Representation and Referenda in U.S. States

• In Event: Representation and Direct Democracy in the US

Abstract

Do state legislators represent their constituents' preferences? This is an important question for assessing representation in state politics and policy. Ballot referenda provide a unique opportunity to study this question, since the exact same item is put to both legislators and voters. Decisions are made in a governing context, so officials and voters have a stake in responding carefully and thoughtfully. We might expect alignment between how legislators and their constituents vote on referenda, and yet referenda do fail at the ballot.

We examine referenda at the institutional-level to determine whether and when the legislature approves items that are disapproved by voters (or vice versa). We also examine this at the district-level, to assess whether and when legislators deviate from their constituents' preferences and what factors may contribute to this dissensus. We are collecting data to study this for a ten-year period in nine states.

Preliminary analysis shows that more than 1/4 of referenda votes fail in Idaho and approximately 1/3 referenda votes fail in Missouri. The biggest political predictor of dissensus between legislators and voters on referenda votes was voter turnout in these two states. Communities with higher rates of participation have more responsive public officials. By collecting data from more states, we will further test the relationship between dissensus and turnout, as well as explore how partisanship, electoral competition, and other political factors affect dissensus in referenda votes.

Authors

• Jaclyn J. Kettler, Boise State University

• Greg W. Vonnahme, University of Missouri, Kansas City