Political competition, learning and the consequences of heterogeneous beliefs for long-run public projects

An incumbent political party, who cares only about voters' welfare, faces future political competition from a similarly well-intentioned party whose beliefs about the consequences of a `long-run' public policy are different from its own. We show that when the incumbent can endogenously influence whether learning occurs (active learning), future political competition gives her an incentive to distort her policy choices so as to reduce uncertainty and disagreement in the future. This incentive pushes all incumbents' policies in the same direction. We demonstrate this mechanism in a two period model of the regulation of a stock pollutant that combines the literature on uncertainty and learning in intertemporal choice with a simple model of political competition. If the interaction between active learning and political competition is strong enough, all incumbents, regardless of their beliefs, will emit more than they would like. Our model thus offers a candidate explanation for the weakness of long-run environmental policy in democracies that applies even in an ideal world in which politicians' objectives are aligned with voters'. The mechanism we identify is likely to apply in many long-run public policy contexts.