Assistant Professor of Political Science, University of Michigan
Multi-party elections are vital to representative politics, shaping relationships between politicians and citizens. Yet, it is difficult to empirically examine how the introduction of elections changes public attitudes toward politicians, because the timing of national-level initial elections is endogenous and there is no within-country variation across citizens. To identify the causal effects of first-ever elections, we exploit plausibly exogenous variation in the staggered timing of initiating village elections in Kazakhstan. Measuring political efficacy by direct questions and anchoring vignettes, perceptions of corruption using double-list experiment, and preferred candidate characteristics by conjoint analysis, we estimate the effects of the first village elections on those attitudes. We find that elections do not make citizens feel more efficacious nor reward leaders’ responsiveness to local needs, but report more frequent bribing of local officials. Our results suggest that introducing elections may foster negative perceptions of government quality, rather than enhancing civic political engagement.