Relative Nash welfarism
Article [Version of Record]
Is part ofCahier de recherche ; #2017-03
Publisher(s)Université de Montréal. Département de sciences économiques.
Relative Nash welfarism is a solution to the problem of aggregating von Neumann-Morgenstern preferences over a set of lotteries. It ranks such lotteries according to the product of any collection of 0-normalized von Neumann-Morgenstern utilities they generate. We show that this criterion is characterized by the Weak Pareto Principle, Anonymity, and Independence of Harmless Expansions: the social ranking of two lotteries is unaffected by the addition of any alternative that every agent deems at least as good as the one she originally found worst. Relative Nash welfarism is more appealing than relative utilitarianism in contexts where the best relevant alternative for an agent is difficult to identify with confidence.