Consequences of Social Risk in Small and Moderate Sized Deliberative Democracies

The voting function that a group adopts is to some extent a reflection of its priorities. Many small to moderate sized communities use consensus voting to aggregate individual preferences in order to ensure that minority voices are heard and valued. In one such group on Stanford University’s campus, unanimous approval is easier achieved than one might expect: votes almost always pass without any argument. In this paper, we gather empirical data to assess the true preferences of voters in this community and find that members often vote against their true preferences. We then present a game-theoretic model which captures the dishonest behavior of voters by taking into account social pressure. Finally, we explore the possibility of using a different voting scheme, quadratic voting, instead of consensus in these small to moderate group contexts. Using data from votes in Columbae, we provide evidence which suggests that QV may be more efficient than consensus voting from a maximin perspective in practice, and therefore may better achieve the goals of communities which aim to value minority voices. In doing so, we contribute to the growing body of work which attempts to bridge normative social choice theory, which considers just the aggregation of preferences, and on deliberative democracy, which formalizes deliberation as a key component of decision-making.