Interstate Conflict Increases the Appeal of Undemocratic Candidates

• In Event: Elite Incentives to (Not) Adhere to Democratic Norms

Abstract

Recent years have seen a rise in undemocratic behaviors from democratically elected leaders, still with such leaders benefitting from widespread public support. This raises the obvious question why it is that citizens prefer undemocratic candidates over democratically compliant alternatives? In this paper we present and test a two-step theory addressing this question; First, candidate undemocratic behavior affects citizens’ trait impressions of candidates such that undemocratic acts increase impressions of dominance. Second, a series of studies find that citizens enhance their preferences for dominant candidates in contexts characterized by interstate conflict and war. In sum, we suggest that citizens come to prefer undemocratic candidates in times of interstate conflict and war, because undemocratic candidates are seen as more dominant. We test our theory across three well-powered and pre-registered studies. Two conjoint experiments (N=3,800) test if respondents in rich information environments with a wealth of information about candidates (e.g., age, gender, party, political positions etc.) form trait impressions about candidates based on whether a candidate complies with democratic principles or behaves undemocratically. Results support the first step of our theory showing that reports of undemocratic behavior enhance citizens’ impressions of candidate dominance (and reduce impressions of warmth and competence). Next, we test the second step of our theory using a vignette-based survey experiment embedded in a large (N=2,500) and representative sample of American respondents. This third study tests whether citizens see candidates engaging in undemocratic behaviors as better capable to lead when they consider situations marked by interstate conflict (priming either the Russian-Ukrainian war or the Chinese-Taiwanese conflict) compared to when they consider peaceful situations. Supporting our theory results show that respondents in peace situations evaluate a democratically compliant candidate much more positive than a candidate who engages in undemocratic behavior, but when assigned to consider one of two interstate conflicts this difference becomes non-significant, and it is effectively reduced to zero. That is, under interstate conflicts candidates who act undemocratically receive evaluations similar to democratically compliant candidates. Given that many societies—the American society in particular, some might say—are constantly involved in some kind of conflict, our results offer an original and important explanation for why citizens do not necessarily turn their back on undemocratic candidates. We end our paper discussing the results in relation to existing theories about candidates acting undemocratically, and citizens’ trait preferences in political candidates as well as the normative questions about democratic backsliding.

Authors

• Lasse Laustsen, Aarhus University

• Kristian Vrede Skaaning Frederiksen, Aarhus University