Managing longevity risks in Singapore : an analysis of returns on variable annuities using the Lee-Carter model, financial time series analysis and Monte Carlo simulation

Abstract

Managing longevity risks is of growing importance in Singapore due to demographic changes and the unique retirement policy. There is a growing demand for variable annuity products to shield against the erosion of inflation and the diminishing annual income as life expectancy continues to improve. However, variable annuities subject the insurers to not only longevity risks but also the volatility of the financial market.

In this research, we use the Lee-Carter Model to forecast the mortality outlook of Singapore retirees aged 62 and above. We also use ARIMA time series model to fit the Straits Time Index (STI) and the current yield of 20-year Singapore government bond. To analyse the combined effect of longevity risk and financial risks, we stimulate the net present values of returns on investment in portfolios across a spectrum of risk exposures using simple Monte Carlo simulation.

Our research reveals that the expected returns on investment in variable annuities are generally positive, however the variance of returns is significant. There are risk-return trade- offs which require the policyholders to consider their risk appetite in choosing a suitable portfolio. We also discover that at older ages, the marginal reward for investing in riskier portfolios diminishes, as the increase in variance of returns, and hence the risk exposure is more than proportionate to the increase in expected returns.