The Democracy Effect: a weights-based estimation strategy

Dal Bó et al. (2010) show experimentally that the effect of a policy may be greater when it is democratically selected than when it is exogenously imposed. In this paper we propose a new and simpler estimation strategy that does not require information on the vote of subjects in the exogenous treatment. The new estimation strategy is based on calculating the average behavior under democracy by weighting the behavior of each type of voter by its prevalence in the whole population (and not conditional on the vote outcome). We derive the distribution of this new weights-based estimator of the democracy effect and show it eliminates selection effects under certain conditions. We apply the new estimation strategy to the data in Dal Bó et al. (2010) and to the data from a new experiment for which we cannot use the previous estimation strategy as we do not have information on how subjects would have voted in the exogenous treatment. We find a significant democracy effect in the former but not on the latter application.