What Happens When Extremists Win Primaries?

Abstract

T his article studies the interplay of U.S. primary and general elections. I examine how the nomination of an extremist changes general-election outcomes and legislative behavior in the U.S. House, 1980House, -2010 a regression discontinuity design in primary elections. When an extremistas measured by primary-election campaign receipt patterns-wins a "coin-flip" election over a more moderate candidate, the party's general-election vote share decreases on average by approximately 9-13 percentage points, and the probability that the party wins the seat decreases by 35-54 percentage points. This electoral penalty is so large that nominating the more extreme primary candidate causes the district's subsequent roll-call representation to reverse, on average, becoming more liberal when an extreme Republican is nominated and more conservative when an extreme Democrat is nominated. Overall, the findings show how general-election voters act as a moderating filter in response to primary nominations.

Crashing the Party? Elites, Outsiders, and Elections

“…Also related to our theoretical predictions, Hall () shows that when a party nominates an ideologically extreme candidate, the party's prospect of winning that seat in the general election decreases by 35–54 percentage points, on average. The majority of the primary contests in the author's data are from districts classified as “safe” or “competitive” for the party—which our model predicts are the districts with both the largest propensity for Outsider entry and electoral penalties.…”

Who Wants to Run? How the Devaluing of Political Office Drives Polarization

“…Second, since these positions are estimated with noise, challengers close to incumbents will be more likely to be mis-classified as "moderate" challengers when they are in fact extremist challengers, or vice-versa. To address this issue, I follow the technique of Hall (2015). Next, I re-estimate the same regressions only using the subset of data in which the distance between the incumbent and her challenger is at or above the median such distance across all races.…”

“…Papers have shown some evidence that this assumption is violated in U.S. House general-election races, where barely winning candidates appear to be disproportionately likely to be incumbents, relative to barely losing candidates (Grimmer et al 2012;Caughey and Sekhon 2011;Snyder 2005). Hall (2015) tests for the validity of the primary RD design and finds no evidence of sorting. Because the paper uses almost exactly the same sample of elections as this analysis, I do not repeat these exercises here-but to sum up, in brief, primaries in which the extremist barely won look no different from those in which the more moderate candidate barely won on a variety of pre-treatment traits including, most importantly, their previous general-election vote shares for one party or the other.…”

“…Among other things, legislators are shown to take more moderate positions-and be rewarded for these positions-in more informed districts. Hall (2015) uses the measure from Snyder and Stromberg (2010) to show that the penalty to nominating extremists is larger-roughly three times as large-in the most informed (i.e., most congruent) district than in the least informed one.…”

Valence uncertainty and the nature of the candidate pool in elections

“…Although many contested races, especially primaries, feature an extremist candidate challenging a more moderate one (Hall, 2015), a nontrivial share of electoral races in the USA remain uncontested. 1 I show that when voters place a higher weight on policy considerations than on the candidates' valence, moderate candidates manage to lead unopposed campaigns.…”