Date of Award

Spring 2011

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Insurance & Risk Management

First Advisor

Olivia S. Mitchell


The first chapter “Investment Patterns in Singapore’s Central Provident Fund System” investigates how plan participants in a national defined contribution system invest their pension accumulations. I find that only a small fraction of participants elects to invest in outside investment products like professionally-managed mutual funds. Simulation results using cost data from over 200 funds demonstrate that the minimum hurdle rate of return a fund must generate is about five percent a year. Accordingly, more policy attention can be devoted to lowering fund commission charges and rationalizing the investment menu offered to participants.

In the second chapter “Longevity Risk Management in Singapore’s National Pension System”, I evaluate the money’s worth of life annuities and discuss the implications of the government entering the insurance market as a public-sector provider for annuities. I find that commercial insurers offer competitively-priced annuities with money’s worth ratios averaging 0.95, which are on par with those in other developed countries. On the other hand, the new annuities launched by government under an annuitization mandate are estimated to provide money’s worth ratios exceeding unity. This will benefit annuitants on average but implies that the annuity mandate will be expensive for the government if current pricing continues.

The third chapter “Beyond Age and Sex: Enhancing Annuity Pricing” assesses how adopting more detailed pricing schemes may help reduce adverse selection in annuity markets. Prices of standard annuity products in the United States do not currently reflect buyers’ personal characteristics other than age and sex. I show that several readily-measurable risk factors can significantly increase explained variability in mortality outcomes in a proportional hazards framework and use them to construct alternative pricing schemes. Simulation results show that more detailed pricing may help reduce adverse selection in annuity markets because shorter-lived groups are made much better off (and thus enter the market) while longer-lived groups are made only slightly worse off (and thus remain in the market).