Prudential Regulation Authority Rulebook

4.1

This chapter sets out the PRA’s expectations of firms regarding the risk identification exercise and the risk calibration and validation of internal models (particularly in respect of the application of the MA within the calculation of the SCR) for illiquid assets. It includes, as an example, several expectations, specific to Income Producing Real Estate (IPRE) loans. Given the heterogeneity of illiquid assets, the PRA expects firms to consider whether these expectations are applicable to other relevant assets within their portfolios. Where firms consider the expectations are not applicable to assets with similar features to IPRE, the PRA expects them to provide, upon request, a justification of why this is the case. The PRA will seek assurance against these expectations in a proportionate way, using similar criteria to those discussed in paragraph 2.6 of this SS.

4.2

IPRE lending refers to a category of funding to real estate where the prospects for ultimate repayment and recovery on the exposure depend primarily on the cash flows generated by the underlying property asset(s).^[24] The primary source of these cash flows would usually be lease or rental payments from commercial tenants (generally for the payment of interest and any amortizing principal) and the sale or refinancing of the asset(s) (generally for the payment of any non-amortizing principal at maturity). The distinguishing characteristic of IPRE (versus other corporate exposures that are collateralised by real estate) is the strong positive correlation between the prospects for repayment of the interest and principal due on the exposure and the prospects for recovery in the event of default. Both primarily depend on the realisation of cash flows generated by a property, whether these are in the form of rental income or sale/refinancing proceeds. Note that the PRA considers this definition to be a useful reference. However, it need not be applied rigidly and the PRA expects the expectations set out below to also be relevant for assets with similar features.

4.3

The PRA’s observation is that IPRE loans are generally structured to isolate the collateral from the bankruptcy and insolvency risks of the other entities that participate in the transaction, eg via an SPV.

4.4

The MA allows firms to adjust the relevant risk-free interest rate term structure for the purpose of calculating the best estimate of a portfolio of MA-eligible insurance or reinsurance obligations. To apply an MA, firms must have an MA permission from the PRA, as per Matching Adjustment 2.1. Firms with an MA permission are permitted to apply an MA for the purposes of determining both TPs and the SCR. The PRA expects firms to have confidence that the level of MA benefit assumed in each of these calculations is fit for purpose. The PRA’s expectations relating to modelling of the MA within the SCR calculation are set out in SS8/18. These expectations primarily apply to the risks arising in respect of corporate bond assets within firms’ MA portfolios. However, the PRA recognises that many of the expectations in SS8/18 would apply regardless of the assets held (see paragraph 1.8 of SS8/18).

4.5

The PRA requires that ‘as regards investment risk, a firm must demonstrate that it complies with the Investments Part of the PRA Rulebook.’^[25] The Investments Part sets out the ‘prudent person principle’ which requires that firms ‘must only invest in assets and instruments the risks of which it can properly identify, measure, monitor, manage, control and report, and appropriately take into account in the assessment of its overall solvency needs in accordance with Conditions Governing Business 3.8(2)(a).’ Firms with an MA permission must also comply at all times with the prudent person principle requirements, as required by Matching Adjustment 2.2(6). See also the PRA’s expectations in SS1/20 (Solvency II – Prudent Person Principle).^[26]

4.6

The PRA reminds firms that the SCR must capture all quantifiable risks to which firms are exposed^[27] whether using the standard formula or internal model, and internal models used to calculate the SCR must capture all material risks that firms are exposed to.^[28]

4.7

In order to ensure that these requirements are satisfied for IPRE loans, the PRA expects a firm to complete a comprehensive risk identification exercise that considers all sources of risks that the firm could be exposed to in relation to its IPRE loans. Due to the bespoke nature of IPRE loans, the risk identification exercise should consider features of individual loans. This applies to standard formula or internal model firms that have IPRE loans.

4.8

The internal rating process (including internal rating models used to inform the FS and the MA benefit attributed to IPRE loans for the purposes of calculating TPs or for calculating the stressed FS) and the SCR should reflect the relevant risks identified in this risk identification exercise.

4.9

The PRA expects a firm to be able to demonstrate that the IPRE loan risks captured by its internal ratings process offer sufficient discriminatory power in determining the credit quality of its assets and that these risks are reflected, as appropriate, in the internal rating models.

4.10

The PRA expects internal model firms to use the risk identification exercise to influence the scope, methodology and calibration of the internal model used to calculate the SCR.

4.11

Whilst the SCR may be calibrated to cover only a subset of the risks identified in the risk identification exercise, eg where some risks have been fully mitigated by a firm, firms are expected to clearly justify and explain any exclusions of risks identified in the risk identification exercise from the SCR calibration. Firms should also allow for any secondary risks introduced through risk mitigation.

4.12

A firm using the standard formula is required as part of the Own Risk and Solvency Assessment to assess the significance of the extent to which its risk profile deviates from assumptions underlying the standard formula.^[29] The PRA expects the output of the risk identification exercise to be considered and incorporated into that assessment. In the event that the standard formula does not reflect the firm’s risk profile, the firm may need to consider whether it should use a partial internal model to calculate the SCR. For internal model firms, the internal model should be validated against the output of the risk identification exercise, noting any model limitations.

4.13

The PRA expects the risk identification exercise to be carried out by persons with the appropriate skills and experience.

4.14

The risk identification exercise should also take into account how the firm’s own or outsourced credit risk management processes may have an impact on the performance, and hence risks, of the assets.

4.15

The PRA does not expect the risk identification exercise to be a one-off exercise. Firms are expected to undertake a risk identification exercise regularly to maintain an up-to-date view of existing exposures, and to capture potential risks arising from the known pipeline of new IPRE loans. Other circumstances that may require a risk identification review include, but are not limited to, changes to the risk appetite, changes in the legal, political or regulatory landscape, a significant change to external market conditions, a change in investment mandates, or the consideration of new IPRE loans.

4.16

In the risk identification exercise, the PRA expects firms to consider all relevant systemic and idiosyncratic risks associated with their IPRE loans.

4.17

The risk identification exercise should consider features of existing individual IPRE loans and those that may be accepted in future in line with a firm’s risk appetite and tolerances set out in its underwriting policy and investment mandates. Grouping of assets by features may be acceptable but care is required to ensure no risks introduced by bespoke features are missed.

4.18

The PRA also expects firms to consider interactions between the risks identified, and how any interdependence may affect both the outcome and impact of risks crystallising. For example, a reduction in the level of rent receivable from a commercial property would increase the probability of default through a reduction of the income coverage ratio, and increase the loss given default, through a reduction in the value of the property on which the loan is secured.

4.19

The risk identification exercise should consider the following high-level areas, as a minimum:

1. (i) external market factors, taking into account property market conditions (eg supply vs demand) and wider economic risks (eg interest rates, economic growth);
2. (ii) cash flow predictability, taking into account for example the tenant(s), lease terms, voids and re-lettings;
3. (iii) collateral, taking into account the characteristics (eg location, design and condition) of the underlying property(s), property and valuation risks, and ability to realise the collateral value within a timely manner and the security package;
4. (iv) loan characteristics (eg leverage, serviceability, pre-payment risk, refinancing risk, covenants or structural protections);
5. (v) risks arising from third parties, such as the strength of sponsor and its willingness to provide support, and parties involved in the servicing and managing of the loan, SPV and/or underlying property(s);
6. (vi) concentration, basis and liquidity risks; and
7. (vii) legal, political and regulatory risks.

4.20

The PRA expects firms to demonstrate that they have appropriate skills and experience to implement the controls and risk management actions assumed in the management of IPRE loan exposures within the internal model. Firms should also demonstrate that these controls and risk management actions can be executed in the timescales assumed. Where this is not possible, firms should consider the extent of any differences between the assumed and actual effect of controls relating to the management of IPRE loan exposures, including timeliness and whether any adjustments are necessary to the internal model as a result.

4.21

The PRA expects a firm’s risk identification exercise and the assumptions underpinning its internal model to reflect the risk profile of the firm’s IPRE loans. This will be influenced by relevant policies and practices of the firm relating to IPRE lending. Such policies and practices are likely to cover the following areas:

1. (i) underwriting of loans, eg the criteria upon which the approval of any loan is based,^[30] and the impact this may have on the risks accepted on IPRE lending;
2. (ii) the due diligence process applied to a loan, including any documented standards, relating to IPRE loans that may be applied to sponsors, borrowers, contracts, collateral property and security package;
3. (iii) agreements between the firm and an internal or external IPRE loan investment manager relating to the obligations of the investment manager,^[31] eg investment mandates;
4. (iv) legal protections required through loan covenants or structural protections;
5. (v) potential conflicts of interest relating to IPRE lending, including identification and management thereof;^[32]
6. (vi) outsourcing of any functions relating to IPRE lending,^[33] where relevant;
7. (vii) ongoing administration, servicing and monitoring of IPRE lending;^[34] and
8. (viii) dealing with any distressed assets,^[35] eg through the workout process.

4.22

The PRA expects a firm’s risk identification exercise to reflect the loan underwriting policies and practices of the firm, and the impact of these on the risks that may be accepted. Firms’ considerations should cover at least the following:

1. (i) IPRE loan features that would be acceptable, such as property, tenant and borrower/sponsor features;
2. (ii) covenants or protections relating to the ongoing management of the IPRE loans that are required as a minimum; and
3. (iii) the sources of concentration risk for IPRE loans and the limits that apply.

4.23

In order to demonstrate that all material risks have been identified, the PRA expects a firm to evidence that underwriting due diligence has been carried out on sponsors, borrowers, contracts, collateral property and security package to the standard required by the firm’s risk appetite and tolerances as set out in its underwriting policy and investment mandates.

4.24

The PRA expects a firm’s risk identification exercise to consider the agreements with investment managers, such as investment mandates for IPRE loans, including any potential risks introduced by discretion available to investment managers. A firm’s risk identification should also consider the potential for the investment manager to act against the firm’s interests.

4.25

The PRA expects the risk identification exercise to consider the impact of the protections provided in the loan agreement by the security package and terms and conditions (eg covenants and structural protections) on the risks accepted on IPRE loans. Any gaps in the protection provided should be captured in the risk identification exercise. In addition, the PRA expects firms to consider any difficulties that may arise in enforcing the legal agreements.

4.26

The PRA expects firms to demonstrate that the IPRE loan risk identification exercise takes account of potential third-party actions and conflicts of interest that may impact the IPRE loan risk profile, and the process around managing any such conflicts. The PRA does not expect that the interests of all parties in an IPRE loan transaction will be fully aligned at all times and firms should therefore consider scenarios when there is likely to be a lack of alignment in determining the risks to which they are exposed via their IPRE loans.

4.27

Where any key functions are outsourced, the PRA expects firms to demonstrate that the outsource providers are able to identify, mitigate and manage any conflicts of interest and report them to the firm.

4.28

The risk identification exercise is expected to consider the processes and policies covering the ongoing maintenance of IPRE loan exposures, including the firm’s capabilities in the management of distressed assets eg the firm’s workout function, which may affect the level of recoveries achievable, and the timeliness of this. The functionality of these processes and policies is expected to be considered in economic downturn scenarios, where multiple loans could become distressed at the same time.

4.29

Firms may choose to design IPRE loan contracts that include options that may be triggered in the event of a risk crystallising, such as technical defaults.^[36] If options are available to a firm in the event of a technical default on IPRE lending, the risk identification exercise is expected to consider the impact and reasonableness of exercising any such option upon technical default.

4.30

For internal models, the PRA does not have a preference on the approach taken by firms to model the risks on IPRE loans, subject to the chosen approach meeting the relevant calibration standards and internal model requirements. For example, firms may choose to model a proxy for the IPRE loan exposures, or to model the underlying IPRE loan risk drivers directly. In deciding an appropriate approach to take, however, the PRA expects firms to consider whether the robustness of the modelling approach is commensurate with the materiality of IPRE loans held, and to clearly explain how the IPRE loan risks (ie arising from the risk identification exercise in the previous section on Risk Identification) have been adequately captured by the model. In addition, a firm should be able to articulate the link between its modelling approach and its internal credit rating approach, and to demonstrate that the approach taken meets the requirements of the Use Test.^[37]

4.31

The PRA is aware that firms may have more limited data for IPRE loans than for other types of asset that are traded more frequently. Consequently, the PRA expects that the model methodology and calibration will make greater use of expert judgement and a qualitative assessment of IPRE loan characteristics. These judgements should be based on the expertise of persons with the appropriate skills and experience. Firms should assess the credibility of expert judgements made in calibrating extreme scenarios where data is limited, and the materiality of these judgements. These judgements should reflect the level of uncertainty within the data in order to demonstrate that firms’ allowance for IPRE loan risks is appropriately calibrated. Firms should ensure that these have gone through appropriate governance, communication, documentation and validation in line with Chapter 4 (Assumption setting and expert judgement) of the EIOPA Guidelines on the use of internal models.^[38]

4.32

The PRA expects firms to re-value IPRE loans in stress, in a manner consistent with the valuation methodology applied to determine the asset value reported in the regulatory balance sheet. However, some modifications may be required. For example, to ensure that the output of the internal model does not include a material model error^[39], adjustments for valuation uncertainty may be required to complete the valuation in stress for the purposes of determining the SCR. In this case, firms should demonstrate that the resulting methodology produces a valuation that is no higher than the valuation based on the valuation methodology without any modifications including any allowance for valuation uncertainty or model error.

4.33

For the purposes of determining the SCR, firms may choose to determine an implied IPRE loan transition matrix by stressing inputs into the internal credit rating model. Such firms should consider the risks that may affect the IPRE loan cash flows in this assessment, including those noted in paragraph 4.19 above.

4.34

The PRA expects a firm to assess the shape (ie distribution by rating) of the IPRE loan transition matrix (where this is used or defined by the firm), if relevant, and validate the appropriateness of the output of the chosen methodology. In applying the IPRE loan transition matrix, firms should consider the binary effect of credit transitions on individual IPRE loans and the impact of name level concentration risks.

4.35

The PRA considers that an assumption that a firm can exercise an option triggered by a technical default (eg option to accelerate loan repayment upon covenant breach) in its internal model constitutes a future management action within the internal model. The PRA therefore expects firms to demonstrate how these assumptions meet the requirements set out in Article 236 of the Commission Delegated Regulation (EU) 2015/35. In particular, if firms assume such actions in their SCR calculations then they should allow for both the benefit and associated costs of the actions within the SCR. Firms should also consider the consistency of assumptions made in respect of these actions in valuing assets and TPs pre-stress and post-stress.

4.36

The PRA expects firms to identify the risks pertaining to their IPRE loans that would be retained in stress, and ensure that these are appropriately reflected in the calculation of the FS in stressed conditions where the MA portfolio includes IPRE loans. When calculating the stressed FS on IPRE loans, firms should consider the risks noted in SS8/18. Specific additional considerations relevant to IPRE loans are set out in this section.

4.37

The PRA expects firms to include all relevant retained credit risks within the calculation of the stressed FS for the purposes of determining the SCR. For IPRE loans, this should draw on the risk identification exercise and it is expected that firms consider at least the following areas:

1. (i) cash flow predictability;
2. (ii) loan characteristics, eg refinancing risk;
3. (iii) concentration risk, which may be more material than in the case of corporate bond holdings; this should also consider the impact of future risks such as political and climate change risks on a concentrated portfolio;
4. (iv) basis risk, which as a consequence of the heterogeneity of these assets, may be more material than in the case of corporate bond holdings;
5. (v) liquidity risk;^[40] and
6. (vi) idiosyncratic risks.

4.38

Within the calculation of the stressed FS on IPRE loans, the PRA also expects firms to consider risks that may impact the value of collateral and security underpinning the loan and therefore the potential recoveries that may be achieved upon default. Of primary importance is property risk and the ability to realise recoveries within a timely manner. Particular components of property risk expected to be included within the stressed FS calculation include (note the below are not necessarily mutually inclusive):

1. (i) stressed property values and valuation risks;
2. (ii) characteristics of the property, eg location, design and condition;
3. (iii) property market conditions, eg supply vs demand for the properties; and
4. (iv) the ability to sell or refinance the underlying property in stress, the time required to complete the sale, and potential haircuts to value to achieve a sale in a timely manner.

4.39

In order to allow for these risks, the PRA would ordinarily expect that a firm would not assume a zero loss given default (LGD) on IPRE loans, if the modelling of LGD is applicable to the firm’s methodology.

4.40

When modelling changes to the FS within the calculation of the SCR, firms should consider the rate of recovery against the collateral and security upon default that is achievable in order to restore compliance with the relevant MA eligibility conditions as set out in the IRPR regulations and the Matching Adjustment Part of the PRA Rulebook. A firm has a two-month window in which to restore compliance with the MA eligibility conditions before its MA (in the absence of any other action) will be reduced as per Matching Adjustment 13.5. The assessment of recovery rate should also allow for the firm’s assumptions about its workout process, eg the assumed recovery amounts in stress, costs of recoveries and timing of recoveries all taking into account the illiquid nature of IPRE lending.

4.41

Within the stressed FS calculation for IPRE loans, firms should consider the impact of a default (eg failure of the borrower to meet interest payments or full repayment at maturity as a minimum).

4.42

The PRA expects the stressed FS calibration to capture the risks retained by a firm, including both the risk of a default event given each asset’s current credit quality and the risk of an increase in the likelihood of a default event (ie downgrade in credit quality), as per the firm’s definition of a default event.

4.43

Firms should continue to compare the levels of downgrades and defaults in their internal models against those seen historically for other relevant credit risky assets. In the case of IPRE loans, firms should specifically look at historic periods of poor commercial property experience. Firms should be able to justify any assumptions that appear materially weak compared to historic experience.

4.44

Firms may alternatively choose to assign stressed FS assumptions for IPRE loans by, for example, applying the stresses applied for corporate bonds of the same duration, rating and sector. The PRA expects, however, firms to justify why this approach is appropriate and how it reflects all of the relevant risks for IPRE loans. In particular, a firm should justify why the risks included in its corporate bond methodology and calibration adequately allow for the risks identified in the IPRE loan risk identification exercise, and that the corresponding SCR is appropriate for IPRE loan exposures.

4.45

Firms may choose to assign stressed FS assumptions for IPRE loans using a stressed credit rating assessment, either directly or indirectly. In this case, the PRA expects firms to consider how the credit rating methodology, including any expert judgements, would apply in practice under stressed conditions. This should include considering the key quantitative and qualitative factors that drive the credit rating.

4.46

The PRA expects firms to maintain a floor (ie a minimum level of FS) based on long term observations of the risk on IPRE lending as part of the modelling of stressed FS in line with expectations set out in paragraph 4.30 of SS8/18. As a minimum, the PRA expects firms to reapply the methodology and calibration for the floor as set out within regulations 6(4) to 6(8) of the IRPR regulations and Matching Adjustment 4.11 to 4.15. Changes to the FS in stress should also include any changes to additions made to the FS used to calculate the TPs, as set out in paragraphs 2.2 to 2.4 of SS8/18.

4.47

For the purposes of determining the TPs, the calibrations for the basic FS used in the MA calculation are published by the PRA in technical information produced in accordance with regulation 3 of the IRPR regulations, and FS additions may further apply. If the stressed FS for an IPRE loan has been derived based on an economic view that results in a different FS at the 50th percentile compared to the FS used in the TP calculations, the PRA expects a firm to consider the reasons for this difference, the materiality and the implications of this for the calculation of the SCR. The SCR should reflect the increase in FS attributed to the risks retained in stress.

4.48

Considerations relevant to rebalancing of the MA portfolio within the SCR calculation are set out in paragraphs 5.5 to 5.14 of SS8/18. Specific considerations relating to rebalancing of IPRE loans are set out in this section.

4.49

Firms may choose to assume that rebalancing of the MA portfolio may be achieved by injecting existing IPRE loans from elsewhere in the business outside of the MA portfolio. In this case, the PRA expects firms to demonstrate that such assets are MA-eligible and have the same features as the assets already in the MA portfolio. The PRA further expects that such an assessment of eligibility and same features may constitute a material exercise, due to the relatively complex and bespoke nature of IPRE loans.

4.50

The PRA would not usually expect firms to assume that IPRE loans may be sold to fund the purchase of assets required to rebalance the MA portfolio in stress, owing to the potential timescale required to achieve such a sale. If a firm is reliant on such an assumption then the PRA expects that strong evidence would be provided to show that the firm has considered, at least:

1. (i) whether a sale would be allowable under the original terms of the loan;
2. (ii) the likely counterparties to the sale and the impact that the stress event may have on the appetite to engage;
3. (iii) the likely timescale in which a sale or sales could be achieved, substantiated by market data if possible; and
4. (iv) the value at which a sale could be achieved, including any allowance for haircuts, substantiated by market data if possible.

4.51

The PRA expects firms would not place reliance on the ability to source new IPRE loans to address a mismatch in the MA portfolio in stress due to the uncertainty around the pipeline and market conditions in stressed conditions.

4.52

The PRA expects a firm’s validation team to carefully consider the challenges of developing the methodology and calibration of IPRE loans based on limited data, including the need to robustly validate the use of proxy datasets (eg Commercial Mortgage-Backed Securities data) and expert judgements. The validation should be mindful of how the firm will demonstrate that the capital requirements resulting from the internal model are appropriate given the potentially greater level of uncertainty of the SCR calculated by the model.^[41] In particular, the validation should review all the key expert judgements made in the calibration process and the range of alternative judgements available, and quantify the impact of changes in these judgements.

4.53

The PRA expects firms with IPRE loans within an MA portfolio to consider, as part of the validation of the stressed MA allowed for within the SCR, a comparison of the stressed FS on IPRE loans with a mechanistic reapplication of the methodology used to assign the FS for the purposes of calculating the TPs. This assessment should consider how the credit quality of the IPRE loan could change in stress. As set out in paragraph 2.5 of SS8/18, the PRA considers that a ‘mechanistic approach’ based on re-application of the approach used to calculate TPs is unlikely to take into account all quantifiable risks to which a firm is exposed. The PRA would therefore expect a firm’s stressed FS to exceed that implied by a mechanistic re-application of the approach used to derive the FS used to calculate the TPs.

4.54

The PRA expects firms to ensure that the MA qualifying conditions in stress can be met in order to support the level of MA on IPRE loans allowed for in the SCR calculation given the risks to which the portfolio is exposed and the interaction between these risks. This expectation also applies in the case of firms using a less bespoke modelling approach to model risks on IPRE loans.

4.55

In validating the appropriateness of the calibration, the PRA expects a firm should conduct back-testing of its calibration for IPRE lending against its own loss experience and appropriate historical data to the extent that such data are reasonably available.

4.56

The PRA recognises that industry benchmarking surveys comparing the calibration and treatment of IPRE loans by firms may be a useful validation tool. However, a firm should consider the extent to which such comparisons are affected by differences in the risk profile of its holdings due to the bespoke nature of IPRE loans, and differences in the materiality of these holdings which may justify the use by some firms of less sophisticated models. Firms should therefore not place material reliance on such benchmarking unless they are sure that comparisons are made on a like-for-like basis.