Predicting Changing Preferences: How People Think About Public Policy Proposals

Abstract

Preference utilitarianism provides a useful normative framework for deciding about public policy. However, preferences are not stable over time, and they may in particular be influenced by the implementation of the public policy itself: people's values might change adaptively, making the policy better than it seemed before implementation; or they might change counteradaptively, making the policy worse than it seemed before.In six experiments, I examined how subjects think about the potential of public policy to change preferences: How do they predict preferences to change, and how does predicted preference change alter what they think of the policy? I also investigated subjects' second-order preferences, that is, what preferences they want others to hold.When we ask subjects to predict in which direction preferences might change, they predict change in their favored direction: subjects who initially think the policy is good predict preferences to change adaptively; those who initially think it bad predict preferences to change counteradaptively. Wishful thinking is a plausible mechanism for this tendency, since subjects also have second-order preferences in the direction of their initial rating: Subjects who think the policy is a good idea want others to like it; those who think the policy a bad idea want others not to like it.With the notable exception of one study, we find that their pattern of predicting preference change leads subjects to polarize their opinions about the policies. When we tell subjects to imagine adaptive preference change in either direction, their second-order preferences influence how they adjust their rating, with the result that subjects who initially rate the policy unfavorably think it worse than before if values were to adapt. I discuss several mechanisms that could be implicated in this phenomenon.