Pricing participating longevity-linked life annuities: a Bayesian Model Ensemble approach.

Area of Science:

• Actuarial Science
• Financial Mathematics
• Demography

Background:

• Participating longevity-linked life annuities (PLLA) provide longevity risk protection and investment potential.
• Existing research focuses on PLLA design and pricing, necessitating advanced modeling techniques.

Purpose of the Study:

• To value PLLAs using a Bayesian Model Ensemble of mortality models.
• To investigate the impact of asset allocation on PLLA valuation.
• To provide insights for cost-effective and efficient risk pooling mechanisms for retirement income.

Main Methods:

• Bayesian Model Ensemble of generalized age-period-cohort stochastic mortality models.
• Longevity option decomposition with risk-neutral simulation.
• Calibration using Taiwanese mortality, yield curve, and stock market data (1980-2019).

Main Results:

• The Bayesian Model Ensemble approach provides robust mortality forecasts.
• Sensitivity analysis reveals the impact of lifecycle asset allocation strategies.
• Empirical results offer significant valuation and policy insights for PLLA.

Conclusions:

• PLLAs can effectively address individual longevity uncertainty and provide retirement income.
• The proposed valuation framework enhances the efficiency of risk pooling mechanisms.
• Findings support the development of cost-effective insurance products for longevity protection.