Compromise and strategyproof social choice under uncertainty. A special case.

Details

Supervisors

Maniquet, François

Faculty

Faculté des sciences économiques, sociales, politiques et de communication

Degree label

Master [120] en sciences économiques, orientation générale

Abstract

The problem of finding domains that admit non-trivial strategyproof functions is an old topic in social choice theory. More than 45 years have passed since Allan Gibbard's and Mark Satterthwaite's impossibility theorem (Gibbard 1973, Satterthwaite 1975) and the literature has gone far in studying all sorts of domains. Therefore, it is surprising to learn that the domain of subjective expected utility preferences has been overlooked. In the new paper Bahel and Sprumont 2020, the two authors model a situation in which a group decision must be taken among uncertain prospects. They show that when the agents' preferences are represented by their subjective expected value associated with these prospects, then there exist some interesting strategyproof social choice functions. Using a very weak unanimity axiom they characterize a class of strategyproof functions that they call locally bilateral top selections. One problem that these functions have is that they only rely on the agents' top outcomes, so that no compromise prospect is allowed even when this would be the natural social choice. In my study I relax the unanimity axiom and characterize a class of strategyproof social choice functions that compromise. I do this in the particular case of two agents, three outcomes and two states of nature. The functions exhibit interesting normative properties, above all anonymity, which was previous violated by the locally bilateral top selections. We argue that our functions constitute a new way of producing group decisions in situations of uncertainty.