PS13/24 – Funded reinsurance

1: Overview

1.1 This Prudential Regulation Authority (PRA) policy statement (PS) provides feedback to responses the PRA received to consultation paper (CP)24/23 – Funded reinsurance. It also contains the PRA’s final policy in the form of the final supervisory statement (SS) 5/24 – Funded reinsurance (Appendix 1).

1.2 This PS is relevant to UK Solvency II firms and insurance and reinsurance undertakings that have a UK branch (third-country branch undertakings) when they hold, or are intending to enter, funded reinsurance arrangements.

Background

1.3 In CP24/23^{footnote [1]} the PRA proposed to set expectations in respect of life insurance firms entering into or holding funded reinsurance arrangements as cedants, covering:

• the ongoing risk management of funded reinsurance arrangements;
• the modelling of the solvency capital requirement associated with funded reinsurance arrangements; and
• how firms should consider the structuring of funded reinsurance arrangements.

1.4 In determining its policy, the PRA considers representations received in response to consultation, publishing an account of them and the PRA’s response (feedback). Details of any significant changes are also published. In this PS, the summary of responses contains a general account of the representations made in response to the CP and the feedback to responses chapter contains the PRA’s feedback.

1.5 In carrying out its policy making functions, the PRA is required to have regard to various matters. In CP24/23 the PRA explained how it had regard to those matters, including a summary of those considered most significant in relation to the proposed policy. The changes to draft policy section of this chapter refers to that explanation, considering consultation responses where relevant.

Summary of responses

1.6 The PRA received 13 responses to the CP. The names of respondents to the CP who consented to their names being published are set out at Appendix 2. Respondents generally welcomed the PRA’s proposals to clarify expectations in this area, and made observations, requests for clarity, and recommendations to specific parts of the proposals, which are set out in Chapters 2 – 5.

1.7 The PRA must consider representations that are made to it in accordance with its duty to consult on its general policies and practices and must publish, in such manner as it thinks fit, responses to the representations.

1.8 The PRA has considered the representations received in response to the CP. Chapters 2 to 5 set out the PRA’s feedback to those responses, and its final decisions.

Changes to draft policy

1.9 This PS takes account of how the policy advances the PRA objectives and of significant matters that the decision maker had regard to. These are as set out in CP24/23.

1.10 The PRA welcomes the responses received to CP24/23 and has carefully considered the feedback and representations made by respondents. Having done so, the PRA has identified several areas where it is appropriate to make adjustments to the draft policy. The most material changes include:

• the addition of an overall clarifying statement confirming that firms may consider diversification between funded reinsurance counterparties and associated risks;
• clarifications on how firms will be expected to set internal investment limits, in particular the limit for recapture from a single counterparty;
• adjustments to the expectations around collateral policies to clarify how the underlying nature of collateral assets should be considered and how these policies should be documented;
• adjustments to the expectations around Board involvement with recapture plans;
• clarifications on how firms would be expected to manage any uncertainty in internal model outputs for funded reinsurance arrangements;
• clarifications on how the PRA considers the time horizon in respect of collateral mismatch risk; and
• amendments to how the PRA will expect firms to use collateral haircuts.

1.11 Further details on all issues raised in responses and any related amendments to the draft policy are set out in the relevant chapters of this PS. The PRA considers the changes made to the draft policy are appropriate and improve the final policy materials in a manner that aligns with the PRA’s statutory objectives.

1.12 The PRA considers that the costs and benefits of the final policy in this PS does not significantly differ from that in the draft policy proposed in CP24/23 and, therefore, the Cost Benefit Analysis (CBA) presented in the CP remains appropriate.

1.13 The PRA does not consider that the final policy in this PS would have a significantly different impact on mutuals as compared with other PRA-authorised firms.

1.14 The PRA considers that the responses to the consultation, and associated changes to the final policy, did not significantly alter its consideration of the matters to which it must have regard in implementing the final policy

Structure of the PS

1.15 The PRA’s feedback to responses received to the proposals in CP24/23 and an explanation of the final policy and rules are structured into the following chapters.

• Chapter 2 – Risk management of funded reinsurance arrangements
• Chapter 3 – Solvency capital requirement (SCR)
• Chapter 4 – Entering into and structuring of funded reinsurance arrangements
• Chapter 5 – Other feedback received

1.16 Appendix 1 of this PS sets out the final policy for the proposals set out in Chapters 2 to 5 of this PS in the form of the final version of SS5/24.

Implementation and next steps

1.17 The PRA has published the final version of SS5/24 alongside this PS. The SS is effective immediately, on publication of this PS on 26 July 2024.

1.18 Alongside this PS the PRA is issuing a letter setting out in more details the implementation approach for the SS and the approach that it expects firms to take in their self-assessment against the expectations of the SS. The PRA expects to seek assurance on firms’ practice in a proportionate way, focusing on the exposures which in the PRA view present the greatest risk. The PRA may consider this as a topic in a firm’s Periodic Summary Meeting (PSM) or, where appropriate, look to commission a Skilled Persons review under s166 of FSMA.

1.19 The PRA will also continue to monitor how market practice evolves and will keep under review whether there is a need for specific supervisory intervention or further policy measures with respect to funded reinsurance. This could include tools to address a potential build-up of sector-wide vulnerability, where this might pose a risk to UK financial stability.

2: Risk management of funded reinsurance arrangements

Introduction

2.1 This chapter provides feedback to responses relating to the proposals in Chapter 2 of the SS and the proposals around ongoing risk management.

2.2 In CP24/23 the PRA proposed to set expectations for firms to:

• establish internal investment limits in relation to funded reinsurance;
• include an immediate recapture metric within these limits, to be calculated before the impact of any management actions;
• set investment limits for single counterparty exposures so that a recapture of business would not threaten their business model;
• set additional limits considering simultaneous recapture from multiple highly correlated counterparties and aggregate funded reinsurance exposures;
• establish collateral policies as part of their risk management, with detailed policies for illiquid assets in collateral pools; and
• establish board approved recapture plans covering steps to take on recapture of a funded reinsurance contract.

2.3 The PRA received 12 responses commenting on the risk management of funded reinsurance arrangements section of CP24/23. These responses are also linked to chapter 2 of the SS.

Feedback to responses

2.4 The PRA has considered the responses received to the proposals around Solvency Capital Requirements (SCR). The responses have been grouped as follows:

• overall feedback;
• singular recapture limit threshold;
• management actions in investment limits;
• SCR-based investment limits;
• investment limit framework;
• diversification allowances;
• highly correlated counterparties;
• board involvement with recapture plan;
• investment limit expectations;
• collateral policy expectations;
• Own Risk and Solvency Assessment (ORSA) stress testing; and
• costs of proposals.

Overall feedback

2.5 Feedback on the PRA’s proposals around expectations for firms’ risk management of funded reinsurance was broadly positive, with nine respondents stating their support of the proposals in whole or in part, and requests for changes or clarifications focused on specific elements of the draft policy. These comments are covered in detail within the following sections.

2.6 The PRA proposed that firms be expected to establish investment limits in relation to funded reinsurance arrangements.

2.7 Seven firms commented that they supported the PRA’s investment limit proposals. Of these, three specifically mentioned their support for the ‘immediate recapture’ metric and two commented that they supported the expectation that firms consider broader factors rather than just relying on external credit ratings of counterparties.

2.8 One of the seven respondents also mentioned their support for various individual elements of the PRA’s proposals, including the expectations that firms should focus on the tail of the loss distribution, ensure they can survive a single recapture event, assume limited recollateralisation during stresses, and consider the nature of collateral backing a transaction.

2.9 Two of the seven respondents also mentioned their agreement with the PRA’s view that material reliance on funded reinsurance with one counterparty may present challenges with regards to compliance with the prudent person principle (PPP), with one additionally stating that they considered large exposures to several highly correlated reinsurers to be incompatible with PPP and another stating that they supported linking the measurement of funded reinsurance risk exposures to the PPP.

2.10 The PRA proposed that firms should be expected to have clear collateral policies in place as part of their risk management policies.

2.11 Five respondents commented that they supported the collateral policy proposals. Two of these also stated that they supported linking the collateral policy to the wider risk management framework and limit setting processes and one supported the expectation for firms to include detailed policies for the management of illiquid assets.

2.12 Three of these respondents also commented that they supported the PRA’s proposal that firms take into account their current matching adjustment (MA) approvals in considering the MA eligibility of collateral assets, and one mentioned their specific support of the proposal that firms assume recapture of assets outside of the MA portfolio unless MA compliance can be demonstrated in both prevailing and stressed conditions.

2.13 The PRA proposed that firms be expected to have a board approved recapture plan covering steps to be taken in the event of the recapture of collateral assets.

2.14 Five respondents commented that they supported the PRA’s recapture plan proposals, with three specifically mentioning their support for the elements the PRA proposed that firms be expected to include.

2.15 Two respondents commented that the PRA should be clear on the specific expectations that apply to funded reinsurance as opposed to other forms of reinsurance, or that the PRA should make a distinction between collateralised and uncollateralised funded reinsurance arrangements due to the security provided by collateral.

2.16 The PRA considers that the SS is clear that the new expectations it sets out are focused specifically on funded reinsurance arrangements. The PRA also considers that while collateralisation can reduce the risks associated with reinsurance arrangements, there are still risks associated with collateralised arrangements, as set out in the SS. Therefore, having considered these responses, the PRA has decided not to make any changes to the draft policy.

Singular recapture limit threshold

2.17 The PRA proposed that firms be expected to set internal investment limits such that a singular idiosyncratic event of a recapture of business from one counterparty does not threaten the firm’s business model.

2.18 Five respondents requested clarification on how the expectation that firms ensure that a recapture of business from one counterparty ‘does not threaten the firm’s business model’ should be interpreted. Some respondents specifically queried whether this meant that firms could still consider the potential for closing to new business and going into run-off in setting limits. While going to run-off would be a threat to business models, respondents noted that it is always a management action available to the firm.

2.19 Having reviewed the responses, the PRA has clarified in paragraph 2.9 of the final SS that it is for firms to determine the level of potential loss that would constitute a threat to their business models in line with its risk appetite, risk management policies, risk tolerance limits and investment strategy alongside its overall business strategy. The PRA has clarified in paragraph 2.9 of the final SS that it would expect firms to avoid taking on single counterparty exposures which could threaten their ability to continue to meet their solvency risk appetite (as detailed in SS1/20 Solvency II: Prudent Person Principle) or which would require significant value destroying management actions to be taken such as closure to new business.

2.20 As set out in the paragraph 2.10 of the final SS, the PRA notes that it considers that a material reliance on funded reinsurance with one counterparty may present challenges with regards to firms’ compliance with the PPP. The PRA considers that taking on new business is an important part of firms’ business models, and that closure to new business would classify as a significant change to a firm’s business model. Moreover, while closure to new business may in some circumstances be an appropriate management action for a single firm in response to an idiosyncratic shock, the PRA does not consider it to be an appropriate assumption for limit-setting purposes when there is a risk of coordinated failures across different firms due to correlated counterparties.

2.21 Additionally, given the complex nature of risks associated with funded reinsurance and the associated uncertainties around valuation and impact of recapture, the PRA would expect firms to pay particular attention to these arrangements and avoid overexposures (as detailed in SS1/20 Solvency II: Prudent Person Principle)^{footnote [2]} or excessive reliance on a particular funded reinsurance arrangement (in line with the Investments – 5.2 section of the PRA Rulebook).

Management actions in investment limits

2.22 The PRA proposed that firms be expected to consider the immediate recapture metric for developing internal investment limits before management actions are taken into account.

2.23 Seven respondents commented that setting investment limits before management actions were taken into account was overly prudent or that these limits should reflect management actions that could reasonably be achieved under stresses. Some respondents specifically mentioned allowing rebalancing on recapture, assuming re-collateralisation or considering other contractual safeguards.

2.24 Having reviewed the responses the PRA has decided not to change the draft policy in this area. The PRA views this limit as capturing the maximum potential loss arising from funded reinsurance arrangements. Therefore the PRA considers that the limit should be presented before considering the mitigating impact of any management actions (some, or all, of which may not necessarily be possible in the event of a recapture). The PRA also notes that firms have flexibility in setting this limit and in analysing additional limits.

SCR-based investment limits

2.25 The PRA proposed that the immediate recapture metric should measure the impact of a recapture on the firms’ SCR coverage ratio, and that internal investment limits should be set to avoid over exposure in periods of high SCR coverage.

2.26 Three respondents commented on these proposals, stating that the volatile nature of the SCR coverage ratio makes it unsuited for setting such limits, and one also stated that limits should consider only prevailing economic conditions rather than referencing periods of high SCR coverage.

2.27 Having considered the responses the PRA has decided not to change the draft policy in this area. While the PRA accepts that setting limits in relation to SCR coverage will need to allow for the volatility of this ratio, this is consistent with the expectation that firms should avoid over-exposure in periods of high SCR coverage, noting the long-term nature of these transactions. In addition, such limits can also be calculated using existing models, capture the key risks of funded reinsurance exposures, and are easily understandable.

2.28 Following on from responses to CP24/23, the PRA has clarified expectations around considering periods of high SCR coverage in paragraph 2.11 of the final SS, stating that this analysis should consider the impact of funded reinsurance exposures on firms’ long term target SCR coverage ratios.

Investment limit framework

2.29 One respondent stated the proposed aggregate funded reinsurance exposure limit would not be useful to help avoid build-ups of systemic or concentration risks, which could instead be covered by considering two recapture metrics for idiosyncratic and systematic defaults. Another respondent commented that too many limits had been proposed which would increase complexity and reduce their usefulness to senior management.

2.30 Having considered the responses the PRA has decided not to make any changes to the draft policy in this area. The PRA considers that firms should consider the potential operational impact of recapturing significant amounts of assets and liabilities which the aggregate limit will help evaluate. The PRA also considers the aggregate limit to be in line with established PPP expectations on monitoring concentrations of risk by asset class.^{footnote [3]} The PRA considers the suggested alternative idiosyncratic and systematic default limits to be materially similar to the limits set out in the draft SS (ie single counterparty default and simultaneous recapture from multiple highly correlated counterparties). The PRA considers that each of the three limits set out in the SS capture a different and important aspect of exposure.

Diversification allowances

2.31 Three responses commented that the SS should more explicitly encourage firms to consider diversification between funded reinsurance counterparties and between the different risk exposures associated with funded reinsurance arrangements.

2.32 Having considered the responses, the PRA has clarified in paragraph 1.5 of the final SS that there may exist diversification benefits in these forms of funded reinsurance contracts. However, the PRA has also clarified in paragraph 1.5 of the final SS that it considers that firms should be particularly prudent in recognising these benefits in its risk management arrangements given the high level of uncertainty associated with funded reinsurance arrangements and the potential size of losses.

Highly correlated counterparties

2.33 Three respondents commented that the SS should clarify how firms should determine whether counterparties are highly correlated, or that the PRA should consider the extremely remote risk of multiple highly correlated counterparties defaulting in setting expectations.

2.34 Having considered the responses the PRA has decided not to make any changes to the draft policy in this area. As set out in the SS, the PRA is concerned that counterparty risks may be underestimated and considers that it is important for firms to monitor their exposure to simultaneous recapture from multiple highly correlated counterparties, even if this risk is considered remote. The SS currently states that firms should assess the similarities in the risk profile of counterparties in determining whether counterparties are highly correlated. The PRA considers this level of guidance to be appropriate but will monitor firms’ approaches in this area to determine if further guidance is required.

Board involvement with recapture plan

2.35 Seven respondents commented on the PRA’s expectations for board approval of recapture plans, stating that while some level of board oversight may be appropriate, expecting this in all cases would be inflexible. The respondents suggested that approval of recapture plans should lie with firms’ management instead. One respondent also recommended that the PRA’s expectations be adjusted to clarify that recapture plans should consider the governance path for decisions to be made in recapture scenarios, rather than setting out a formulaic method for making decisions.

2.36 Having considered the responses the PRA has decided to adjust the proposed policy in this area, clarifying in paragraphs 2.20 and 2.21 of the final SS that board involvement is expected in setting the high level principles underlying recapture plans and reviewing inherent uncertainties, rather than there being an expectation of explicit board approval of every element of each individual recapture plan. The PRA has also clarified that it would expect the level of board involvement in reviewing and approving the recapture plans to be proportional to the level of risk being taken. This should reflect how funded reinsurance investment limits have been set and what potential impacts on a firm’s business model have therefore been accepted.

Investment limit expectations

2.37 The PRA proposed that firms be expected to consider the risks beyond the 1 in 200 confidence level over one year in assessing their funded reinsurance exposures and stated that this could be performed using a tail value at risk (TVaR) approach.

2.38 Four respondents commented on this proposal, stating that a TVaR metric would be of limited value given the lack of data available for credible analysis, with three respondents also stating that they considered stress and scenario testing to be more appropriate.

2.39 The PRA notes that the reference to a TVaR metric was provided as one example of an approach that could be applied to consider risks beyond the 1 in 200 point, and has clarified in paragraph 2.8 of the final SS that alternative approaches such as stress and scenario testing could also be used.

2.40 The PRA proposed that firms assuming collateral would be recaptured within the MA portfolio be expected to assume a reduced MA spread to reflect prudent rebalancing costs and assume the recapture of a ‘worst-case’ portfolio of assets.

2.41 Two respondents commented on these proposals; one recommended that rebalancing costs could be allowed for directly rather than in the MA spread, and the other commented that it was not clear why reinsurers would always deliver ‘worst-case’ assets in terms of cashflow matching. Three other respondents noted their support for the expectation that firms assume that a ‘worst-case’ portfolio of assets is received from counterparties in the event of a recapture, with two of these stating that this was their existing approach.

2.42 As noted above, the PRA considers that the intention of the investment limit expectation is to capture the maximum potential loss arising from funded reinsurance arrangements. The PRA considers that it is appropriate to expect firms to consider the potential loss arising from a ‘worst-case’ portfolio of assets being received rather than relying on potential upsides that are not contractually guaranteed. The PRA also notes that this is in line with some firms’ existing approaches and that firms can apply additional analysis on top of this if desired. Having considered this feedback, the PRA has maintained the expectation that firms consider the risks that they recapture a ‘worst-case’ collateral portfolio, but has also adjusted paragraph 2.7 of the final SS to clarify that rebalancing and trading costs could be allowed for separately from the MA spread.

2.43 The PRA proposed establishing the expectations that firms consider broader factors beyond just the counterparty’s credit rating when setting investment limits and establish continuous monitoring processes.

2.44 Two respondents commented on this expectation; one noted that internal credit rating processes already consider broader factors, and the other recommended that this assessment also considers support provided by collateral and that the expectations in this area were unclear in terms of what broader factors should be considered and how truly continuous monitoring could be carried out.

2.45 Three respondents commented that the expertise of external credit rating providers, and their access to non public information make external credit ratings a useful piece of information for firms to consider, and that these should not be disregarded.

2.46 Having considered this feedback the PRA has clarified in paragraph 2.12 of the final SS that it does not expect firms to rely solely on external credit ratings which may take longer to react to underlying changes in risk than internal assessments. The PRA accepts that external credit ratings provide useful information for firms to consider and would not expect firms to ignore this information, but has maintained the position that firms be expected to supplement these ratings with additional analysis for funded reinsurance arrangements.

2.47 The PRA considers that the current policy text would already allow for firms to consider collateral as one of the broader factors mentioned and that it is for firms to determine what other broader factors may be appropriate to consider and has not adjusted the SS text further on that point. Likewise, where a credit rating (internal or external) does not reflect concentration or correlation considerations, those are likely to be relevant broader factors to consider. The PRA has also clarified in the SS that monitoring should be performed on an ongoing basis rather than being continuous.

2.48 One respondent recommended that firms be expected to break down the immediate recapture metric into its constituent parts ie considering the impact on the best estimate liabilities, risk margin and SCR, rather than just the SCR coverage ratio.

2.49 Having considered this response the PRA has decided not to change the policy. The PRA considers the impact of a recapture on SCR coverage to be sufficiently informative and that setting an expectation for a breakdown of this figure to be analysed in all cases may dilute the focus it is given. The PRA also notes that firms are free to consider a breakdown of the SCR coverage impact if they consider it useful.

2.50 One respondent recommended that the PRA clarify the references in paragraph 2.5 of the draft SS to make it clearer which expectations apply to the immediate recapture metric and which are broader on limit setting. Considering this feedback the PRA has updated the references in 2.5 of the final SS to clarify this point.

Collateral policy expectations

2.51 Five respondents commented on the overall proportionality and materiality of the collateral policy expectations. Feedback included comments that the expectations may be disproportionate where counterparties are high quality, where collateral is liquid or has large haircuts applied, or where holdings are not material. Two of these respondents also commented on the expected frequency of ongoing monitoring of matching adjustment eligibility, and that this should reflect the nature of the underlying assets and the materiality of exposures.

2.52 The PRA notes that the draft policy stated that the detailed collateral policy expectations apply for illiquid asset holdings specifically. Having considered the feedback, the PRA has clarified in paragraph 2.16 of the final SS that the level of detail in the policy should reflect the materiality of exposures. As stated in the SS, the PRA considers that counterparty risks in funded reinsurance arrangements may be underestimated and as such has maintained the expectation that firms establish a collateral policy even if counterparties are considered high quality.

2.53 Having considered the feedback around ongoing monitoring of collateral, the PRA has also clarified in paragraph 2.17 of the final SS that ongoing monitoring should reflect the characteristics and materiality of the collateral assets.

2.54 The PRA proposed that firms be expected to develop supporting analysis to demonstrate that recapture from a counterparty would not result in a breach of MA conditions in both base and stressed conditions.

2.55 One respondent queried what the PRA meant by base conditions in this context. The PRA has adjusted the text in paragraph 2.19 of the final SS to clarify that the base analysis is expected to reflect prevailing economic conditions at the time of the analysis.

2.56 The PRA proposed that firms be expected to have a detailed collateral policy in place for illiquid assets in collateral pools given the increased risk associated with these.

2.57 One respondent stated that it was unclear what the PRA considered as illiquid assets. The PRA has included some additional detail in paragraph 2.16 of the final SS on what factors could lead to an asset being considered illiquid, and would also refer firms to the existing SS3/17 - Solvency II: illiquid unrated assets.

2.58 The PRA proposed that collateral policies be expected to include investment management approaches on recapture.

2.59 One respondent commented that they considered it unfeasible for the collateral policy to define a single set of management actions for how collateral would be recaptured which could be relied on in all situations given the uncertainty around the economic and market conditions at recapture, and that a range of scenarios should be considered.

2.60 Having considered this feedback, the PRA has clarified in paragraph 2.16 of the final SS that the collateral policy would be expected to consider potential investment management approaches under different circumstances, including specific consideration of how assets may be managed long term if necessary (eg if assets are received in a stressed scenario and it is not possible to sell them for the expected price, even taking into account any collateral haircuts).

2.61 The PRA proposed that firms’ collateral policies be expected to be closely linked to their limit setting processes.

2.62 One respondent stated that it was unclear what was meant by these being closely linked in practice.

2.63 Having reviewed this feedback the PRA has clarified in paragraph 2.15 of the final SS that it would expect firms to consider the nature of the collateral they are accepting exposure to, and factor this into their limit setting processes.

2.64 One respondent commented that they would not consider a collateral policy document to be the singular place where all the items mentioned in the SS would be documented, and that in practice it may cross reference other documents.

2.65 The PRA has clarified in paragraph 2.16 of the final SS that the collateral policy may reference existing methodologies or other documentation.

2.66 One respondent recommended that the collateral policy additionally be expected to include a dynamic element, considering what actions could be taken if it seemed that a counterparty was close to failure, for example more frequent collateral valuation.

2.67 The PRA notes that the draft policy around the recapture plan already included an expectation that firms establish approaches to monitoring counterparties and defining activities which could be carried out if a deterioration is identified. The PRA considers that this sufficiently covers the substance of the recommendation and has not made any further adjustments to the SS.

ORSA stress testing

2.68 The PRA also proposed that firms be expected to carry out annual stress testing around funded reinsurance in their own risk and solvency assessment (ORSA).

2.69 Three respondents commented that that this expectation may be disproportionate where funded reinsurance exposures and associated risks are immaterial and could be performed as part of overall stress and scenario testing rather than as a standalone exercise. One respondent supported the expectation that testing around funded reinsurance be included in the ORSA.

2.70 Having considered this feedback, the PRA has adjusted the expectation in paragraph 2.23 of the final SS, such that firms be expected to analyse their funded reinsurance exposures annually and only include specific stress testing around funded reinsurance in the ORSA report where exposures are material.

Costs of proposals

2.71 One respondent commented that the expectations around the recapture plan may require significant work for firms. The PRA has considered the costs associated with the expectations as part of the consultation process and noted that some firms may incur additional costs to bring their practice in line with the PRA’s expectations. The PRA does not expect these costs to be significant for most firms and considers that the benefits of the proposals in the SS outweigh the costs. The PRA has therefore not made any adjustments to the draft policy in response to this comment.

3: Solvency capital requirement (SCR)

Introduction

3.1 This chapter provides feedback to responses relating to the proposals in Chapter 3 of the SS and the proposals around SCR.

3.2 In CP24/23 the PRA proposed to introduce new expectations:

• setting out the risks that should be reflected in internal models to calculate a counterparty SCR for funded reinsurance at a sufficient level of granularity to inform the decision making process when it comes to deciding whether to enter into a funded reinsurance arrangement as a risk mitigation technique;
• setting expectations for the elements that need to be considered in probability of default (PD) including data, stressed PD, termination clause and dispute, counterparty’s solvency ratio, as well as forward looking and non public information;
• setting expectations for the elements that need to be considered in loss given default (LGD) including stressed liabilities, stressed counterparty default adjustment, risk margin on recapture, management actions and wrong way risks;
• setting expectations considering risks related to the quality of the collateral, including a need to look through to the underlying asset portfolio, the need to consider collateral mismatch risk, potential issues in re-collateralisation and reduced effectiveness of the risk mitigation technique in stress; and
• set an expectation that firms assume that assets and liabilities associated with funded reinsurance contracts be recaptured outside of an MA portfolio, unless they clearly demonstrate that such an inclusion would not result in non-compliance with the MA conditions under both prevailing and stressed scenarios.

3.3 The PRA received 10 responses to the solvency capital requirement chapter of CP24/23. These responses are also linked to Chapter 3 of the SS.

Feedback to responses

3.4 The PRA has considered the responses received to the proposals around solvency capital requirements. The responses have been grouped as follows:

• overall feedback;
• proportionality and materiality;
• potential positive diversification from engaging with diversified reinsurers;
• insufficient data to allow counterparty solvency modelling in stress;
• probability of recapture, linked to dispute and termination clauses;
• absence of adequate forward looking and non public information;
• recognition of diversification of non MA assets;
• day one gains validation cannot be adequately modelled; and
• other responses.

Overall feedback

3.5 Feedback on the PRA’s proposals around expectations for aspects of solvency capital requirement for funded reinsurance was broadly positive, with four respondents stating their support of the proposals in whole or in part, and requests for changes or clarifications focused on specific elements of the draft policy. These comments are covered in detail within the following sections.

3.6 One respondent noted that the expectations seemed to be reasonable and proportionate, noting that several of the approaches suggested by the PRA had been adopted by the respondent in other parts of its internal model.

3.7 One respondent noted that they generally agreed with the proposals, in particular by asking firms to be methodical in setting out how the PD is calculated and what information they use. The respondent also noted that it may encourage reinsurers to provide solvency ratio sensitivities on a more frequent and comprehensive basis.

3.8 One respondent agreed with most of the proposed expectations in this chapter, noting that the expectations were reasonable and could be included as part of their annual validation processes.

Proportionality and materiality

3.9 The PRA proposed specific expectations in respect of measuring the counterparty credit risk capital charge in a firm’s’ internal models or partial internal models, with the aim of capturing all material and quantifiable risks and taking into account the effects of funded reinsurance as a risk -mitigation techniques.

3.10 Four respondents stated that the principles of proportionality and materiality should apply in the modelling expectations for the internal model. Two respondents stated they would not expect to include risks that do not materially affect the losses on recapture. Another respondent noted that the expectations add significant computational complexity and expert judgement to the modelling and will likely have an immaterial impact on the final SCR. One respondent noted that the expectations were too explicit and that they should be considered in proportion to their informative value. One respondent stated that the approach adds computational complexity and diverts resource and attention from more valuable risk mitigants such as robust recapture planning, or evolving the limits methodologies. One respondent noted that the wording may impose disproportionate modelling expectations for insurers and prescribe overly cautious limits frameworks and capital requirements.

3.11 The PRA is proposing to adjust the expectations to take into consideration several responses (in this and subsequent sections of this chapter) that focus on the uncertainties in the modelling of the risks associated with funded reinsurance. The final SS sets out the expectations on how firms can meet the standards for the counterparty risk module of their SCR. The adjustments emphasize that, if in doing so, firms identify areas of uncertainty in their modelling, they would be expected to consider this in their risk management system, including limiting their exposure to factors which increase the level of uncertainty. This could be in the form of tighter funded reinsurance investment limits or reduced risk-taking in contractual structures or collateral portfolios.

3.12 Senior management are expected to be informed by a firm’s internal model of the economics of the individual transaction as part of considering its risk return. This level of granularity at the transaction level, expected specifically in the context of funded reinsurance, means firms need to demonstrate that the level of modelling is sufficiently robust at the individual contract level. The PRA considers that this will better reflect the proportionate approach that firms are expected to take and will lead to better decision making. These updates set out in paragraph 3.5 of the final SS reflect the important use-test principle that models must be sufficiently granular to play an important role in the relevant management decisions of the firm.

Potential positive diversification from engaging with diversified reinsurers

3.13 The PRA proposed specific expectations in respect of measuring the counterparty credit risk capital charge in a firm’s’ internal models or partial internal models,

3.14 Three respondents stated that insurers should capture broader factors than those proposed in the draft policy, including positive forms of diversification. Two respondents stated that this includes:

- diversification between the reinsurer and insurer’s business model and risk profile;

- diversification between a global reinsurer that is exposed to both property and casualty (P&C) and life risks, with a low exposure to corporate credit and other non-government credit; and

- the absence of wrong way risk where there is diversification of the collateral portfolio with the reinsurer’s business model and risk profile.

3.15 The PRA recognises the importance of various forms of diversifications to prudent risk management. The inclusion of such elements could incentivise prudent risk taking and contracts with lower levels of uncertainty. However, the final SS focusses on the additional risks from the correlated counterparties that should be reflected in the capital charges for transactions with such counterparties, and the size of limits on exposure to them. The PRA has therefore decided not to change draft policy that firms consider the heightened risk from the more correlated counterparties and the wrong way risks. A point in time assessment of these diversification benefits for the rapidly evolving business models of the counterparties may not be a prudent approach to take.

Insufficient data to allow counterparty solvency modelling in stress

3.16 The PRA proposed that firms analyse how the solvency ratio of their counterparties changes under various market stresses, and how this could inform their stressed PD. The PRA proposed in paragraph 3.9 of the draft SS to capture the wrong way risk and proposed in paragraph 3.11 of the draft SS to consider risk that their counterparty to the funded reinsurance arrangement might not be willing or able to replenish the collateral portfolio in certain stressed conditions due to a breach in solvency risk appetite.

3.17 Four respondents stated that the expectations on analysing the counterparty solvency ratio in stress were unreasonable. One respondent noted that it was not practicable given the resources required to build a suitable model that could cater for extreme stresses. Three respondents noted that the information required to perform this calculation of impact of stresses to a high level of granularity is commercially sensitive or not sufficiently disclosed in public or private disclosures. One respondent noted on the other hand that the expectations may encourage reinsurers to provide this information on a more frequent and comprehensive basis (likely in public disclosures to avoid the potential for disclosing market sensitive information, especially for listed reinsurers).

3.18 After considering these responses, the PRA has decided not to change the draft policy in paragraph 3.6 of the final SS. The PRA is not requesting firms to build an internal model module for their counterparties and believes the expectations as set out are informative and proportionate. For example, firms should be able to use public disclosure of solvency ratio sensitivities to carry out a high level analysis of how, under various stresses and scenarios, a counterparty’s solvency ratio changes, and how this could make recapture or default more likely in various stress environments.

Probability of recapture, linked to dispute and termination clauses

3.19 The PRA proposed that firms consider the occurrence of any credit event set out in the transaction documentation when calibrating their PD. This may include when certain contractual triggers (eg termination clauses) are breached.

3.20 Four respondents noted that the expectations on considering the risk of recapture ahead of the default of the counterparty were unreasonable. One respondent argued that this scenario should be captured separately rather than being accounted for through an adjustment to PDs. The PRA considers that this is still possible within the approach consulted on. Two respondents noted that such an expectation may generate a perverse outcome if firms were disincentivised from seeking these credit mitigants such as termination clauses set at prudent levels due to an overly onerous expectation to model in granular detail the potential for exercising these rights. Two respondents noted that it was unlikely that a firm would take action to enforce a termination clause if this resulted in a loss under stressed conditions.

3.21 Having considered the comments raised, the PRA has decided not to change paragraph 3.6 of the final SS but to add a clarifying sentence to note that, given the beneficial nature of these clauses, the PRA expects that firms are able to recognise their benefits in stress. Firms do not need to calibrate their internal model to explicitly model the interaction between solvency ratio of the counterparty and the termination option triggers getting in-the-money. The expectation only sets out that firms should consider how those two interact and how it can inform the calibration of their stressed PD, consistent with the approaches detailed in the recapture policy. This analysis should inform the assessment of the level of confidence in the output of the model. For some idiosyncratic shocks to a single counterparty, the PRA agrees that the availability of a termination trigger clause acts as a valuable tool to manage the potential losses to firms. However, in a more credit focussed event affecting multiple counterparties simultaneously, actioning termination triggers may generate an immediate loss, but avoid the firm suffering a more catastrophic loss if they waited. This approach needs to be consistent with how the firms have documented their approach to termination triggers in the recapture plan.

Benefits of MA ineligible assets

3.22 The PRA proposed that firms should consider the risks associated with accepting MA ineligible assets in the collateral portfolio.

3.23 Two respondents commented on the benefits of MA ineligible assets in the collateral pool as a form of diversification against the firm’s balance sheet assets which may be helpful under stress. One respondent noted that firms should not be operationally penalised for accepting a diverse range of collateral assets, particularly where MA eligibility could reasonably be expected to be achieved within reasonable timescales after recapture. One respondent noted that illiquid assets can in fact improve diversification, reducing correlation and the impact of a credit event.

3.24 Having considered these responses, the PRA has decided not to change paragraphs 3.14 and 3.15 of the final SS. The PRA agrees that in some instances a wide range of assets in a portfolio can be beneficial if it unlocks diversification. However, where these are non MA eligible, the benefits need to be carefully balanced against the potential increased risks on recapture linked to the realisable value or the cost of transforming into an adequate amount of assets. Given the material uncertainty in the costs and risks associated with these non MA eligible assets, the PRA expects that firms will be prudent in allowing these in contracts or have adequate mitigants.

Validations

3.25 The PRA proposed that firms should develop validation processes to specifically explain the sources of any day one new business gain generated by entering a funded reinsurance arrangement.

3.26 Four respondents stated their concerns with the expectations to develop validations to explain the sources of day one new business gain. One respondent noted that this would be difficult because reinsurers do not disclose their pricing bases, or their investment and capital strategies to the market, and consequently it would not be possible in practice to reach a fully reasoned view based on publicly available information. Three respondents questioned the relevance to the internal model SCR.

3.27 Having considered these responses, the PRA has decided to provide some clarifications as to how this validation exercise could be carried out in paragraph 3.6 of the final SS. One way this validation can be carried out is by comparing the premium charged by the reinsurer with the premium that would have been charged by the insurer and reconciling the difference. The difference can be explained by various elements such as differences in gross investment spread given different investment and hedging strategies (eg based on the returns of the collateral portfolio), differences in expenses, and differences in the cost of capital. While this analysis is not intended to be exhaustive or perfect, given its reliance on a range of data sources, it may point to an unexplained source of day one gain which represents additional counterparty credit risks that are not sufficiently captured in the internal model SCR output.

Other responses on the SCR

3.28 The PRA proposed that firms consider forward looking information and non-public information in their assessment of the PD. Five respondents noted the risks in using forward looking information or non-public information in setting the PD of the counterparties. One respondent noted that rating agencies that have large teams of specialist dedicated resources regularly monitor these reinsurers. Three respondents noted that the forward-looking information is unlikely to be sufficiently comprehensive and that the credit default swap (CDS) data is limited given the illiquid nature of the market. Three respondents however noted that this form of information should be allowed where appropriate to reduce the PD.

3.29 The PRA notes that firms can use a range of approaches to set their PD and the expectations are set up such that firms consider a broad range of data sources. The PRA does consider this information to be potentially informative. The PRA has decided not to change the draft policy in these areas.

3.30 The PRA proposed that firms to apply prudent assumptions in setting recovery rates. Two respondents noted that the recovery assumptions should consider the nature of the recapture and the severity of the stress. One noted that the recoveries from recapture prior to default would be expected to be higher than recoveries due to default events. Two respondents were supportive of the need to set prudent recovery assumptions to reflect how the severity of the stress could mean that the counterparty is not willing or able to replenish the collateral. However, one also noted the need to recognise the benefits of contracts where the reinsurers would be required to replenish the collateral portfolio post the stress.

3.31 The PRA is open to considering alternative approaches and believes that the responses are not inconsistent with the current drafting of the policy. Therefore, having considered the responses, the PRA has decided not to change the draft policy in paragraph 3.11 of the final SS.

3.32 The PRA proposed that firms should adopt a prudent approach and not take into account management actions in their internal models or partial internal models where there is insufficient data to demonstrate availability of pricing of such actions. Three respondents noted that the expectations regarding management actions on recapture should be consistent with existing rules. Two respondents agreed with the proposal but one respondent noted that the PRA should be more explicit by what it means by management actions that can reasonably be expected to be carried out.

3.33 Having considered the responses, the PRA proposes not to change the draft policy in paragraph 3.9 of the final SS. Management should make their own assessment of their ability to carry out specific management actions and should be prudent about actions where there is limited or no historical precedent to justify the action.

3.34 The PRA proposed that firms should stress the collateral portfolio on a look through basis to reflect the risks that the firm would ultimately be exposed to on recapture. Three respondents stated that the look through approach to modelling the collateral in stress may not be the most informative approach. Two respondents noted that the substitution rights means that the counterparty could change the collateral portfolio quickly ahead of recapture. Two firms presented alternative approaches including the use of the worst-case portfolio (ie the limit of the investment guidelines agreed with the counterparty) or the use of valuation haircuts.

3.35 While the PRA agrees that substitution rights generate risks that firms need to manage, it also believes that a clear modelling of the collateral portfolio based on the nature and timing of the cashflows of the actual assets is important to help firms understand the risks they are actually exposing themselves to. Simplified template portfolios may understate the actual issues with the actual portfolio and may result in firms not fully identifying all the potential sources of risks that have not been mitigated as part of the structuring processes. Having considered these responses the PRA has decided not to change the draft policy in paragraph 3.11 of the final SS.

3.36 The PRA proposed the elements that need to be considered when assuming a recapture within the MA portfolio. Three respondents noted that the PRA should be explicit that the expectations on recapture within the matching adjustment portfolio (MAP) are consistent with SS8/18 Internal Models Modelling the Matching Adjustment^{footnote [4]}. One respondent noted that the compliance in line with SS8/18 was clear from the draft SS. Having considered these responses, the PRA has decided not to change draft policy as it is already sufficiently clearly referenced within the SS including paragraph 3.16 of the final SS.

3.37 One respondent noted that where the liabilities are already included in the MAP, the firms should be permitted to assume automatic recapture into the MAP. Two respondents however agreed that to recapture within the MAP, firms needed to demonstrate that compliance is achieved post recapture.

3.38 The PRA has decided that there is insufficient evidence provided to support the argument that automatic costless recapture into the MAP could be assumed in the internal model. Having considered these responses the PRA has decided not to change draft policy.

3.39 One respondent noted additional validations could be spelled out in the PRA expectations. This included the need to compare the counterparty default adjustment to the fundamental spread of the collateral portfolio and the need to compare the capital requirements of funded reinsurance with that of a combined longevity swap and secured financing transaction.

3.40 The PRA agrees that appropriate additional validations would be helpful but does not believe that this is appropriate for all firms given the nature of their funded reinsurance arrangements. Having considered the response the PRA has decided not to change draft policy as it may not be appropriate for all firms.

4: Entering into and structuring of funded reinsurance arrangements

Introduction

4.1 This chapter provides feedback to responses relating to the proposals in Chapter 4 of the SS and the proposals around entering into and structuring of funded reinsurance arrangements.

4.2 In CP24/23 the PRA proposed to:

• set out an expectation for the cedant to have a quantitative risk assessment process for funded reinsurance arrangements to identify and measure the specific risks it might incur when negotiating the funded reinsurance arrangements for the purpose of determining internal limits and risk management processes;
• set out an expectation that this quantitative risk assessment process will support the structuring process to allow the implementation of adequate safeguards in the funded reinsurance arrangements to mitigate the risks generated;
• set out expectations on how firms at a minimum should allow for basis risk and collateral mismatch risks;
• set out an expectation to have internally approved minimum guidelines on contractual features for funded reinsurance transactions; and
• set out an expectation to have clear risk based collateral haircuts linked to the risk being addressed.

Feedback to responses

4.3 The PRA received 8 responses to the entering into and structuring of funded reinsurance arrangements chapter of CP24/23. These responses were also linked to Chapter 4 of the final SS.

4.4 The PRA has considered the responses received to the proposals around entering and structuring funded reinsurance arrangements. The responses have been grouped as follows:

• overall feedback;
• haircut policy;
• maximum accepted loss at contract level;
• only expect consideration of material risks in risk assessment;
• counterparty strength;
• frequency of collateral rebalancing;
• quantitative assessment over the lifetime of a funded reinsurance arrangement;
• basis risk on termination;
• use of internal model rather than 4 step framework for quantitative assessment;
• diversification; and
• contractual mitigations.

Overall feedback

4.5 Three respondents were broadly supportive of the proposals of entering into and structuring of funded reinsurance arrangements. One respondent noted that the proposals around basis risks, collateral mismatch risks, time horizon, and contractual mitigations are aligned to their current approach. Another respondent agreed with the PRA’s proposal to have quantitative risk assessments and limits for funded reinsurance. The respondent saw the risk assessments and stress and scenario testing as an essential part of an insurance company’s risk management toolkit. The third respondent believes that funded reinsurance agreement is a complex structure and having minimum “pass or fail” criteria can oversimplify the transaction. Hence, the respondent agreed that risk identification should be used to support the structure of the transaction and appropriate safeguards should be determined. They agreed that PPP expectations should apply to funded reinsurance.

4.6 One respondent commented that as a detailed risk assessment was likely to be carried out on a reinsurer that it would cause reinsurers to be more proactive in providing materials to support the risk assessment. They also welcomed the focus on additional secondary risks that can arise from these transactions and linking these to a firm’s internally approved risk appetite framework.

Haircut policy

4.7 Contractual protections can be a powerful tool to manage risks and incentivise the right behaviour on the part of the counterparty. The PRA proposed that firms have internally approved minimum guidelines on contractual features for funded reinsurance transactions which they would apply when deciding whether to enter a funded reinsurance arrangement. The PRA proposed firms document the rationale for the choice of the minimum guidelines adopted in their policies.

4.8 The PRA also proposed firms use clear risk based collateral haircuts linked to the risk being addressed and allow for over collateralisation in steps 1 and 2 of the assessments set out in paragraph 4.4 of the draft SS.

4.9 One respondent agreed with the PRA’s statement that haircut calibrations or collateral policies to mitigate risks should be set by considering price volatility, currency risk, and market risks of eligible collateral which should be assessed using a long time period.

4.10 Two respondents noted that haircuts are fixed at transaction closing and it is not possible for haircuts to move in line with key risk drivers. They further noted that funded reinsurance should not be considered comparable to a derivative credit support annex (CSA). One respondent noted that in negotiation of funded reinsurance it may not be feasible from a legal perspective to include mechanisms that have a dynamic linkage to a firm’s own risk policies which themselves can change over time.

4.11 Two respondents suggested that cashflow matching is not a material risk because of potential to access to the interest rate market. One of these respondents stated that having sensible policies in place to manage duration mismatch risk, and to periodically validate the risk of cashflow matching alongside the need to restore MA compliance in a recapture event as part of the SCR expectations is sufficient. They also stated that haircuts which factor in cashflow matching requirements are likely to be overly complex and disproportionate to the risk it is trying to manage. One respondent noted that funded reinsurance represents a small proportion of overall balance sheets, is captured in the detailed recapture plan, and reserved for in the SCR, so cashflow matching should not be considered within any haircut policy.

4.12 Four respondents suggested that haircuts should not be required where firms intend to retain the assets or that haircuts should not be standardised but should be bespoke to the risks in the specific contract. They noted that firms should be responsible for setting haircut policy as the current wording is too prescriptive. One respondent suggested that there should not be an expectation of both haircuts and over collateralisation.

4.13 One respondent suggested that haircuts would not change reinsurer behaviour towards investment decisions as they make decisions based on their own credit assessments and risk appetite. They also disagreed with adopting what they described as a one size fits all approach.

4.14 One respondent noted that determining haircuts involves negotiation so calibrating haircuts at a high confidence level using historical data is not practical but should be done if possible.

4.15 Having considered the above responses, the PRA has made several changes to the wording in paragraph 4.13 of the final SS. The revised wording states that the PRA also expects firms to use clear risk based collateral haircuts or over collateralisation linked to the risk being addressed. It goes on to list circumstances where haircuts may be more appropriate and where over collateralization may be more appropriate. The PRA has added wording to clarify that haircuts and over collateralisation should be calibrated to ensure that the risk of a shortfall in the realisable value of collateral in the event of default relative to the total amount due from the reinsurer to the cedant is within the level of confidence required by the approved internal contractual risk appetite statement (as defined in paragraph 4.6 of the SS). The PRA has altered the wording around key risk drivers such that haircuts and over-collateralisation should allow for the expected volatility of key risk factors that drive of the movement under stressed conditions in the value of the collateral assets and the total amount due from the reinsurer. This reflects the fact that haircut and over collateralisation terms can be set at the outset of the contract. The PRA has expanded the wording around broader risk considerations to add clarity.

4.16 The PRA considers that the rest of the expectations in this area are appropriate as they represent risks that funded reinsurance presents and has retained wording in respect of this. The PRA views that cash flow mismatches should be considered as part of over collateralisation. The PRA notes it is up to individual firms to set their own haircut and over-collateralisation policy which allows firms to be commercially flexible where appropriate and the wording in the final SS achieves this.

Maximum accepted loss at contract level

4.17 As part of the assessment of risks, the PRA proposed that firms, when negotiating the funded reinsurance arrangements, to undertake a quantitative assessment to identify and measure the specific risks they might incur for the purpose of determining their internal limits and risk management processes.

4.18 The PRA proposed for firms to have an approved internal contractual risk appetite statement setting out the maximum acceptable loss at the individual funded reinsurance contract level.

4.19 Three respondents suggested that it is better to set maximum accepted loss limits at an aggregated level rather than single contract level. One respondent noted that a better approach is to consider the aggregate counterparty exposure level and any diversification of risks where different risks have been ceded. Another respondent doubted that a limit would be effective in practice and would undermine other limits that in place. The respondent recommended the removal of paragraph 4.6 of the draft SS and no individual contract limit for funded reinsurance.

4.20 Having considered the responses, the PRA considers it appropriate to expect firms to have a maximum acceptable loss at the individual funded reinsurance contract level and has decided not to change its draft policy. This reduces the need for the PRA to set expectations around individual clauses in reinsurance contracts. This also supports other limits that are expected from firms as part of the SS.

Only expect consideration of material risks in risk assessment

4.21 As part of the assessment of risks the PRA proposed firms, when negotiating the funded reinsurance arrangements, should undertake a quantitative assessment to identify and measure the specific risks it might incur for the purpose of their internal limits and risk management processes.

4.22 Two respondents suggested that when carrying out the quantitative risk assessment for counterparty risk firms should only consider material risks and not all basis risks, collateral mismatch risks, and contractual mitigations. One respondent proposed to add the word material in all the areas within the SS where the PRA has listed individual risks.

4.23 One respondent agreed that collateral should be monitored for material mismatch but at a high level only.

4.24 Having considered the responses, the PRA has decided not to change its draft policy. The PRA considers it appropriate for firms to consider all the risks that are in a funded reinsurance contract as part of the risks assessment before any assessment of materiality is considered. The PRA considers it appropriate for firms when responding to risks to take a proportionate approach to the nature, scale, and complexity of risks and the wording reflects this.

Counterparty strength

4.25 The PRA proposed that firms should apply the expectations on entering into and structuring funded reinsurance arrangements to all contracts that a firm enters into.

4.26 One respondent commented that by focusing on collateral, firms could unintentionally overlook counterparty strength which is an important consideration when entering into a funded reinsurance transaction. The respondent proposed to the PRA to include a statement noting that counterparty strength is an important consideration when assessing the risks associated with a funded reinsurance arrangement, and the potential contractual protections that should be sought.

4.27 Having considered the responses, the PRA has decided to retain the proposed wording and not include specific expectations on firms to consider counterparty strength. The PRA expects firms to ensure that there are strong contractual protections in place as part of any funded reinsurance transaction.

Frequency of collateral rebalancing

4.28 The PRA proposed that for the collateral mismatch risk, shortfall can emerge both from how the collateral terms are defined and from the frequency of margining. Where the margining is undertaken only on an infrequent basis (for example quarterly), the PRA proposed that it expects firms to consider the risk that large shortfalls emerge at recapture. Where ad hoc revaluation options are available firms should only allow for this if they have clear approved policies on how they would use this power.

4.29 Two respondents requested the PRA to clarify what is meant by the term infrequent. They believe that monthly collateral margining is reasonable and should be sufficient as frequent collateral calls are not practical given the need to update member movements and produce revised cashflows in stress scenarios.

4.30 Having considered the responses, the PRA has decided to remove the example wording for infrequent rebalancing in paragraph 4.11 of the final SS. The PRA does not believe that it is appropriate to specify a frequency for rebalancing in a SS that would be considered infrequent. The PRA expects firms to justify the frequency of rebalancing when considering whether there is a risk that large shortfalls emerge at recapture and to take account of the potential size of those shortfalls in capital and limits.

Removing the focus from the prescriptive test

4.31 The PRA proposed that firms undertake a quantitative assessment to identify and measure the specific risks it might incur when negotiating the funded reinsurance arrangements for the purpose of their internal limits and risk management processes.

4.32 One respondent noted that there are some merits from quantitative assessment of risks through a hypothetical contract in which a firm sets its risk appetite and seeks to secure funded reinsurance terms in line with this contract. However the respondent recommended that the PRA consider a broader articulation of risk assessment that relies less on prescriptive tests to be more appropriate and effective as actual contracts are more bespoke and heavily negotiated.

4.33 Having considered the responses the PRA has decided to maintain the wording set out in paragraphs 4.3 to 4.11 of the final SS. The SS provides firms the opportunity to set their own limits and risk appetite and therefore gives firms sufficient commercial flexibility.

Quantitative assessment over the lifetime of a funded reinsurance deal

4.34 The PRA proposed firms perform a quantitative assessment under plausible stress scenarios, both for the full life of the contract and at potential contract termination, ahead of contract completion.

4.35 One respondent suggested the PRA should allow firms to use a qualitative assessment after the contract is established for the lifetime of the funded reinsurance contract rather than a quantitative assessment which might be a disproportionate numerical analysis.

4.36 Having considered the response the PRA has decided to maintain the wording set out in paragraph 4.9 of the final SS. The PRA believes that a quantitative assessment is appropriate for the lifetime of the funded reinsurance contract.

Basis risk on termination

4.37 The PRA proposed a firm’s identification of risks to include, at a minimum, an assessment of possible shortfalls between expected reinsurance cover and actual cover.

4.38 One respondent broadly agrees that simplifications are sometimes made when agreeing reinsurance benefits and should be considered when valuing liabilities. Two respondents suggested that the basis risk for mortality and longevity risk does exist, however, it can be recovered in the future. They suggested that if a dispute arises in terms of the shortfalls then the independent expert dispute processes present in funded reinsurance contracts adds protection and can be utilised.

4.39 Having considered the responses the PRA has decided to maintain the wording set out in paragraph 4.7 of the final SS. The responses indicated that basis risk can still be a risk with a funded reinsurance recapture and therefore it is appropriate that these are taken into account in the quantitative assessment.

Four-step framework for quantitative assessment

4.40 The PRA proposed the use of the four step framework described in paragraph 4.4 of the draft SS could help firms to ensure that their approach to identifying, measuring, monitoring, managing, controlling, and reporting covers all material and quantifiable risks to which they would be exposed if they entered into the funded reinsurance arrangement. The SS expects firms either to use this framework or satisfy themselves that if they take a different approach it covers these areas.

4.41 One respondent recommended changing the wording to state that using a suitably designed internal model for quantitative assessments is acceptable. The respondent believes a suitably robust internal model can be appropriate for carrying out a quantitative assessment of a new funded reinsurance agreement.

4.42 Having considered the response, the PRA has decided to maintain the wording set out in paragraph 4.4 of the final SS. The expectations in paragraph 4.4 of the final SS are only an example of an approach that firms can take, and other appropriate approaches may be used. Firms may find it useful to use the internal model as part of a quantitative assessment but the PRA would not expect this to be the only tool that is used as firms may find that their internal models lack sufficient capability to meet all the expectations in paragraph 4.3 of the final SS.

Diversification

4.43 The PRA notes that the proposed assessment is to be performed at a contractual level and does not consider the diversification with other risks.

4.44 One respondent believes that it is appropriate to determine whether the new arrangement falls within the internal contractual risk appetite by assessing identified risks in a stress event with an allowance for risk diversification. The respondent suggests that allowance for risk diversification is acceptable in the analysis.

4.45 Having considered the response the PRA has decided to maintain the wording set out in chapter 4 of the final SS. The PRA believes that contractual risk and mitigants should be considered in isolation for each funded reinsurance contract.

Contractual mitigations

4.46 The PRA proposed that firms have internally approved minimum guidelines on contractual features for funded reinsurance transactions which they would apply when deciding whether to enter into a funded reinsurance arrangement. Firms are expected to document the rationale for the choice of the minimum guidelines adopted in their policies.

4.47 One respondent agreed with the PRA’s proposal of firms having internally approved minimum guidelines on contractual features for funded reinsurance transactions. One respondent agreed that minimum guidelines can be useful but suggested changing the policy wording as they anticipate that imposing minimum expectations for each element of the contract would prevent firms from being able to negotiate a better overall proposition.

4.48 Having considered the response, the PRA has decided to maintain the wording set out in paragraph 4.12 of the final SS. It is appropriate for firms to consider the risks and set contractual mitigants to reflect the risk. The SS does not prescribe specific expectations for minimum levels, and this allows firms to set minimums that allow for the contractual features in other areas.

5: Other feedback received

Introduction

5.1 This chapter provides feedback to responses which do not directly relate to any of the individual chapters of the draft supervisory statement (SS).

Feedback to Responses

5.2 Six respondents commented that the proposals should distinguish between risky monoline counterparties and arrangements with well-established reinsurers, consider the impact of counterparties based in other jurisdictions, or focus more on offshore arrangements.

5.3 Having considered this feedback, the PRA has clarified in paragraph 1.5 of the final SS that firms may consider benefits resulting from well diversified funded reinsurance portfolios or diversification between the cedent and the counterparty’s risk profile or asset holdings into their risk management and modelling. The PRA has not made any further adjustments to the draft policy to reflect the differing nature of potential counterparties, as it considers that many of the risks arising from funded reinsurance arrangements would arise regardless of the nature of the counterparty.

5.4 Two respondents commented that it was unclear to what extent firms are expected to re-examine existing contracts or arrangements.

5.5 The expectations in the SS around risk management and SCR modelling apply to all funded reinsurance arrangements. Firms are expected to assess the risks of the arrangements they have already entered into as set out in chapter 4 of the final SS if such an analysis has not already been carried out. This will ensure that they are appropriately aware of any resulting risks for risk management and internal model purposes. The PRA clarifies that it does not expect firms to renegotiate the terms of contracts already entered into as a matter of course but firms should consider continued compliance with existing rules and requirements as well as their own risk appetite.

5.6 Two respondents recommended that the scope of the PRA’s expectations exclude intra group transactions.

5.7 Having considered the responses the PRA has not made any further adjustments to the draft policy. The PRA expects firms to assess the risks associated with intra group funded reinsurance transactions considering the design of the reinsurance arrangements, the nature of collateral assets, the group structure, and how stresses may impact the separate entities involved in the transaction.

5.8 Six respondents commented that, taken together, the expectations set out in the SS may be disproportionate when applied to all firms, that the expectations should reflect the extent to which a firm uses funded reinsurance arrangements, or that the PRA should clarify that standard principles of proportionality and materiality still apply.

5.9 The PRA agrees that proportionality and materiality are relevant considerations and has made a number of adjustments to the draft policy highlighting this in specific areas of the proposals including the expectations around collateral policies and SCR modelling in the sections above.

5.10 One respondent stated that they considered that the universal application of the expectations to all firms led them to disagree with the PRA’s conclusion in the have regards analysis relating to differences in the nature of, and objectives of, businesses carried on by different persons that the proposals adequately recognise the differences in firms’ businesses.

5.11 The PRA considers that the expectations set out in the draft policy adequately reflect the differences in firms’ businesses, with the expectations around risk management, modelling, and arrangement structuring expected to be easier to meet for firms making limited use of funded reinsurance arrangements than for firms using these arrangements extensively. The PRA also considers that the adjustments made to areas of the draft policy to clarify that firms should take proportionate approaches. The PRA considers that the proposals appropriately recognise differences in the nature of and objectives of businesses carried on by different persons.

5.12 Two respondents commented that the proposals should put greater emphasis on expectations which would reduce the likelihood of a default occurring rather than mitigating the impact of a default.

5.13 The PRA considers the focus of the proposals on mitigating the impact of a loss in the event of a default to be appropriate. The PRA also consider that the proposals as a package will help ensure that firms are appropriately considering and modelling the full range of risks associated with funded reinsurance arrangements. Having considered these responses the PRA has decided not to change the draft policy but will continue to monitor the market to assess whether further policy could assist in this area.

5.14 One respondent commented that the proposals taken as a whole may result in higher costs for firms. Four respondents commented that the PRA should consider the benefits of Bulk Purchase Annuity (BPA) deals and funded reinsurance on the pension scheme market, the potential impact of the proposals on the design of funded reinsurance arrangements, and the potential impact on the growth and capacity of the funded reinsurance and BPA markets.

5.15 The PRA has considered the potential costs to firms to meet the expectations set out in the SS and the impact of these expectations on the funded reinsurance and BPA markets carefully in the development of the policy. The PRA considers that the proposals would help ensure that firms are properly considering the risks associated with the use of funded reinsurance arrangements and directly advance the PRA’s primary objectives. The PRA also considers that while some firms may incur additional costs to bring their current practice in line with PRA expectations, these costs should not be significant to firms, and there should not be a significant impact on firms’ business models. As stated in paragraph 1.12 the PRA considers that the costs and benefits of the final policy in this PS does not significantly differ from that in the draft policy proposed in CP24/23 and, therefore, the CBA presented in the CP remains appropriate.

5.16 Four respondents noted that the PRA should allow more time for implementation of the SS. One noted an extremely challenging target. One respondent noted that a practical approach to determining a sensible ‘immediate recapture metric’, including frequency and sophistication of production and proportionate to the insight provided, will require an implementation date of 31 December 2024. Another respondent noted that implementation in 2025 was preferred to allow sufficient time for review, development and governance of internal models and risk frameworks. One respondent noted that while they consider many elements of their existing framework to be in line with the intent and principles of the SS, certain expectations (including the additional internal model expectations) could require significant development activity and an appropriate lead time to take through internal governance. The respondent therefore asked the PRA to allow firms sufficient lead time for implementation. One respondent noted that firms will find implementation by Q2 2024 extremely challenging given the need to balance business as usual activity including ongoing BPA business with the extent of change required from other regulatory changes introduced by the recent Solvency UK consultation, with a suggestion of implementation by year end 2024.

5.17 Having considered the feedback received, the PRA has decided not to change the implementation date of the SS as it believes that it is necessary to introduce the expectations quickly given the growing use of funded reinsurance by firms and necessary for the supervisors to understand firms’ compliance with existing rules and requirements. See paragraphs 1.18 to 1.20 for details of the implementation approach.