitted by the authorised person.

(3)

References in those sections to the failure by a relevant firm to comply with a requirement applicable to the carrying on by it of any activity include anything done, or omitted to be done, by it in carrying on the activity—

(a)

which is in breach of a duty or other obligation, prohibition or restriction; or

(b)

which otherwise gives rise to the availability of a remedy or relief in legal proceedings.

(4)

It does not matter whether—

(a)

the duty or other obligation, prohibition or restriction, or

(b)

the remedy or relief,

arises as a result of any provision made by or under this or any other Act, a rule of law or otherwise.

(5)

References in sections 404 to 404B to a relevant firm include—

(a)

a person who was at any time a relevant firm but has subsequently ceased to be one; and

(b)

a person who has assumed a liability (including a contingent one) incurred by a relevant firm in respect of a failure by the firm to comply with a requirement applicable to the carrying on by it of any activity.

(6)

References in those sections to the carrying on of an activity by a relevant firm are, accordingly, to be read in that case with the appropriate modifications.

(6A) References in sections 404 and 404E to an “electronic money issuer” are references to a person mentioned in paragraph (a), (b), (c), (d), (h) or (i) of the definition of “electronic money issuer” in regulation 2(1) of the Electronic Money Regulations 2011.

(7)

If the FCA varies a permission or authorisation of a person so as to impose requirements on the person to establish and operate a scheme which corresponds to, or is similar to, a consumer redress scheme, the provision that may be included in the permission or authorisation as varied includes—

(a)

provision imposing requirements on the person corresponding to those that could be included in rules made under section 404; and

(b)

provision corresponding to section 404B.

(8)

In subsection (7) the reference to the variation of a permission or authorisation by the FCA is a reference to—…

(aa) the imposition or variation of a requirement under section 55L, …”

Section 415A provides:

“415A Powers under the Act

Any power which the FCA, the PRA or the Bank of England has under any provision of this Act is not limited in any way by any other power which it has under any other provision of this Act.”

Having set out the relevant statutory provisions, the UT Decision considered the history and development of FSMA at [226]-[250]. It identified that as originally enacted FSMA provided four means by which clients or customers of authorised firms could obtain redress where breaches of regulatory rules were established, and that these remained the four means of redress notwithstanding subsequent amendments.

The first was a civil claim, which s. 150(1) FSMA as originally enacted provided was available as an action for breach of statutory duty for any contravention, at the suit of a private person “who suffers loss as a result of the contravention”, subject to s. 150(2) which provided that rules made by the FCA could disapply s.150 to specific provisions of the rules. Those provisions were replaced in materially identical terms by s. 138D(1) and (2) by the Financial Services Act 2012 (‘FSA 2012’). The Rules made by the FCA have always excluded a breach of the Principles as a contravention for which a civil claim lies under s. 150(1)/138D(1): PRIN 3.4.4R. The explanation given by the FSA before FSMA was enacted was:

“We have designed the Principles as a set of regulatory expectations, not as a set of legal rights at large. The high level at which they are expressed makes it important that their interpretation and application should be in harmony with the overall body of FSA rules and guidance and declared authorisation, supervisory and enforcement policy. This might be put at risk if civil litigation between private parties were to become the engine driving the interpretation of the Principles. The investor protection need can be amply met (as it is at present) by providing for civil actionability below the level of Principles in more specific rules.”

The second means of redress was that pursuant to ss. 382-384 FSMA, which have remained in materially identical form, the court (s. 382) and the FCA (s. 384) could make restitution orders. By s. 382(1) and 384(1) such an order may be made if the authorised firm has contravened or been knowingly concerned in the contravention of a “relevant requirement” and “profits have accrued to him as a result of the contravention” or “one or more persons have suffered loss or been otherwise adversely affected as a result of the contravention”. The definition of “relevant requirement” was in wide terms and included breaches of FCA Rules, including the Principles, as well as criminal offences. Sections 383 and 384(2) extended the jurisdiction to cases of contravention of the market abuse regulation by insider dealing, unlawful disclosure of inside information and market manipulation. By s. 382(2) and s. 384(5) the restitution takes the form of “such amount as appears to be just” having regard to the profit made by the authorised person or the extent of the loss or adverse effect suffered by a person as a result of the contravention.

The third means of redress identified was that, pursuant to s. 404 FSMA as originally enacted, the Treasury (not the FCA) had the power to impose a consumer redress scheme. Such a scheme could be imposed where the Treasury was satisfied that there was “evidence suggesting—(a) that there has been a widespread or regular failure on the part of authorised persons to comply with rules relating to a particular kind of activity; and (b) that, as a result, private persons have suffered (or will suffer) loss in respect of which authorised persons are (or will be) liable to make payments (‘compensation payments’)”. The FCA was required to produce a report setting out the details of the alleged failure and of the scheme proposed (s.404(4)). We were told that the Treasury would then obtain Parliamentary approval for the scheme. The Treasury could authorise the FCA to establish and operate a scheme for “(a) determining the nature and extent of the failure; (b) establishing the liability of authorised persons to make compensation payments; and (c) determining the amounts payable by way of compensation payments” (s.404(2)). This provision was substantially amended by the Financial Services Act 2010 (‘FSA 2010’).

The fourth means of redress was that, pursuant to Part XVI FSMA, the Financial Ombudsman Service (‘the FOS’) was, by its ‘compulsory jurisdiction’ empowered to resolve complaints which relate to an act or omission of a person in carrying on a regulated activity specified by the FCA (s.226(1)-(4)). Complaints are to be determined on the basis of what, in the opinion of the ombudsman, is “fair and reasonable in all the circumstances of the case” (s.228(2)). As to financial redress, the FOS is empowered to make “an award against the respondent of such amount as the ombudsman considers fair compensation for loss or damage … suffered by the complainant” (s.229(2)(a)). The loss and damage for which such an award can be made is limited to (a) financial loss, or (b) any other loss, or any other damage, of a kind specified in the FOS rules (s.229(3), (11)), which includes distress, inconvenience, pain and suffering and damage to reputation. All compensation that can be awarded is subject to a monetary limit as specified from time to time (s.229(6)); this limit was originally £100,000 and after various increases now stands at £430,000. In addition to such money award the ombudsman can:

The UT Decision concluded at [286] that each of these four powers of redress as originally set out in FSMA required the FCA, or the FOS, to establish (at a minimum) the Loss, Causation and Duty Conditions; and additionally, both the s. 404 consumer redress scheme power and s. 150 action for damages required that the relevant wrong fulfil the Actionability Condition.

As to the authorisation regime in FSMA as originally enacted, s.43 FSMA provided that when granting a permission, the FCA might include “such requirements as the [FCA] considers appropriate” (s.43(1)); and those requirements might include “a requirement to take…or refrain from taking specified action” (s.43(2)). Section 43(3) further provided that “A requirement may extend to activities which are not regulated activities”. Sections 44 and 45 FSMA, as originally enacted, provided the FCA with the further power to vary a permission or to cancel it, once granted, either on the application of the authorised person (s.44) or on the FCA’s own initiative (s.45). The FCA could exercise its own initiative power, inter alia, where it considered it “desirable to exercise that power in order to protect the interests of consumers or potential consumers” (s.45(1)(c)). The FCA’s power to vary a permission included a power to include any provision that could have been included if a fresh permission were being sought (s.45(4)), thereby effectively incorporating an ability to impose any condition permitted by s. 43.

The Upper Tribunal then identified the changes introduced by FSA 2010 and FSA 2012. The main relevant amendment to FSMA made by FSA 2010 was to the power to impose multi-firm consumer redress schemes under s. 404. The primary purpose of the amendment was to provide the FCA itself with the power to impose such consumer redress schemes without the need for the scheme to be approved by the Treasury. This required both the amendment of s.404 and the inclusion of a number of new provisions (at ss.404A to G) which together gave the FCA a new rule-making power to impose rules on certain firms for the purposes of establishing a consumer redress scheme. These are the provisions which remain currently in force in relation to market wide redress schemes in s.404 to s. 404F.

The UT Decision emphasised that the new provisions also placed significant restrictions on the FCA’s use of that power and drew attention to two aspects. First, the new sections provided that the FCA could only make rules for such a redress scheme under s. 404(3) where the conditions in s. 404(1) were satisfied, namely where the FCA considered that (a) there had been “widespread or regular failure by relevant firms” (plural) to comply with requirements; (b) “it appears to it that, as a result, consumers have suffered (or may suffer) loss or damage in respect of which, if they brought legal proceedings, a remedy or relief would be available in the proceedings”; and (c) it considers that it is “desirable” to impose the scheme to secure redress for consumers. Thus, the Upper Tribunal emphasised, the FCA was permitted to impose such a scheme only if all four elements of Loss, Causation, Duty and Actionability were established; and breaches of Principles are not actionable for this purpose. The second relevant change brought about by FSA 2010 was that ss. 404F(7) and (8) were introduced for the first time.

In FSA 2012 the existing power of the FCA to impose a requirement on a firm when determining an application for permission (under s.43) was retained in materially the same terms by the introduction of s.55L(1). The FCA’s existing own initiative power to impose conditions by way of variation of existing permission (under s.45) was retained in materially the same terms by way of s.55L(2); and section 404F(8) was amended to refer to s.55L in place of s. 45. The requirements available on the exercise of the FCA’s own initiative power which had been contained in s. 43 (by reason of s. 45(4)) were reproduced in s. 55N(1)-(4). None of this involved any substantive change.

The main substantive change made by FSA 2012 which is relevant in the present context was the introduction of s. 55N(5) with its new specific provision that the own initiative requirements available included requirements by reference to past conduct.

After this historical survey, the Upper Tribunal recited the rival arguments, and the applicable principles of statutory construction, before setting out its reasoning and conclusion on this issue at [273]-[336]. It started its analysis by saying (at [280]) that the FCA’s interpretation of FSMA would create a surprising result in this case (and in others) because it would mean that the FCA could require a firm to pay something which the FCA refers to as “redress” to its customers (in this case in a sum estimated to be over US$700 million) without there being any statutory requirement on the FCA to establish in respect of those customers any of the four elements of loss, causation, duty and actionability; and would also mean that it could do so without needing to show that the firm had breached a regulatory requirement at all or that the redress proposed even related to regulated activities. The Upper Tribunal concluded that the power in s. 55L must be constrained so as to avoid such unreasonable and absurd results; and that it is constrained by “the terms of s. 404F(7) and the rules under section 404, namely section 404A” so as to require the FCA to establish the Four Conditions (at [283]).

At [281]-[282] it made clear that it had not found it necessary to determine the merits of BCMUK’s arguments that the FCA’s case would interpret s. 55L in a way which would violate BCMUK’s rights under A1P1 or that the common law has a similar effect on the question of interpretation.

It considered that the conclusion was supported by the history of the FSMA provisions which “originally provided for the four carefully tailored means of redress … (action for damages, restitution, consumer redress schemes and the FOS scheme) each of which require specific statutory hurdles to be satisfied (such as a failure to comply with a duty – a breach, together with causation and loss)” ([285]). Thereafter, it was unlikely that Parliament also intended to provide the FCA with a further general power under section 55L, or previously under s.45, to impose a redress scheme on a single firm by way of a requirement without any restriction or statutory hurdles beyond the FCA considering it desirable to do so for its operational objectives. Such a largely unrestricted power might render both the FCA’s restitution power and consumer redress power otiose ([286]).

It would also, the Upper Tribunal said, give rise to a striking anomaly in giving the FCA the power to require redress to be paid by single firms to consumers without significant restriction when a multi-firm consumer redress scheme cannot be imposed without establishing the conditions of causation, loss, actionability and duty under section 404 FSMA ([287]).

A necessary implication of the FCA’s case was that its power to impose a redress scheme on a firm if desirable for operational objectives had always been available ever since FSMA was first enacted, under s.45(1)(c) which was replaced by s. 55L. The Upper Tribunal was satisfied that that could not be correct, first because as originally enacted, s. 404 required the FCA to seek specific permission from the Treasury before implementing a redress scheme under s.404, and it cannot reasonably have been the intention of Parliament that, notwithstanding such onerous oversight, the FCA could of its own initiative simply have imposed consumer redress schemes on a firm-by-firm basis (and thus on as many firms as it liked), bypassing all restrictions beyond considering it to be desirable to do so for operational objectives; secondly, because after s.404 was amended by FSA 2010 to permit the FCA itself to impose redress schemes without reference to the Treasury, it again cannot reasonably have been the intention of Parliament to impose restrictions on the use of that power for collective schemes but to have allowed the FCA to implement single firm consumer redress schemes on a firm-by-firm basis without any of those restrictions; and, thirdly, because section 404F(7)(a) introduced the power for the FCA to include in a firm’s permission, provision imposing requirements on the person corresponding to those that could be included in rules made under section 404 for multi-firm consumer redress schemes: if it were the case that the FCA already had an unfettered power to impose a consumer redress scheme on individual firms absent s.404F(7), there would be no need for subsection 404F(7)(a) of that provision at all; it would be redundant; the FCA could simply make requirements that corresponded to those set out in the s.404 rules if it wished under the unfettered variation power in s.55L or previously s. 45 ([289]-[293]).

On its proper construction, s.404F(7) is the provision which enables the FCA to impose on a single firm a requirement under s.55L(3) which corresponds to or is similar to a consumer redress scheme, and provides that if the FCA does so, the requirements that the FCA is permitted to impose on that firm are those corresponding to the rules for consumer redress schemes against multiple firms as set out in s.404 and s.404A ([295], [327]). Section 404F(7) ensures, in essence, that the statutory criteria are aligned whichever statutory route is chosen: the single or multiple firm route ([296]). The use of the word “may” in s. 404F(7) is not permissive; rather it defines the circumstances in which any form of redress scheme can be imposed ([298]): s. 404F(7) requires a single firm scheme to be similar to or correspond to a s. 404 market wide redress scheme ([294]-[298]).

At [301]-[302] the Upper Tribunal derived support for its conclusion from a passage in the July 2009 Treasury White Paper which preceded FSA 2010 which stated:

“The FSA has existing powers to impose redress schemes on a firm-by-firm basis where a large number of consumers are affected. The power of the Treasury to initiate a collective redress scheme on a wider basis is set out in section 404 of FSMA. The Government believes there should be new powers for the FSA to require a firm or firms to make redress either on an industry wide or firm-by-firm basis as appropriate.”

At [304]-[306] the Upper Tribunal said that its construction was also consistent with subsequent delegated legislation which has referred to s.404F(7) as providing a separate power under which a s.55L requirement (specifically, a requirement for a consumer redress scheme) may be imposed. The Upper Tribunal gave as examples Article 58(6) of the Financial Services and Markets Act 2000 (Regulated Activities) (Amendment) (No. 2) Order 2013/1881; and Article 82(12) of the Financial Services and Markets Act 2000 (Claims Management Activity) Order 2018/1253. It was also consistent with how the FCA itself considered the power and the guidance it gave as to its potential use of the power, which treated s. 404F(7)(a) as a new power to “implement rules contained in ss. 404 and 404A which it otherwise [would not] be permitted to do” ([307]-[311]).

At [314]-[316] the Upper Tribunal derived further support for its construction from the introduction in 2012 of s. 55N(5). This was after s. 404F(7) had been introduced by FSA 2010; and so was added when the FCA already had the power to impose a requirement which corresponded to or was similar to a consumer redress scheme. Accordingly, “If s.55L(3) were really as unlimited a power as is contended, there would be little purpose whatsoever to be served by s.55N(5). In reality, as the Explanatory Notes to FSA 2012 (referred to above) make clear, s.55N(5) was intended to afford the Authority a new and additional power.” The Explanatory Notes at paragraph 206 say that new section 55L replicates the powers in s. 43 to impose requirements as condition of permission save for the new provisions in s. 55N, identifying 55N(4) and (5) and saying of the latter that it could be used to require a person to carry out a review of its past conduct to identify customers who have been treated unfairly. Section 55N(5) conferred a new additional power because it referred to “remedial action” which is different from redress. The Upper Tribunal regarded remedial action as intended “to denote such activities as a review of past business” and relied in that respect on examples set out in the FCA’s own guidance. It went on to hold in the alternative that even if s. 55N(5) did provide for or clarify a power to require payment of redress, then the reference to “remedial action” equally requires the FCA to establish the four elements of loss, causation, duty and actionability “which are integral to the concept of remedying a wrong done by the firm to an affected consumer”.

At [317] the Upper Tribunal rejected the FCA’s argument that s. 415A supported its case, saying that s. 415A merely provides that one power under FSMA does not limit any other. The provision has no other relevant effect – it is not seeking to imply restrictions or limitations otherwise as to how relevant provisions are to be interpreted. BCMUK was not seeking to argue, against the legislative framework, that s. 404F(7) has a narrow ambit because of a broader power.

For all these reasons the Upper Tribunal concluded at [318]-[319] that on the proper construction of FSMA, the source of the power to impose a redress requirement on a single firm is by virtue of s. 55L, read together with and restricted by, the terms of s.404F(7). There is no free-standing power under s.55L(2)(c) and (3) for the Authority to impose a redress requirement simply by being satisfied for rational reasons that the consumer protection objective is met. The requirements which the FCA is permitted to impose when requiring a single firm to conduct a consumer redress scheme under s.55L and s. 404F(7)(a) are only those that may be imposed by way of rules made under section 404, which are provided for in s.404A.

The Upper Tribunal then considered the conditions which attached to the ability to make such a market wide redress scheme and concluded that it included the Four Conditions by reference to those requirements being contained in s. 404A ([319]-[332]).

Its conclusion on this issue was expressed in [334]:

“Our conclusion is that the four statutory conditions are required to be satisfied by virtue of ss.55L, 407F(7) and 404A before a single firm consumer redress requirement can be imposed, namely: loss, causation, duty and actionability must therefore be established. These correspond to and are similar to the four necessary conditions that are required to be satisfied before multi-firm consumer redress schemes can be imposed by virtue of s.404.”

The FCA’s argument can briefly be summarised as follows, although the compression fails to do it justice. Section 55L, like its predecessor s.45, identifies the relevant constraints on the exercise of the power; they are those set out in subsection (2), including, so far as here relevant, that it appears desirable to the FCA to exercise the power in order to advance one of its operational objectives. There is the further constraint that public law principles apply, including a requirement of rationality and that powers must not be exercised for an improper purpose. Although the FCA has a wide measure of discretion within those constraints, the Upper Tribunal was wrong to characterise its argument as being that it had an unfettered power. Moreover there were safeguards in the decision being capable of challenge in the Upper Tribunal. There is no warrant for interpreting s. 55L as subject to additional constraints which are not contained in Part 4A which is concerned with permissions. This is reinforced by s. 55N(5) which makes clear that a requirement may require a firm to take remedial action for past conduct. Remedial action does not here have some technical or defined meaning but is used in its ordinary sense which would include redress of the kind imposed by the FSN in this case. The power to impose a single firm redress scheme is derived from s. 55L, not s. 404F(7), whose language expressly recognises that the power exists elsewhere and makes provision as to what the FCA may do in exercising the power. Section 415A means that the power in s. 55L cannot be restricted by reference to powers expressed in s. 404F(7). In the latter, the “may” is permissive, not exclusive: what is set out in sub-paragraphs (a) and (b) are powers which are conferred, but which do not comprise an exclusive definition of the powers which exist. Moreover the powers conferred by (a) and (b) do not themselves import requirements of loss, causation, duty or actionability. Accordingly the Upper Tribunal was wrong to treat s. 404F(7) as imposing the Four Conditions for the imposition of a single firm redress scheme because the power to do so was derived from s. 55L, not s.404F(7); alternatively, if derived from 404F(7), the powers are not limited to those in (a) and (b) from which the Upper Tribunal derived the Four Conditions; alternatively, if the power is to be found in s. 404F(7) and the powers are limited to those for which the FCA could make rules under s. 404A, the Upper Tribunal was wrong to say that s. 404A imported the Four Conditions.

BCMUK’s argument can briefly be summarised as follows, again without doing it full justice as a result of compression. The UT Decision was correct for the reasons it gave. The FCA’s construction would produce a surprising result, which would be all the more surprising given that FSMA provided for the four carefully tailored means of redress each of which has demanding and specific statutory hurdles. Section 415A does not prevent the ordinary process of statutory construction in which provisions are to be interpreted in the context of other provisions of the same statute and does not have any bearing on whether the power in s. 55L is unfettered or to be read as constrained by s. 404F(7). The FCA’s construction would produce a striking anomaly between multiple firm schemes and single firm schemes, an anomaly which is all the more striking in the historical context of FSMA which did not give the FSA power to impose multi-firm redress schemes but did, on the FCA’s case, empower it to impose a series of single firm redress schemes. Further, if the FCA were correct, s. 404F(7) would be redundant. BCMUK’s construction was supported by the FSA 2010 White Paper and the subsequent legislation and FCA guidance to which the UT Decision referred. Moreover the enactment of a new power in s. 55N(5) in FSA 2012 militated against the FCA’s construction.

There was no real dispute about the principles applicable to statutory construction, following the recent Supreme Court decisions in R (O) v Secretary of State for the Home Department [2022] UKSC 3 [2023] AC 255; and Potter v Canada Square Operations Ltd [2023] UKSC 41 [2023] 3 WLR 963. So far as relevant they can be summarised as follows:

I have reached the clear conclusion that the FCA is correct in its construction and that the power to impose a single firm redress scheme derives from s. 55L, is not constrained by s. 404F(7) and contains none of the Four Conditions. My reasons are as follows.

The own-initiative power to impose requirements as a variation of an existing permission is expressly provided for in s. 55L. The redress scheme imposed in this case is within the wide wording encompassed by s. 55L, provided it is a rational decision which the FCA considers to advance the objective of securing an appropriate degree of consumer protection. The statutory context is that this section is part of a suite of provisions in Part 4A of FSMA which deal in detail with the granting and maintenance of permission to carry out regulated activities and the conditions which may be attached to such permission. They expressly address the scope of conditions which may be imposed, and include at s. 55N(5) an ability to impose conditions by reference to past conduct. I shall return to the wording of s. 55N(5) later but the point to be made at this stage is that if it were intended to impose a constraint on the power so as to confine it to cases in which the Four Conditions were fulfilled, one would expect such constraint to be found expressly in Part 4A, not tucked away in an “Other definitions etc” section in a different part of the Act which is not concerned with permissions, and then only by implication.

There is nothing in Part 4A (or, as I shall explain, elsewhere) which would limit the permission requirements which may be imposed to those for which the FCA or FOS otherwise has powers to impose compensation or other forms of redress. This is not surprising bearing in mind that granting and maintaining permission to carry out regulated activities serves a different function from the imposition of sanctions, compensation or redress. Permission requirements are concerned with whether firms meet and will continue to meet the threshold conditions in Schedule 6 (s. 55B), both on initial application, and on variation of existing permissions, at the forefront of which by Schedule 6 para 2E is whether they are and remain fit and proper persons to carry out regulated activities. That is an essentially prospective assessment of present and future fitness, but past conduct is used as an evidential basis to make the assessment. Thus questions of honesty and integrity, which are qualities central to the prospective fitness of an authorised person, are judged by reference to past conduct. The consumer protection objective which the FCA is required to take into account is, at least in the context of Part 4A, primarily the protection of future consumers of regulated activities, and that protection can be engaged by imposing a requirement of redress or compensation for past conduct in two ways. In part it may protect the consumers of the regulated activities to be provided in the future by the authorised person themselves: if a person has indulged in conduct which demands some form of compensation or redress, they are unlikely to be regarded as fit and proper if they do not pay such compensation or make such redress. That is a matter of protection of future consumers of their regulated activities, not merely those to whom compensation or redress is due. It applies to past conduct which may be unrelated to regulated activities, just as past conduct unrelated to regulated activities may be relevant to questions of honesty and integrity. The consumer protection objective may also extend to consumers of regulated activities provided in the future by other firms, because the regulator’s treatment of individual cases can have an important deterrent and incentivisation effect on others in the market. That is one of the well-recognised aspects of regulatory control of a services industry and informs, for example, the exercise of its disciplinary powers. It is expressly recognised in s. 55T of FSMA.

There is thus a false premise underlying that part of BCMUK’s argument, and the reasoning of the UT Decision, which asserts or assumes that powers to impose conditions on permission are linked to and constrained by powers to impose compensation or redress as a separate sanction. The premise is a false one for each of three reasons. First, there is nothing in the wording of the statute which so limits them and as the Court of Appeal said in R (Rudewicz) v Secretary of State for Justice [2012] EWCA Civ 499, [2013] QB 410 at [30]: “it is, at least in the absence of special circumstances, inappropriate for the court to treat a statutorily conferred discretion with no express limitations or fetters as being somehow implicitly limited or fettered”. Secondly, as already explained sanctions perform a different function from permission requirements. Thirdly sanctions must relate to regulated activity whereas permission requirements need not. Mr Herberg was not inclined to accept the last of these, but it seems to me self-evident, for example, that where an authorised firm becomes subject to a court judgment, the FCA could rationally impose an own-initiative requirement that the judgment be satisfied as a condition of continued permission; and that that is so whether the judgment debt arises out of regulated or unregulated activity. If it is the latter, it is not a form of redress which the FCA could itself impose under any of its other powers, but it is capable of being an appropriate own-initiative permission requirement.

Even if the premise were sound, it seems to me impossible to derive a requirement that all of the Four Conditions are required by reference to other forms of redress when two of the other forms of redress require none of them:

It is also significant that the forms of redress are not mutually exclusive and that a particular form of redress may be authorised by more than one power. Indeed it is striking that in this case the form of redress attached by the FCA as a permission requirement in exercise of powers under s. 55L could just as easily be characterised as a redress requirement by way of restitution, which could have been imposed as redress under s. 384, as it could be characterised as a single firm consumer redress scheme which corresponds to or is similar to a s. 404 consumer redress scheme. The FSN describes the notice as imposing “a requirement that [BCMUK] pays redress” in the form and amount then set out. It could as easily have used the word ‘restitution’ as ‘redress’. There is no doubt that both at the time, and during the argument before the Upper Tribunal and in this court, the FCA treated the redress as corresponding to or being similar to a (market wide) consumer redress scheme for the purposes of s. 404F(7); and the FSN itself refers to s. 404F(7) as a relevant statutory provision, albeit not the one conferring the power. However that does not prevent the redress requirement imposed being one which could have been imposed as restitution in accordance with s. 384. BCMUK’s argument, therefore, is one of form not substance.

Against that background, it would need very clear wording in s. 404F(7), which is tucked away in an “Other definitions etc” section, if it were to be a provision which confers the power to impose a redress requirement which would not otherwise exist and constrains the circumstances in which it can be exercised. However s. 404F(7) does not bear that meaning as a matter of its plain language; on the contrary its language is inconsistent with such a meaning in two respects.

First, the grammatical structure of the section is “If [the FCA imposes a requirement which resembles a consumer redress scheme] the provision that may be included in the permission or authorisation as varied includes….” This clearly presupposes that the power to impose the requirement is to be found elsewhere and this provision merely governs something which may happen if it exercises that power. BCMUK’s construction, to the effect that this is the subsection creating the power, would more naturally be expressed along the lines: “The FCA may impose a requirement that...”. Not only does s. 404F(7) presuppose that the power is to be found elsewhere, but s. 404F(8) specifically identifies that the power is to be found in s. 55L.

Secondly, the natural meaning of the words “the provision that may be included … includes…” are inclusionary and non-exclusive, rather than being, as BCMUK’s construction requires, exclusionary and by way of definition of the scope of the power.

Section 404F(7) therefore simply does not say, as a matter of its plain language, that which the Upper Tribunal interpreted it as meaning, namely that it was the subsection conferring the power to make a single firm redress scheme and that the s. 404A powers it referred to defined both the scope of those powers and the conditions which attached to when they could be exercised.

Nor in any event does its language attach threshold conditions to the exercise of the power; it simply provides for the same powers as may be imposed by rules under a market wide scheme. In the case of a s. 404 market wide scheme the Four Conditions are contained in s. 404(1)(b), (6) and (7). Contrary to BCMUK’s submissions, they are not to be found in any of the provisions in s. 404A which are the provisions which may appear in rules made pursuant to s. 404 (and in particular not in s. 404A(1)(c) and 404A(3) which were those relied on by BCMUK and the Upper Tribunal). Yet it is the s. 404A rules which are referred to in s. 404F(7) as variation requirements available for a single firm redress scheme, not the s.404 threshold conditions. Section 404F(7) does not contain any express language attaching threshold conditions.

Nor can such a conclusion be reached by implication. The fact that the statute provides for threshold conditions for a market wide scheme in s. 404 cannot give rise to the necessary implication that the same threshold conditions are applicable to exercise of s. 404F(7) powers which are concerned with permission to continue to practice. Neither logic nor language justifies treating the threshold conditions for the application of those redress scheme requirements as applicable by way of implication to a variation requirement, merely because the requirements which may be imposed for each are similar or even identical. If a statute provides that you can do X (market wide scheme) only if threshold conditions are met, in which case you can impose requirements Y (s. 404A provisions), and also that you can also do Z (single firm scheme), in which case you can also impose requirements Y, there is no logical reason why the threshold conditions which are expressed to attach to X should also attach to Z, to which they are not expressed to attach. That is so even where X is similar to Z, but in this case X is not similar to Z in a relevant sense. True it is that s. 404F(7) arises in respect of a scheme which “corresponds to or is similar to” a consumer redress scheme under s. 404. But s. 404 is concerned with a consumer redress scheme on a market wide basis which is unavailable as a form of redress unless there is a problem affecting more than one firm; whereas s. 404F(7) is concerned with a “similar” scheme but solely by way of a permission variation requirement, which applies to a single specific firm and involves a different regulatory function aimed at that firm’s continued fitness to carry out regulated activities.

BCMUK’s case therefore distorts rather than gives effect to the plain language of s. 404F(7) and (8).

The FCA’s interpretation does not involve s. 404F(7) being redundant. It enables the FCA to make provision corresponding to s. 404B, which is concerned in the case of market wide schemes with binding the FOS to reach decisions in accordance with the scheme. This the FCA would have no power to do by imposing requirements on the authorised person under s. 55L. So far as subsection (a) is concerned, which permits the FCA to impose requirements on the authorised person corresponding to those it could require under s. 404A, there was more force in BCMUK’s argument that this would add nothing if the FCA’s construction is correct. Mr George suggested that it would enable redress to be enforceable as a debt as could be ordered under s. 404A(1)(m); however s. 404F(7)(a) only authorises requirements to be imposed on the authorised person, that is to say steps which that person must or must not take, as opposed to independent direct provision (which s. 404F(7)(b) allows). It therefore confers no greater power, on the FCA’s case, than the wide power to impose such requirements under s. 55L. It could achieve the result achieved in s. 404(1)(m) in a market wide scheme rule by imposing a requirement that the authorised person undertake by deed to make the redress to each consumer, and that would exist under s. 55L as much as under s. 404F(7)(a). However such (limited) “redundancy” does nothing to undermine the other pointers towards the correct construction. There is much in the statute which is clarificatory, and it would not be surprising if this subsection were clarificatory of individual powers which were examples of the more generally expressed power expressed in s. 55L.

Section 55N(5) is a good example of such clarificatory drafting. As originally enacted, the variation power to attach requirements found in s. 45 did not have its own explanatory subsections as to aspects of such requirement powers. In FSA 2012 they were spelled out in s. 55N, as an adjunct to s. 55L which replaced s. 45. In subsections (1) to (4) of s. 55N it included a repeat of what was in s. 43 as applied to first time applications for permission, which it is common ground applied to s. 45 variations although not in express terms as such. Section 55N was to that extent purely clarificatory. There is no reason to treat s. 55N(5) differently; it is simply another example of an explanation of the scope of the powers conferred elsewhere, for the avoidance of doubt. The Explanatory Notes to FSA 2012 are not inconsistent with such an interpretation.

Indeed this seems the inevitable conclusion from the timing of its introduction in FSA 2012. Section 404F(7) had been enacted two years earlier in FSA 2010 and recognised (on the FCA’s case) or conferred (on BCMUK’s case) a power to impose a single firm redress scheme as a variation requirement under s. 45 (when introduced in 2010), which is a requirement by reference to past conduct. It would in any event have been very surprising if from inception in 2000 s.45 did not include the power to impose a condition in relation to past conduct, as my earlier example of a court judgment illustrates.

It follows that what the FCA did in its FSN was an exercise of powers expressly conferred by s. 55N(5), not only by s. 55L. “Remedial action in respect of past conduct” is not a defined term or term of art, and in its natural meaning is wide enough to comprehend a single firm redress scheme. The Upper Tribunal’s view that it was intended to cover only a review of past conduct is inconsistent with review being referred to in the subsection separately from remedial action. Nor was there any warrant for its alternative unreasoned holding that s. 55N(5) must itself be subject to the Four Conditions.

Once it is recognised that the FCA’s decision to impose the FSN was in exercise of its powers under s. 55L and s. 55N(5), and that s. 404F(7) is not the source of its power to do so, s. 415A comes into play and precludes any argument that such powers are limited or restricted by powers expressed in s. 404F(7).

It is convenient here to deal with BCMUK’s arguments that the Four Conditions should be imposed because a single firm consumer redress scheme should be aligned with a market wide consumer redress scheme; and that otherwise a market wide scheme could be imposed by the device of a number of single firm schemes so as to circumvent the statutory thresholds.

As to the alignment argument, it is met, in my view, by the fact that there are a number of differences between single firm schemes and market wide schemes which mean that there is no necessary correlation between the threshold conditions which must attach to each. First, as I have observed, one is purely a form of redress, the other a form of regulation of permission to continue to undertake regulated activities. Secondly, the fact that a consumer redress scheme under s. 404 requires widespread or regular failures by multiple firms reflects the fact that it is intended to address a market wide or sector wide concern and enables investigatory obligations to be imposed on a collective basis irrespective of any grounds for believing that each and every firm which is subjected to it has been in contravention of any requirement. Indeed although s. 404(4) defines a consumer redress scheme as one which contains any one of the following features, one of which, in s. 404(7)(b), is the making of redress, the sequence of subsections (5) to (7) and the language that “the firm must first investigate” and that “the next step” is determining loss suggests that the paradigm case of a market wide redress scheme will involve all these steps, starting with an investigation by individual firms as to whether they have contravened any requirements. Such consumer redress schemes can be used to protect public confidence in sectors of the market as a whole. By contrast s. 404F(7) is concerned with single firms by reference to their own fitness to continue to carry out regulated activities, which is a narrower focus. Thirdly, it is significant, to my mind, that (market wide) consumer redress schemes are to be made only where it is desirable having regard to other ways in which consumers may obtain redress (s. 404(1)(c)); they may well not be appropriate where, for example, restitution on a firm by firm basis would be a satisfactory alternative. By contrast, variation of permission requirements are not constrained by whether any similar form of redress can be ordered under a different power, and one would not expect them to be: if the circumstances justify redress by restitution under s. 384, for example, to impose a requirement that carrying out such redress is a condition of continued authorisation is obviously capable of meeting the consumer protection objective, and there is no constraint equivalent to s. 404(1)(c) in the wide powers in s. 55L or elsewhere. A single firm scheme may be similar to such other forms of redress as well as being similar to or corresponding to a “consumer redress scheme” as defined for the purposes of ss. 404 and 404F(7), especially when one bears in mind that the latter need have only one of the requirements in s. 404(5)-(7), such as a payment requirement.

As to the circumvention argument, it is met by the fact that any power of the FCA is subject to the public law requirement that a power not be used for an improper purpose, which is a fact-sensitive inquiry which does not arise in the present appeal. The principle that a public authority is acting unlawfully if it exercises a statutory power with the intention and effect of achieving a result which is regulated by clear and particular procedures of another statutory power so as to circumvent the provisions of the legislation was recognised in R v Liverpool City Council ex parte Baby Products Association [2000] BLGR 171 per Lord Bingham at p.178; and in R (British Bankers Association) v FSA [2011] Bus L R 1531 as potentially applicable to s. 404 (in its original form) in the context of an argument that proceeding by way of a policy statement and letter of standards in relation to PPI mis-selling breached the principle, although the decision in that case was that the FSA had not contravened the principle: see [189] and [255]-[261].

The FCA’s argument is not properly to be characterised as leading to absurd or surprising results. I would readily accept that where none of the Four Conditions are fulfilled, it will rarely be the case that the FCA would be able to justify the imposition of a redress requirement as rational. I would not, however, accept that it is an empty class. The FOS, whose function is equally one of consumer protection, may impose redress in circumstances where none of the Four Conditions is fulfilled. If the FCA took the same view in a case fulfilling criteria which would justify a FOS award, there is no good reason why it should not be able to impose the redress which FOS could award as a condition of continued permission. Nor should equivalent conduct in relation to unregulated activities necessarily fall outside the scope of the FCA’s ability to order redress as a permission requirement. If an authorised person has engaged in inappropriate but not illegal behaviour outside the regulated activity workspace, such as rudeness or bullying which has caused distress or inconvenience, the FCA might rationally take the view that some redress was required if the person were to continue to be regarded as a fit and proper person. If an ombudsman had determined that an authorised person should do something other than pay money, as it is empowered to do under s. 229(2)(b) even if a court could not order the steps to be taken, for example issue a public apology, it would be surprising if the FCA could not make the taking of such steps a condition of continued permission.

By contrast, BCMUK’s case would have very surprising commercial consequences because it requires all Four Conditions to be fulfilled and involves an inability to impose a requirement which is similar to a consumer redress scheme (i.e. contains only one of the potential constituent elements identified in s. 404(5)-(7)) where three of the Four Conditions are fulfilled and only one not. That would arise, for example where an authorised firm had made a profit through an actionable breach of duty owed to consumers who have suffered no loss. On BCMUK’s case the FCA could rationally conclude that such conduct justified a restitution order requiring disgorgement of the profit; and that it justified a cancellation of permission under s. 55J (which may be made by reference to the same objective of consumer protection); but that it could not, instead of cancelling permission, resort to a less onerous option of imposing disgorgement as a requirement of continued permission.

Moreover the logical consequence of BCMUK’s argument, as Mr Herberg accepted in argument, was that prior to 2010 there was no power under s. 45 to make an own initiative requirement which required payment of money; the power was only created for the first time in 2010 by the enactment of s. 404F(7). That would be a most surprising lacuna in the FCA’s powers under s. 55L to pursue the consumer objective.

BCMUK argued that its characterisation of the FCA’s argument as giving it unfettered power was a fair one, and not to any significant extent met by a recourse to public law constraints: given the breadth of the discretion afforded, by reference to a subjective judgement of what the FCA considers an appropriate degree of consumer protection, itself a wide and ill-defined concept, a rationality challenge would in practice be very hard to make out; in truth the discretion simply gave the FCA an ability to exercise a subjective judgement arbitrarily, and the safeguard of an ability to mount a challenge in the Upper Tribunal did not change the fact that public law grounds for such a challenge were so remote as to provide no or no sufficient protection.

This argument is of limited weight because, even if accepted to its full extent, it would equally apply to many aspects of the FCA’s undoubted power under s. 55L, including its ability to impose restitution requirements which it could impose as a sanction under s. 384. However its weight is further diminished by two considerations.

First, it is commonplace in regulation of complex market activity to have rules and powers which are expressed in general terms and by reference to high level objectives, and to leave the discretion as to how they are to be fulfilled to the expertise of an experienced regulatory body of experts. That is especially necessary in the field of financial markets activity covered by the FCA’s regulatory remit, which will potentially involve a myriad of different factual circumstances in a complex market with constantly evolving and novel products and services, something which is positively to be welcomed on a macro-economic level. The FCA is obliged by s. 1B(1)(a) to act compatibly with its strategic objective of ensuring that the relevant markets work well, and by s. 1B(4) to discharge its functions in a way which promotes competition on the interests of consumers. By s.1E(2)(e) the FCA must in pursuing its competition objective have regard to how far competition is encouraging innovation. The narrower and more prescriptive the terms in which its powers and rules are expressed, the less likely they are to provide an effective tool for regulating financial market activity to achieve these objectives. It is therefore neither surprising nor objectionable that the FCA, as the specialist and expert regulatory body, should be afforded by Parliament a wide measure of subjective discretion in seeking to achieve the defined statutory objectives.

Secondly the remedies available to authorised persons are greater than simply those available in a judicial review of other statutory exercises of discretion because of the powers conferred on the Upper Tribunal when a supervisory notice imposing a variation requirement is referred to it. Section 133(6) and (6A) FSMA provide that in such a case:

Unlike a judicial review, therefore, the Upper Tribunal can investigate and make its own findings as to the facts and is not bound to accept the FCA’s findings subject only to a rationality challenge, as a court would be in a normal public law judicial review. In this case, for example, the Upper Tribunal will hear witness evidence and be able to investigate and reach its own conclusions as to the existence and scope of the conflict of interest, the reasonableness of steps taken to meet it, the extent of any loss, and an appropriate methodology for quantification of any redress. The same is true of issues of mixed fact and law, covering the evaluative assessments which are often at the heart of discretionary decisions: issues of fact and law and mixed fact and law are matters for the Upper Tribunal’s own determination. This is the effect of (6A)(a). Under (6A)(b) the Upper Tribunal can decide what matters should or should not be taken into account. Mr Herberg submitted that this goes further than the public law scrutiny which can take place in a judicial review, and involves the Tribunal determining for itself what matters should be taken into account and in what way, not merely considering whether the FCA took into account an irrelevant matter or failed to take into account a relevant matter. That may be so, but I express no concluded view about it. The “procedural or other steps to be taken in connection with the making of the decision” which the Upper Tribunal may decide under (6A)(c) are not restricted in any way, and are wider than powers on a judicial review; the Tribunal could in an appropriate case, for example, provide that the step to be taken was to consult and follow recommendations of some other person or body, such as the Treasury; or requiring some binding reference to mediation, arbitration or expert opinion of another person or tribunal.

Before us the parties agreed that the jurisdiction was wider than that of a normal judicial review and referred to it as ‘JR plus’. It is important to emphasise that the “plus” is a very considerable additional safeguard which in important respects is by way of an independent determination of the relevant facts, issues and procedures by the Upper Tribunal irrespective of the view taken by the FCA in the relevant notice or the process prior to the reference.

I have derived no assistance from the extraneous materials upon which BCMUK relies.

In R (O) v Home Secretary [2023] AC 255 Lord Hodge DPSC said at [30]:

“External aids to interpretation therefore must play a secondary role. Explanatory Notes, prepared under the authority of Parliament, may cast light on the meaning of particular statutory provisions. Other sources, such as Law Commission reports, reports of Royal Commissions and advisory committees, and Government White Papers may disclose the background to a statute and assist the court to identify not only the mischief which it addresses but also the purpose of the legislation, thereby assisting a purposive interpretation of a particular statutory provision. The context disclosed by such materials is relevant to assist the court to ascertain the meaning of the statute, whether or not there is ambiguity and uncertainty, and indeed may reveal ambiguity or uncertainty: Bennion, Bailey and Norbury on Statutory Interpretation, 8th ed (2020), para 11.2. But none of these external aids displace the meanings conveyed by the words of a statute.”

The July 2009 Treasury White Paper which preceded FSA 2010 does not clearly disclose the mischief aimed at or the legislative intention lying behind the statutory language. Nor does it clearly assist BCMUK’s case. Each side spent a little time arguing that it favoured its own construction, which only served to reinforce the conclusion on a first reading that it does not clearly point in either direction.

As to the two examples of subsequent delegated legislation relied upon, “the courts have frequently stressed the need for delegated legislation to be roughly contemporaneous with the passing of the Act to be persuasive” as a guide to interpretation of an Act: see Bennion, Bailey and Norbury on Statutory Interpretation (8th ed) at [24.18]. The two statutory instruments relied on are not roughly contemporaneous with the introduction of s. 404F(7) in the FSA 2010. They were passed in 2013 and 2018 respectively.

As to FCA guidance, this has no special status in relation to the interpretation of its statutory powers; it is for the court not the FCA to determine the correct construction of the statute: Chief Constable of Cumbria v Wright [2006] EWHC 3574 (Admin) [2007] 1 WLR 1407 at [17]; Grays Timber Products Ltd v Revenue and Customs Commissioners [2010] UKSC 4 [2010] 1 WLR 497 at [55]. In any event, the guidance is equivocal on the point. BCMUK can point to passages in some documents which are more consistent with its construction, for example 2016 guidance found in CONRED (the consumer redress section of the Handbook) paragraph 1.8.1(G) which provides:

“Section 404F(7) of the Act empowers the FCA to require a firm ‘to establish and operate a scheme which corresponds to, or is similar to, a consumer redress scheme’ established under section 404 of the Act.”

The reverse is true of other parts of guidance documents. For example CONRED paragraph 1.8.3(G), which comes two paragraphs after 1.8.1(G) quoted above, says:

“The relevant triggers for determining whether the FCA can require an authorised person with a permission to establish and operate a scheme which corresponds to, or is similar to, a consumer redress scheme are different to those that apply for an ‘industry wide’ consumer redress scheme established under section 404 of the Act. Rather than considering the test set out in section 404(1) of the Act, the FCA has to consider the relevant legal triggers for varying a permission or varying or imposing a requirement on a firm (see sections 55H, 55J and 55L of the Act).”

These points were argued with conspicuous skill by the juniors on each side, Mr Ratan for the FCA and Mr Burgess for BCMUK.

BCMUK relied on A1P1 in two ways. First it argued that as a matter of statutory interpretation, its construction was to be preferred because s. 3 of the Human Rights Act 1998 (‘HRA’) required FSMA to be interpreted in a way which is compatible with Convention rights, and the FCA’s construction was necessarily inconsistent with A1P1 rights. Secondly, it argued that the FSN in this case breached its A1P1 rights in contravention of s. 6 HRA which obliges the FCA to act compatibly with such rights. The two arguments are conceptually distinct: the first depends on whether it can be said that the FCA’s construction necessarily involves a breach of A1P1 rights, which is an inquiry which looks to the potential application of the FCA’s construction on hypothetical facts, rather than the circumstances of the present case. The second argument, by contrast, is fact specific to the contents of the FSN in this case. I will take each in turn, dealing first with the s. 3 HRA argument of construction.

It was common ground that A1P1 rights are qualified rights which may properly be invaded by a provision where three conditions are fulfilled, namely that the provision is (1) in the public interest; (2) subject to conditions provided for by law, and (3) proportionate in striking a fair balance between the interests of the community and the property rights of the individual.

As to public interest, BCMUK’s argument was that the broad public interest for the discretion in s. 55L/404F(7) is the satisfaction of one of the regulatory objectives, which in this case is identified as consumer protection. It submits that it is incompatible with that objective to be able to exercise the power where there is no loss to consumers by reason of the firm’s wrongdoing. The focus of the argument was on the absence of a requirement to demonstrate recoverable loss, but it logically applies to a case where none of the Four Conditions is present, because the FCA’s interpretation can only be A1P1 compliant if an exercise of the power to deprive a firm of property is capable of serving the public interest in at least some hypothetical instances where that is the case. This aspect of BCMUK’s argument is, in substance, the argument that the FCA’s construction leads to absurd results, which I have already rejected for the reasons given above: there are hypothetical circumstances in which variation of a permission by the imposition of a requirement may satisfy the public interest requirement where none of the Four Conditions exist.

The emphasis on the requirement of loss, which characterised BCMUK’s argument on this point, reinforces this conclusion. It is plainly in the public interest for a firm to be required to disgorge profit made as a result of wrongdoing irrespective of loss to consumers. It is equally in the public interest for contravening firms to make redress where there is no loss as such but distress, inconvenience or damage to reputation. Indeed it seems to me that if BCMUK’s argument were correct that recoverable financial loss is an essential component of what may properly be in the public interest, it would lead to the surprising conclusion that the FOS scheme as a whole is unlawful as contravening A1P1 rights, which I would unhesitatingly reject.

As to the second requirement, of “conditions provided for by law”, this is explained in a number of the authorities, as meaning that the provisions be sufficiently accessible, precise and foreseeable. The applicable principles were discussed by Lord Sumption in R(P) v Secretary of State for Justice [2019] UKSC3 [2020] AC 185 at [17]-[41], reviewing both the Strasbourg and domestic jurisprudence. Without attempting a summary, the essence is that an excessively broad discretion cannot be in accordance with law unless there are sufficient safeguards, on known legal principles, against the exercise of that discretion so as to make its application reasonably foreseeable; safeguards against arbitrary interference with Convention rights are required by the rule of law in order to protect against the abuse of imprecise rules or unfettered discretionary powers.

BCMUK’s argument is that a power to impose a redress scheme without having to establish at least the Duty, Causation and Loss Conditions could not possibly be sufficiently precise and foreseeable to meet these requirements; on the contrary the power would be entirely at large and a firm would have no means of foreseeing in what circumstances it would fall foul of such a power, especially in the absence of anything providing relevant clarification and guidance issued by the FCA. I am unable to accept this argument. Section 55L is not, on the FCA’s construction, an unfettered power. It may only be exercised for defined statutory objectives which are set out in some detail in ss. 1C (consumer protection), 1D (integrity) and 1E (competition). It is subject to well established public law principles which constrain exercises of discretion. It is further subject to the safeguards of judicial oversight in the ‘JR plus’ jurisdiction of the Upper Tribunal. These are sufficient safeguards against it being exercised arbitrarily, which in the words of Lord Bingham in R (Gillan) v Commissioner of Police of the Metropolis [2006] 2 AC 307 at [34] means acting on a personal whim, caprice, predilection or purpose other than that for which the power is conferred. Nor does it lack sufficient foreseeability. A firm will know that it is subject to making redress for contravention of regulatory rules where there is a rational connection between the contravention and what the firm is required to pay as a condition of being regarded as fit and proper to continue to carry out regulated activities. It is not suggested that the power to make restitution under s. 384 lacks the relevant condition of law, or that where it can properly be exercised the same obligations cannot be attached as a condition of continued permission to practice under s. 55L; yet the power there arises without the need to establish loss.

BCMUK advanced a similar argument by reference to common law principles that property or economic interests should not be impaired except under clear authority of law, relying on Swindon Borough Council v DB Symmetry [2022] UKSC 33, [2023] 1 WLR 198 at [41]. This adds nothing to the A1P1 argument and fails for the same reasons. Insofar as BCMUK sought to rely on a principle of construction against doubtful penalisation, it was misplaced: section 55L is not concerned with penalty but with redress in the context of continued fitness to practice; and in any event there is nothing ‘doubtful’ about the application of the discretion if it does not have the Four Conditions as threshold requirements.

As to proportionality, BCMUK’s argument again focusses on loss. It is said that requirements imposed by the FCA are incapable of complying with the proportionality requirement in circumstances where the redress does not bear a reasonable relationship of proportionality to the loss caused by the wrongdoing. This argument too founders by reason of the fact that a legitimate objective of the exercise of the s. 55L power may be to require redress where there is no loss to the consumer. Whether the redress ordered is a proportionate measure to the objective which is sought to be achieved in any individual case is not the question here under consideration. BCMUK’s case under s. 3 HRA is, as it must be, that it can never be a proportionate response if there is no loss. That is unsound for the reasons I have endeavoured to explain.

Turning to the second argument, under s. 6 HRA, that the FSN involves a breach of BCMUK’s A1P1 rights in this case, the submission relied entirely on BCMUK’s case, rejected by the UT Decision, that the FCA’s pleaded case was not capable, if made out, of constituting recoverable loss for the alleged wrongdoing.

In this respect it is important to keep in mind the exercise being undertaken by the Upper Tribunal. It was deciding preliminary issues which had been carefully defined after adversarial argument in the directions order made by UT Judge Herrington of 14 November 2022 (‘the Directions Order’). By paragraph 31 of the Directions Order the questions of law fell to be determined on the assumption that the FCA could prove all the facts upon which it relies in its S/C and draft Rejoinder. The draft Rejoinder at paragraph 18.2 identified the divergence between the results achieved by the Internal and External Funds for the years 2011 to 2015. If proved, they amount to a very dramatic difference in performance. For example, in 2012 the Internal Fund made a gain of about 112% compared to a gain by the External Fund of about 6%. The FCA recognises that it cannot say the divergence in performance can be directly linked to the wrongdoing which is alleged by reference to conflicts of interest and supply of information. Instead the case pleaded at para 173 of the S/C and amplified in the draft Rejoinder is that the service provided by BCMUK involved inadequate arrangements to protect against the conflict of interests and inadequate disclosure, and that those failings were such that the level of service provided to investors was therefore below that which could reasonably be expected to be commensurate with the fees paid by the investors. The quantification of the difference is alleged to be captured by the redress required in the FSN, which is to be calculated so as to deprive BCMUK of management fees to the extent they exceeded those which would have been payable by institutional investors in an alternative investment; and of performance fees save to the extent that the External Fund achieved a higher rate of return than an industry average.

The Directions Order went on to provide at paragraph 32(2) that the facts which were assumed to be proved for the purposes of the issues of law to be decided included the assumptions that a sub-standard service had been provided, and specifically that the failings by BCMUK were such that the level of service received by the investors was below that which could reasonably be expected to be commensurate with the fees paid by those investors.

The common law recognises that in a contractual claim where services provided fall short of the services promised, damages may be awarded for skimped service by reference to the value of the deficient service provided rather than any further loss to the consumer. Chitty on Contracts (35th ed) states at para 30-064:

“if the services provided are incomplete or of less than the contracted quality, the buyer is prima facie entitled to the difference in value between the service promised and the service delivered, whether or not he has suffered any further loss and whether or not he has obtained substitute services from another supplier.”

This is well established at the highest level as a recoverable head of loss: see Attorney General v Blake [2001] 1 AC 268 per Lord Nicholls at 286; White Arrow Express Ltd v Lamey’s Distribution Ltd [1995] CLC 1251 per Sir Thomas Bingham MR at 1256; Morris-Garner v One-Step (Support) Ltd [2018] UKSC 20 [2019] AC 649 per Lord Reed PSC at [80]. An application of the principle is to be found in Giedo van der Garde BV v Force India Formula One Team Ltd [2010] EWHC 2373 (QB).

BCMUK disputed the applicability of these principles to the circumstances of the redress ordered in the FSN. It submitted that it does not follow from the fact that a service is compromised by failures to comply with a firm’s regulatory obligations that the value of the service received by the customer is lower than it would have been if the regulatory obligations had been complied with. However, by reason of paragraphs 31 and 32 of the Directions Order, that was the basis which was to be assumed for the purposes of the issues to be decided by the Upper Tribunal.

BCMUK’s main argument on this point was that the principle only applies where the claimant has suffered some financial loss, and the difference between market value and actual value of the services is then used as a means of determining the quantum of such loss. This argument is unsound. As the authorities referred to above make clear, the principle is that the receipt of the sub-standard service is itself loss, which is quantified by the difference in value of the services, irrespective of any other financial loss which the claimant may or may not be able to establish. The Upper Tribunal correctly analysed the law at [352] to [358] and concluded at [359] to [363] that the case pleaded in the S/C and draft Rejoinder gave rise to a reasonable prospect of success that loss could be made out. There was no error in that conclusion. The conclusion was reached on the assumption, which I would treat as mistaken, that what was required was loss which could be recovered in the courts. It is all the more justified if, as I would hold, the FCA is entitled in an appropriate case to impose redress for loss of a kind which is not recoverable in the courts. It will be a matter for trial before the Upper Tribunal exercising its ‘JR plus’ jurisdiction, following a fact-dependent inquiry, whether the particular nature of the redress imposed by the FSN, which is the mechanism which quantifies what BCMUK must pay, can be justified by the consumer protection objective by reference to such loss. The current issues arise on a strike out application, and BCMUK would have to establish that it would not be open to the Upper Tribunal to conclude that the redress could be justified on the basis pleaded. I have little hesitation in rejecting such a proposition.

For this reason BCMUK’s s. 6 HRA argument by reference to A1P1 rights or common law principles cannot succeed.

The Respondent’s notice point that the Upper Tribunal should have held that the Loss, Causation and Duty requirements were not fulfilled does not arise because those are not conditions to the exercise of the powers in s. 55L.

For these reasons I would allow the appeal on Ground 1, dismiss the strike out application and allow the rejoinder application.

FSMA confers upon the FCA a wide range of statutory powers to undertake the authorisation and supervision of firms and individuals engaged in financial services, which broadly speaking may be characterised as covering authorisation, supervision, and enforcement. Where the FCA proposes to take action in the course of any of these three core functions, it is normally required to give the subject written notice of its proposal or decision. FSMA, and the associated guidance in the FCA’s Handbook, distinguish between two principal categories of decisions which are subject to different decision-making procedures, namely:

The facts of the present case span both categories: the decision to impose a financial penalty on BCMUK was proposed in a warning notice and taken in a decision notice in accordance with ss.206-208 FSMA which provide that in such a case a warning notice and decision notice “must be given”: s. 207(1) and 208(1). The decision to impose a redress requirement was taken by way of supervisory notice in accordance with s.55L and 55Y FSMA.

The first stage is the issue of a warning notice which will follow an investigation in which the FCA has access to documents and information from the firm under the wide ranging investigatory powers contained in Part XI of FSMA. Section s. 387(1)(a) FSMA provides that a warning notice “must…state the action which [the FCA] proposes to take” and s. 387(1)(c) that it must “give reasons for the proposed action”. The warning notice must also specify a reasonable period (of not less than 14 days) within which the person concerned may make representations to the FCA in response to the notice (s. 387(2)). Where s. 394 applies (which it does in the present case) the FCA must give the firm access to the material upon which it relied to reach the decision and any secondary material which might undermine that decision (s. 387(1)(d) and (e)).

If (having considered the representations) the relevant decision-maker within the FCA decides to take the proposed action, this must be set out in a decision notice under s.388 FSMA. The decision notice must “give the reasons of [the FCA] for the decision to take the action to which the notice relates” (s. 388(1)(b)) and also give an indication of any right to refer the matter to the UT (s. 388(1)(e)). By s.388(2) “the action to which the decision notice relates must be under the same Part as the action proposed in the warning notice”. Following a decision notice there is an opportunity to refer the matter to the Upper Tribunal, which must deal with it in accordance with s. 133 FSMA (to which I shall return). Again if s. 394 applies (which it does in the present case) the FCA must give the firm access to the material upon which it relied to reach the decision and any secondary material which might undermine that decision.

Subject to any reference to the Upper Tribunal, the decision taken by the FCA will then be recorded in a “final notice” and become operative under s.390 FSMA.

The FCA has introduced an ‘expedited reference procedure’ set out in the FCA’s Handbook: the Decision Procedure and Penalties Manual (“DEPP”) at DEPP 5.1.8E to J, by which a firm may, if it wishes, challenge the proposed action before the Upper Tribunal without engaging with the FCA’s internal decision-making process. That is what BCMUK chose to do in respect of the WN in this case. Accordingly it made no representations on the WN to the FCA and the DN was issued in materially identical terms to the WN.

This notice process, and the Upper Tribunal’s statutory reference jurisdiction under s.133 FSMA, is not confined to the enforcement context. As set out in Annex 1 in Chapter 2 of DEPP, there is a large and diverse range of actions for which warning notices and decision notices are required (some one hundred instances under FSMA alone), including not only enforcement action but also matters concerning the grant or refusal of authorisation and the ongoing supervision of authorised firms. By way of example, in each case where the FCA refuses an application for permission to carry on a regulated activity or a controlled function it must issue a warning notice and a decision notice to the firm or individual concerned, who will then have the right to refer the matter to the Upper Tribunal if they so wish.

The term “supervisory notice” is defined in s.395(13) FSMA which provides a list of all of the written notices required under FSMA falling within this description. Once again, FSMA sets the parameters for a staged decision-making process in respect of supervisory notices and this is further detailed in Chapter 2 of DEPP. There are over a hundred different types of decision, identified in DEPP Chapter 2 Annex 2, to which this procedure applies.

The first stage is the issue of a first supervisory notice. Where this notice concerns the exercise of the FCA’s power under s.55L, it must give details of the action that is being taken (s.55Y(5)(a)) and “state the regulator’s reasons for the variation of the permission or the imposition or variation of the requirement” (s.55Y(5)(b)). The first supervisory notice may take effect immediately (if the notice so specifies) or after a prescribed period of time (s.55Y(2)). The firm may then refer the matter to the Upper Tribunal (which is what happened in this case), or make further representations, or both, and the notice must inform the firm of those rights (s. 55Y(5)(c) & (e)). If the FCA receives representations in response to a first supervisory notice, it will consider those representations and decide whether to give a second supervisory notice (DEPP 2.3.1). In the context of the power to impose a requirement under s.55L, if the FCA concludes that the same form of requirement proposed in the first supervisory notice remains appropriate, it will embody this in the second supervisory notice and must inform the subject of their right to refer the matter to the Upper Tribunal (ss.55Y(7) & 55Y(9)). If the FCA decides to impose a different requirement, the second supervisory notice must again set out the reasons for the imposition of this requirement (ss.55Y(8)(b) & 55Y(10)), inform the person concerned of their right to make representations to the FCA in response to the notice (s.55Y(5)(c)), and inform the person of their right to refer the matter to the Upper Tribunal (s.55Y(5)(e)).

FSMA provides for a degree of internal separation in the FCA’s decisions to issue statutory notices, including warning notices, decision notices and supervisory notices. In particular, s.395(2)(a) FSMA requires that such decisions must be taken by a person not directly involved in establishing the evidence on which the decision is based, or a body which includes at least one such person. In practice the FCA does not involve any investigators in the decision to issue statutory notices. The FCA has instead established a process whereby the decision to issue a statutory notice will be taken by an internal decision-making committee, the Regulatory Decisions Committee (‘the RDC’), whose procedures are set out in Chapter 3 of DEPP; or alternatively by senior staff of the FCA under the FCA’s “executive procedures” or “settlement decision procedure”, as set out in Chapters 4 and 5 of DEPP respectively. In this case the decisions were made by the RDC.

The RDC is a committee of the board of the FCA and is chaired by an employee who is otherwise separate from the FCA’s Executive. The other members of the RDC are entirely independent of the FCA, and comprise a mixture of financial services practitioners and other lay members. The decision-making process of the RDC (and that of the FCA’s senior staff under the executive procedures or settlement decision procedure) is administrative in nature: these bodies do not (and are not intended to) satisfy the requirement for an “independent and impartial tribunal” under Article 6 ECHR. The Upper Tribunal reference process (as further addressed below) is therefore an essential part of the statutory procedure in order to ensure compliance with the right to a fair hearing before an independent and impartial tribunal under Article 6 ECHR, as is well recognised: see e.g. Financial Conduct Authority v Carillion plc [2021] EWHC 2871 (Ch) [2022] Ch 162 at [94].

The UT Decision, and the parties in argument before us, treated the scope of the Upper Tribunal’s jurisdiction on a reference as being defined in s. 133 FSMA. Although it makes no difference to the analysis, to my mind it is equally to be found in the relevant sections which confer the ability to refer something to the Upper Tribunal. In this case it is found, for the FSN, in s. 55Z3(2) FSMA, which provides: “An authorised person who is aggrieved by the exercise by either regulator of its own-initiative variation power or its own-initiative requirement power may refer the matter to the Tribunal”; and for the s. 206 disciplinary penalty it is found in s. 208(4)(b) FSMA which provides that “If a regulator decides to… impose a penalty on an authorised person under s. 206….the authorised person may refer the matter to the Tribunal”. In each case, therefore, the right to refer arises when the decision is made, upon which what may be referred is “the matter”. This is the language used in all provisions of FSMA which confer the right on the firm (but not the FCA) to initiate a reference to the Tribunal.

What the Upper Tribunal may do upon such a reference is identified in s. 133 FSMA. Although we are concerned only with “references” from the FCA, the section also governs appeals from FCA decisions, and appeals and references from decisions of the Prudential Regulation Authority and the Bank of England. The nature of the decisions in those other cases, and their decision-making processes, were not explored in argument before us and nothing I say should be assumed necessarily to apply to such other cases.

I have already set out subsections (6) and (6A) of s. 133 but it is convenient to repeat them here within the other relevant parts of the section:

The distinction between “disciplinary references” which are to be dealt with in accordance with subsection (5) and other references which are to be dealt with in accordance with subsections (6) and (6A) was introduced by FSA 2012. Prior to that time all references were to be dealt with in accordance with subsection (5). Subsection (7A) defines what constitutes a “disciplinary reference” for this purpose and it includes a penalty imposed under s. 206 FSMA.

I have already identified the nature of the ‘JR plus’ jurisdiction for non-disciplinary references, for which subsections (6) and (6A) provides. As originally enacted, when the jurisdiction in all cases was that contained in s. 133(5) for the Tribunal to determine the appropriate action, s. 133(7) imposed an express limitation that on a supervisory notice reference it could not direct the FCA to take action which required a decision notice. Section 133(7) has been repealed but the limitation is essentially preserved for non-disciplinary references, which cover all supervisory notice references, by the remission to the FCA in such cases under subsection (6)(b) to reach a decision in accordance with the findings of the Tribunal, which it could not do, following a supervision notice reference to the Tribunal, by taking an action governed by the decision notice procedure, without giving a warning and decision notice in accordance with the mandatory requirements of FSMA.

For other references the jurisdiction under subsection (5) is wider. It is for the Upper Tribunal to decide what action is appropriate and to direct the FCA to take it. The only express constraint is in s. 133A which provides that the Tribunal may not direct the FCA to take action it would not have had power to take because of the restriction in s. 388(2), in other words imposing an action under a different Part of the Act from that proposed in the warning notice.

Procedure in the Upper Tribunal is governed by UT Rules as updated from time to time. Rule 2 contains the overriding objective of the Rules, by reference to which they are to be interpreted and applied, which is expressed as being to enable the Upper Tribunal to deal with cases fairly and justly. By Rule 2(2), dealing with cases fairly and justly includes, amongst other things, avoiding unnecessary formality and seeking flexibility in proceedings, using any special expertise of the Upper Tribunal effectively and avoiding delay so far as compatible with proper consideration of the issues.

Rule 5 contains case management powers which by rule 5(1) are powers to regulate its own procedure subject only to any provisions of the Tribunals, Courts and Enforcement Act 2007 (‘TCEA’). Rule 5(3) sets out non-exclusive examples of such powers, including a power in rule 5(3)(c) to permit or require a party to amend a document.

Rule 26B and Schedule 3 make provision for financial services cases, in which the procedure on a reference is initiated by a reference notice by the applicant. The reference notice must identify the issues which it wishes the Upper Tribunal to consider (para 2(3)(d) of Schedule 3). It is then for the respondent (the FCA) to plead first by a statement of case, which must, amongst other things, state the reasons for the referred action and set out all the matters and facts upon which the respondent wishes to rely to support the referred action (para 4(2)). The respondent must at the same time give disclosure of documents on which it relies in support of its case and any material which in its opinion might undermine the decision to take the referred action. The applicant (here BCMUK) must then serve a reply which not only responds to the statement of case but also states the grounds on which the respondent relies in the reference (para 5(2)).

It is to be noted that the general jurisdiction of the Upper Tribunal may be narrowed in any individual case either by the parties, or by the Upper Tribunal itself exercising its case management powers. It may be narrowed by the firm in framing the issues in its reference notice or reply. It may be narrowed by the FCA in framing its statement of case. It may be narrowed by the Upper Tribunal exercising its powers in Rule 5(3)(c) to require the respondent to amend its statement of case to remove allegations which it would not be in accordance with the overriding objective to permit it to pursue; or the applicant to amend the reply on similar grounds. It was common ground between the parties that this would allow it not merely to refuse amendments as a matter of discretion, but to impose deletions for the same discretionary reasons. By this means the Upper Tribunal’s jurisdiction contains a wide discretion to refuse to allow the FCA to pursue allegations in the reference if it would not be fair and just to allow it to do so, notwithstanding that they fell within its gateway jurisdiction of “the matter” referred. Defining the scope of that gateway jurisdiction is not, therefore, the only means of limiting the scope of what the Upper Tribunal may decide upon, or even the primary means of so limiting it: whatever the scope of the gateway jurisdiction, the Upper Tribunal retains the ability to restrict it by reference to what is just and fair in relation to each individual case applying its powers flexibly.

The Upper Tribunal started by observing that the words “matter” and “subject-matter of the reference” in s. 133 FSMA were not defined but, in their broadest terms read with s. 208(4)(b), the subject-matter of the reference in relation to the DN was the FCA’s decision to impose a financial penalty on BCMUK. However the authorities, it said, give it a narrower interpretation ([48]).

The authorities considered were, with one exception, decisions of previous constitutions of the Upper Tribunal or its predecessor the Financial Services and Markets Tribunal (‘the FSMT’). Taking them in chronological order they were Jabre v Financial Services Authority (10 July 2006, unreported); Allen v Financial Services Authority (5 June 2013, unreported); Financial Conduct Authority v Hobbs [2013] EWCA Civ 918, [2013] Bus LR 1290; Khan v Financial Conduct Authority [2014] UKUT 186 (TCC); and Seiler, Whitestone & Raitzin v Financial Conduct Authority [2023] UKUT 0133 (TCC). I will address them and what the UT Decision said about them below.

The UT Decision also treated the notice regime as important, requiring reasons to be given both for the warning notice and the decision notice. The warning notice requires such reasons so that its subject may make effective representations as to why the action should not be taken; the same applies following a decision notice: the affected person has reasons which enable it to contest the matter by referring the decision notice to the Tribunal knowing clearly what the allegation is ([58]-[59], [171]).

The UT Decision expressed its conclusion in the following terms:

“75.

We are satisfied that the passages from Jabre and the later authorities are to be read as follows. In order for such allegations to form part of the subject matter of the reference and be considered or determined by the Tribunal, they must be of the same nature and based upon the same factual background as the allegations made to the RDC and contained in the warning and decision notices, even if no findings are made upon them therein.

76.

If the Tribunal is of the view that the new matters on which the Authority seeks to rely do fall within the scope of the subject matter of the reference, even though not contained in the warning notice, then the Tribunal has a discretion as to whether to allow matters concerned to be pleaded, exercised through its case management powers. In that regard, see the discussion at [988] to [1012] of Whitestone, which we refer to in more detail below. In those circumstances, as that discussion indicates, the burden on the Authority to satisfy the Tribunal that the new allegations should be pleaded will be a heavy one.

77.

Clearly, if the Tribunal takes the view that the matters concerned do fall within the scope of the allegations made in the warning notice as well as the facts and matters on which the Authority relied in support of those allegations, then it will simply be a case of the Tribunal considering whether, exercising its case management powers in accordance with principles referred to below, it should permit the matters concerned to be pleaded.

78.

In relation to the FSN, which as we have said, is the subject of a non-disciplinary reference, Parliament, in introducing the distinction between disciplinary and non-disciplinary references by way of the Financial Services Act 2012, considered that “matter” was limited to allegations which had been properly ventilated before the RDC, such that there could be an intelligible and meaningful review of their lawfulness and reasonableness on a non-disciplinary reference. Parliament, in other words, cannot have intended or considered “matter” (in either subsection) to have a wider meaning than it was given in the Jabre case back in 2006, requiring the allegation to have been ventilated before the RDC.

79.

This in no way prevents the Authority, on a reference to the Tribunal, from relying on fresh evidence that was not capable of being contained within the Decision Notice in relation to allegations or findings that are contained in the Decision Notice. Section 133(4) of the Act specifically contemplates this and empowers the Tribunal to make findings on a reference based upon evidence that was not available to the decision-maker (in this case the RDC of the Authority) at the time of the decision so long as the evidence relates to the subject matter of the reference.

…

189.

The “matter referred” and thus the ambit of the Tribunal’s jurisdiction is the “allegations made in the decision notice and the circumstances on which these are based” (Jabre at [28]).

When applying this test to Amendments 3 & 4, the Upper Tribunal accepted that the new reliance on Principle 7 and COBS to a large extent relied upon “the same body of disclosure” to that pleaded in support of the Principle 8 allegation, and relied on the same regulated activity. It said that despite the need not to take too narrow an approach, “there must come a point at which a hard edge is applied to the Tribunal’s jurisdiction”. ([168]). It sought to define that edge by reference to the question whether the amendment proposed embraced an allegation which was one of the key reasons for the regulatory action proposed in the warning notice ([169]). In order to form part of the subject-matter of the reference the amendments must not only be based on the same factual background as the allegations made to the RDC but also be of the same nature. Where the allegation relies on a different regulatory provision it cannot be considered to be of the same nature ([170]). Moreover the amendments seek to rely on additional facts not previously pleaded in the context of the allegation of breach of Principle 8, namely the allegation at para 174 of the draft amended S/C relating to the knowledge of individuals to whom the FCA seeks to attribute responsibility for the alleged misconduct ([173]). The matter referred, and thus the ambit of the Tribunal’s jurisdiction, is the allegations made in the decision notice and the circumstances on which these are based (Jabre at [28]); the alleged breaches of Principle 7 and/or COBS were not advanced or referred to in the DN; and they were not contained or foreshadowed in the WN nor were they raised by the RDC of its own initiative, nor otherwise discussed during the RDC process ([189]). Those amendments were therefore outside the scope of the Tribunal’s jurisdiction, and could not be allowed. The UT said it did not need therefore to consider how it would have exercised its discretion if it had jurisdiction ([200]).

Amendment 2, by contrast, was within the jurisdiction of “the matter” referred ([140]-[151]; and as a matter of discretion should be allowed ([152]-[162]).

The FCA’s primary argument is that the jurisdiction in s. 133 FSMA is confined only by the express limits in FSMA. In particular, in relation to decision notice references it is constrained only by s. 388(2) and s. 133A so that the action which the Tribunal can direct the FCA to take is limited to action under the same Part of FSMA as that proposed in the warning notice; and in the case of a supervisory notice reference, it is confined to remitting it for the FCA to take action which it would be entitled to take by way of supervisory notice but not an action which required a warning and decision notice. Beyond that, anything which the FCA seeks in the reference is within the Tribunal’s jurisdiction as constituting the matter referred. This is sufficiently moderated and balanced by the Tribunal’s wide discretionary case management powers.

The FCA’s secondary and alternative argument is that “the matter” encompasses anything which arises from the same factual situation which gave rise to the regulatory action in the statutory notice referred to the Tribunal, or is otherwise connected with the circumstances, the evidence and/or the allegations, whether factual or legal, which were before the FCA’s decision-maker.

The FCA’s further and tertiary alternative is that however narrowly it is appropriate to construe “matter”, it extends to legal relabelling of existing factual allegations, which is in substance all that Amendments 3 & 4 do in this case.

On any of these three alternatives, the FCA contend that the Upper Tribunal had jurisdiction to allow Amendments 3 & 4. If we accepted that conclusion, it would be open to this court either to remit the matter to the Upper Tribunal for it to perform the exercise of discretion which it held it was unnecessary to embark upon, or to remake the decision by exercising its own discretion, by reason of s. 14(2)(b)(ii) TCEA. After some equivocation the FCA asked us to take the latter course, to which BCMUK also agreed if it arose.

BCMUK’s primary case on jurisdiction is that the UT Decision defined its jurisdiction too widely. The “matter” was limited to both the factual and legal content contained in the relevant notice which constituted the decision. Anything else was excluded, although further evidence of such breaches is permitted by s. 133(4). The statement of claim on a reference could only, in effect, replicate the contents of the decision notice or supervisory notice.

Its secondary and alternative case was that the Upper Tribunal was right on the scope of its jurisdiction for the reasons it gave.

Within these two alternatives BCMUK placed particular emphasis, as had the UT Decision and the Upper Tribunal in Seiler, on the statutory scheme which required the FCA at each stage to give its reasons and disclose the relevant documents. It also submitted that its construction was supported by a consistent line of authority against the background of which FSA 2012, and the Financial Services Act 2021 (‘FSA 2021’) and the Financial Services Act 2023 (‘FSA 2023’) were enacted, continuing the language of “matter”, such that the Barras principle (Barras v Aberdeen Steam Trawling and Fishing Co Ltd [1933] AC 402) required them to be interpreted in the way contended for. BCMUK also argued that a different approach was justifiable to questions of honesty and integrity in forward looking decisions about fitness to practice from those involving backward looking decisions to impose redress or penalties, which explained a number of the authorities.

On either basis Amendments 3 & 4 were outside the Upper Tribunal’s jurisdiction. So too was Amendment 2. Alternatively all the amendments should have been refused as a matter of discretion.

We were referred to a number of cases in addition to those referred to and relied on in the UT Decision (which made passing reference to some of them in addition to the five I have identified above). Three are Court of Appeal decisions. The others are decisions of the Upper Tribunal or its statutory predecessor the FSMT. Although the latter are not binding on us, they comprise a body of case law of tribunals with specialist expertise and experience in this area which commands respect, and for that reason I will include a summary of those of them which involve some relevant reasoning or where the decision points in a particular direction. It is convenient to address them in chronological order, not least because it assists in addressing BCMUK’s reliance on the Barras principle.

In Parker v Financial Services Authority 26 October 2004 (unreported) the reference was of a decision notice imposing a penalty for market abuse by the misuse of information. The FCA applied to amend its statement of case to introduce a new allegation of market abuse comprising the sale of shares. The Tribunal held that it had no jurisdiction to allow the amendment because the new allegation did not form part of the matter referred. It held that it was the decision notice which was the matter referred, being the action taken and the allegations which form the reasons for it, and it was not possible in effect to amend the decision notice to introduce a new allegation, for three reasons. First market abuse was a serious allegation and the procedures laid down by FSMA should be followed. Secondly, FSMA prescribes the use of a warning notice which should not be circumvented or dispensed with. Thirdly, the Tribunal Rules do not envisage that such an allegation can be introduced after the respondent has served its reply to the statement of claim.

This approach is that for which BCMUK contends in its primary argument. It is not how the FSMT and Upper Tribunal jurisprudence has developed, and is not in my view sound, for the reasons I identify below.

Next comes Jabre v Financial Services Authority (2006) which has been treated in the Tribunal jurisprudence as a seminal foundational guide. In that case the warning notice sought to withdraw Mr Jabre’s approval but the RDC, after submissions from Mr Jabre, determined in the decision notice to impose a financial penalty without withdrawing Mr Jabre’s approval. In the reference, the FCA sought to maintain its case in the warning notice that approval should be withdrawn, alternatively impose a prohibition order. Mr Jabre argued that that was not within the matter referred. The FSMT, chaired by Stephen Oliver QC, rejected Mr Jabre’s objection and held that there was jurisdiction. The primary reasoning at [28] and [29] was that the decision itself was not strictly speaking a relevant consideration for the tribunal, but rather it was the allegations made in the decision notice and the circumstances on which these are based which fall to be considered and evaluated, and which comprise the matter referred, together with any further evidence falling within s. 133(4); and that the matters upon which the FCA wished to rely had all been before the RDC when making the decision reflected in the decision notice, either as recorded in the decision notice itself or as referred to as matters on which the RDC made no decision.

The UT Decision treated Jabre as establishing that “the matter” was circumscribed such that the foundations for it must be contained in the decision notice ([55]). It cannot, however, be treated as a case in which the jurisdiction is confined to the allegations which formed the basis for the relief in the decision notice, because the FCA was permitted in Jabre to advance allegations which it had made to the RDC but on which the RDC had made no findings. It therefore suggests that the jurisdiction is wide enough to go beyond both the action, and the allegations providing the reasons for action, in the decision notice. The reasoning in Jabre included, although was not wholly based on, a passage at [30]-[31] which echoed the FCA’s primary submission to us, by reference to s. 388(2) and s. 133(7) as “a clear code” that the Tribunal’s jurisdiction enables it to depart from the FCA’s decision referred if in the case of a decision notice reference it does not impose an action which is not in the same Part of FSMA as the warning notice, and in the case of a supervisory notice reference, is action which could be taken by a supervisory notice rather than a decision notice preceded by a warning notice. That too is an approach which does not commend itself to me for reasons explored below.

It was a case which on its facts involved allegations which had been advanced in the warning notice and before the RDC. It did not, therefore, cut across what BCMUK suggests is the fundamental importance of the warning notice procedure. However for reasons I will explain, changes from the warning notice stage are not a jurisdictional bar, although they may be relevant matters in deciding whether to exercise the jurisdiction (as the Upper Tribunal chaired by Judge Herrington itself thought in Seiler at [995]-[999]).

In Nazia Bi v Financial Services Authority 22 September 2011 (unreported), the notices withdrew permission and imposed financial penalties for breach of APER 6 (exercise of due skill care and diligence). In the Tribunal reference the FCA sought to rely on conduct comprising providing false and misleading information in the investigation as a breach of APER 4 (duty to deal with the FCA in an open and cooperative way). At the hearing the Tribunal raised a doubt whether s. 133 authorised it to determine that the FCA could rely on a failure to comply with APER 4 which would potentially involve different and more onerous sanctions, and opined that it would have to be satisfied that the warning and decision notice procedure had been satisfied. It did not need to decide the question because having asked for written submissions the FCA did not pursue a case that separate penalties should be imposed in respect of the alleged breaches of APER 4. It does not therefore represent a decision on the scope of “the matter” after adversarial argument.

Allen (2013) was a case in which the FCA had issued a decision notice under s. 56 FSMA against Mr Allen on the basis that he was not fit and proper to perform any function in relation to regulated activities because he lacked honesty and integrity. The allegation was that in arranging insurance for a client firm of solicitors he had overcharged by adding fees to their premiums and concealed it; and had misappropriated money due to the firm; and that additionally he had misappropriated money due to his employer. In unrelated proceedings in Australia the judge found that the witness upon whose evidence the FCA had relied for the allegations of Mr Allen’s lack of honesty and integrity was not a witness of truth, and consequently the FCA did not seek to rely upon those allegations before the Tribunal. Instead the FCA sought to rely on other findings of the judge against Mr Allen in the Australian proceedings, of forging or causing to be forged a signature on a document; giving an untrue account of a key meeting; colluding with witnesses in the submission of false evidence; producing and submitting false documents and evidence to bolster his case; and repeatedly lying to the court. These were wholly unrelated to the allegations which had been relied on in the decision notice. The FCA sought to amend its statement of case to rely in the proceedings before the Tribunal on these new allegations to support the prohibition order. The Tribunal allowed the new allegations to be pleaded and treated them as within the scope of the matter referred because the issue referred was simply whether Mr Allen lacked honesty and integrity, and it was at that level of generality that “the matter” fell to be defined.

The UT Decision explained Allen as a case applying s. 133(4) and therefore as one allowing new “evidence” relating to the subject matter of the reference ([65]) and emphasised that the material had not been available at the time of the warning notice or decision notice ([70]). Nevertheless, it suggests a broad construction of “the matter”. Although it based itself on Jabre as permitting “allegations” which were in the decision notice, it defined the allegation at a sufficiently high level of generality (i.e. that Mr Allen lacked honesty and integrity), as to permit what were allegations of wholly new conduct to be advanced.

R (Willford) v Financial Services Authority [2013] EWCA Civ 677 was a case in which the FCA had issued a decision notice imposing a penalty of £100,000. Mr Willford did not seek to refer the decision but rather brought judicial review proceedings on the grounds that the decision lacked adequate reasons. Silber J gave leave for judicial review and quashed the decision on the grounds that the reasons were inadequate. The FCA appealed on the grounds that judicial review was inappropriate because there was an adequate alternative remedy in the form of a reference to the Tribunal. The decision did not therefore directly concern the scope of the Tribunal’s jurisdiction where there is some departure from the position advanced in warning and decision notices, but is of interest for some dicta by Moore-Bick LJ, with whom Black LJ agreed (Pill LJ giving his own reasons for the decision).

Having recited the statutory procedure by way of the warning notice, decision notice and reference to the Tribunal, Moore-Bick LJ concluded at [6] that;

“On a reference to the tribunal the FSA may, if it thinks fit, broaden the scope of its allegations (provided they relate to the same matter) and the tribunal has power to direct that a more severe penalty be imposed than that which the FSA previously proposed….”

And at [37] when rejecting the argument that Mr Hobbs needed fuller reasons in order to be able to decide whether to make a reference to the Tribunal he said:

“Moreover, I am not persuaded that he needs the RDC’s reasons to be stated more fully in order for him to make an informed decision whether to refer the matter to the tribunal. The risks that the FSA might put forward additional allegations and that the tribunal might increase the penalty are there and have to be taken into account.”

This suggests that “the matter” is not so narrowly confined as to be limited to things which are in the warning and decision notice; and nor is the FCA constrained by the decision notice so as to prevent it from seeking a more severe penalty.

At [21] Moore-Bick LJ said:

“It is important, in my view, to begin by considering the regulatory scheme established by the Act. I have already described its essential characteristics, which I need not repeat. The disciplinary procedure involves, first, an administrative process under which the FSA decides whether to impose a penalty on a person for whom it has regulatory responsibility. Although the function of the RDC carries with it an obligation to act fairly and to give fair consideration to any representations made to it, the RDC remains an organ of the FSA and the giving of a Decision Notice is the final step in a disciplinary process conducted by the FSA. The statutory right to refer the matter to the Upper Tribunal enables the person subject to the disciplinary procedures to remove the matter from the sphere of the FSA for a fresh decision by an expert tribunal exercising a judicial function. That is the context in which the question falls to be decided. Although separate from the FSA both in terms of its constitution and function, the tribunal is nonetheless an integral part of the regulatory scheme established under the Act.” (my emphasis)

At [37] he also said:

“The purpose of establishing the FSA to regulate the financial services industry and associated markets was to place responsibility for ensuring the maintenance of high standards in the hands of an expert body. It is not surprising, therefore, that the statutory scheme included provision for disputes relating to decisions taken by the FSA in the exercise of its regulatory functions to be referred to an expert tribunal. The opportunity to refer an investigation that has culminated in a disputed Decision Notice to the tribunal for a full re-hearing forms an integral part of the statutory scheme. It would be surprising, therefore, if Parliament had intended that disputes relating to the procedure adopted by the FSA should be reviewed by the courts, save in the most exceptional cases.…..The argument that the tribunal is incapable of giving Mr. Willford the remedy he needs in this case is, I think, overstated. It is true that the tribunal cannot quash the Decision Notice and remit the matter to the RDC for it to give better reasons, but it can reconsider the whole matter afresh and thus deal with the substance of the allegations against him.” (my emphasis)

In Hobbs (2013) the decision notice imposed a prohibition order on Mr Hobbs on the basis that he had engaged in market abuse and had lied to his employer and the FCA in the course of the investigation. Mr Hobbs referred the matter to the Tribunal which allowed his reference, deciding that his trading did not amount to market abuse. The Tribunal also found that Mr Hobbs had lied to the Tribunal about why he had undertaken the trades in question. The FCA successfully appealed against the Tribunal’s decision on the basis that, whilst it did not challenge the decision that Mr Hobbs had not been guilty of market abuse, the Tribunal ought to have considered its alternative case that the prohibition order was justified by Mr Hobbs’ lies both during the course of the investigation, and during the hearing before the Tribunal. It is the lies to the Tribunal which are of importance to the jurisdiction question: the lies in the course of the investigation were not new, having been included in the warning notice and decision notice, and the issue which applied to them was whether the decision notice and statement of case relied on those lies as a separate ground from market abuse as giving rise to a lack of honesty and integrity, which the court concluded at [33] to [37] that it had. As to the lies to the Tribunal, the court recognised that the issue was one of statutory construction of whether they constituted part of “the matter” [32]. It noted (at [32]) counsel’s agreement as to the broad meaning of the matter and that “‘The matter’ includes the facts and evidence referred to in the decision notice on the basis of which the authority concluded that the person in question was not fit and proper and that a prohibition order was appropriate”. This inclusive definition would not in fact have encompassed lies to the Tribunal, which were ex hypothesi not facts or evidence referred to in the decision notice. Nevertheless the court treated the lies to the Tribunal as something which it ought to have considered [39]. Its reasoning on this aspect is contained in [38] where Sir Stanley Burnton (with whom Rimer and Ryder LJJ agreed) said:

“Furthermore, in my judgment it is important for the tribunal to consider all the facts and evidence put before it on a reference under section 57. There are two reasons for this. The first is that its consideration of a reference is not ordinary civil litigation. There is a public interest in ensuring, so far as possible, that persons who are not fit and proper persons to perform functions in relation to a regulated activity are precluded from doing so. A narrowing of the inquiry by the tribunal that excludes relevant material from its assessment of an applicant is to be avoided, provided, of course, that the applicant is given a fair opportunity to address the authority’s case. In Mr Hobbs’ case, it could not be suggested, and was not suggested, that he did not have a fair opportunity to address the allegations that he had been guilty of repeated and persistent lying. The second reason is that if the tribunal incorrectly restricts its determination, it may be difficult for the authority to rely on the excluded facts in future in assessing, for example, whether the applicant is a fit and proper person, or should be granted an authorisation he seeks to engage in a regulated activity. To take the present case as an example, I can see that it might be arguable that on Henderson v Henderson grounds ((1843) 3 Hare 100) the authority should not be permitted to rely on allegations that it put before the tribunal but which the tribunal did not accept demonstrated that Mr Hobbs was not a fit and proper person. Such a situation should be avoided.”

Hobbs is therefore Court of Appeal authority that “the matter” is wide enough to cover something which was not itself relied on at any stage of the process prior to the reference, in that case because it had not yet happened. Any construction of the matter must be wide enough to accommodate this conclusion. Moreover, this passage recognises that it is undesirable that the jurisdiction should be narrowed to exclude material which is relevant to the action the Tribunal is being asked to order, or the considerations that the Tribunal should take into account in remaking the decision, provided always that fairness is not impaired. This militates in favour of a wide jurisdictional gateway, with case management powers regulating what is allowed in the circumstances of each particular case. Further, it may be unsatisfactory for the FCA to be forced to recommence a process in which it risks being met by arguments of time bar or Henderson v Henderson abuse. This is a further reason for not narrowing the threshold jurisdictional gateway.

In Khan (2014) the FCA had imposed a financial penalty of £80,000 on Mr Khan for breach of Principle 1 (failing to act with integrity) in relation to personal mortgage applications. Relevant for present purposes is that there was an additional allegation that on two mortgage applications for customers Mr Khan had certified that passport photos were true likenesses of the customers, but had not verified that fact by a face to face comparison. That was an allegation in the decision notice which supported a finding by the RDC that in that respect Mr Khan had been in breach of Principle 6 (failure to act with due skill, care and diligence). On the reference to the Tribunal Mr Khan applied to disallow reliance by the FCA on the certification allegations as a breach of Principle 1 (lack of integrity), on the grounds that such a finding was not part of the Decision Notice. On a preliminary application, UT Judge Herrington rejected the application on the grounds that the dishonest certification allegations fell within the Jabre criterion of matters which were referred to in the decision notice and had been the subject matter of the warning notice. At the trial it appeared that dishonest certification had not in fact been part of the warning notice allegations and the RDC had not been asked at the decision notice stage to make any findings of dishonesty on that issue. The Upper Tribunal, chaired by UT Judge Herrington, expressed the view that such allegations were outside its jurisdiction, although the allegations of dishonesty were rejected on the evidence. The UT Decision at [71]-[72] treats Khan (by reference to what the Tribunal also chaired by Judge Herrington said about it in Seiler) as establishing that where the FCA seeks to make an allegation in its statement of case which was not pursued before the RDC, but could have been, it is necessary that the allegation was raised during the RDC process (sic); and that unless matters occurred after a warning notice, they could not be relied on without having been the subject of a full investigation and the warning notice procedure. I have not found this entirely easy to follow. The distinction between matters which arose before and those which arose after the completion of the disciplinary process was no doubt designed to accommodate the decisions in Allen and Hobbs; but in my view it is unprincipled and illogical to treat “the matter” as capable of being fulfilled by reference to circumstances which occur or are discovered after the matter has been referred, but not wide enough to cover the same matters if they arose or were known before the reference. That would involve widening the scope of the word “matter” for things happening after the decision from its narrower scope in relation to the same things happening before the decision. It appears to be driven by an a priori assumption that the warning notice procedure is sacrosanct and must be complied with where possible. However, that is a mistaken assumption for reasons I address below.

ITV Plc v Pensions Regulator [2015] EWCA Civ 228 [2015] 4 All ER 919 concerned the Upper Tribunal’s jurisdiction on a reference from a determination by the Pensions Regulator. The regulator issued notices to the appellant companies warning of its intention to take regulatory action to protect members of the companies’ occupational pension fund, on the ground that a joint venture transaction, whereby businesses had been transferred to the companies and the fund established, had not been an arm’s length sale to, or investment in, an unconnected third party. A determinations panel of the regulator, pursuant to the standard procedure under section 96 of the Pensions Act 2004, considered detailed representations by the regulator and the companies and concluded that it would be appropriate for the regulator to issue the companies with financial support directions, based on outcomes and not on any culpable conduct. The companies referred the decision of the panel to the Upper Tribunal. The regulator submitted a statement of case, adopting the determinations panel’s reasons for the cause of the scheme’s deficit; but, following the companies’ response, the regulator in reply laid the ground for a case of misconduct or lack of due diligence by asserting that there was in substance no independent advice on the terms of the transaction. The companies applied to strike out such allegations, contending that the regulator should not be permitted to advance a new case. Dismissing the application, the Upper Tribunal held that the test for allowing the regulator to introduce allegations not made in the warning notice was whether the facts and circumstances were within the scope of the determination, and that the regulator was entitled to take a different view in the Upper Tribunal from that taken by the determinations panel, given that the core allegations before the panel were not affected and it was undesirable to exclude allegations in the context of a request for a financial support direction. On the appeal to the Court of Appeal the companies contended that the regulator had to show good reason for departing from its case in the warning notice or in the determination of the determinations panel. That jurisdictional argument was rejected: the absence of any provision in the Pensions Act 2004 constraining the conclusions that the determinations panel or the Upper Tribunal could reach in the absence of a relevant ground in the warning notice indicated that the question of whether they could do so was left to their discretion. At [60] Arden LJ, with whom Floyd and Christopher Clarke LJJ agreed, said:

“But it is significant that the 2004 Act does not go on to say that either the determinations panel or the Upper Tribunal are constrained in the conclusions they can reach by the absence of a relevant ground in the warning notice. In my judgment, the absence of a provision to that effect firmly indicates that Parliament left the question whether the determinations panel or the Upper Tribunal could do so to their discretion.”

The appeal was, however, allowed on the basis that the Tribunal’s exercise of its discretion was flawed and should be retaken on a remission.

There are similarities between the relevant procedure in the FSMA and pensions context with both involving warning notices and references of the decision/determination to the Tribunal, which by s. 103(4) of the Pensions Act 2004 is to determine the appropriate action for the Regulator to take “in relation to the matter referred to it.” The jurisdictional argument in that case was not that there was a complete jurisdictional bar to the regulator relying on matters which were not in the warning notice or determination, but rather that it could not do so without justification by “good reason”. Contrary to Mr Herberg’s submissions to us, that was a jurisdictional objection, not merely one of case management discretion. The rejection of that argument at [60] and [63], on the grounds that nothing in the 2004 Act imposed such a constraint, took as its unspoken assumption that things which were not in the warning notice or determination could fall within “the matter” referred. That was an unspoken assumption because Mr Pannick KC on behalf of the companies had not argued to the contrary. It is a decision which is premised on the proposition, without adversarial argument, that in the pensions context matters which are not in a warning notice or determination notice are capable of being a “matter” referred. Arden LJ’s reasoning at [63] also treated the fact that on a reference the Tribunal can permit further evidence to be filed and fresh arguments to be advanced as supporting the conclusion that in an appropriate case it must be open to the regulator to adduce additional grounds for its proposed regulatory action on a reference to the Upper Tribunal.

However the decision is of limited relevance to FSMA references because the pensions context is different and Arden LJ’s reasons at [61]-[62] relied on protections of the target which are specific to the pensions context. Moreover at [72] Arden LJ suggested that no assistance was to be derived from the FSMA cases cited to the court. Its significance, in my view, is that the Court saw nothing surprising or objectionable in a regime under which something which had not been raised in the warning notice or determination notice would not for that reason alone fall outside the jurisdiction of the Upper Tribunal as a “matter referred”, but whether the regulator should be allowed to raise it in a particular case could adequately be dealt with by the Tribunal’s exercise of its discretion. It therefore supports a broad jurisdictional construction of “the matter” referred, albeit in the pensions context which has both similarities and differences.

In Markou v Financial Conduct Authority [2023] UKUT 00101 (TCC) the Upper Tribunal refused to allow the FCA to pursue an allegation in its statement of case which involved an allegation of an entirely different nature from that put to and considered by the RDC and contained in the decision notice ([163]). The Tribunal heard no argument on the jurisdictional authorities, but considered for itself Jabre, Allen, Hobbs and Khan and concluded (at [144]) that in order for the FCA’s allegations to form part of the subject-matter of the reference and be considered or determined by the Tribunal, they must be of the same nature and based upon the same factual background as the allegations made to the RDC and contained in the warning and decision Notices, even if no findings are made upon them. At [434] it said:

“While [counsel for the FCA] points out that it is in the public interest for the Tribunal to make relevant findings on all matters under consideration, this should not usurp the Authority’s function to decide, with clarity and certainty, the regulatory case that it wishes to pursue. The starting point should be that if the Authority wishes to pursue an alternative or lesser case it should plead this from the outset of enforcement proceedings before the RDC and then the Tribunal itself. Pleadings on a reference to the Tribunal are in no way akin to an indictment in criminal proceedings or particulars of claim in civil proceedings. A reference is a continuation of a regulatory process that has begun by way of a Warning Notice and enforcement proceedings before the RDC. In those proceedings the Applicant is entitled to know the full nature of the allegations, findings and decisions made against him by the Authority in order to consider whether to contest the regulatory action proposed or whether to make a reference to the Tribunal.”

Seiler (2023) is another decision of the Upper Tribunal chaired by UT Judge Herrington. The decision was that on the evidence the FCA had failed to make out its case that the applicants had acted recklessly and consequently with a lack of integrity. Its relevance for present purposes is its treatment, in an Appendix, of an issue as to whether it had jurisdiction to consider the FCA’s reliance on what was termed the Third FX Transaction. The decision (unsurprisingly) treated Jabre, Allen, Hobbs and Khan as establishing the propositions attributed to them in the UT Decision. Its conclusion was that the authorities demonstrated that in relation to references concerning prohibition orders the Tribunal has jurisdiction to consider allegations based on facts and matters which were considered before the RDC, whether or not those facts and matters were referred to in the relevant warning notice ([985]); but that as a matter of discretion, rather than jurisdiction, the starting point was that the Tribunal should not consider facts and circumstances not relied on by the FCA in its warning notice unless, in its discretion it decides it would be appropriate to do so, and that there would need to be an application by the FCA to do so even if the matters were referred to in the decision notice ([998]-[999]). The Tribunal considered that because of the mandatory requirement in s. 387 FSMA to give reasons in a warning notice, the FCA would bear a considerable burden in satisfying the Tribunal that it was appropriate to rely on a matter which had not been properly formulated and clearly stated in the warning notice ([1000]). The integrity of the warning procedure notice is important and its objectives would be compromised if the FCA does not use its best endeavours to ensure that all relevant matters are placed on the table at the warning notice stage; the FCA should not be tempted into thinking that if there are deficiencies in its case at the warning notice stage then these can be remedied later in the proceedings ([1001]-[1007]).

Finally there is Burdett & Goodchild v Financial Conduct Authority [2024] UKUT 00156 (TCC), a decision which postdates the UT Decision in this case. The relevant issue was whether the FCA should be allowed to amend its statement of case to add an additional alternative allegation of negligence to the allegation of lack of integrity which was the subject of a decision notice to impose a prohibition and a penalty, based on the same conduct. The Deputy UT Judge reviewed the authorities and recognised a tension between the approach to what is a “matter” in cases under the Pensions Act and those under FSMA, and a tension between Jabre, and Khan and Seiler; and that it was not easy to choose between the approaches based on policy:

“73.

On the one hand, just as in the tax jurisdiction of the tribunals there is a “venerable principle” that the task of the tribunals is to determine the right amount of tax due, not to decide who has the better argument (see, for example, Shinelock Limited v HMRC, [2023] UKUT 00107 (TCC)), there is here a public interest in getting to the “right” regulatory outcome, in particular ensuring, so far as possible, that persons who are not fit and proper persons to perform functions in relation to a regulated activity are precluded from doing so. On the other hand, Parliament has established a process for resolving disputes between a subject of enforcement action and the Authority, which is based on [thorough], fair and effective administrative decision-making procedures, which are less formal, less expensive and quicker than tribunal or court proceedings. These involve a process of investigation, followed by warning and decision notices. Those objectives will be compromised if the Authority does not ensure that all relevant matters are placed on the table at the warning notice stage. That is why it has been observed, on several occasions (see, for example, Seiler at [1010]), that it is generally to be expected that the Authority will have completed its investigation before the commencement of regulatory proceedings and will carry forward the same case both through the regulatory proceedings and into the Tribunal.”

The Deputy UT Judge concluded that he should follow the most recent, carefully reasoned, judgment of the UT Decision in this case, and on that basis held that there was no jurisdiction to make the amendment.

I shall return to consider the policy considerations raised in these authorities. First I will consider BCMUK’s argument based on the Barras principle.

The relevant aspect of the principle was affirmed in the recent Supreme Court decision Centrica Overseas Holdings Ltd v Commissioners for His Majesty’s Revenue and Customs [2024] UKSC 25 where Lady Simler JSC, with whom the other Justices agreed, said at [52]:

“The principle established in Barras v Aberdeen Steam Trawling and Fishing Co Ltd [1933] AC 402 and re-affirmed in R (N) v Lewisham London Borough Council [2014] UKSC 62, [2015] AC 1259 applies, and as Lord Hodge JSC explained at para 53:

‘… where Parliament re-enacts a statutory provision which has been the subject of authoritative judicial interpretation, the court will readily infer that Parliament intended the re-enacted provision to bear the meaning that case law had already established …’”

The principle is limited to “authoritative judicial interpretation over a period” and is no more than a presumption which can apply where it is reasonable to assume that the intention of Parliament was to give the term the meaning established by such authoritative guidance: B v Secretary of State for Work & Pensions [2005] EWCA Civ 929 [2005] 1 WLR 3796 at [35].

The re-enactments relied on by BCMUK are the recasting of s. 133(5)-(6A) in FSA 2012; amendments to add s. 133(5A) by Schedule 11 para 4 of FSA 2021, which applies the disciplinary notice powers to a cancellation or variation of permission for failure to undertake permitted regulatory activity; and the FSA 2023 both in extending the scope of s. 133 (by amendment of s. 133(1)(c)) to decisions made pursuant to FSA 2023; and adding s. 55Z3(2A) which extends the right to refer the matter to the Tribunal upon the exercise of the new power conferred by s. 55NA(5)(b) for the FCA of its own initiative to vary or cancel permission to approve certain communications.

It is wholly unrealistic to think that Parliament addressed its mind to the meaning of “matter” when making the minor extensions in the 2021 and 2023 Acts; and no reason why it should have done so in distinguishing powers in disciplinary references from those in other references in the 2012 Act. In any event FSA 2012 received Royal Assent in 2012, at which stage of the authorities to which we were referred only Parker, Jabre and Nazia had been decided. These were Tribunal decisions which did not speak with one voice and there were no higher court decisions providing any guidance. This falls a very long way short of the authoritative judicial interpretation over a period which is required to bring the Barras principle into play.

I start with a number of considerations, which to some extent overlap, which suggest that the scope of the Upper Tribunal’s jurisdiction should be widely construed.

First, the word “matter” has a broad meaning. It is a noun which, as Turner J observed in R (Khan) v Secretary of State for the Home Department [2013] EWHC 601 (Admin) [2013] 3 All ER 499 at [9], is apt to include as broad a section of relevant material as possible.

Secondly, it is not the decision or exercise of the power which is referred, but “the matter”. This must mean the matter to which the decision relates rather than the decision itself. This is reinforced by the breadth of s. 133(4) which permits the Upper Tribunal to consider any evidence relating to the subject-matter of the reference. This suggests that the jurisdiction is wider than what is identified or relied on in the decision or supervisory notice itself.

Thirdly, a reference is not an appeal from a regulatory process which concludes with the FCA decision, but rather a continuation of that process by an independent specialist judicial tribunal. This is reflected in the language of the matter being referred and a reference rather than an appeal. The significance of the Tribunal reference being part of the regulatory process was emphasised by this court in Willford at [21] and [37]; and in Hobbs at [38]. The reference before the Tribunal transcends the private interests of the parties in ordinary civil litigation because it is concerned with the regulatory objectives which affect consumers and the market. The policy that the Tribunal should potentially be able to take into account all relevant material, subject of course to issues of fairness which can be addressed by the exercise of discretion on a case by case basis, is a powerful one.

BCMUK sought to distinguish the relevance of what was said in Hobbs on the grounds that it was a s. 57 decision, namely a reference in relation to prohibition orders against those who are not fit and proper persons under s. 56; and that this was a forward looking power, by contrast with the redress and penalty in the present case. However it is mistaken to treat the s. 55L power as retrospective merely because it requires payment by reference to past conduct. It is just as prospective as the exercise of the prohibition power under consideration in Hobbs, because it is concerned with a requirement attached to continued permission and therefore continued fitness to practice, for the purposes of protecting future consumers. Moreover, as Mr George told us in argument and Mr Herberg accepted, and as is evident from the authorities cited to us, it is very common for the same conduct to be the subject matter of both a prohibition/cancellation of permission and enforcement/penalty, and this would have been readily foreseeable when FSMA was enacted. It would be highly disruptive if the subject matter scope of the jurisdiction varied between the two actions, and Parliament cannot be taken to have so intended. I can see no good policy reason for distinguishing between forward looking notices and backward looking notices, both of which may be justified by the consumer protection objective, and there is nothing in the language of s. 133 to justify the distinction; on the contrary the only distinction drawn is between disciplinary notices and other notices. For these reasons I would reject BCMUK’s argument that “the matter” can mean something different depending on whether notices are forward or backward looking.

Fourthly, and consistently with the Upper Tribunal’s role as an integral part of the regulatory process, the nature of the decision-making jurisdiction which s. 133 confers on the Upper Tribunal points towards a wide construction of the matter. The powers of the Upper Tribunal are not heavily constrained by the decision reached by the FCA. In the case of disciplinary references, they are not so constrained at all: the matter is essentially at large before the Upper Tribunal, which reaches its own conclusions. That was the position for all references when FSMA was enacted and until FSA 2012 introduced a difference in powers for non-disciplinary references. This appears to have been a conscious decision to confer a wide jurisdiction. The original draft bill, by clause 68(3), proposed an appellate body to hear appeals. The FSMT’s role was changed from that of an appellate body at the instance of the Treasury so as to be “a tribunal of first instance, fully able to consider the merits and facts of each case, and to substitute its own conclusions for those of the FSA” (as recorded in the Joint Committee on Financial Markets First Report (27 April 1999) paragraph 182). But however that may be, the very breadth of s. 133(5) means that, as Moore-Bick LJ put it in Willford at [37], the matter referred will be “reconsidered completely afresh”. For non-disciplinary references, since 2012 there is some constraint on the Tribunal’s decision making power by reference to what is in the notice, but the ability of the Tribunal to reach its own independent conclusions under its ‘JR plus’ jurisdiction is a considerable one.

Fifthly, s. 133(4) recognises that the Tribunal may take into account evidence relating to the subject-matter of the reference. The Tribunal jurisprudence, and the argument before us, assumed that the drafter intended “subject-matter” to mean the same as “the matter” referred, and I will assume that that is so, although the use of different language does not make it self-evident. Mr Herberg’s argument treated the subsection as if it were confined to fresh evidence which arises or becomes known after the decision notice, and it is sometimes so treated in the Tribunal jurisprudence, but this is mistaken: its terms make clear that it extends to evidence which existed and was known to the FCA during the process. The important point is that as a matter of jurisdiction, and subject to the procedural safeguards in its case management powers, the Tribunal is allowed to take into account evidence which was in existence during the process before the FCA which was not relied on in either the warning notice or decision notice and which the firm has the opportunity to address only for the first time in the part of the process which takes place before the Tribunal. This somewhat undermines the policy arguments repeatedly invoked in the Tribunal jurisprudence about the importance of the warning notice and the opportunity to address it, but the point goes further.

Mr Herberg emphasised that s. 133(4) only applies to evidence. It is true that there is a conceptual difference between evidence and allegations, but it is often not an easy distinction to draw in practice because it depends at what level of granularity one defines the “allegation”. In Allen it was defined at the highest level of generality as being that Mr Allen lacked the honesty and integrity to be a fit and proper person, so that allegations of conduct completely unrelated to the conduct relied on throughout the process before the FCA were permitted to be relied on before the Tribunal. If that were the correct approach, the allegation in this case could be framed at a general level as being that BCMUK should be subject to a redress requirement or a penalty for failing properly to manage conflicts of interest in relation to the two funds and provide sufficient information about them. All the rest would then be evidence. On the other hand, if an “allegation” is defined as a particular instance of conduct or state of mind relied on, or fact from which such state of mind is to be inferred, it would often be possible to define the conduct or fact with such granularity that almost anything new would fall into the category of allegation rather than evidence. It will often be difficult to see where the line is to be drawn between separate conduct and separate evidence of conduct, which will depend upon a largely arbitrary decision on the degree of granularity to be applied.

The granularity required by FSMA is at a high and general level. The requirement in relation to warning and decision notices is to give “reasons”. This does not prescribe any particular level of detail. By contrast, when it gets to the Tribunal stage, paragraph 4 of Schedule 3 of the Upper Tribunal Rules requires the respondent in its statement of case not only to state the reasons for the referred action, but also the “matters and facts upon which the respondent relies to support the referred action”. The undesirable consequence of BCMUK’s narrow jurisdictional argument is that it would provide an incentive for the FCA to express its reasons in the warning notice and decision notice at the highest level of generality and without detail, which, whilst permitted by FSMA, is clearly undesirable.

Sixthly, the Upper Tribunal has wide powers of case management to control the scope of any reference so as to ensure that the overriding objective is achieved in ensuring justice and fairness. This provides a significant safeguard which supports the threshold jurisdictional gateway not being narrowed any more than is necessary, and any such narrowing would need to be justified by considerations which could not be accommodated by the exercise of case management powers.

Seventhly, Hobbs is Court of Appeal authority, binding on us, that “the matter” is wide enough to cover allegations and reasons which were not relied on at any stage of the process prior to the reference (in that case because they had not yet happened). Any construction of “the “matter” must be wide enough to accommodate this conclusion. It cannot be justified, as the Tribunal in Khan and Seiler sought to justify it, by the suggestion that a special rule applies to circumstances which arise after the decision notice.

Last, there is perhaps a small further pointer in the fact that “the matter” is not something which falls to be identified for the first time after a decision notice and only for the purposes of a reference. Section 391 FSMA deals with publication of notices. Subsection (1)(c) provides that in the case of certain warning notices (including those given in relation to penalties under s. 207), “after consulting the persons to whom the notice is given or copied, the regulator giving the notice may publish such information about the matter to which the notice relates as it considers appropriate”. At this stage the matter cannot be confined to what is in the warning notice itself because the subsection addresses the wider concept of the matter to which the warning notice relates. This is a less weighty factor because it is entirely possible for “the matter” to have a quite separate meaning here from its use in relation to Tribunal references. But the use of the same term in a statute should if possible be given the same meaning; and it can be given a consistent meaning if in both places it connotes a connection with the subject matter of process, at each stage. The content of what constitutes the matter may change with the course of the process, so as to cover something different at each stage, but the meaning of the expression remains the same. I will return to this concept of connection with the subject matter of the process below.

By contrast, I find the practical and policy reasons advanced for narrowing the jurisdictional gateway to be distinctly less convincing or weighty.

Foremost amongst them as relied on by BCMUK, and grounded in statements repeatedly made in the Tribunal jurisprudence, is the importance of the mandatory warning notice and its requirement to provide reasons and disclosure as an important part of the process by which the firm knows what case it has to meet and what representations to make about it. The same is said about the first supervisory notice where that applies, which may be the equivalent of a warning notice if representations are made resulting in a second supervisory notice, or the equivalent of the decision notice if the firm chooses to go directly to the Tribunal. For simplicity I will refer simply to the warning notice/decision notice procedure.

With due respect to experienced members of constitutions of the Upper Tribunal who have invoked this consideration, it is in my view overstated. I have already referred to the fact that the point is to some extent undermined by s. 133(4) FSMA, which allows the FCA, as a matter of jurisdiction, to rely in the Tribunal on evidence which it had at the time of the warning notice but did not then rely upon. But the difficulty with the point is more fundamental. The warning notice is only the first step in the process. The very purpose of the representations which the firm is to have the opportunity to make is to enable the FCA to change its position at the next stage of the process. It would be contrary to this purpose if it could not adjust its decision, either by reference to the facts relied on or the action to be taken, as a result. It cannot sensibly have been intended that the warning notice was set in stone, so that at the decision notice stage the only two options would be to accept the matter as it was framed in the warning notice or dismiss it. On the contrary the warning notice and decision notice are part of a continuous evolving process of adversarial engagement. Section 388(2) expressly recognises that the FCA may in the decision notice impose a different action from that in the warning notice, subject only to the limitation that it must be of a kind which arises under the same Part of the Act. There is no prohibition on the FCA imposing a harsher sanction or requirement. Indeed Mr Herberg expressly disavowed any contention that the FCA was not entitled to change the allegations in the warning notice when it came to the decision notice, and accepted that allegations, evidence, reasons and proposed action could all change between the warning notice and the decision notice, provided the firm was given a fair opportunity to address the changes.

The same is true of a first supervisory notice which contemplates that a further supervisory notice may be issued in different terms to reflect representations made by the firm in relation to the first supervisory notice, subject only to the constraint that the requirement or action is one within the supervisory notice regime rather than the decision notice regime.

The point might at first sight appear to have greater cogency if focussed on the decision notice rather than the warning notice, but in my view it does not. This too is undermined by s. 133(4) which allows new evidence to be adduced in the reference notwithstanding the absence of any reliance on it before the RDC or in the decision notice. Moreover, the evolving process between warning notice and decision notice is also characteristic of what is to happen after the decision notice in a reference, which is an integral part of the regulatory process. In a disciplinary reference the matter is reconsidered afresh by a specialist judicial tribunal. In other references the ‘JR plus’ jurisdiction allows the specialist tribunal to exercise a considerable measure of independent judgement, including, for example forming its own view of the facts. The process therefore envisages moving on from the warning notice or first supervisory notice stage through the decision notice or second supervisory notice stage (if the firm wants it) to the Tribunal stage with each stage involving the decision maker exercising a judgement afresh in the light of the arguments and evidence put before it. The Tribunal will hear witnesses who are cross-examined, and adversarial argument, in a forensic examination which makes it very likely that new things will emerge. That is a common experience of civil and criminal litigation. The Tribunal’s jurisdiction must have been intended to take account of this aspect of the process. There is no restraint on a firm’s ability to rely in the reference on new facts and contentions which it did not raise prior to the reference. It would be unfair to the FCA if it could not refine its case in the light of such new material.

The imperative that the FCA should be encouraged to get its house in order when first proposing action against a firm, and should not be tempted to think that it can simply remedy failings at a later stage, is a real one. It may well play a part in the exercise of discretion when case management powers are to be exercised. It is not, however, to my mind, a powerful reason for narrowing the jurisdictional gateway. The jurisdictional gateway, if narrowly construed, takes no account of the culpability or otherwise of any failings or, significantly, the consequences if it is treated as an absolute bar preventing something new being raised in the Tribunal. As Sir Stanley Burnton emphasised in Hobbs at [38], the Tribunal is concerned with pursuit of the statutory objectives, which may be frustrated by a narrow jurisdictional gateway. If allowing a new consideration to be raised in the Tribunal would cause no procedural unfairness or prejudice to the firm, but would preclude it ever being relied on by the FCA because starting again at the warning notice stage of the process would be met by a Henderson v Henderson objection, that would run counter to the purpose of the Tribunal in pursuing the regulatory objectives of protecting consumers etc. All this points towards the desirability of a broad jurisdictional gateway with constraints imposed by reference to what is fair and just on a case by case basis by the exercise of the wide case management powers which enable the Tribunal to determine what case the FCA can or cannot advance before it by its control over the content of the statement of case. The recent Burdett case provides a potent illustration of the difficulties caused by a narrow approach.

A further objection to a wide jurisdictional gateway is that even if the Tribunal can prevent the FCA raising something new by the exercise of case management powers, the firm will then have been put to the time and cost of having to participate in the Tribunal proceedings to secure their dismissal. This is of limited weight because it is balanced, if the gateway is narrow, by the waste of time and cost in a case which would require the FCA to start again with the warning and decision notice procedure when that would inevitably lead to an additional conjoined reference, with the first reference stayed to allow it to catch up. The latter is undesirable not only from the point of view of the parties but also contrary to the public interest in the efficient administration of justice and promotion of the statutory objectives: delay in regulatory action being implemented may itself harm these objectives by having an adverse effect on consumers.

The point was put by Mr Herberg as one which did not merely involve potentially wasted time and cost but would have a chilling and deterrent effect on firms in deciding whether to initiate a reference, for fear that there could be a new case raised by the FCA in the Tribunal. There are three answers which diminish the force of this suggestion. First, a wide jurisdictional gateway does not necessarily permit a new case to be advanced; it will be permitted under the case management powers only if it is not unfair or unjust to allow it. Secondly, when a firm is deciding whether to initiate a reference to the Tribunal, it is an inevitable part of that decision that doing so risks a different and harsher action being imposed by the Tribunal. Any deterrent effect of something new being raised, which can only be something which does not unfairly prejudice the firm in being able to deal with it, is of a lesser order. Thirdly, this argument was rejected by Moore-Bick LJ in Willford at [37] where he said that making a reference involved the risk of new allegations being raised (as Allen and Hobbs illustrate).

Mr Herberg also relied on prejudice in such a case from the loss of privacy in the Tribunal which a decision to refer involves, by contrast with the process before the FCA which is in private. However the firm has no absolute right to privacy at the earlier FCA stages of the process in relation to decision notices, or in many cases warning notices: s. 391 FSMA.

Mr Herberg made a further point that third parties who may be affected by a decision have protective rights under FSMA to take part at the RDC stage, and after a decision notice may themselves initiate a reference. His submission is that they would be prejudiced if the FCA were not limited to reasons identified during the FCA stage of the process. However there is no reason why such affected parties should not be able to participate in Tribunal proceedings if an issue affecting them is raised for the first time before the Tribunal, in an appropriate case.

Taking into account all these considerations which dictate how widely or narrowly the jurisdiction gateway should be set, they clearly and decisively favour a wide construction of “the matter”. That is required by the wide language used, the nature of the regulatory process of which a Tribunal reference forms part, the other policy considerations I have identified, and the decision of this court in Hobbs.

What then is the test? I have not found any of the five different formulations proposed by the parties satisfactory. BCMUK’s primary case is clearly too restrictive, and so too, in my view, is its secondary case which adopts the UT Decision’s approach. I have explained why I do not think that a test which distinguishes between prospective and retrospective decisions can be justified.

Conversely, the FCA’s primary case is too wide. It would, as BCMUK’s argument pointed out, have startling consequences. It would recognise the Tribunal’s jurisdiction to allow a firm which had referred a decision notice or supervisory notice to be faced with an entirely new case based on different allegations of different contraventions of different rules arising from different facts relating to a different time period, different customers and different products in order to impose a different form of sanction or other requirement, subject only to the limits in s. 388(2), 133A and 133(6) of it being allowed under the same Part of FSMA (for a decision notice case) and by a supervisory notice procedure (in a supervisory notice case). Mr George said that such a case would be excluded by the exercise of case management discretion, but in my view it is contrary to the language of FSMA as a matter of jurisdiction. The sections of FSMA which confer the right for a firm to initiate a reference to the Tribunal provide that when the decision is made, or the power is exercised, “the matter” may be referred. The “matter” must therefore bear some relationship to the decision (which I use as a shorthand also for the exercise of the power). The FCA’s primary case would strip the words “the matter” of any meaning at all. This is reinforced by s. 133A(4) which provides that the action specified in the decision notice must not be taken “during the period within which the matter to which the [decision] relates may be referred” (my emphasis). This means that what may be referred is a matter which must be related to a decision notice. The issue is what amounts to a sufficient relationship, to which I shall return.

The FCA’s tertiary case is not really a test at all, but an argument that however narrow the test, Amendments 3 & 4 fall within it.

The FCA’s secondary case comes closest to an appropriate test, namely that “matter” encompasses anything which arises from the same factual situation which gave rise to the regulatory action in the statutory notice referred to the Tribunal or is otherwise connected with the circumstances, the evidence and/or the allegations, whether factual or legal, which were before the FCA’s decision-maker, but is in my view still too narrow.

What is clear is that there must be some sufficient relationship between the matter referred and the decision which triggers the right to refer, and the critical question is: what is required by the concept of sufficiency in this context? The answer is to be found in the fact that the decision is a stage in the regulatory process, and the Tribunal reference a further stage in that process. The logical answer is therefore that something is sufficiently related to the decision which triggers the reference to amount to or be included in “the matter” if it has a real and significant connection with the subject matter of the process, in the sense of its procedural or substantive content, which has culminated in the decision notice or supervisory notice. Such connection must be real and significant, not fanciful or tenuous. But if so, that is sufficient. It need not be something upon which the FCA has specifically relied during the process, provided that it has a real and significant connection with the subject matter of the process. What is required when the FCA seeks to rely on something new in the Tribunal is an examination of what is new, and of the procedural or substantive content of the process culminating in the decision or supervisory notice, and the establishment of a real and significant connection between them. If what is new has this connection it is within the Tribunal’s jurisdiction. It is a separate question whether the FCA should be permitted to rely upon it in any particular case, which is a matter for the exercise of the Tribunal’s case management powers as to whether it would be just and fair.

If it be objected that this is not hard-edged, I would respond that it is undesirable to seek to define it more prescriptively because it must be flexible enough to take account of over 300 types of decisions, set out in DEPP Chapter 2 Annexes 1 and 2, which may give rise to a reference to the Upper Tribunal. However, it is consciously and deliberately a very wide gateway, for the reasons I have discussed. Accordingly, I would expect it to be a rare and obvious case which fell outside it so as not to come within the Tribunal’s jurisdiction.

It would not be appropriate on this appeal to seek to define the factors which the Upper Tribunal should take into account in exercising its discretion in individual cases as to whether the FCA may be permitted to rely on something new, or the weight to be attached to them. That is something to be determined by the Upper Tribunal. I should make clear, however, that I do not agree with suggestions in the Tribunal jurisprudence that the FCA always bears a heavy burden if it wishes to rely on something which was not in the warning notice. That places an undue emphasis on the importance of the warning notice in the process, for the reasons I have tried to explain.

Amendments 3 & 4 are essentially limited, as Mr George submitted, to relabelling the conduct which was already alleged as a breach of different regulatory rules, namely Principle 7 and COBS 4.2.1. That falls within the matter referred. The UT Decision treated what was contained in the allegation at para 174 of the draft amended S/C relating to the recklessness of individuals to whom the FCA seeks to attribute responsibility for the alleged misconduct as a new allegation. If it is correctly so categorised, it too falls within the scope of the matter referred, although I would in any event have treated it as falling on the evidence side of the line and therefore permissible under s. 133(4): the allegation of recklessness was already made in the DN, and this merely identified by way of further particularisation the individuals whose knowledge and recklessness supports the allegation.

It follows that the UT Decision involved an error of approach in relation to Amendments 3 & 4 in deciding it had no jurisdiction to allow them. We can remake the decision by ourselves considering whether they should be allowed, in the exercise of a discretion upon which the UT Decision did not embark, and both parties asked us to do so rather than remitting it to the Tribunal.

I have little hesitation in saying that they should be allowed. They arise out of the same facts and matters as were contained in the DN and the WN. They do not change the action which the FCA asks the Upper Tribunal to confirm by dismissing the references. They can readily and without any prejudice be addressed by BCMUK in the course of the reference. I am unable to accept BCMUK’s submission that they should not be allowed because they are a response to BCMUK’s argument on an Actionability Condition advanced in its Reply. Mr George frankly accepted that that was what gave rise to them, but asked us to determine whether they should be allowed even if we rejected the Actionability Condition (as I have). There is nothing inherently improper about the FCA raising something to meet an argument advanced during the course of the reference. That does not of itself render it unfair or unjust that the FCA should be permitted to rely on it.

BCMUK seeks to cross appeal against the UT’s decision to allow Amendment 2, both as a matter of jurisdiction and discretion. As to jurisdiction, the wide definition I have given to “matter” clearly encompasses Amendment 2.

As to the exercise of discretion, an appeal from a Tribunal decision only lies to this court on an error of law: s. 13(1) TCEA. I very much doubt whether the grounds advanced amount to alleged errors of law within the section (cf Financial Conduct Authority v Seiler & Whitestone [2024] EWCA Civ 852). But in any event there is no detectable error of principle in the way the discretion was exercised, and the conclusion was well within the range of decisions open to the Upper Tribunal in its evaluative assessment. Indeed I would have reached the same conclusion. The existing unamended pleading clearly averred that the investors in the External Fund were BCMUK’s customers and clients, as did the DN and WN. The amendment merely spells out by reference to the relevant FCA rules and the Regulated Activities Order why that is so. The amendment introduces no new factual allegations. It is made at an early stage of proceedings in which BCMUK has ample opportunity to address and respond to the amendment without suffering any prejudice.

For these reasons I would allow the appeal on Ground 2 and remake the UT Decision by allowing Amendments 3 & 4; and would dismiss the cross-appeal in relation to Amendment 2.

I agree, and am very grateful to Popplewell LJ for his comprehensive explanation and analysis of the issues.

If one stands back from the inevitable detail, the first question resolves itself into this (all references are to FSMA). By s. 55L(2)(c) and (3)(a) the FCA has power “to impose a new requirement” on an authorised person if it appears to the FCA that it is “desirable to exercise the power in order to advance one or more of the FCA’s operational objectives”. By s. 1B(3)(a) the FCA’s operational objectives include “the consumer protection objective”, and by s. 1C(1) the consumer protection objective is “securing an appropriate degree of protection for consumers”. By s. 55N(1)(a) a requirement imposed under s. 55L may “require the person concerned to take specified action”; and by s. 55N(5) a requirement “may refer to the past conduct of the person concerned (for example, by requiring the person concerned to review or take remedial action in respect of past conduct).” The issue is this: is there any reason to read limitations into this power of the FCA to impose a requirement that the person concerned take remedial action in respect of past conduct?

There is of course the limitation that the requirement can only be imposed if it appears to the FCA that it is desirable to exercise the power to advance one of its operational objectives, including the consumer protection objective; and like any other public law power, it is subject to the usual public law constraints such as that it must be exercised rationally and for the purposes for which it was conferred. But there is nothing in the statutory language imposing further limitations, and for the reasons explained by Popplewell LJ, there is no reason to imply them. I agree that on the first issue the FCA’s appeal should be allowed and the other orders made as proposed by Popplewell LJ in paragraph 111 above.

The second issue resolves itself into this: what is “the matter” that is referred to the Tribunal? At the risk of simply rephrasing the question, the answer, it seems to me, is that the matter is what the case is about. But for the reasons given by Popplewell LJ, that must be expressed at a high level of generality. In the present case for example one could without inaccuracy say that the case is about an alleged breach of Principle 8, but that would in my view be too narrow as an identification of “the matter”. At a higher level of generality, what the case is about is the alleged way in which BCMUK managed the Internal and External Funds, and managed (or failed to manage) the conflict of interest between them. That is quite wide enough to include the way in which it communicated (or failed to communicate) information about this to its clients and hence to embrace an alleged breach of Principle 7 and COBS 4.2. There is therefore no jurisdictional bar to the amendments.

I should make it clear that I do not intend in this short judgment to qualify or differ from the much fuller way in which Popplewell LJ has dealt with this question and I therefore agree with the way he has expressed it in paragraph 202 above.

I agree that the appeal should be allowed on Ground 2 and the other orders made as proposed by Popplewell LJ in paragraph 210 above.

I agree with both judgments.