ABSTRACT
Governments must often make decisions about the allocation of public funds among competing projects of uncertain value. With the benefits and costs of public projects evaluated in terms of consumers’ willingness to pay, consistent accounting requires that the social cost of a project’s risk be assessed according to the same metric. In an influential paper, Arrow and Lind (1970) argue that, when this approach to evaluating public projects is adopted, the social cost of project risk in a large economy is negligible, and projects should be evaluated solely on the basis of their expected values. The argument is that, as the number of consumers becomes large, the private risk borne by any one consumer becomes small and, more decisively, the aggregate (social) cost of risk becomes small as well – a result known as the Arrow-Lind Theorem.
