New Affluence Index helps governments and pension and annuity providers to improve forecasts of mortality rates
Research could have wide-ranging implications for governments and private-sector pension and annuity providers
Academics at Cass Business School, City University London, and Heriot-Watt, Aarhus and Durham Universities have developed a new way to model socio-economic differences in mortality rates – using an affluence index.
The work by Professor David Blake, Professor Andrew Cairns, Professor Kevin Dowd, Dr Malene Kallestrup-Lamb and Carsten Rosenskjold could have wide-ranging implications for governments and private-sector pension and annuity providers – as well as institutions buying out pension liabilities in the new longevity risk transfer market.
The research, Modelling Socio-Economic Differences in the Mortality of Danish Males Using a New Affluence Index, can be applied to any country with reliable measures of income and wealth at high ages, including the UK. Denmark has particularly high-quality data on the Statistics Denmark national register.
The researchers:
- Investigated and modelled how the mortality of Danish males aged 55-94 has changed over the period 1985-2012.
- Divided the population into ten socio-economic sub-groups using the new measure combining wealth and income that they call the affluence index.
The data showed that individuals could switch annually between the adjacent subgroups, but the researchers found that the best modelling results arose if individuals were locked-down into a particular sub-group at retirement age until death. This finely granular dataset was then used to project future death rates. The improved forecasts would be very useful for both policy makers planning national pension budgets and private-sector providers of financial products.
Professor Blake said: “The affluence index, in combination with sub-group lockdown at age 67, is shown to provide consistent sub-group rankings based on crude death rates across all ages and overall years in a way that improves significantly on previous studies that have focused on life expectancy. The gap between the most and least affluent is confirmed to be widest at younger ages and has widened over time.”
Professor Cairns said: “We introduce a new multi-population mortality model that fits the historical mortality data very well and generates smoothed death rates that can be used to model a larger number of smaller sub-groups than has been previously possible without losing the essential character of the raw data. The model produces bio-demographically reasonable forecasts of mortality rates that preserve the sub-group rankings at all ages.”