Senior Lecturer in Actuarial Studies
Department of Economics, Centre of Actuarial Studies
The University of Melbourne
An annuity and a conventional tontine can produce such similar income streams that a retiree gains little from holding both. We introduce a power tontine and show how to choose its payout profile so that combining the two products improves the trade-off between expected income and risk.
Using Australian data, we model two ways to help households buy a home. Lower deposit requirements can delay homeownership for lower-income households when supply is tight. Early access to pension savings improves access across income groups, but can leave buyers with less retirement wealth.
When an annuity’s reserves differ from what is needed, how much of the difference should the policyholder bear at each age? We derive a closed-form approximation to the optimal sharing rule and examine when the policyholder and the insurer would both choose it over guaranteed benefits.
This study compares neural networks with established methods for detecting underwriting cycles. In simulations, the networks identify cycles even when outliers or structural breaks distort the data. The method is then applied to Brazilian motor insurance prices for individual risk profiles.
An annuity can pass some of the cost of unexpectedly long lifetimes back to policyholders through adjustments to benefits. Under mean-variance preferences, the paper identifies which combinations of premium loading and transferred risk make this arrangement preferable to a guaranteed annuity for both parties.