Trends in the UK mortality experience

Student thesis: Master's ThesisMaster of Philosophy


The increase in life expectancy has profound implications for the pensions and life investment industry. It is, therefore, of paramount importance to accurately model longevity risk so that asset-liability management is as effective as possible. We identify several observable economic and sociological time series that are relevant to the overall UK mortality experience including income, government spending, educational attainment, time and unemployment. These are regressed on three longevity risk factors derived from the Aro and Pennanen (2011) mortality model, calibrated on UK data from 1922 to 2014. We found no factors that were significant in modelling youth longevity for either sex. Income was weakly significant in modelling male middle aged longevity but there was no relationship with the female middle aged longevity factor. Conversely, all factors except unemployment provided good models for elderly longevity. We found that individual income provided the strongest model, being preferred to a multi-factor model. These results have big implications for the investment industry and can be used to create simulation models for liability-managed portfolios.
Date of Award1 Oct 2018
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorTeemu Pennanen (Supervisor)

Cite this