No abstract

For decades, political scientists have argued that competition is a fundamental component of a responsible party system, such that when one party dominates politics, legislative coalitions destabilize and democratic accountability suffers. In this paper, I evaluate these predictions in an important but largely unexplored legislative environment: American local government. Using an original collection of roll-call records from 151 municipal councils, I show that legislative behavior is more one-dimensional when elections are partisan and the electorate is evenly balanced between the parties. When either of these features is absent, however, elite behavior remains unstructured, with coalitions shifting over time and across issues. These differences across institutional and competitive contexts suggest that partisan elections—and the party organizations that nearly always come with them—are critical for translating electoral insecurity into organized government, raising questions about the capacity for electoral accountability in a growing set of one-party dominant governments across the country.

Objective: A growing literature argues that national issues and partisanship structure local-level conflict in the United States. This argument contrasts starkly with the traditional view of local politics as fundamentally nonpartisan and nonideological. We reconsider these diverging arguments. Methods: We use a large-scale survey of municipal officials to examine elite preferences on a series of policy tradeoffs. We identify latent dimensions that underlie elite preferences across cities and on a common scale. We examine correlates of these dimensions including individual-and city-level demographics and electoral support from constituent groups. Results: Our results demonstrate that elite preferences in local politics-unlike in national politics-are multidimensional, with two underlying cleavages: one based on partisanship, the other a market orientation to the provision of local services. These latent dimensions align with indicators of constituent group support, suggesting that each dimension reflects substantively meaningful features of conflict in local electoral politics. Conclusions: These findings suggest that local politics are not completely structured by national-level partisan politics. Future research should aim to measure and understand local conflict along both partisan and market-orientation dimensions.A growing literature documents the nationalization of urban and local politics in the United States, with evidence suggesting that local election outcomes, elite preferences, and policy are all increasingly oriented around partisanship (