Earlier today, while out walking, I experienced a moment of appreciation for democracy—a moment of gratitude and celebration. Even as democracies appear to be struggling around the world, my feeling reflects a common sentiment amongst those living in consolidated democracies, who still generally prefer democratic systems over other regime types.

In ideal circumstances, democracy can be likened to a well-functioning team, where each member plays an important role in achieving collective goals. Just as successful teamwork requires communication, cooperation, and respect for differing viewpoints, democracy thrives on the principles of participation, dialogue, and accountability among its citizens, public administrators, and politicians. In this collaborative context, each person is called upon to uphold democratic norms and engage constructively with others. Naturally, this is challenging for everyone involved.

Authoritarian Behaviour

For example, within any given group, authoritarian behaviour in all its guises will invariably challenge teamwork and democratic processes. In a democratic political system, authoritarian behaviour can manifest in a variety of ways, including the rejection of democratic norms and rules, denying the legitimacy of others’ views, attempting to limit the freedom of “rivals,” and so on. A recent study by Hobolt and Osnabrügge highlights a curious problem: Although people living in consolidated democracies generally appreciate democracy, they may, at the same time, tolerate a certain amount of authoritarian behaviour amongst the politicians they elect. They may rationalize these undemocratic behaviours based on their own personal policy preferences and, indeed, show relatively more support for undemocratic political leaders if they perceive them to be more “competent” than their democratic political opponents.

So how might we counter authoritarian behaviour in democracies? One way is to simply not vote for authoritarian politicians. If we have a choice between a candidate who displays authoritarian behaviour and a candidate who does not display authoritarian behaviour, we choose the latter. To examine this in more detail, Hobolt and Osnabrügge used a very clever survey-based conjoint experimental strategy to analyse people’s preferences for candidates involved in various “controversies.” A sample of 4,012 participants living in the United Kingdom were each presented with five separate conjoint political candidate decisions (i.e., a total of 40,120 decisions were sampled). In each case, participants were asked to choose between two hypothetical candidates for election. Importantly, each candidate was described as being involved in a controversy of some type, and this could be either some form of authoritarian behaviour (e.g., they argued that government should exclude journalists from press briefings, or they advocated ignoring the courts in times of crisis) or a nonauthoritarian misdemeanour (e.g., the candidate misused public funds).

Across the different scenarios, Hobolt and Osnabrügge also varied the reaction from other politicians (MPs) to the candidate’s controversial behavior. They hypothesised that participants would be less likely to choose candidates displaying authoritarian behavior when those candidates are criticized or countered in some way by other politicians and that these counterreactions would be more effective if they came from politicians in the same political party as the candidate as opposed to politicians from a different party. Finally, Hobolt and Osnabrügge varied how politically costly the counteractions were for the MPs: from a baseline of “no reaction,” to mere verbal criticism, to more politically costly responses such as refusing to work with the candidate or calling for the candidate’s expulsion from the party. It was hypothesised that more costly counterreactions from MPs would have a greater negative impact on the candidate’s appeal compared to less costly actions.

Findings

The findings from this study are quite fascinating. First, averaged across all the scenarios, controversies related to authoritarian behaviour do not necessarily decrease support for a candidate more strongly than other misdemeanours. For example, arguing that the government may ignore the courts in times of crisis only decreased the probability of supporting candidates in pairwise choices by 2.9 percentage points compared to the controversy of being involved in an extramarital affair. When comparing the effects, Hobolt and Osnabrügge suggested that the key factor was not so much the comparison between authoritarian behaviour and other misdemeanours but, rather, whether the controversies related to actual actions rather than the candidate’s arguments (i.e., their verbal behaviour); for instance, encouraging online harassment of a politician from a different party resulted in a 22.9 percentage point reduction in support for the candidate, while arguments related to possible government actions like ignoring courts only diminished support by 2.9 percentage points, indicating that voters view actual controversial actions, such as harassment, as far more threatening to democracy than authoritarian rhetoric or other offenses, such as extramarital affairs, which are perceived as less threatening overall.

Importantly, counteractions from other politicians in response to controversial behaviours were seen to influence participants’ choice of candidates, and this effect was seen for both authoritarian behaviours and other misdemeanours. Hobolt and Osnabrügge found that these counteractions from other politicians do reduce the appeal of candidates, decreasing voter likelihood of supporting candidates by approximately 2.9 to 7.4 percentage points. Although not as strong an effect, counteractions were more impactful when they came from politicians in the same political party as the candidate, resulting in a 1.8 percentage point decrease in the probability of selecting the candidate compared to the scenario where criticism came from politicians in a different political party. Finally, as hypothesised by Hobolt and Osnabrügge, more costly counterreactions from MPs have a greater negative impact on the candidate’s appeal compared to less costly actions: When compared with the “no reaction” condition, a call for expulsion resulted in a 7.4 percentage point drop in the probability of selecting the candidate, refusal to work with the candidate resulted in a 5.8 percentage point drop, and merely criticizing the candidate’s behaviour resulted in a 2.9 percentage point decrease.

Overall, the study provides valuable insights into the deeper ways in which citizens engage with and appreciate democracy. When given the opportunity to vote, citizens recognise and are influenced by the counteractions of politicians to candidates who engage in authoritarian behaviours. In observing and responding to these counteractions—particularly the more costly counteractions of politicians—citizens are recognising and reinforcing the critical role of accountability in preserving democratic norms. Politicians need to be seen to hold other politicians to account. Furthermore, in the longer-term of election cycles typical in systems of representative democracy, by responding in particular to the costly counteractions by politicians, citizens’ voting behaviour is likely to counter authoritarian behaviours, specifically by lowering collective preferences for authoritarian candidates. In the broader societal context of teamwork, which so many of us value, all of this makes perfect sense: Just as effective teams rely on each member’s commitment to uphold collaborative norms, democracies, too, thrive on the active engagement of politicians and citizens alike to hold each other accountable and reinforce the principles that sustain democratic integrity.