PS5/23 – Risks from contingent leverage

1. Overview

1.1 This Prudential Regulation Authority (PRA) policy statement (PS) provides feedback to responses to consultation paper (CP) 12/22 – Risks from contingent leverage. It also contains the PRA’s final policy, as follows:

• updated supervisory statement (SS) 31/15 – ‘The Internal Capital Adequacy Assessment Process (ICAAP) and the Supervisory Review and Evaluation Process (SREP)’ (Appendix 1);
• amendments to the Reporting (CRR) Part of the PRA Rulebook (Appendix 2);
• introduction of reporting templates LV49-52 (Appendix 3);
• updated ‘Instructions for reporting on leverage’ (Appendix 4); and
• updated SS45/15 – ‘The UK leverage ratio framework’ (Appendix 5), to add LV49-52 to the list of leverage reporting templates.

1.2 This PS is relevant to banks, building societies, and PRA-designated investment firms, and their qualifying parent undertakings (referred to as ‘firms’) that perform the Internal Capital Adequacy Assessment Process (ICAAP)^{footnote [1]} as far as the updates to SS31/15 are concerned, and to firms subject to a leverage ratio minimum requirement^{footnote [2]} (LREQ firms) as regards to the reporting rules, template, instructions, and SS45/15.

Background

1.3 Certain types of financing have a lower exposure measure value than other economically similar transactions in the calculation of the leverage ratio (and are therefore more capital-efficient). The capital-efficient nature of these forms of financing depends; however, on a set of conditions persisting, as set out in CP12/22. Contingent leverage risk arises when those conditions no longer hold, for example in a stress, and so a firm can no longer rely on the associated capital efficiencies.

1.4 In CP12/22, the PRA proposed to:

• update the PRA’s supervisory expectations for firms undertaking an ICAAP in relation to the risks from contingent leverage (the ICAAP expectations); and
• introduce a new data reporting requirement (the reporting requirement) for LREQ firms, collecting data on trading exposures where these risks may most likely arise.

1.5 The PRA considered that those proposals would help the PRA and firms better understand the significance of these risks, and better manage them should they materialise.

Summary of responses

1.6 The PRA received one response to the CP. This response broadly welcomed the measures proposed, while seeking clarification on aspects of the proposed ICAAP expectations and reporting requirement, and suggesting targeted changes to the reporting requirement. The points raised are set out in Chapter 2.

Changes to draft policy

1.7 The PRA has added a clarification on the business lines and trade structures in scope of the ICAAP expectations, and how to assess the materiality of contingent leverage risks to a firm’s business.

1.8 The PRA has also made the following minor changes to the reporting requirement (via changes to the proposed reporting rule, templates, and instructions in Appendices 2-4):

• postponed the entry into force of the requirement to 1 January 2024, with a first reporting reference date of 30 June 2024, to give more time for firms to build and test their reports;
• removed a row from reporting template LV-52 asking that written credit derivatives backed by level 1 High Quality Liquid Assets (HQLA) be reported separately. The PRA agrees with the respondent that the quality of the underlying assets is not relevant to assessing contingent leverage risks in the specific case of credit derivatives; and
• made minor drafting changes to the reporting rule, templates, and instructions for clarity and consistency.

1.9 More detail on the changes to the ICAAP expectations and reporting requirement is provided in Chapter 2.

1.10 The PRA considers that the changes to the ICAAP expectations and the reporting requirement do not materially alter the cost benefit analysis (CBA) presented in CP12/22. The PRA does not consider that the final policy will have a significantly different impact on mutuals, or on other PRA-authorised firms, relative to the proposals in CP12/22.

1.11 When making rules, the PRA is required to comply with several legal obligations, including considering responses to consultation and publishing an explanation of the PRA’s reasons for believing that making the proposed rules is compatible with its objectives and with its duty to have regard to the regulatory principles.^{footnote [3]} In addition, when making CRR rules or rules applying to certain holding companies, the PRA must consider certain additional matters, and publish an explanation of the ways in which that consideration has affected the proposed rules.^{footnote [4]} In CP12/22, the PRA provided these explanations in Chapter 2.  

1.12 In carrying out its policy making functions, the PRA is required to have regard to several matters, as set out in CP12/22 in Appendix 5. In the CP, the PRA explained how it had had regard to the most relevant of these matters in relation to the proposed policy. Below, the PRA provides relevant updates to that explanation, taking into account the consultation response.

1.13 The PRA considers that the changes to the ICAAP expectations enhance transparency, by clearly identifying the trade structures relevant to contingent leverage risks and providing guidance on the circumstances in which these risks are expected to be material to a firm’s business.

1.14 The PRA considers the changes to the proposed reporting requirement are consistent with the PRA’s public law duty to act proportionately. They ensure that firms will have more time to prepare for implementation of the new requirement, and will not have to report a data point that is not relevant to the PRA’s or firms’ ability to monitor and manage contingent leverage risks.

1.15 The PRA has considered whether the consultation response received, or subsequent changes made, have affected the other explanations given in CP12/22. It has concluded that those explanations remain appropriate.

1.16 The PRA must also publish a summary of the purpose of the proposed rules.^{footnote [5]} The purpose of the rules is to introduce a new data reporting requirement for LREQ firms, collecting data on trading exposures that may be sources of contingent leverage risk.

Implementation

1.17 The ICAAP expectations for firms undertaking an ICAAP will take effect on publication of this PS. The reporting requirement for LREQ firms will take effect on 1 January 2024, with a first reporting reference date of 30 June 2024.

Interactions with the Banking Data Review

1.18 In the PRA’s future approach to policy discussion paper (DP4/22), the PRA announced it was launching a ‘Banking Data Review’. The review will look at what data the PRA collects currently and what data it needs now and in the future.

1.19 The PRA does not currently expect to revisit the reporting requirement introduced through this PS, or to redesign the associated templates, as part of the Banking Data Review.

2. Feedback to responses

2.1 Before making any proposed rules, the PRA is required by FSMA to have regard to any representations made to it, and to publish an account, in general terms, of those representations and its feedback to them.^{footnote [6]}

2.2 The PRA has considered the response received to the CP. This chapter sets out the PRA’s feedback to this response, and its final decisions.

2.3 The points in the response received have been grouped as follows:

• potential PRA actions in response to contingent leverage risks;
• scope and materiality of contingent leverage risks;
• other clarifications of the ICAAP expectations;
• frequency and commencement of the reporting requirement for LREQ firms;
• reporting on written credit derivatives;
• ‘daily averaged’ basis of reporting;
• potential overlap with other reporting; and
• other reporting clarifications.

Potential PRA actions in response to contingent leverage risks

2.4 In CP12/22, the PRA proposed to insert expectations on the risks of contingent leverage into SS31/15 to update the PRA’s expectations of firms with regards to identifying, monitoring, and mitigating contingent leverage risks in their ICAAPs.

2.5 The respondent asked for clarification of the scenario in which the PRA would undertake targeted action under the Pillar 2 framework in response to contingent leverage risks, questioning the appropriateness of such intervention. Further, the respondent sought clarity on the methodology for calibrating additional capital to cover against contingent leverage risks.

2.6 As set out in the CP, the objective of the ICAAP expectations is to help firms identify, monitor, and manage contingent leverage risks, and to improve the PRA’s ability to monitor the evolution of these risks. These ICAAP expectations will not automatically lead to additional capital requirements for PRA firms with respect to contingent leverage risks. The PRA has therefore not provided a methodology for calibrating any such additional capital requirement.

2.7 The PRA considers, however, that firms’ assessment of their risk of excessive leverage, including contingent leverage risks, will be subject to the same rules and supervisory scrutiny applying to any part of the ICAAP. In particular, under PRA rules, firms must at all times maintain overall financial resources, including own funds and liquidity resources, which are adequate both as to amount and quality, to ensure there is no significant risk that its liabilities cannot be met as they fall due.^{footnote [7]} Added to this, under SS31/15, the PRA expects a firm to take appropriate actions or steps at an early stage to prevent any future potential failure to meet its prudential regulatory requirements as part of the SREP where this is deemed necessary.^{footnote [8]}

Scope and materiality of contingent leverage risks

2.8 The respondent asked for clarification as to which business lines and trade structures should be considered in scope for the purposes of the PRA’s new ICAAP expectations on contingent leverage risk. The respondent suggested greater alignment with the templates for the new reporting requirement for LREQ firms, and commented that the focus of the ICAAP expectations should be on those trades that may be subject to contingent leverage risks in a stress.

2.9 Having considered the comment, the PRA has decided to clarify, in the final ICAAP expectations, which business lines and trade structures the PRA considers to be potential sources of contingent leverage risk. At the minimum, this includes the trade structures that the PRA will require LREQ firms to report data on, to align with the reporting requirement. Were a firm to identify any further trades that might lead to contingent leverage risks besides those set out by the PRA, the firm should also consider how those trades might affect its leverage ratio as part of its ICAAP. This is consistent with the PRA’s objective for the ICAAP expectations to help the PRA and firms better understand the significance of these risks, and better manage them should they materialise.

2.10 The respondent argued that trades which may benefit from netting under the leverage ratio framework, such as repurchase agreements (repos), do not create material contingent leverage concerns due to the short-term nature of those trades and the high quality of collateral used.

2.11 The PRA continues to consider that netted repo trading is a relevant source of contingent leverage risk. Loss of regulatory netting benefits in stressed market conditions, if of material size, may lead to breaches of the leverage ratio requirements or supervisory expectation even where the trades are short-term or backed by high quality collateral.^{footnote [9]}

2.12 The respondent indicated a preference for firms to be able to set their own materiality standards in assessing contingent leverage risks, and to consider these risks at an aggregate rather than individual trade structure level.

2.13 The PRA agrees with the respondent that, consistent with the PRA’s wider approach to ICAAPs, each firm should assess the materiality of contingent leverage risks to its business – while noting that, where a firm has identified a risk as not being material, it should also be able to provide evidence of the assessment process that determined this and discuss why that conclusion has been reached.^{footnote [10]}

2.14 The PRA has also decided to elaborate in its ICAAP expectations on the circumstances in which contingent leverage risk is expected to be material to a firm’s business. Factors should include the extent to which the firm engages in the relevant trade structures (especially where the trades are subject to contractual obligations, franchise risk, or liquidity management considerations), and the size of the impact on the firm’s leverage ratio should the firm lose the capital optimisation benefits from the trade structure.

2.15 The PRA has also decided to maintain the expectation for firms to set out their assessment of contingent leverage risks by each individual trade structure, while adding clarity that this is only expected for trade structures judged by firms to be a material source of contingent leverage risk. Assessment at the aggregate level only would run counter to the PRA’s objective for the ICAAP expectations, as it might prevent firms or the PRA from identifying, monitoring, and managing vulnerabilities in specific trade structures.

Other clarification of the ICAAP expectations

2.16 The respondent requested clarification on the definition of franchise client, and if intra-group exposures are in scope of the ICAAP expectations.

2.17 Franchise clients are those that may give rise to franchise risk as defined in the final ICAAP expectations.^{footnote [11]} The PRA also confirms that intra-group exposures to other entities within the wider group are intended to be in scope of the ICAAP expectations, to the extent that these entities do not fall within the basis (whether solo, sub-consolidated, or consolidated) on which the firm is calculating its leverage ratio. To aid consistency, the PRA has also decided to align the definitions of franchise client and ‘intra-group activity’ in the final reporting instructions (Appendix 4) more closely to those in the ICAAP expectations.

2.18 There is a typographical error in the first sentence of paragraph 2.44 of the draft ICAAP expectations published alongside CP12/22, which referred to ‘lower’ rather than ‘higher’ ‘leverage ratio exposure measure values’. This sentence has been corrected in the final ICAAP expectations, to state: ‘firms may have to continue exposure to transactions or trade structures which receive higher leverage ratio exposure measure values, for a variety of reasons, including contractual obligations, franchise considerations, liquidity management, or other commercial reasons’.

2.19 The PRA has also, of its own initiative, provided further definitions within the ICAAP expectations, to aid clarity.

Frequency and commencement of the reporting requirement for LREQ firms

2.20 The respondent agreed with the PRA’s proposal that the reporting frequency should be six-monthly. The respondent suggested, however, the reporting requirement should take effect from 1 January 2024, instead of 1 July 2023 as proposed in CP12/22.

2.21 The PRA considers it proportionate to allow firms more time to prepare for implementation of the new reporting requirement, and has therefore decided that the requirement will enter into force on 1 January 2024, with a consequent 30 June 2024 first reporting date. The reporting requirement will follow standard remittance dates for six-monthly reporting (11 August and 11 February).^{footnote [12]}

2.22 As noted in CP12/22 and recognised by the respondent, in recent years, the PRA had conducted a number of ad-hoc data collections to monitor contingent leverage risks – the latest of which covered the second half of 2021. To cover the gap between this latest ad hoc data collection and implementation of the new reporting requirement, and to enable the PRA to continue monitoring contingent leverage risks in the intervening period, the PRA will run further ad hoc data collections on contingent leverage risks covering 2022 and 2023. Supervisors will communicate these to the affected firms.

2.23 Firms will not require new reporting systems and infrastructure in place to complete these exercises – as with previous ad-hoc data collections, it will be possible to submit the results via an Excel spreadsheet. Therefore, the PRA does not expect these ad-hoc data collections to undermine the benefits for firms from the delay in implementation of the reporting requirement.

Reporting on written credit derivatives

2.24 The respondent noted that, in respect of written credit derivatives, data on level 1 HQLA assets is not relevant to contingent leverage risks, and asked for the corresponding row in template LV52 to be removed. The PRA agrees with this response. The quality of the underlying asset is not relevant to assessing contingent leverage risks for credit derivatives – though, as set out in CP12/22, it is relevant to the PRA’s ability to judge the materiality of contingent leverage risks for other trade structures through internal scenario assessments. The PRA has removed the relevant row from the final LV52 template and the associated instructions accordingly.

‘Daily averaged’ basis of reporting

2.25 In CP12/22, the PRA proposed that firms would report data on both an end-period and averaged basis, and that period averages would be calculated using daily values in the case of on-balance sheet items and securities financing transactions (‘daily averaged reporting’). The respondent questioned the prudential benefits of daily averaged reporting, suggesting that period averages based on month-end values would be sufficient for the PRA to assess contingent leverage risks, while reducing the operational burden on firms.

2.26 Having considered the response, the PRA has decided to confirm the requirement for daily averaged reporting for on-balance sheet items and security financing transactions. Period averages based on daily values are an important input into the PRA’s and firms’ ability to monitor contingent leverage risks. Unlike daily values, month-end values may not be reflective of overall activity over the reference period, given the short-dated nature of the relevant trades and the natural volatility of trade activity. Relying on month-end values would therefore run counter to the PRA’s objective of obtaining a better understanding of contingent leverage risks through the new reporting requirement.

2.27 Daily averaged reporting would also facilitate comparability between the new contingent leverage templates and pre-existing leverage reporting templates, where LREQ firms already report period averages based on daily values (at a much more aggregate level than in the contingent leverage templates) for on-balance sheet items and security financing transactions.^{footnote [13]} Consistent with the existing supervisory expectations outlined in SS45/15, the PRA considers that ‘best estimates’ for daily averaged reporting within the contingent leverage templates, including for the additional breakdowns of trade activity, would be acceptable as long as they are measured consistently and prudently^{footnote [14]}.

2.28 The respondent suggested that the proposals in CP12/22 would require collateral swap data to be reported as averages based on month-end values. In fact, as specified in the draft amendments to the instructions in CP12/22, and confirmed in the final reporting instructions, for on-balance sheet assets and security financing transactions – which include collateral swaps – the data required is the arithmetic mean of the relevant data element on each day in the half-year period ending on the relevant reporting reference date. This is consistent with the existing averaging requirements for LREQ firms mentioned above. To aid clarity, the PRA has decided to reference these existing requirements in the final instructions on contingent leverage reporting.

Potential overlap with other reporting

2.29 The respondent noted that the proposed reporting templates overlapped significantly with liquidity templates and suggested that some of the data from the Liquidity Coverage Ratio (LCR) template LCR 75.01 in particular could be re-used to avoid unnecessary burden and duplication. The respondent also proposed that validation rules be introduced to enforce consistency across liquidity and contingent leverage templates.

2.30 The PRA considers that there are significant differences between the contingent leverage data template and LCR 75.01, such that capturing certain data points from LCR 75.01 as suggested by the respondent would not be a viable option for assessing contingent leverage risk. For example, LCR 75.01 only covers monthly data for collateral swaps that are due to mature within 30 days (the time horizon for the Liquidity Coverage Ratio), net of flows resulting from unwinding associated hedges. Therefore, it would provide the PRA with only a sub-set of the relevant trades and data required for assessing contingent leverage risks. Because of these differences, the PRA has also decided that the implementation of validation rules across the two templates would not be meaningful or helpful.

Other reporting clarifications

2.31 The PRA has, based on the response received to the CP, added a definition of internalisation within the reporting instructions.

2.32 The PRA has also, of its own initiative, modified the reporting instructions for the ‘internalised’ columns within LV52 (treatment of internalised trades). This modification is to avoid double-counting of certain internalisation activity.