Is There a Demand for Pain and Suffering Coverage?

Abstract

There is a large literature on this question, almost all of which avers that economic theory rules out the possibility of demand for insurance coverage of nonpecuniary losses on a priori grounds. There is little hard evidence, and much controversy over the interpretation of what evidence there is (see, e.g., Croley and Hanson, Nonpecuniary Costs of Accidents: Pain-and-Suffering Damages in Tort Law, Harvard Law Review, 1995). Nevertheless, the American Law Institute's Reporter has accepted that view into the recommendations for the new Restatement of Torts. This paper argues that the existence or nonexistence of demand for pain and suffering coverage is a purely empirical issue and policy recommendations should view it as such. The underlying economic theory is too weak to support such a recommendation without, at least, some supporting evidence. I argue first that even taken on its own terms, orthodox theory omits various possible contingencies, such as pre- and post-mishap curves that cross, that some writers such as Ellen Smith Pryor, The Tort Law Debate, Efficiency, and the Kingdom of the Ill: A Critique of the Insurance Theory of Compensation, 79 Va. L.R. 91 (1993), point out are sometimes valid. Second, I argue that as a matter of economic theory, it is odd to rule out demand for anything without evidence, inasmuch as economics generally holds that there is no accounting for taste. It turns out, on close inspection, that the literature generally descends from Cook and Graham, The Demand for Insurance and Protection: The Case of Irreplaceable Commodities, 91 Q.J.Econ. 143 (1977), whose equalized marginal utility view in turn descends directly from the controversial independence axiom implicit in the original work of Von Neumann and Morgenstern and made explicit by other authors in the 1940s. A result with such a disputable base, and no other evidence, is a poor basis for a policy recommendation. Still further, I offer two economic models in which the demand for pain and suffering coverage cannot be ruled out. The first (see appendix, below) is based on the expected model, allowing for the lag between entering an insurance agreement and the realization of the random event. The second is a simpler diagrammatic version, which bypasses the expected model altogether, inspired by Mishan, Choices Involving Risk: Simple Steps toward an Ordinalist Analysis, 86 Econ. J. 759 (1976). In both models, it may be optimal for a person to over- or underinsure pain and suffering relative to the strict position taken by the orthodox view. The point is not that to claim that there is or is not demand for this insurance, but merely to demonstrate that economic theory, by itself, cannot answer the question.