Date: Wednesday 7th May at 2PM
Title: Tontines: diverse cohorts and automated adjustments (a grant application)
Abstract: Tontines gained attention as sustainable alternatives to guaranteed pensions. They deal with an uncertain future by adjusting retirees' income over time. In particular, tontines remove costly guarantees from pensions, but its members bear the risk of an uncertain future income.
Predicting and controlling the fluctuation is crucial for the success of tontines. However, making long-term predictions about income payments decades in advance is a mathematical challenge. Specification errors can have a cumulative effect on the income payments. In particular, dealing with differences in mortality among members within one tontine is an open problem.
We aim to optimally adapt the income when we assume one mortality distribution but face a different one; we aim to understand how the fluctuations of different age cohorts affect each other; and we aim to apply probabilistic limit theorems and stochastic analysis for precise and robust predictions about the fluctuations decades in the future. In the talk, we discuss those problems with specific examples and demonstrate proof of concepts.