Optimizing Insurance Product Portfolios Under Longevity Risks: A Multistage Stochastic Programming Approach

Abstract

This paper uses an asset liability management framework to derive an optimal product mix for insurance portfolios facing systematic longevity risk. Mortality dynamics are simulated using the Lee-Carter model and then subsequently optimized via a multistage stochastic program. Numerical examples illustrate how various characteristics affect the natural hedge potential of a product and consequently change the optimal product allocation strategy for the insurer. Our findings conclude that insurers who utilize natural hedging between annuities and life insurance in their product composition are able to achieve lower longevity risk exposures for and better returns on their portfolios