This research program is based on the idea that good voting systems should take into account the frequency with which different choice problems arise. Traditional social choice theory requires properties over a fixed domain of choice problems but does not offer the possibility of trading off the properties of a decision at one problem against those at another. Professor Green, together with Katherine A. Baldiga, is studying a set of social choice rules that they call "assent maximizing rules". These rules produce a rational social ordering in all cases, and therefore can be used to make social welfare judgments. Each rule corresponds to a probability distribution over the set of all possible problems.

The incentive properties and the qualitative axioms needed to characterize these rules are also being investigated in this project.

Applications include the study of polarization of preferences and the rationalization of irrational choice rules.