4

Best estimate cash flows and matching

4.1

[Deleted]

Best estimate cash flows for assets with HP cash flows

4.1A

The PRA expects firms to model a projection of the best estimate asset cash flows to assess the quality of matching and to calculate the MA. For assets with HP cash flows, the PRA notes that firms may need to make a number of additional assumptions in order to determine the best estimate cash flows.

4.1B

The PRA also notes that the cash flows used to calculate the BEL are determined using a probability-weighted methodology. For consistency, the PRA expects that such an approach should be the default methodology for the matching assets. However, given the scarcity of data in some instances and the size of some holdings, this may not be practical or proportionate for the entirety of the assets within firms’ MA portfolios.

4.1C

In deciding on a methodology, the PRA considers that firms may want to draw a distinction between assets exposed to economic variability and assets exposed to ‘event’ (or non-economic) variability.

4.1D

Examples of economic variability include:

  1. i. optionality over redemption dates (eg callable bonds); and
  2. ii. amount variability where the amounts are expected to (but may not) change in line with an index.

4.1E

Examples of non-economic variability include:

  1. i. event-driven variability (eg pre-payment on construction failure); and
  2. ii. amount variability that is dependent on meeting operational targets.

4.1F

For assets exposed to economic variability, there may be sufficient relevant and credible data that shows how the payment profile is likely to vary under different economic conditions. For assets exposed to event risk there may be very limited data and hence significantly more expert judgement is likely to be necessary.

4.1G

Regardless of the approach taken, the PRA expects firms to:

  • maximise the use of relevant observable data;
  • assume that their counterparties are economically rational (where justifiable, the PRA considers that some allowance for frictional constraints, such as operational or reputational considerations, may be made); and
  • consider the size and materiality of the exposure when selecting a methodology.

4.1H

The PRA considers that for smaller exposures to callable bonds, it may be reasonable for firms to adopt a ‘yield to worst’ methodology. However, as the exposure increases, the PRA considers that this approach is less likely to be appropriate, and as such would expect firms to consider using a probability-weighted approach that models the risk of changes to the call date.

4.1I

For assets with event-driven variability, uncertainty in event estimation may mean that it is not practical to derive a probability-weighted estimate of future asset cash flows. In these circumstances firms may adopt a deterministic approach where the cash flows represent the firm’s median best estimate outcome.

4.1J

If a deterministic approach is taken, the PRA expects firms to be able to set out and justify:

  • the limitations of the approach and hence whether any further increase to the FS addition is required;
  • any expert judgements underlying the cash flow profile;
  • the materiality of, and triggers to reassess, the expert judgements, including any potential correlations with other assets and the wider economic environment; and
  • how frequently the cash flows will be reassessed.

4.1K

The PRA requires the same level of rigour over expert judgements in the asset projection as elsewhere in the Solvency II balance sheet (Matching Adjustment 5.4(3)). This could include monitoring experience over time to demonstrate that the estimation process is not biased.

4.1L

The PRA requires firms to maximise the use of market data consistent with the economics of the asset (Matching Adjustment 5.4(2)). This could include using market measures of expected economic variables and their volatility rather than historical information and ensuring that the present value of deferred possession of property is less than the value of immediate possession.

4.1M

Irrespective of the methodology, the PRA would expect that in most circumstances the cash flow profile would be consistent with that used for fair valuation of the assets under International Financial Reporting Standards. Firms should be able to justify any deviations from this.

Demonstration of matching

4.2

When demonstrating that (as required in regulation 4(7) of the IRPR regulations) the expected cash flows of the assigned portfolio of assets replicate each of the expected cash flows of the portfolio of insurance or reinsurance obligations in the same currency, firms should carry out a quantitative cash flow-based projection assessing the extent of any cash flow surplus or deficit arising in each future period.

4.3

When demonstrating that (as required in regulation 4(8) of the IRPR regulations) any mismatch between the expected cash flows does not give rise to risks that are material in relation to the risks inherent in the insurance or reinsurance business to which the MA is applied, the PRA expects firms to undertake a quantitative assessment of the interest rate, currency exchange rate, inflation rate or other relevant risks that arise as a result of any cash flow mismatch and an assessment of the materiality of these risks when compared to the risks of an MA portfolio as a whole.

4.3A

Where a firm invests in assets with HP cash flows, the PRA expects that the firm should assess, and be able to demonstrate compliance with, the requirement of regulation 4(9)(a)(i) of the IRPR regulations that the risks to the quality of matching are not material.

4.3B

For assets with HP cash flows, firms should quantitatively assess the extent of any cash flow mismatch that could arise from changes to the expected payment amounts and/or the timing of those payments.

4.3C

Where such mismatches arise, the PRA expects that this will crystalise as either reinvestment risk or liquidity risk. Firms should, in their assessment of the materiality of these risks, consider the consequential impacts on their liquidity plans and MA management policies.

4.4

The PRA recognises that some firms’ liabilities may be significantly longer-dated than the assets generally available to match them, or can increase in line with an inflation index for which there are currently no specific matching assets available. In such cases, the PRA expects firms to be able to provide evidence to justify how these liabilities are matched in accordance with the requirements in regulations 4(7) and 4(8) of the IRPR regulations.

4.5

For the purpose of assessing the overall level of matching, one possible method is to split the relevant portfolio of assets into the following components:

  • component A – assets where cash flows replicate the expected liability cash flows after being adjusted for the component of the FS that corresponds to the probability of default (PD) (taking account of differences in credit quality by rating notch if possible and appropriate to do so);
  • component B – additional assets that, when added to component A, result in the value of components A and B combined being equal to the BEL within an MA portfolio (when discounted at the risk-free rate plus MA); and
  • component C – further assets that are deemed ‘surplus’ for the purpose of meeting the best estimate liabilities, but that may or may not still be needed to demonstrate compliance with the other MA eligibility conditions.

4.6

To assist the PRA to take a consistent approach to assessing whether any mismatch gives rise to risks that are material in relation to the risks inherent in the insurance business to which the MA is intended to be applied, or (in relation to assets with HP cash flows) where the risks to the quality of matching are not material, firms should be able to provide cash flow and statistical information for each MA portfolio, in the form of specified ‘tests’ (‘PRA Matching Tests’) (see Appendix 1 of this SS for the tests).

4.6A

The PRA expects all firms with MA portfolios to apply PRA Matching Tests 1, 2 and 3. Firms holding assets with HP cash flows in their MA portfolios are also expected to apply PRA Matching Tests 4 and 5.

4.7

The PRA Matching Tests seek to assess:

  • the extent to which firms may be forced sellers of assets to meet liability cash flows;
  • the materiality of any mismatch in relation to interest rate, currency or inflation risks;
  • whether firms are materially under-matched;
  • the impact on the MA if cash flows are received in a manner that reduces the MA benefit that may be earned; and
  • the increase in the extent to which firms may be forced sellers of assets to meet liability cash flows, where cash flows are received later than expected, or are a lower amount than expected.

4.8

The PRA has also calibrated a set of indicative thresholds for each PRA Matching Test, which is aimed at identifying material mismatches. The PRA expects firms to monitor compliance against the thresholds on a regular basis. Where a firm does not fall within the threshold in any one of the tests, it should notify the PRA immediately. In this case, the PRA would expect the firm to demonstrate how it will restore compliance with the MA eligibility conditions, in particular regulations 4(7) and 4(8) of the IRPR regulations.

4.9

[Deleted]

4.10

The PRA also expects firms to be able to explain how they have treated each asset type (including reinsurance assets and derivatives) within the PRA Matching Tests and in particular what reinvestment assumptions they have made (if any) in the cash flows presented. However, for the purposes of projecting future cash flows to assess cash flow matching, the PRA expects firms:

  • not to assume any future management actions. This includes items such as entering into derivative contracts at some future point in time or selling assets to meet cash flow eligibility conditions;
  • for assets other than those considered to have HP cash flows, to assume that all asset cash flows arrive on their contractual date - any surplus assets cannot be assumed to be reinvested and realised at a future date. This implies that, where cash is used to demonstrate matching, the cash balance should be assumed to be realised in full in year 1 of the cash flow projection; and
  • for assets with HP cash flows, to use the same best estimate projection as used in the MA calculation.

4.10A

For assets with HP cash flows, firms may optionally include a reinvestment spread above the risk-free rate in both the PRA Matching Test 4 result and the methodology for determining the FS addition. Any reinvestment spread above the risk-free rate should be limited to that used for determining the adequacy of modified Spens clauses, as set out in paragraph 2.39 of this SS, less the FS the replacement assets would incur.

4.10B

The PRA recognises that under PRA Matching Test 4, which involves assessing the lowest MA benefit for each asset with HP cash flows using a cash flow profile permitted under the contractual terms, the cash flow profile that results in the minimum MA benefit may not be the cash flow profile that results in the greatest level of reinvestment risk. In such cases, firms should consider whether a further assessment of the quality of matching is required.

4.11

The PRA expects firms to carry out the PRA Matching Tests on a ‘net of reinsurance’ basis for all applicable tests (including both the numerator and denominator) and to consider separately the extent to which an MA portfolio’s reinsurance assets and liabilities are appropriately matched.

4.12

Where assets are grouped or paired, as referred to in paragraphs 2.8 to 2.11 of this SS, firms should be able to explain:

  • how cash flows from the component A hedging assets are treated in the assessment of matching, particularly in relation to PRA Matching Test 1;
  • whether the cash flows of the underlying asset(s) in a pairing or grouping have been hedged based on their contractual cash flows or expected cash flows. If the latter, firms should be able to explain what they are taking as ‘expected’ cash flows: for example, cash flows that have been de-risked for the default component of the FS; and
  • how the paired or grouped assets have been mapped to FSs, and in particular whether the mapping is done for the combined asset or individually. For example, a floating rate note (FRN) or interest rate swap pair could be mapped as one fixed cash flow asset, or the FRN and the swap could be mapped individually, with different FSs then potentially applying to each part.

4.13

The PRA expects that defaulted assets should not be used to match liabilities within component A. Given the uncertainty around potential recovery value, it may also not be appropriate for such assets to be held in component B. The exact treatment of any defaulted asset will depend on the type and severity of the default event; for example default triggered by the failure of the borrower to meet its contractual payments to the lender(s) could be treated differently to a technical event of default where payments are still expected to be made in future. With that in mind, the PRA expects firms to develop their own definitions of default together with the associated consequences of different types of default event occurring in practice, including implications for the MA portfolio.