Strategic Report Governance
Financial Statements
131
Key audit matter
How our audit addressed the key audit matter
Valuation of insurance contract liabilities – Annuitant mortality assumptions (Group) Refer to Group Audit Committee Report, Accounting policy 1.7 Insurance contracts and note 22 Insurance contracts and related reinsurance. Annuitant mortality assumptions are an area of significant management judgement due to the inherent uncertainty involved. Annuity liabilities are sensitive to the choice of best estimate annuitant mortality assumptions due to the large volume of annuity business. The best estimate annuitant mortality assumption has two main components: 1) Base mortality assumptions This part of the assumption is mainly driven by internal experience analyses, but judgement is also required. For example, in determining the most appropriate granularity at which to carry out the analysis; the time window used for historic experience, or whether data should be excluded from the analysis; and in selecting an appropriate industry mortality table to which management overlays the results of the experience analysis. 2) Rate of future mortality improvements This part of the assumption is more subjective given the lack of data and the uncertainty over how life expectancy will change in the future. The allowance for future mortality improvements is inherently subjective, as improvements develop over long timescales and cannot be captured by analysis of internal experience data. The extent to which mortality rates may remain elevated in future, as a result of COVID-19 and other trends in the UK, is subject to considerable uncertainty. Judgement is required in estimating the allowance for expected high future mortality rates in the long term. The Continuous Mortality Investigation Bureau provides mortality projection models which are widely used throughout the industry and contain a standard core set of assumptions calculated by the CMIB based on the most recent available population data. Risk adjustment for longevity risk In addition, under IFRS 17, an allowance for risk in excess of the best estimate and representing the view of compensation for non-financial risk that management require is held (known as the risk adjustment). The primary component of the risk adjustment is annuity mortality risk and the selection of the distribution and associated stresses is a matter of judgement.
We performed the following audit procedures to test the annuitant mortality assumptions (including base mortality assumptions, rate of future mortality improvements and the risk adjustment): • Tested the design and operating effectiveness of controls in place over the performance of annuitant mortality experience analysis studies, approval of the proposed assumptions and implementation within the actuarial model; • Tested the reasonableness of the methodology used to perform the annual experience studies. This involves the assessment of key judgements with reference to relevant rules, actuarial guidance and by applying industry knowledge and experience; • Tested completeness and accuracy of experience analysis data. For a sample, agreed experience analysis data used to source documentation; • Assessed the appropriateness of expert judgments used in the development of the mortality improvement assumptions, including the selection and parameterisation of the CMI model (e.g. the choice of the smoothing parameter, initial rate, long term rate and tapering at older ages); • Assessed management’s considerations in relation to the long term impacts of COVID-19 and other trends in the UK on results of experience analysis and in the allowance made for potential changes in current and future expected rates of annuitant mortality; • Assessed management’s risk adjustment methodology relative to the compensation required by management for non-financial risk, including the selected confidence level and calibration, and the impact of Partnership Life Assurance Company Limited (PLACL) being on the PRA-approved Solvency II internal model, as well as testing management’s controls over the processes; and • Compared the annuitant mortality assumptions selected by management against those adopted by peers using our annual benchmarking survey of the market (to the extent available). Based on the work performed and the evidence obtained, we consider the assumptions used for annuitant mortality to be appropriate.
Powered by FlippingBook