Appendix 1 : PRA Matching Tests

In previous communications with firms, the PRA has described other versions of these tests. The tests described below are the most recent versions.

Test 1: Accumulated Cash Flow Shortfall Test

A description of this test is as follows. Firms should:

  • project best estimate liability cash flows in an MA portfolio at annual (or more frequent) intervals;
  • project cash flows from assets in component A, after being adjusted for that part of the FS that corresponds to the PD, at annual (or more frequent) intervals;
  • calculate any cash flow surpluses and shortfalls arising in each time interval and accumulate them at the risk-free rate;
  • note the highest accumulated shortfall from all future time intervals in the projection; and
  • calculate the present value of liabilities in an MA portfolio (at the valuation date) discounted at the risk-free rate.

The frequency of the time intervals used for the cash flows in this calculation should be consistent with the method the firm uses to conduct its matching.

Threshold rate: the maximum accumulated shortfall in any time interval of the projection should not exceed 3% of the present value of liabilities.

Firms should carry out this this test on a regular basis (monthly if they are writing new business in the fund and quarterly otherwise).

Test 2: 99.5th Percentile Value at Risk (VaR) Test

A description of this test is as follows:

  • firms should calculate the 99.5th percentile 1-year VaR of an MA portfolio for each of the following risks: interest rate, inflation and currency. For assets with HP cash flows, the calculation for each risk should be conducted using stressed cash flows that are consistent with the scenario being modelled;
  • the calculations should consider the change in the value of both the assets and the liabilities within the portfolio as a result of each stress;
  • the PRA expects firms to calculate undiversified capital requirements corresponding to a confidence level of 99.5% over a 1-year period for each of the risks specified in the first bullet point above. Where firms split a risk into components (such as might be the case for interest rate and currency risk), the PRA asks firms to aggregate these components into a single capital number for that risk, and to explain the approach adopted in determining this single number;
  • the PRA expects firms to determine the best estimate liabilities of an MA portfolio, calculated by discounting at a rate equal to the relevant basic risk-free interest rate plus the MA;
  • firms should then compute six statistics: the undiversified 99.5th percentile 1-year VaR capital requirement for an MA portfolio for each of interest rate, inflation and currency risks, and the result of dividing each of these capital requirements by the best estimate liabilities of that MA portfolio; and
  • for the purposes of this calculation, the assets to be included are those hypothecated to components A and B, ie those that are required to cover the best estimate value of the liabilities.

Threshold rate: the undiversified 99.5th percentile 1-year VaR capital requirement should not exceed 1% of the firm’s calculated best estimate liabilities for any of the three risks.

Firms should carry out this test on a regular basis (at least quarterly in line with SCR calculations).

Test 3: Notional Swap Test

The aim of this test is to establish by how much the MA would change if the firm were able to eliminate any surplus or shortfall in its net (asset less liability) cash flows by investing in a ‘notional swap’ that simulated a perfectly matched position.

Firms are asked to set out:

  • the notional MA calculated by using the actual assets hypothecated to component A only (ie firms should state the amount of MA in bps);
  • the notional MA calculated by scaling the market value and cash flows (after being adjusted for that part of the FS that corresponds to the PD) of the assets in component A either up or down by a single factor until the present value of the future surpluses and shortfalls is zero when discounted at the basic risk-free interest rate (also referred to as the ‘notional swap approach’); and
  • the market value of the assets in component A after they have been scaled in accordance with the above.

The frequency of the time intervals used for the cash flows in this calculation should be consistent with the method the firm uses to conduct its matching.

Threshold rate: there is no specific hurdle rate set for this test but the PRA would expect firms to explain where the scaling factor as calculated above showed a ratio above 100% or below 99%.

Firms should carry out this test on a regular basis (at least quarterly in line with SCR calculations).

Test 4: MA Loss Test for assets with HP cash flows

The aim of this test is to establish by how much the MA would change if the cash flows on assets with HP cash flows were to be received in a manner that minimises the MA benefit that may be earned.

A description of this test is as follows. Firms should:

  • for each asset with HP cash flows, determine the cash flow profile, consistent with the contractual terms, that results in the lowest possible MA benefit;
  • (optionally:) where the cash flows are now expected to be received earlier than in the base case, assume that the expected proceeds are reinvested for the balance of the original term in assets with the same FS sector and credit quality at a prudent reinvestment spread above the risk-free rate, less the FS the replacement assets would incur;
  • sum across the portfolio the potential loss of MA benefit; and
  • divide the total potential loss in MA benefit by the MA benefit being claimed on the entire MA portfolio.

Threshold rate: the maximum loss in MA benefit should not exceed 5% of the MA benefit being claimed.

Firms with assets with HP cash flows should carry out this this test on a regular basis (monthly if they are writing new business in the fund and quarterly otherwise).

Test 5: Modified Accumulated Cash Flow Shortfall Test

The aim of this test is to establish the increase in the extent to which firms may be forced sellers of assets to meet liability cash flows where HP cash flows are received later than expected, or are of a lower amount than expected.

A description of this test is as follows. Firms should:

  • project best estimate liability cash flows in an MA portfolio;
  • project cash flows from assets in component A, after being adjusted for that part of the FS that corresponds to the PD;
  • for assets with HP cash flows, assume that the cash flows are extended to the latest date possible under the contract, taking credit for any coupons (including coupon step-ups) that arise from the extension;
  • calculate any cash flow surpluses and shortfalls arising and accumulate them at the risk-free rate;
  • note the highest accumulated shortfall from all future periods in the projection; and
  • calculate the present value of liabilities in the MA portfolio (at the valuation date) discounted at the risk-free rate.

The frequency of the time intervals used for the cash flows in this calculation should be consistent with the method the firm uses to conduct its matching.

Threshold rate: the maximum accumulated shortfall in any period of the projection should not exceed 5% of the present value of liabilities.

Firms with assets with HP cash flows should carry out this this test on a regular basis (monthly if they are writing new business in the fund and quarterly otherwise).

Reporting of PRA Matching Test results

Where relevant, the results, expressed as a percentage, of PRA Matching Tests 1–5 as set out above, should be reported in the MALIR (see paragraph 8.1A of this SS). Results should be as at the effective date of 31 December each year and shown on a portfolio basis as far as possible. Where the PRA Matching Tests are performed at an aggregate level (ie firm level) rather than a portfolio level they should be reported in the MALIR for the largest MA portfolio (by market value). If a test threshold has not been met then an explanation should be provided in the appropriate section of the MALIR.