Date of Award

Summer 8-7-2012

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Risk Management and Insurance

First Advisor

Martin F. Grace

Second Advisor

Glenn W. Harrison

Third Advisor

Zhiyong Liu

Fourth Advisor

E. Elisabet Rutstrom

Fifth Advisor

J. Todd Swarthout


Accident forgiveness, often considered as a type of "premium insurance," protects the insured against a premium increase if an at-fault accident occurs. Although accident forgiveness has received considerable attention in the auto insurance industry, there is little or no literature on accident forgiveness in the actual contract. In this dissertation, I use dynamic modeling to examine optimal insurance contracts with an accident forgiveness option and further use a structural modeling approach to investigate the impacts of risk and time preferences on the accident forgiveness contract purchase. This study attempts to improve our understanding of the implications of this new type of insurance option. The first essay develops an asymmetric learning model in which insurers compete to attract policyholders. When information about previous at-fault accidents is not shared perfectly by insurers in the market, information asymmetries arise between the initial insurer and the rival insurer, as well as between the insured and the insurer. I design an auto insurance contract with accident forgiveness that charges policyholders higher-than-market premiums according to their risk types in the first period and then experience-rates both types in the second period contingent on their previous at-fault accidents. Contrary to the prior literature, which elicits competition as the reason to temper the effects of experience rating, this model is built such that accident forgiveness is the device to temper experience rating. This contract attracts policyholders since it "forgives" at-fault accidents and provides "rewards" in terms of coverage and premiums for those who remain accident-free. Risk and time preferences influence a variety of economic behaviors. In the field of insurance economics, attitudes toward risk and time are likely to affect the insurance purchase decision. As can be observed in the auto insurance market, when offered an optional accident forgiveness policy from insurers, the insured shows different purchase patterns, regardless of driving behavior. The question of whether and how individual risk aversion and discount rates affect the accident forgiveness purchase decision is critical to understanding contract design. In the second essay, by conducting a unique experiment under controlled laboratory conditions, I examine the role of risk and time preferences in accident forgiveness contract purchase and determine that individual discount rates and product prices are significant factors. Interestingly, I also find evidence that less risk-averse policyholders generally behave more like risk-neutral agents when making insurance decisions. Risk attitudes affect insurance decision-making only among those with a relatively high degree of risk aversion.