The Electoral Challenge of Balancing Primary and General Electorates (Pre-Recorded)

• In Event: Constituency-Level Politics

Abstract

Elected officials must appeal to different subsets of the electorate to win reelection (Fenno 1978). Legislators in the U.S. face a two-stage election with a primary election and a general election, each with different sets of voters. Thus, they must appeal to both their base – the primary electorate – and their broader constituency – the general electorate. They may face tradeoffs between appealing to these constituencies. Legislative positions that win support among the primary electorate may cost support among the general electorate.

In contemporary politics, scholars, pundits and the public often presume that legislators are responding most to the interests of their base rather than to the broader electorate. There are theoretical reasons to believe that when the base and general electorates disagree, it is more advantageous to be aligned with the primary electorate than the general electorate. For example, in primary elections, party is not on the ballot making it more likely that voters consider factors like the incumbent’s roll-call record instead. Empirical scholarship has not determined whether the gain from siding with primary voters outweighs the potential costs legislators face in the general election.

For this project we conduct survey experiments to learn about how much voters reward incumbent legislators for their roll-call positions in Congress both in primary elections and general elections. We assess whether the probability that a voter in the primary election will vote for the incumbent changes more based on the legislator’s roll-call position that does that same probability for a general election voter – whether legislators face a primary premium. We also assess the mechanisms that drive differences between the responses of the primary and general electorates - whether differences are driven by changes in voter behavior across primary and general elections or by differences in the composition of the electorate in the two types of elections. After estimating the size of the primary premium, we conduct simulations that incorporate the primary premium, the distribution of voters’ preferences on various issues, and the incumbent’s prior vote share. This allows us to evaluate how often incumbents have strong incentives to support the position of the primary electorate even when doing so requires a position in opposition to the preferences of a majority of voters in the general election.

Authors

• Sarah Anderson, University of California, Santa Barbara

• Daniel M. Butler, WUSTL

• Laurel Harbridge-Yong, Northwestern University