Probabilistic Projection of UK Kinship

Joe Butterick, University of Southampton
Jason Hilton, University of Southampton
Jakub Bijak, University of Southampton
Peter W. F. Smith, University of Southampton
Joanne Ellison, University of Southampton
Erengul Dodd, University of Southampton

Forecasting how kinship networks unfold in the future provides a rich framework important to many policy relevant issues. For example, kin-supplied informal care and social security are affected by, and thus evolve with, demographic change. Establishing the expected numbers and ages of kin available to care by the year 2070 is of great interest. In this paper, by drawing from a matrix projection model, we provide the first steps for estimating future kinship networks in the UK. We apply a Bayesian approach to forecast male and female age-specific-mortality and-fertility as model inputs. This innovation advances on previous methods which approximate male fertility by female rates, thereby yielding more accurate inferences of kinship. We provide uncertainty intervals on future expected numbers of kin, which naturally arise from uncertainty in the statistical rate forecasts. To assess the predictive accuracy of our model, we validate the results using data taken from the English Longitudinal Survey of Ageing. We also investigate the possibility of extending the framework to incorporate parity-specific fertility rates. Our findings will be the first fully sex-specific projections of kin, advancing the reliability of our understanding of familial change and informing future social care policy.

Keywords: Mortality and Longevity, Population projections, forecasts, and estimations, Families, Unions and Households, Population Ageing

See extended abstract.