Stéphane Zuber

How can social prospects be evaluated and compared when there may be a risk on i) the actual allocations that people will receive, ii) the existence of these future people, and iii) their preferences? This paper investigates this question, which can arise when considering policies, such as climate policy, that affect people who do not yet exist. We start from the observation that there is no social ordering that meets minimal requirements of fairness, social rationality, and respect for people’s ex ante preferences. We explore three ways around this impossibility. First, if we drop the ex ante Pareto requirement, we can obtain fair ex post criteria that take an (arbitrary) expected utility of an equally-distributed equivalent level of well-being. Second, if the social ordering is not an expected utility, we can obtain fair ex ante criteria that evaluate uncertain individual prospects with a certaintyequivalent measure of well-being. Third, if we accept that interpersonal comparisons rely on VNM utility functions even in absence of risk, we can construct expected utility social orderings that satisfy of a version of Pareto ex ante.

Suppose that, for whatever reason, it is decided that inequalities within countries are more offensive than inequalities between countries, and that inequalities between populations living together are more offensive than inequalities between generations living in different times. Can a social welfare function express that preference? We show that it is actually difficult to incorporate such a localist preference into a social welfare function, except in a limited way (i.e., from a situation of specific similarity between countries). We also show that in order to obtain such preferences, the relative size of inequality aversion within and between countries may be counter-intuitive in some relevant cases, in the sense that a greater inequality aversion may happen to be required across countries than within countries. This research highlights new social welfare functions that aggregate the outcomes of evaluations over pairs of agents.

We examine utilitarian criteria for evaluating profiles of well-being among infinitely many individuals. Motivated by the non-existence of a natural 1-to-1 correspondence between people when alternatives have different population structures, with a different number of people in each generation, we impose equal treatment in the form of Strong Anonymity. We demonstrate how a novel criterion, Strongly Anonymous Utilitarianism, can be applied in the Ramsey model, leading to an efficient and sustainable stream. We show how the criterion is the result of combining Strong Anonymity with other regularity axioms (Monotonicity, Finite Completeness, and continuity axioms) as well as axioms of equity, population ethics, sensitivity, and separability. We relate it to other strongly anonymous utilitarian criteria.