The Continuous Mortality Investigation (CMI) Self-administered Pension Schemes (SAPS) Committee has published the ‘S4’ Series pension scheme mortality tables.
The Continuous Mortality Investigation (CMI) Self-administered Pension Schemes (SAPS) Committee has published the ‘S4’ Series pension scheme mortality tables.
This blog summarises key features of the tables. CMI subscribers can find more detail in Working Paper 185, which accompanies the final tables and Working Paper 181, which describes how the tables are constructed.
Actuaries need assumptions about mortality rates to manage the finances of defined benefit pension schemes that commit to pay members a pension throughout their retirement.
The S4 tables provide information on the recent mortality of such pension schemes in the UK, which tends to be lower than for the general population. It also sets out how mortality in these schemes differs by age, sex, retirement type, pension amount, and socio-economic status.
This is a useful starting point when setting assumptions, although actuaries will typically also consider the mortality experience and/or socio-economic status of a specific pension scheme to adjust the rates.
We collect data for self-administered defined benefit pension schemes (that is, not those managed by an insurer) with more than 500 pensioners. This is provided by actuarial consultancies, the Pension Protection Fund (PPF), and the Government Actuary’s Department (GAD) and covers both private sector and public sector schemes.
The S4 tables are based on data for 2014 to 2019, so they exclude the impact of the COVID-19 pandemic. We expect that users will tend to allow for the pandemic through their mortality improvement assumptions. We comment on pension scheme mortality experience over 2020 to 2021 in Working Paper 169 and plan to update this to 2022 in our next working paper.
The S4 Series tables have 42 tables, each relating to a subset of the total dataset. The tables are defined by combinations of the following, although not every possible combination is included:
We have included 12 new tables that vary by IMD in the S4 Series – we did not include any tables that varied by IMD in the previous S3 Series, as we only started to collect this data more recently. We found that allowing for a combination of IMD and pension amount made it possible to distinguish groups of members with a wider range of mortality rates than when only using pension amount. This is particularly true for females.
The other 30 tables do have equivalents in the S3 Series. However, the S4 tables are based on updated data, and some of our methods for producing the tables have changed. Anyone moving from the S3 tables to S4 tables should consider these impacts before using a new table.
In general, life expectancies for S4 tables tend to show a small fall compared to the equivalent S3 table projected to a consistent date using the CMI Mortality Projections Models. Further details and analysis are set out in the working paper.
We continue to collect pension scheme data and publish annual analysis of mortality experience. The next, for 2015 to 2022, is expected in late March 2024.
In normal conditions, we would typically expect an interval of around 5 years between mortality tables, but this is complicated by the pandemic. We will review emerging mortality experience and form a view on when, and how, to produce the ‘S5’ Series.
