Unintended Consequences: Anticipation of General Election Outcomes And Primary Election Divisiveness

Abstract

This article offers the first theory to explain the relationship between primary election divisiveness and general election outcomes that is grounded in candidates' own behavior. Conventional wisdom holds that divisive primaries cause candidates to do poorly in general elections. I show that primary divisiveness does not cause this or any other pattern of general election results. Rather, expectations about general election results cause primaries to be divisive. Non-incumbents enter races they think they can win, and they think they can win where the incumbent is vulnerable. More candidates enter those races than others, splitting the vote among them. This stampede creates divisive primaries in which incumbents are most likely to do poorly, and challengers well, in the general elections. As a result, divisiveness is associated with (but does not cause) better general election performances among challengers and worse performances among incumbents. In this manner, primary divisiveness is an unintended consequence of behavior directed towards the goal of winning the general election. I tested these propositions using data from major-party House primaries between 1976 and 1998 and found that (a) candidate expectations of victory determine when and where divisive primary elections occur, (b) those expectations drive the correlation between primary divisiveness and general election results, and (c) primary divisiveness correlates with incumbents doing poorly, and challengers well, in general elections.