2023 Australasian Actuarial Education and Research Symposium

Lintao Wang

University of Melbourne

Evaluating climate change impacts on mortality, life insurance and annuities

This is joint work with Rui Zhou, Andrew King, Ping Chen

Recent studies provide compelling evidence of the association between non-optimal temperature and excess mortality. With the increasing frequency, duration and intensity of extreme temperature events due to climate change, understanding its effects on human mortality — and the consequent implications for life insurance and annuity pricing and valuation — becomes paramount. This study introduces a comprehensive framework to project human mortality and assess insurance and annuity liabilities under various Shared Socioeconomic Pathways (SSP) scenarios. We refine current environmental epidemiological approaches to establish the relationship between weekly age-specific mortality and daily mean temperature. This refinement includes integrating exposure-to-risk into the distributed lag non-linear model (DLNM), accommodating mixed-frequency sampling, and mitigating overfitting concerns. We also devise a method to consolidate age-specific annual mortality data with weekly mortality data for broader age groups, yielding a detailed age-by-age temperature-mortality relationship. By combining pattern scaling — a prevalent climate science technique — with time series analysis, we incorporate the impact of global warming into the simulated future local temperature paths. These, when paired with temperature-mortality relationships, facilitate the projection of future mortalities under assorted SSP scenarios. Through an analysis of two Spanish regions, we illustrate the shifts in mortality patterns and quantify the differences in insurance/annuity liabilities across various SSP scenarios.