Accountability in a Federal System

Abstract

Theories of political accountability suggest that officeholders should be electorally punished when they perform poorly or fail to represent their constituents. This dissertation evaluates the extent to which these theories apply to state legislatures by addressing the question: do elections hold state legislators accountable for their own performance? In analyses of state legislative challenger decision-making, party performance, and representatives' roll-call activity, I find little evidence of electoral accountability in state legislatures. Over a third of state representatives do not face major party challengers in the general election, and when state legislators face competition, voters do not appear to reward or punish state legislators for state-level policy outcomes, their legislative records, or their general performance. Instead of serving as a referendum on state legislators' own actions, state legislative elections are dominated by national politics. State legislators affiliated with the president's party - especially during unpopular presidencies - are the most likely to be challenged, and compared to individuals' assessments of the state legislature, changes in presidential approval have at least three times the impact on voters' decision-making in state legislative elections. Thus, while state legislatures wield considerable policy-making power, elections appear relatively ineffective in holding state legislators accountable for their own performance and lawmaking.