Positional scoring rules optimize various voting welfare measures

Optimal Voting: Beyond a Utilitarian View

Computer Science and Game Theory

Summary

Deciding the best way to choose a winner by voting can depend on how you value different people's happiness. The authors studied many ways to measure overall happiness, beyond just adding up everyone's scores. They found that voting rules using a certain scoring system pick winners that maximize these happiness measures over time. They also discovered that the simple plurality rule does best when randomness is allowed, and that sometimes the best scoring choices look different from known methods. This helps us understand how to design better voting systems for different fairness goals.

What this means in practice

  • For election system designers: Design voting schemes that maximize different fairness measures beyond total happiness by choosing optimal score vectors tailored to voter preferences.
  • For collaborative decision-making platforms: Implement voting rules that improve collective outcomes by selecting winners according to varied welfare criteria in group decision settings.

Authors

Ben Armstrong, Nicholas Mattei

Abstract

A common measure of the quality of a voting rule is the social welfare of the alternatives it elects. Prior work shows that, when voter utilities are drawn from a neutral distribution, expected utilitarian welfare is maximized by a positional scoring rule. We extend this approach to the power mean family of welfare functions, which includes utilitarian, Nash, and egalitarian welfare as special cases. We show that, across the power mean family, a deterministic positional scoring rule (PSR) is asymptotically optimal for expected social welfare among all social choice functions and we characterize this rule for a broad class of utility distributions. We further show that, among randomized PSRs, plurality uniquely maximizes expected social welfare for all power means. Experimentally, we find that optimal score vectors often yield substantially higher welfare than canonical score vectors. On highly structured utility distributions, optimal score vectors take on novel shapes unlike existing canonical score vectors, highlighting open questions for future analysis.