Abstract:
This paper examines the causal impact of citizen-initiated ballot measures, defined as initiatives and veto referendums, on state government revenues and expenditures. Across U.S. states, direct democratic institutions have been introduced at different points in time, creating a setting that allows for causal identification. Leveraging a newly compiled panel dataset spanning from 1890–2008 with detailed contextual information, we apply a partially pooled synthetic control method that constructs counterfactuals based on both within-state and cross-state pre-treatment imbalances. The results indicate that the adoption of citizen-initiated ballot measures has no significant effect on state fiscal outcomes. We further show that signature requirements for launching ballot measures are uncorrelated with their actual frequency, helping to explain why we do not find a difference between direct and indirect fiscal effects. The findings are robust across model specifications and institutional contexts. This research offers new insights into the fiscal implications of citizen-initiated legislation.
