The political consequences of candidate polarization: Quasi-experimental evidence from two-round elections
P14-3
Presented by: Sergi Ferrer
A long-lasting question in political science is how candidate polarization conditions political stability and citizens’ behavior. Although several authors have addressed this question the results are far from being conclusive. In this study I focus on two facets of political stability –the vote for incumbent candidates and their reelection rates– and test whether candidate polarization conditions the support for incumbents. To do so, I apply a Regression Discontinuity Design to data from 5,036 legislative and cantonal electoral districts in France. These elections use a two-round system in which the two most voted candidates in the first round automatically qualify for the second round. Following the logic of ‘close race elections’, I analyze cases in which incumbents qualify first in the first round and use the vote share of candidates who rank second in the first round as the qualifying threshold that third candidates would need to reach to qualify for the second round. I then compute the ideological distances between incumbents, second and third candidates to determine whether the presence of second candidates increases/decreases the level of polarization in the second round. Overall, results show that polarization, understood as the degree of ideological divergence between incumbents and runner-up candidates, clearly benefits incumbents in terms of vote share and re-election rates. Moreover, I show that this effect is particularly strong when I restrict the analyses to cases in which incumbents face an extreme-left/right candidate in the second round.