Quadratic voting

Quadratic voting is a collective decision-making mechanism in which voters can purchase votes for or against a proposal at a cost equal to the square of the number of votes purchased — one vote costs one credit, two votes cost four, three cost nine, and so on. Proposed by Steven Lalley and E. Glen Weyl, the mechanism aims to solve the tyranny of the majority by making it expensive to express very strong preferences, while still permitting minorities to concentrate their influence on issues they care about deeply. The quadratic cost function is not arbitrary: under certain assumptions about independent voter valuations, it elicits truthful reporting of preference intensity and produces the socially optimal outcome in equilibrium.

The mechanism is theoretically elegant and practically fraught. The assumption of independent private valuations rarely holds in political contexts, where preferences are correlated, strategic, and shaped by social influence. More critically, quadratic voting encodes a specific theory of democratic legitimacy — that preference intensity should matter alongside preference direction — without defending that theory against the objection that intense preferences are not necessarily more legitimate than mild ones. A wealthy minority can buy overwhelming influence if its members coordinate, and a mechanism designed to empower minorities can become a tool for plutocratic capture when wealth is concentrated.

Despite these concerns, quadratic voting has been experimentally deployed in corporate governance, community fund allocation, and even political polling. Its advocates argue that any voting system already weights preferences through participation costs, information access, and media influence; quadratic voting merely makes the weighting explicit and mathematically tractable. Its critics respond that making plutocracy explicit is not an improvement over concealing it. The debate is not about mechanism design but about political philosophy: what should democratic aggregation maximize, and who decides?