The value of statistical life for adults and children: Comparisons of the contingent valuation and chained approaches

Abstract

Estimates of the Value of Statistical Life (VSL) provide a vital input to a variety of policy decisions ranging from health provision to transportation planning. However, the bulk of VSL research has focussed on estimating average values rather than taking account of the potential variation in VSL across groups. Policymakers are particularly concerned that using estimates based on data concerning adults might provide poor proxies of the values associated with preventing child fatalities. We investigate this empirical problem while also addressing methodological critiques of standard contingent valuation (CV) approaches to VSL estimation which ask survey respondents to value an outcome described in terms of both the probability of occurrence and the health impact of an event. A prior lab experiment confirms fundamental problems in subjects’ abilities to provide internally consistent valuations of such compound goods. Given this we compare CV approaches with the ‘chaining method’ of Carthy et al. (1999) which splits the valuation task in two, assessing the probability of an event and the disutility of that event separately and then ‘chaining’ responses together to obtain a VSL estimate. We provide a first application of this method to the estimation of the VSL for children and contrast this with values for adults. Results confirm prior expectations that VSL values for preventing child fatalities significantly exceed those for adults. Finally, we carry out the first replication of the chaining approach in a large and nationally representative sample of parents. We identify many advantages of chaining over CV approaches, however, through a novel variant of a validation test suggested by Carthy et al., we reveal anomalies in the estimates produced by the chaining method suggesting that a robust method for VSL calculation is yet to be refined.