How to Make a Public Choice About the Value of a Statistical Life : The Case of Road Safety

Cost-benefit analysts involved in evaluating projects influencing the risk of death and injury have access to a wide group of studies that provide a large range of estimates of the value of a statistical life (VOSL). It is of course a difficult task to pick the right estimate. This paper discusses the potential avenues available to analysts looking for values of a statistical life and of injuries to be used in cost-benefit analyses of Quebec projects involving changes in road safety. Actually, the discussion is conducted in the context of Quebec, but most of it could easily apply to the rest of Canada. First, we discuss the relevance of looking for an original set of estimates involving a new study and the collection of new data. We present many arguments in favour of such a strategy. Second, if the time or the resources necessary to conduct a new study are not available, we offer an analytical framework that allows one to make a choice of estimates (or of a range of estimates) from existing studies. We conclude that a VOSL of 5 million dollars (CAN $, 2000) would be acceptable. Another contribution of this paper is to present, to our knowledge, the most up-to-date survey of studies on the value of a statistical life covering more than 85 papers