Traditional economic models of accident law are static and assume homogeneous individuals under perfect information. This paper relaxes these assumptions and presents a dynamic unilateral accident model in which potential injurers differ in their probability of accident. Information about individual risk-type is hidden from the social planner and from each potential injurer. We ask how negligence standards should be optimally tailored to individual risk-type when this is imperfectly observable. We argue that information about past accident experiences helps to efficiently define negligence standards, narrowing the distance between first-best standards perfectly tailored to individual risk-type and third-best averaged standards. We finally show that negligence standards refined on the basis of past accident experiences and of individual risk-type do not undermine private incentives to undertake due care.
Guerra, Alice and Hlobil, Tobias M, Tailoring Negligence Standards to Accident Records (August 10, 2015).