Steven P. Lalley
Department of Statistics, University of Chicago
E. Glen Weyl
Microsoft Research; Yale University
December 16, 2016
We propose Quadratic Voting (QV) as a method for binary collective decision-making: individuals buy votes for their preferred alternative, paying the square of the number of votes purchased. Quadratic cost uniquely makes the marginal cost proportional to votes purchased, encouraging voting proportional to the value and thus maximizing welfare. Similar heuristic arguments and experiments suggest it is more robust than other efficient mechanisms and it grows naturally from a variety of ideas in disparate literatures. Nonetheless, formal analysis beyond simple examples has proved illusive. By explicitly characterizing the subtle statistical structure of equilibrium, we prove convergence towards optimality in large populations in a canonical environment.
The online appendix for "Quadratic Voting" may be found here: http://ssrn.com/abstract=2790624.
Number of Pages in PDF File: 34
Keywords: social choice, collective decisions, large markets, costly voting, vote trading
JEL Classification: D47, D61, D71, C72, D82, H41, P16
Date posted: February 13, 2012 ; Last revised: December 19, 2016