A Democratic Emergency After a Health Emergency? Exposure to COVID-19, Perceived Economic Threat and Support for Anti-Democratic Political Systems.

Objectives: The urgency of the COVID-19 pandemic has led governments to impose restrictions on individual freedom and required citizens to comply with these restrictions. In addition, lockdowns related to COVID-19 have led to a significant economic crisis. We aimed to study how the pandemic and related economic threats have impacted support for anti-democratic political systems. Method: We analyzed data from a quota panel of the Italian adult population (N = 1,195), surveyed once before and once during the pandemic.
Results: A hierarchical regression model showed that exposure to COVID-19 and perceived economic insecurity were associated with support for anti-democratic political systems, independent of participants' predispositions toward a strong leader.
Conclusion: An authoritarian personality is not a necessary precondition for individual anti-democracy: when facing severe personal threats, anyone could restore a subjective sense of control over the social world by becoming anti-democratic, independent of their initial predisposition to support anti-democratic political systems.

The majority of the world population is facing threats stemming from the COVID‐19 pandemic, a virus for which there is not yet a cure or vaccine. No fundamental scientific information on COVID‐19 is available: we do not know enough about its contagiousness and speed of mutation nor whether those who have recovered from COVID‐19 are immune to reinfection. This crisis has forced millions of people to cope with a severe and previously unknown existential threat to their survival and that of the people they love. The lack of reliable forecasts regarding when and how the threat will end has added further psychological distress.

Many nations have tackled this health emergency by adopting strict lockdown measures. Citizens’ compliance with governmental restrictions on their individual liberties implies that they perceive these authorities as empowered to impose limitations on their civil rights. This climate of urgency, which has led to the imposition of measures that restrict freedom, raises questions about the impact of the current situation on citizens’ proneness to submit to authorities. Moreover, the pandemic is concurrent with what could be the greatest economic crisis since the Great Depression of 1929. The suspension of a large portion of production activities for weeks or even months has provoked a significant drop in many countries’ GDPs and has dramatically worsened the economic security of millions of people (Maliszewska, Mattoo, and Van der Mensbrugghe, 2020).

How does this convergence of health and economic emergencies impact public opinion? In this study, we aimed to answer this question based on the social psychology literature on support for anti‐democratic political systems. This literature indicates strong relationships between social threat and support for anti‐democratic political systems, conceived as an individual's authoritarianism, support for anti‐democratic governmental systems, or intention to vote for extreme right‐wing political parties (e.g., Altemeyer, 1996; Russo, Roccato, and Mosso, 2019). Social threat refers to situations where an out‐group threatens the in‐group—through, for example, criminality and terrorism (e.g., Asbrock and Fritsche, 2013)—and where an out‐group is not easily visible: archival research has shown that social indicators of authoritarianism rise in times of economic crisis (Sales, 1973). In general, the idea of support for anti‐democratic political systems as a collective response to social threat is well established (Onraet et al., 2013).

A convincing explanation for this link is that the feeling of living in a random, chaotic, and uncontrollable social world—as is the case with social threats—leads to high arousal and anxiety and reduces individuals’ perceived control over their social world (Kay et al., 2009). When individuals feel unable to exert personal control over their social environment, they are driven to restore subjective order and predictability by resorting to compensatory control mechanisms, such as religion, government, and—most importantly in the context of this study—submission to anti‐democratic authorities (Kay et al., 2011). Consistent with this idea, some studies have shown that nonauthoritarians react to threats with the strongest endorsement of authoritarian attitudes. This conditional shift toward authoritarian attitudes is dependent on the use of authoritarianism as a coping strategy to deal with the uncertainty of threats. Previous studies have observed this pattern in response to different types of societal threats, including criminality, malicious out‐groups threatening social order, and exposure to natural disasters (Mirisola et al., 2014; Russo et al., 2020).

While this stream of research has focused on perceived collective threats, much less is known about the effects of personal threats, that is, threats to an individual. Even the literature on the links between psychological reactions to crime and authoritarianism mainly focused on participants’ worry about crime as a social problem more than on their individual fear of crime (e.g., Dallago and Roccato, 2010). To the best of our knowledge, there is no information about the effect of a large‐scale health emergency—such as the COVID‐19 pandemic—on the endorsement of support for anti‐democratic political systems.

The Present Study

In this study, we analyzed the endorsement of support for anti‐democratic political systems in response to COVID‐19 exposure and perceived economic threat. Based on previous research showing a positive link between perceived personal economic threat and authoritarianism (Rickert, 1998), we hypothesized a positive relationship between perceived economic threat related to the COVID‐19 pandemic and support for anti‐democratic political systems (H1). Terror management theory (Solomon, Greenberg, and Pyszczynski, 1991) posits that mortality salience pushes individuals to cope with the loss of subjective control by endorsing authoritarian leaders and worldviews. Thus, we hypothesized that exposure to COVID‐19, which heightens mortality salience, would be positively associated with support for anti‐democratic political systems (H2). We also explored the possibility that the link between threat and support for anti‐democratic political systems is conditioned upon initial authoritarian disposition. We speculated that, if personal threats function similarly to societal threats, the association between COVID‐19 exposure and perceived economic threat, on the one hand, and support for anti‐democratic political systems, on the other, should be observed only (or more markedly) among participants with low predisposition toward a strong leader prior to the pandemic. On the contrary, if personal threats drive both low and high authoritarians to restore subjective control with the same intensity, we should observe no statistically significant interactions.

The study was performed in Italy, a country that is an extraordinary natural laboratory for investigating the public opinion consequences of the present emergency, due to both the severe impact of the virus on the Italian population's health (Dowd et al., 2020) and the dramatic economic consequences of Italy's lockdown (Fernandes, 2000). When we wrote this article (August 2020), Italy had the fourth highest number of COVID‐19 cases in Europe, with more than 250,000 people ill, and the second highest number of deaths, with more than 35,000 dead (Dowd et al., 2020; 〈http://www.salute.gov.it/portale/nuovocoronavirus/dettaglioContenutiNuovoCoronavirus.jsp?lingua=italiano&id=5338&area=nuovoCoronavirus&menu=vuoto〉). The COVID‐19 emergency brought the Italian health‐care system to its knees (Remuzzi and Remuzzi, 2020), especially concerning the availability of intensive care unit beds (Lauterio et al., 2020). The Italian government tackled the emergency by resorting to a radical lockdown strategy. For roughly two months, Italians were not allowed to leave their homes except to shop for food in the shops closest to their homes. This dramatically affected Italians’ quality of life and economic situation.

This health and economic emergency developed in a country often considered a “laboratory of populism” (Blokker and Anselmi, 2019). In the early 1990s, after decades of stability, the Italian political system crashed: traditional parties and leaders were substituted by new ones, Italian's trust in institutions became weaker and weaker (Chiaramonte et al., 2018). In the 2000s, the political crisis was exacerbated by a severe economic crisis and by a growing number of immigrants (Caiani, 2018). Consistent with this picture, in recent years, Italy has experienced intense voter mobility, stronger than that of other European countries (Kriesi and Pappas, 2015), and a growing success of populist parties (Vassallo and Shin, 2018).

Method

Participants and Procedure

On two occasions, we surveyed via email a quota panel of the Italian adult population (N = 1,195; women = 50.7 percent; M age  =  49.83; SD  =  14.56), stratified by gender, age, geographical area of residence, and size of area of residence. The first survey (T 0¸ N = 1.504) was conducted between May 26 and June 1, 2019 and the second (T 1, N = 1,195) between April 17 and April 26, 2020. Unless otherwise indicated, the measures below were assessed using data from T 1.

We checked if the dropout from 2019 to 2020 was related to gender, age, education, predisposition toward a strong leader, and perceived economic threat. We ran a logistic regression analysis to establish whether sample attrition (dropout = 0, retention = 1) was systematic. No significant differences emerged, with the exception of a positive effect of age (Wald = 22.50, p < 0.001), indicating that as age increased, the respondents were more inclined to participate also in the second wave. Generally, a low Nagelkerke R 2 (0.02) confirmed that the differences between who participated in both assessments and who participated only in the first assessment were not substantial.

Measures

Dependent Variable

Based on Russo, Roccato, and Mosso (2019), we administered participants two four‐category items (1 = awful system, 4 = excellent system) asking them to report how good or bad the following political systems would be for governing Italy in a period such as the present: (a) a strong leader who does not have to bother with parliament and elections; and (b) a military government (r = 0.51; p < 0.001). Previous research (e.g., cf. Finkel, Sigelman, and Humphries, 1999; Weil, 1989) shows that the combination of these two items is a valid operationalization of participants’ preference for anti‐democratic political systems. We computed a mean index for these items, with higher scores indicating preference for anti‐democratic political systems.1

Independent Variables

We measured participants’ exposure to COVID‐19 using the following question, developed ad hoc for this study: “Did you or some person close to you (such as relative or close friends) contract COVID‐19?”, to which participants could respond 1 (No), 2 (Not me, but some people close to me did), 3 (Yes, I did, but none of the people close to me did), or 4 (Yes, both me and some of the people close to me).2 We measured participants’ perceived economic threat using the following item from the European Social Survey: “Which of the following descriptions comes closest to how you feel about your household's income nowadays?,” to which participants could respond 1 (living comfortably on present income), 2 (coping on present income), 3 (finding it difficult on present income), or 4 (finding it very difficult on present income).

Control Variables

We controlled for participants’ gender (1 = woman), age, and years of education. Moreover, based on social psychological research showing that predisposition toward a strong leader leads to preference for anti‐democratic political systems (Altemeyer, 1996), we controlled participants’ predisposition toward a strong leader at T 0 by averaging the following five‐category items: “Some people think that the Parliament as a whole best represents the interests of society. Others think that the will of the people can be carried out only by having a strong leader. Where would you place yourself between these opposing opinions?” and “Some people think that in politics you need a strong leader to guide the people. On the other hand, others think that having a strong leader would be dangerous for democracy. Where would you place yourself between these opposing opinions?” (reverse‐scored; r = 0.29, p < 0.001) (Roccato et al., 2019). Finally, we controlled for participants’ perceived economic threat at T 0 using the same item on perceived economic threat described above.

Results

Table 1 reports the descriptive statistics for the variables we used and the correlations among them. Consistent with the idea that the COVID‐19 pandemic heightened the Italians' perceived economic threat, perceived economic threat mean level significantly rose between 2019 and 2020, t(1192) = −14.65, p < 0.001.

TABLE 1

Descriptive Statistics of Our Variables and Bivariate Correlations Between Them

	Mean	SD	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.
1. Female gender	0.51	0.50	1	0.05	−0.11***	0.06*	0.10*	−0.04	0.11***	−0.03	0.02	0.07*
2. Age	49.83	14.56		1	−0.26***	0.01	−0.03	−0.14***	−0.00	−0.02	−0.01	−0.10***
3. Years of education	13.85	3.62			1	−0.17***	0.18***	0.08**	−0.14***	−0.01	0.01	−0.14***
4. Predisposition toward leader at T 0	3.12	0.99				1	−0.02	−0.05	0.03	−0.01	0.03	0.37***
5. Perceived economic threat at T 0	2.65	0.65					1	−0.01	−0.56***	0.00	−0.05	0.03
6. Exposure to COVID‐19	1.27	0.64						1	−0.02	−0.06	−0.01	0.12***
7. Perceived economic threat at T 1	2.86	0.72							1	−0.02	−0.06*	0.08**
8. Exposure to COVID‐19 × Predisposition toward leader at T 0	−0.03	0.53								1	0.235	−0.06*
9. Perceived economic threat at T 1 × Predisposition toward leader at T 0	0.02	0.74									1	0.02
10. Support for anti‐democratic political systems	1.93	0.83										1

note: The means of the variables we used in the interaction are computed before centering them. The “mean” of the gender variable is the proportion, on a 0–1 range, of women. When gender is involved, the point‐biserial correlation is involved.

p < 0.001;

p < 0.01;

p < 0.05.

As a preliminary step, due to the nested structure of our data and the differences in COVID‐19 spread throughout Italy, we checked whether our dependent variable varied across Italian counties and regions, using an unconditional multilevel model to examine variance in support for anti‐democratic political systems, partitioning it into between‐individual and between‐county or between‐region variances. The variance of the dependent variable was not significant between regions or counties. Thus, despite the nested nature of our data, there was not room to proceed with multilevel analysis.

Using SPSS 26, we subsequently tested our hypotheses using a three‐step hierarchic moderated regression model, aimed at predicting support for anti‐democratic political systems as a function of the control variables (Model 1), of exposure to COVID‐19 and of perceived economic threat (Model 2) and of the interactions between exposure to COVID‐19 and perceived economic threat on the one hand and predisposition toward a strong leader at T 0 on the other (Model 3).

Table 2 shows that, among the control variables, participants’ gender was not related to support for anti‐democratic political systems, while their age and education had a negative association with it. Participants’ predisposition toward a strong leader had a positive effect on the dependent variable while the effect of perceived economic threat at T 0 was not significant (see the first three columns of the table). The fit of Model 2 (which included the control and the independent variables) was significantly higher than that of Model 1 (which included only the control variables). Consistent with H1 and with H2, exposure to COVID‐19 and perceived economic threat at T 1 showed a positive association with support for anti‐democratic political systems (see Columns 4–6 in the table). Finally, the fit of Model 3 (which included the control variables, the independent variables, and the interaction between the independent variables and participants’ predisposition toward a strong leader) was statistically equal to that of Model 2. Consistent with this, both the interactions between exposure to COVID‐19 and perceived economic insecurity at T 1 with the predisposition toward a strong leader at T 0 had a nonsignificant association with the dependent variable (see the last three columns of the table).

TABLE 2

Prediction of Attitudinal Anti‐Democracy

	Model 1			Model 2			Model 3
	B	SE	beta	B	SE	beta	B	SE	Beta
Intercept	2.61***	0.18		2.71***	0.18		2.71***	0.18
Female gender	0.07	0.05	0.04	0.06	0.04	0.04	0.06	0.04	0.04
Age	−0.01***	0.00	−0.14	−0.01***	0.00	−0.12	−0.01***	0.00	−0.12
Years of education	−0.03***	0.01	−0.11	−0.03***	0.01	−0.11	−0.03***	0.01	−0.11
Predisposition toward leader at T 0	0.29***	0.02	0.35	0.29***	0.02	0.35	0.29***	0.02	0.35
Perceived economic threat at T 0	−0.00	0.04	−0.00	−0.05	0.04	−0.04	−0.05	0.04	−0.04
Exposure to COVID‐19				0.17***	0.04	0.13	0.16***	0.04	0.13
Perceived economic threat at T 1				0.10*	0.04	0.08	0.10*	0.04	0.08
Exposure to COVID‐19 × Predisposition toward leader at T 0							−0.02	0.04	−0.01
Perceived economic threat at T 1 × Predisposition toward leader at T 0							0.00	0.03	0.00
Fit of the model	Adj. R2 = 0.16 F(5,1174) = 44.46, p < 0.001			Adj. R2 = 0.17 ΔF(2,1172) = 14.39, p < 0.001			Adj. R2 = 0.17 ΔF(2,1170) = 0.13, p = 0.88

p < 0.001;

p < 0.01;

p < 0.05.

Adj. R = 0.16

Adj. R = 0.17

Adj. R = 0.17

Discussion

The COVID‐19 pandemic has undoubtedly created challenges in people's lives. Some have contracted the virus themselves, others have had people in their social networks fall ill, and still others have stayed healthy but faced an unprecedented severe existential threat. Everyone has had to cope with the dramatic life changes resulting from the lockdown and many have had to face an additional serious economic threat. In this study, we analyzed the increase in support for anti‐democratic political systems as a political consequence of this unique convergence of health and economic personal threats. Exposure to COVID‐19 and perceived economic threat were positively associated with the endorsement of anti‐democratic political systems, and such associations were not moderated by participants’ prepandemic predispositions toward a strong leader. The dependent variable did not show contextual variations. Thus, its prediction was a matter of social psychological variables and of personal threat more than of epidemiological variables and of threat to society.

These results are interesting because they show that the COVID‐19 pandemic may have undesirable political effects in addition to its evident health and economic consequences. For example, in Hungary, the urgency of the pandemic gave Prime Minster Orbán the opportunity to take extraordinary power with no end date and the certainty of public support. In Italy, this event was applauded by Matteo Salvini (the leader of the League, an Italian right‐wing extremist party), and some commentators highlighted the advantages of managing the crisis in a nondemocratic state like China. Our findings substantiate these observations by revealing that in the COVID‐19 crisis, people became more open to anti‐democratic leaders independent of their initial predispositions toward such leaders. Moreover, this study contributes to the literature on the relation between threat and support for anti‐democratic political systems. In line with previous studies showing that living situations involving societal threats lead people to endorse potentially anti‐democratic preferences and behaviors (e.g., Asbrock and Fritsche, 2013), our study highlighted that personal threats may also lead people to favor anti‐democratic political systems, independent of their predisposition toward a strong leader.

When we performed our survey, the COVID‐19 emergency was almost uniquely the focus of the Italian mass media, which systematically reported dramatic information about the growing numbers of the infected, the hospitalized, and the dead. Emblematic pictures, such as those showing dozens of coffins piled in the Bergamo Cathedral, went viral and shocked the Italian public. Based on Ben‐Zur, Gil, and Shamshins (2012), we believe that a direct exposure to COVID‐19 produced a primary traumatization and that those who remained healthy and did not have infected individuals in their social networks suffered a secondary traumatization. Russo et al. (2020) showed that secondary traumatization fosters support for anti‐democratic political systems. Our study showed that primary traumatization incontrovertibly adds to secondary traumatization, which supports the relevance of the link between personal threat and support for anti‐democratic political systems.

Our findings raise several new questions that could be tackled in future research aimed at addressing the limitations of this study. First, we performed our study in Italy, which paid one of the highest prices in terms of health, economy, and lifestyle because of the COVID‐19 pandemic. Moreover, Italy is traditionally characterized by low civic sense and low institutional trust (Gasperoni, 2013), and in the last decades has become one of the European countries where the populist parties have become most successful (Blokker and Anselmi, 2019). Future studies might perform a cross‐national replication of this research in countries less severely impacted by COVID‐19 and with different political cultures, as it is plausible that the rise of support for anti‐democratic political systems could translate into actual anti‐democratic behaviors and votes as a function of political culture and the efficacy of political measures used to tackle the effects of the pandemic (Klingemann, 2018). Second, it is unclear how long the effects we detected will persist. A new wave of the study, performed when the most severe phase of the COVID‐19 pandemic is over, could undoubtedly add new knowledge regarding the dynamics we have studied.

We believe that the limitations of our study are counterbalanced by some strong contributions. First, mortality salience is typically analyzed in the context of lab studies, where participants are asked to think about their own death or their own death is subliminally primed (e.g., Pyszczynski et al., 1996). Our field study focused on real‐world existential threats, which increases its ecological validity. Second, our longitudinal approach allowed us to control for level of economic threat and initial predispositions toward anti‐democracy, which ensures that the effects observed are actually due to threats related to COVID‐19.

In conclusion, we showed that an authoritarian personality is not a necessary precondition for individual anti‐democracy: when facing severe personal threats, anyone could restore a subjective sense of control over the social world by becoming anti‐democratic. People tend to react to severe personal threats in the same way, independent of their initial predisposition to support anti‐democratic political systems.