The unlikely president:

the populist captain and his voters

Maria Hermínia

Tavares de Almeida

mhbtdalm@usp.br

Centro Brasileiro

de Análise e Planejamento

 

Fernando Henrique Guarnieri

fhguarnieri@gmail.com

Universidade do Estado

do Rio de Janeiro

 

Brasil

 

Recibido: 04/01/2020

Aprobado: 25/02/2020

 

Resumen

En este artículo utilizamos datos de encuestas para comprender el electorado de Bolsonaro y, al hacerlo, discutimos interpretaciones significativas sobre las condiciones del surgimiento de los políticos populistas. En la primera sección, presentamos y discutimos interpretaciones macro que recurren a procesos de cultura política o socioeconómica para explicar el populismo. En la segunda parte, describimos brevemente la carrera política de Jair Bolsonaro y mostramos por qué es un miembro auténtico de la familia populista. En las secciones tercera y cuarta, presentamos datos de encuestas sobre los votantes de Bolsonaro, sus características socioeconómicas y sus creencias con respecto a los regímenes políticos y las instituciones democráticas, y las cuestiones morales en conflicto. Argumentamos que los datos muestran que las bases sociales del presidente no son las que la literatura actual relaciona con los líderes populistas y que sus actitudes hacia la política y las cuestiones morales no difieren significativamente de las que se encuentran entre los partidarios de otros candidatos que se postulan para la presidencia en 2018. conclusiones, sugerimos algunas explicaciones alternativas al inesperado éxito electoral de Bolsonaro.

 

Palabras clave

Electorado de Bolsonaro, Cultura Política Brasileña, Populismo.

 

 

 

Abstract

In this paper we use surveyed data to understand Bolsonaro’s electorate and, in doing so, we discuss significant interpretations about the conditions to the rise of populist politicians. In the first section, we present and discuss macro interpretations that resort either to socioeconomic or political culture processes in order to explain populism. In the second part, we describe, briefly, Jair Bolsonaro’s political career and show why he is an authentic member of the populist family. In the third and fourth sections, we present survey data on Bolsonaro’s voters, their socioeconomic features, and their beliefs regarding political regimes and democratic institutions, and contending moral issues. We argue that data show that the president social bases are not those the current literature relates to populist leaders and that their attitudes towards politics and moral issues do not differ significantly from those found among supporters of other candidates running for the presidency in 2018. In the conclusions, we suggest some alternative explanations to Bolsonaro’s unexpected electoral success.

 

Keywords

Bolsonaro Electorate, Brazilian Political Culture, Populism.

 

 

 

Introducción

In October 2018, a majority, of the Brazilian voters have chosen Jair Messias Bolsonaro, an extreme-right populist politician, president of Brazil. The election was the culmination of five years of political turmoil. It began with massive street demonstrations, in 2013, followed by a polarized presidential election in the next year, the disclosure of corruption scandals that brought down the elected president allowing for the rise of a discredited and unpopular vice-president, in 2016, amid a deep economic crisis compromising the government fiscal capacity. The crisis has buried a long-lasting political arrangement, dating from the mid-1990s, and organized around centripetal electoral competition between center-left and center-right party coalitions.

Besides bringing to power an extreme-right Messiah, the elections brought several new-comers to govern the states and produced extreme party fragmentation in Congress.

 

 

 

Current

explanations for the rise of populism

The concept of populism has a long story in the political studies, dating from the late XIX Century. Throughout this period, its meaning has changed following the different phenomena it was supposed to capture. Populist was the Jacksonian ordinary people’s democracy. Populists were the radical intellectuals searching for the real soul of Mother Russia among poor peasants. Populists came to be labeled the Latin American modernizing leaders that promoted state-led industrialization and the political inclusion of the popular masses -such as the Mexican Lazaro Cárdenas, the Brazilian Getúlio Vargas and the Argentinian Juan Domingo Perón-. Last but not least, populist were called the economic policies that did not conform to the hegemonic liberal orthodoxy and were fiscally permissive, besides allowing for redistributive policies thought to have nefarious inflationary consequences (Dornbush & Edwards,1991).

More recently, the term populism has referred to the rise of leaders and movements, either from the left or from the right, that seem to be challenging the customary workings of contemporary liberal democracies.

Different as they may be according to their political origins, trajectories, or organized support, neo-populists differ from other democratic politicians by their anti-establish­ment, anti-pluralism, anti-elite rhetorics. Challenging the status quo, they present themselves as the unique and authentic representatives of the people whose genuine aspirations and traditional values they claim to exclusively embody. The claim to exclusive representation is not an empirical one; it is always distinctly moral (Müller, 2017: 47). Since the people are understood as a unique and unified social body, its recognition entails the existence of those who do not belong to it, the anti-people: the elites and the foreigners of any kind (Muller, 2017, Mounk, 2018, Mudde & Kaltwasser, 2017, Filchenstein, 2019).

The rise of contemporary populism would be related to political polarization and the shrinking and the weakening of the political center in different societies, either by the radicalization of existing political parties or by the emergence or strengthening of extremist parties and movements (Abramowitz, 2010, Donahue & Heck, 2019, Mann & Ornstein, 2018, Ginsburg & Huq, 2018, Levistky & Zimblatt, 2018). According to Moffit (2016), populism is a political style whose global incidence was made possible by changes in the technologies of communication and by the increase and scope of information networks.

Political scientists and pundits agree that are tense the relations between populism and the democratic regime and see the rise of populist leaders as the sign of democracy’s discontent, if not of crisis. However some argue that populism contributes to balance representative institutions’ inherently elitism and class-biased (Laclau, 2005, Mouffe, 2019).

Although the definition of populism and its relationship to polarization is consensual among scholars and analysts, there is a substantial discussion regarding what favors the ascent of populist leaders and movements around the world.

Risking oversimplification, we can, for analytical purposes, classify current explanations into three groups that sometimes overlap.

The first, see populism expressing the anger and fears of ordinary citizens displaced by the economic changes brought about by globalization and the rapid technological change entailed by it (Rodrik, 2018). Support for populism would come typically from the characters similar to those of the praised documentary American Factory: white American blue-collar workers whose jobs have disappeared with the reallocation of firms or technological innovations. Or from their European equivalents whose jobs are coveted -and supposedly “stolen”- by non-European immigrants. Or even, from large sectors of the middle classes from the developed Western World whose income, according to Milanovic (2010) have decreased or stagnated as the result of globalization, while the conspicuous concentration of income and wealth grew remarkably. Another version of this argument emphasizes the unequal and powerful impacts of the 2008 global crisis (Judis, 2016)

The central idea in those different versions of the same script is one of the people feeling displaced by sudden economic shifts and forgotten by the political establishment (Hochschild, 2016, Donahue & Heck, 2019, Judis, 2017). Economic insecurity, fear regarding the future, rage against callous economic and political elites have been a recurrent explanation of Trump’s surprising electoral success in 2016 among scholars and pundits (Vance, 2018).

The second set of explanations dwells into cultural cleavages focusing on values, beliefs, and on attitudes regarding race/ ethnic groups, religion or traditional moral values. Con­servative reactions towards secularization, internationalization, multiculturalism and gender politics would supposedly threaten the lifestyles of traditional national communities, elusive as they may be (Norris & Inglehart, 2019, Lilla, 2017, Siles, Tesler & Vevreck, 2017, Major, Blodorn & Major Blascovich, 2016, Green & McElwee, 2019).

The third set emphasizes the representation gap that seems to be part and parcel of the institutionalized democratic politics (Frank, 2016). According to those views, liberal technocracy’s extensive powers and increased distance between parties and/or professional politicians and their constituencies would open space for populist leaders speaking straight to the hearts of ordinary voters. In this sense, Müller (2018: 101) correctly remarks that populism is the permanent shadow of representative politics.

In this paper, we acknowledge the importance of previous political polarization in Brazil, dating from at least 2013, to favor the ascent to the presidency of a backbencher that did not have either the ear or the respect of his fellows at the Congress or significant political support from voters.

On the other hand, we aim at contributing to the discussion about populism based on the case of Brazil. We will use survey data to assess the weight of socioeconomic features, political attitudes, religion, and moral values on electoral preference for Jair Bolsonaro among Brazilian voters. We will also try to describe in those terms the core group of his most faithful supporters.

We argue that explanations linking the rise of extreme right populist politicians to macro processes such as globalization effects on jobs and income, downward mobility, cultural backlash or representative democracy crisis seem not to fit the survey data we have from Brazil. They show that Bolsonaro voters were far from being economic losers and that their political opinions and values are not that different from political opinions and values of those that have chosen other candidates. Therefore, we tend to support explanations pointing to contingent factors related to voter’s decisions under political and economic crises. Before that, we argue why populist is a suitable label to describe Jair Bolsonaro’s political style.

 

 

 

An accidental

populist president

In August 2018, Bolsonaro had the preference of 23% of the Brazilians, according to electoral polls, and very few people thought he could go much further. Against all the odds, two months later, he was the 38th elected President of Brazil, receiving 55,13% of the valid ballots against 44,87% of his opponent.

His was not a career leading to the presidency. Jair Messias Bolsonaro graduated, in 1977, from the Military Academy and served the Army in the artillery and paratroops units where he came to be known as an unruly officer. In 1986, he was arrested and detained for 15 days, after publishing an article at a weekly magazine criticizing the officers´ low salaries. One year later, he was accused of planning to plant bombs in military units to protest once more against the military salaries. After been convicted by a military court he has been acquitted by the Brazilian Supreme Military Court in 1988 and sent to retirement with the rank of captain (Maklouf, 2019).

In the same year, he was elected to the Rio de Janeiro City Council and two years later as one of the state’s representatives to the National House, to where he has been re-elected six times.

Throughout his 27 years career in Congress, he has been an obscure backbencher who never presided any of the House committees and had only one law proposal approved. At the House, he came to be known as a bizarre outspoken politician that championed for the military interests, praised the 1964 military coup, the ensuing authoritarian regime, and advocated torture against the democratic opposition. He also stood up for a deeply conservative agenda, opposing same-sex marriage and gay rights, abortion, affirmative action, drug liberalization, and arms control.

During his political career, Bolsonaro changed parties six times and became the presidential candidate of a small party, the PSL (Social Liberal Party), that in 2018 had only two representatives at the Congress.

In Rio de Janeiro, he and three of his sons that have followed his steps in politics are known for their close connections to private militia groups that exert control over significant portions of the poorest neighborhoods.

Friend to well-known evangelical protestant leaders, he has been married to his third wife by a minister of the largest neopentecostal denomination in Brazil -Assembleia de Deus- in 2013, and took baptism vows from another pastor, at the Jordan river, during a visit to Israel, in 2016.

A few months before, he had announced his decision to run for president, arousing contempt and disbelief. Eventually, he joined the PSL and launched his presidential campaign in 2018, having a right-wing retired military in his slate as vice-president and counting on the support of other few but prominent retired generals.

Bolsonaro presented himself not only as of the alternative to the leftist Workers Party (Partido dos Trabalhadores) but also to the “corrupt” political elite that had plunged Brazil into an economic and moral crisis. Since long a professional politician, he introduced himself as an outsider running against the elites, the political establishment, and the “rotten” party system. His motto was vaguely nationalist and strongly religious: Brazil above everybody, God above everything.

Family values and security issues were at the core of his political rhetoric. Nevertheless, he also championed liberal pro-market economic policies. He picked up an ultra-liberal economist operating in the financial markets, with a PhD degree from Chicago, as his advisor and promised future Minister of Economy.

Anyway, he did not have the opportunity to detail his proposals since he was stabbed, in September, while campaigning and therefore spared from participating in any presidential debates and network interviews. A blessing, taking into account his limited intellectual and rhetorical skills.

Bolsonaro’s TV propaganda was quite uncouth. However, he used social media very effectively to convey himself as an ordinary, simple man confronting the political elites, the leftist cosmopolitan intellectuals, and, above all, the PT “communists”. It worked. By now, we should ask who were the voters that gave him the presidency.

 

 

 

Core voters and electoral landslide

We have used electoral polls series from the institute Datafolha to better understand the profile of Bolsonaro’s voters. We have focused on the surveys done in June, just before the campaign started, and one day before election day in October.

Table 1 shows the results of logistic regression of the voting intention for Bolsonaro in June 2018 and October 2018 on a series of sociodemographic and attitudinal variables. The regression coefficients allow us to examine the likelihood that different social profiles would declare their intention to cast a vote for him.

 

Table 1. Results of logistic regression for prediction of voting intention in Bolsonaro

in the begining of electoral campaign (june 2018) and one day before election day (first round, october 2018)

 

 

Voting intention in Bolsonaro

 

(07/06/2018)

(03/102018)

(Intercept)

-0.34 (-0.80, 0.11)

0.31*** (0.11, 0.50)

age 25 to 34 years old

0.13 (-0.17, 0.43)

0.21*** (0.06, 0.35)

age 35 to 44 years old

-0.57*** (-0.90, -0.25)

-0.01 (-0.16, 0.14)

age 45 to 59 years

-0.87*** (-1.20, -0.53)

-0.05 (-0.19, 0.10)

age 60 years or older

-0.87*** (-1.24, -0.51)

-0.0001 (-0.15, 0.15)

women

-0.97*** (-1.18, -0.76)

-0.57*** (-0.66, -0.49)

Income 2 to 3 minimum wages (m. w.)

0.36** (0.08, 0.64)

0.64*** (0.52, 0.76)

Income 3 to 5 m. w.

0.85*** (0.57, 1.13)

1.09*** (0.97, 1.21)

Income 5 to 10 m. w.

0.96*** (0.63, 1.29)

1.32*** (1.17, 1.46)

Income 10 to 20 m. w.

1.26*** (0.67, 1.85)

1.19*** (0.97, 1.41)

Income 20 to 50 m. w

1.67*** (0.44, 2.90)

2.09*** (1.64, 2.53)

Income more than 50 m. w.

-11.36 (-593.71, 570.98)

1.98*** (1.12, 2.84)

Black

-0.31*** (-0.54, -0.08)

-0.32*** (-0.41, -0.22)

Yellow

0.11 (-0.45, 0.67)

0.30** (0.05, 0.56)

Indigenous

-0.35 (-1.06, 0.35)

0.07 (-0.25, 0.40)

Other

-0.24 (-0.79, 0.30)

-0.44* (-0.89, 0.01)

Evangelical Pentecostal

0.13 (-0.23, 0.48)

-0.27*** (-0.42, -0.11)

Evangelical Neo Pentecostal

0.18 (-0.49, 0.85)

-0.22 (-0.53, 0.09)

Other Evangelical

0.17 (-0.54, 0.88)

-0.15 (-0.51, 0.21)

Umbanda, Candomblé

or other Afro-Brazilian religions

-1.59*** (-2.75, -0.44)

-1.56*** (-1.95, -1.17)

Kardecist spiritist, spiritualist

-0.91*** (-1.58, -0.25)

-0.76*** (-1.00, -0.52)

Catholic

-0.43*** (-0.73, -0.13)

-0.77*** (-0.89, -0.65)

Jewish

0.49 (-1.19, 2.18)

-0.60 (-1.40, 0.20)

Other religion

-0.24 (-1.10, 0.63)

-0.84*** (-1.22, -0.46)

Has no religion / agnostic

-0.40* (-0.84, 0.03)

-0.91*** (-1.12, -0.71)

Atheist / doesn’t believe in God

-0.74 (-1.83, 0.35)

-2.17*** (-2.75, -1.59)

South

0.10 (-0.20, 0.39)

0.22*** (0.10, 0.35)

Northeast

-0.46*** (-0.75, -0.18)

-0.66*** (-0.78, -0.55)

Midwest

0.40** (0.03, 0.77)

0.04 (-0.13, 0.20)

North

0.25 (-0.12, 0.63)

0.22*** (0.06, 0.39)

Democracy is the best

form of government

-0.35*** (-0.56, -0.14)

-0.63*** (-0.72, -0.53)

 

Observations

2,824

10,930

Log Likelihood

-1,241.82

-6,412.63

Akaike Inf. Crit.

2,549.63

12,891.25

Note:

*p<0.1; **p<0.05; ***p<0.01

Source: Datafolha

In both moments, Bolsonaro’s core voter -who would be more likely to choose him- was a young man (25 to 34 years old), white, with high family income between 20 and 50 minimum wages, evangelical and resident of the North, South and Midwest regions of Brazil. The likelihood that someone with this social profile declared his intention to vote for Bolsonaro was 89.5%, in June 2018, while the overall average was 19%. In October 2018, the probability rose to 95%, v.g., almost every person of that group would vote for the PSL candidate.

The social type least likely to vote for Bolsonaro was a woman aged 45 to 59 years old, black, with a family income under two minimum wages, umbandista, self-employed, living in the Northeast Brazil. In June 2018, only 1,5% of people with this profile would vote for Bolsonaro. In October, the probability had risen to 10%, an impressive increase, but still well bellow the average voting intention at that moment (35%). It is needless to say that both groups represented a small part of the population and the electorate.

The six characteristics we have chosen in order to model the vote for Bolsonaro (sex, age, income, religion, region and color/race) allow 24.750 combinations. Some profiles are non-existent in the real world, and others are actually found. Therefore, although we can say which profile was more or less likely to declare his/her intention to vote for Bolsonaro, we cannot say that he has won due to the support of any specific group. What the data show is that both contextual and demographic factors help explain Bolsonaro’s victory. On one hand, since a higher income is directly related to preference for the extreme-right candidate and the Northeast region support for him was lower, we can speculate that his bad electoral performance -and a greater support for Fernando Haddad -in that region were due to PT governments public policies aimed at poor constituencies. On the other hand, in demographic terms, the higher propensity of blacks and women to reject the right-wing candidate points out to the strength of the identity agenda in these elections.

Core voters made Bolsonaro a competitive candidate but were not sufficient to secure victory. It eventually arrived as a result of the electoral competition, as voters increasingly came to prefer him to other center-to-the-right candidates.

We have examined nine waves of polls, carried out by Datafolha before the first round, to see how different segments of voters changed their propensity to choose Bolsonaro. We have compared the coefficients of the same regression model applied to the nine waves. These coefficients correspond to categorical variables. Therefore, we will see how the categories of a particular variable behave in relation to a reference category.

Using the coefficients of the complete model we can calculate the evolution of the predicted probability of different voter profiles to choose Bolsonaro. Figure 1 shows the evolution of the probability of Catholic voters with an income below two minimum wages according to sex, color, and region. In all the charts, we can see a jump starting in September (after the beginning of the TV campaign and the stabbing episode) and another jump in the second ballot. The assassination attempt brought greater visibility to the PSL candidate, which may explain the first leap. In the second ballot, the jump can be explained by antipetism (Samuels and Zucco, 2018).

Between the jumps, there are no significant changes, making the graphics take on the appearance of a staircase with three steps. This shape may indicate the absence of a contagion process. The idea of contagion, dear to several Brazilian analysts, denotes that Bolsonaro’s victory was due to the successful use of social networks allowing the support for Bolsonaro spread like a virus. When there is contagion, the increases are gradual and the graph shows a ramp shape. This is not what happened here.

The graphs reaffirms the differences between the Southeast and Northeast regions, between men and women, blacks and whites. It is interesting to note that color/race, gender, and region interact to make the northeast the region where Bolsonaro got his lowest support rate. Among low-income voters, almost the majority of black men in the southeast and white men in the northeast supported the extreme-right candidate. Black or Northeastern women, on the other hand, always gave little support to him.

 

 

Figure 1. Evolution of predict probabilities of vote in Bolsonaro for low-income Catholic voters

from a logistic regression model

 

 

Imagen2105.PNG 

Source: Logistic regression model runned in nine round of Datafolha surveys

(07/06/2018 ,21/08/2018, 19/09/2018, 28/09/2018, 02/10/2018, 04/10/2018, 18/10/2018, 25/10/2018, 27/10/2018)

 

 

 

Figure 2 shows the predicted probability of voting for Bolsonaro among neopentecostal evangelical voters with an income below two minimum wages, according to sex, color, and region. We see that being evangelical attenuates the effect of gender and region, except for black women from the Northeast. The ladder in Figure 2 is already starting to look more like a ramp, which could indicate contagion as we will see by analyzing the profile of those earning between 20 and 50 minimum wages.

 

 

Figure 2. Evolution of predict probabilities of vote in Bolsonaro for low-income New-Penthecostal voters

 

 

 

9299.png 

Source: Logistic regression model runned in nine round of Datafolha surveys

(07/06/2018 ,21/08/2018, 19/09/2018, 28/09/2018, 02/10/2018, 04/10/2018, 18/10/2018, 25/10/2018, 27/10/2018)

 

 

Figure 3 shows what happens with neopentecostal evangelical voters with high income, between 20 and 50 minimum wages. Now the ramp shape is definite, indicating contagion among the wealthiest, which can be probably explained, at least in part, by the use of social networks.

 

Figure 3. Evolution of predict probabilities of vote in Bolsonaro for high-income New-Penthecostal voters

 

 

 

9356.png 

Source: Logistic regression model runned in nine round of Datafolha surveys

(07/06/2018 ,21/08/2018, 19/09/2018, 28/09/2018, 02/10/2018, 04/10/2018, 18/10/2018, 25/10/2018, 27/10/2018)

 

Interestingly, the same contagion effect appears with Catholics, challenging the interpretations that stress the importance of the evangelical church for Bolsonaro’s election, at least among the wealthier. Another interesting finding is the inflection of the candidate’s support curve in the second round, common to all groups analyzed here.

The data shows that an array of factors accounted for Bolsonaro’s election. Identity issues related to gender and race, retrospective voting, political factors interacted to convey the candidate’s support. The non-evangelical, the poor, and the lower middle class, black and northeastern women, were the bastion of resistance to Bolsonarism. High-income men and women supported him. The PT incapacity to gain the majority of poor whites in the northeast and poor blacks in the southeast, especially if they were evangelicals, may have meant the difference between victory and defeat.

In the end of the day, almost fifty-seven million eight hundred Brazilians voted for Bolsonaro in the second ballot giving him an advantage of 10.67 percentual points over the PT candidate.

How different were Bolsonaro voters’ opinions from those who have chosen Haddad or other candidates in the first round?

 

 

 

Brazilian voters’

political attitudes and moral values

In order to answer that question, we have used data from LAPOP 2019 regarding different political and moral issues to compare attitudes of those who have declared voting for Bolsonaro and other candidates in the first election’s round.

Table 2 shows that around 60% of respondents think that although democracy has problems, it is still the best form of government .

There is no significant difference between the two groups of the electorate: 61% of Bolsonaro’s voters and 64% those that had other candidates value the democratic regime positively.

On the other hand, Brazilians, in general, are not very satisfied with the way democracy works in their country: around 70% say they are dissatisfied or very dissatisfied. Bolsonaro’s voters show more satisfaction with democracy than others: 48% versus 36% (Table 3). This result is not surprising: since their preferred candidate won the election, the system reflects their will. Finally, 44% of the extreme-right voters agree that those who govern the country show interest in what people like them think, against 29% of those who voted for the losers.

 

Table 2. Proportion of respondents who believe that democracy is the best form of government among those who voted and did not vote for Bolsonaro

 

Democracy is the best form of government

I feel that the Government hear me

 

Others

Bolsonaro

Others

Bolsonaro

Disagree very much

0.08

0.07

0.26

0.18

Disagree

0.03

0.03

0.13

0.08

Disagree a little

0.08

0.09

0.16

0.14

Indiferent

0.18

0.19

0.14

0.17

Agree a little

0.22

0.19

0.14

0.18

Agree

0.14

0.16

0.08

0.12

Agree very much

0.28

0.26

0.09

0.14

Total

1.00

1.00

1.00

1.00

Source: LAPOP 2019

 

 

Table 3. Proportion of respondents who are satisfied with democracy

among those who voted and did not vote for Bolsonaro

 

Satisfaction with Democarcy

Others

Bolsonaro

Very satisfied

0.07

0.08

Satisfied

0.29

0.40

Unsitisfied

0.48

0.43

Very unsitisfied

0.17

0.09

Total

1.00

1.00

Source: LAPOP 2019

 

 

Although more satisfied with democracy, 46% of those voting for Bolsonaro think that too much crime justifies a coup d’état. Among those who voted for other candidates, the percentage is 31%. The difference between bolsonaristas and non-bolsonaristas is greater when it comes to corruption. 47% of those who voted for Bolsonaro, against 25% of those who did not, believe that a coup d’état is justified in the face of too much corruption. This difference of more than 20% is the same when people are asked if, in difficult times, the Supreme Court should be closed, although the difference drops to 10% when it comes to shutting down the Congress (29% vs. 19%). Table 4 and 5 show these results.

 

Table 4. Proportion of respondents who think a coup is justified in the case of _________ among those who voted and did not vote for Bolsonaro

 

Great Violence

Much Corruption

Others

Bolsonaro

Others

Bolsonaro

The military would be justified to take power by a coup d’etat.

0.31

0.46

0.25

0.47

The military wouldn’t be justified to take power by a coup d’etat.

0.69

0.54

0.75

0.53

Total

1.00

1.00

1.00

1.00

Source: LAPOP 2019

 

 

Table 5 - Proportion of respondents who think it is justified to close _________

in very difficult situations, among those who voted and did not vote for Bolsonaro

 

Congress

Supreme Court

Others

Bolsonaro

Others

Bolsonaro

It is justified

0.19

0.29

0.27

0.47

It is not justified

0.81

0.71

0.73

0.53

Total

1.00

1.00

1.00

1.00

Source: LAPOP 2019

 

 

It is worth emphasizing that the majority of both bolsonaristas and those who voted for other candidates, although dissatisfied with democracy, still think that it is the best form of government and mostly reject a coup or the closing of the STF or Congress.

In addition to being more satisfied with democracy, Bolsonaro’s voters, according to LAPOP 2019, are also more satisfied with political institutions. The majority (51%) say they respect the country’s political institutions as opposed to 45% of those who preferred other candidates. Table 6 shows the distribution of respondents who, on a scale of 1 to 7, spell their degree of trust in some political institutions. We consider grades 5 to 7 on the scale as trust in institutions. Conversely, confidence degrees between 1 and 3 signal mistrust.

While approximately two-thirds of respondents (63%) who voted for other candidates do not trust Congress, less than half of the Bolsonaristas (49%) have the same position. Although few respondents trust the political parties, confidence is greater among bolsonaristas (15% against 8%).

On the other hand, Bolsonaro’s voters also reveal greater confidence in security institutions such as the Armed Forces (82% vs 58%) and the Military Police (62% vs 44%).

These data indicate that, in general terms, Bolsonaro’s voters do not distinguish themselves for being particularly anti-democratic or averse to free political institutions, at least while his favorite candidate is in power. There is - and there has always been - a significant portion of Brazilians that do not praise the democratic regime or are indifferent to it and a huger one that distrusts democratic institutions. Nevertheless, dissatisfaction and distrust regarding them do not distinguish Bolsonaro’s electorate from those that casted their vote for other presidential candidates.

 

Table 6 - Proportion of respondents who trust and don’t trust selected institutions

among those who voted and did not vote for Bolsonaro.

(Lines don’t add up to 100% since they don’t include indifferent)

 

Trust

Don’t trust

National Congress

Others

0.23

0.63

Bolsonaro

0.34

0.49

Political Parties

Others

0.08

0.82

Bolsonaro

0.15

0.74

President

Others

0.30

0.59

Bolsonaro

0.76

0.13

Supreme Court

Others

0.33

0.50

Bolsonaro

0.54

0.30

Armed Forces

Others

0.58

0.29

Bolsonaro

0.82

0.11

Military Police

Others

0.44

0.39

Bolsonaro

0.62

0.24

Source: LAPOP 2019

 

 

We had shown before that Jair Bolsonaro had a group of core voters, who upheld him before the election campaign began and whose endorsement was well above average. LAPOP 2019 results confirm that this core is formed mainly by white, young men, with upper-middle-income, neopentecostal, living in the South or Midwest regions. The table with the results of logistic regression are in the Appendix 1.

Even among these most entrenched bolsonaristas there are no vast differences in opinion and satisfaction with democracy when compared to those that have chosen other candidates. White, young men, with higher income, neopentecostal residents of the South or Midwest are neither more nor less satisfied with democracy than the other Brazilians.

Nevertheless, ideological self-identification distinguishes Bolsonaro’s voters from those supporting other candidates. While close to half (49%) of the bolsonaristas say they are on the right -- ranking higher than 7 on a scale of 1 to 10, where 1 is to be on the left -- only 19% of voters of other candidates do the same as shown in Table 7.

 

Table 7. Self-classification of respondents on a left - right scale

among those who voted and did not vote for Bolsonaro

(1 = left and 10 = right)

 

Left (1) - Right (10) placement

 

Others

Bolsonaro

1

0.16

0.06

2

0.05

0.05

3

0.12

0.04

4

0.06

0.05

5

0.24

0.15

6

0.09

0.09

7

0.09

0.08

8

0.06

0.17

9

0.03

0.08

10

0.10

0.23

Total

1.00

1.00

Source: LAPOP 2019

 

 

Regardless of being bolsonarista or not, people with some features of the Bolsonaro’s core supporters tend to identify more with the right. White men with income ranging from three to four minimum wages, evangelical and non-evangelical protestants consider themselves more to the right than people with other socioeconomic profiles, both among those who voted for Bolsonaro and those who did not. Tables with the results can be found in the Appendix 2.

Answers to questions such as whether the state should reduce inequality or spend more with the poorest and if rich people should pay more and receive less are very similar for bolsonaristas and other voters for all socioeconomic profiles with the exception, as expected, of those with higher income. In brief, being at the political right here seems to be more related to moral and behavioral issues..

In indeed, while 56% of Bolsonaro’s voters disapprove of same-sex marriage (with 36% strongly disapproving), 39% of other candidates’ supporters share the same opinion (Table 8). Controlling by the vote in the first round, those who most disapproved of gay marriage are the poorest, neopentecostal evangelical, living in the North and Northeast regions.

 

 

Table 8. Position regarding same sex marriage

among those who voted and did not vote for Bolsonaro

 

Same-sex people should have the right to marry

(1 = Frimly desapprove. 10 = Firmly approve)

Others

Bolsonaro

1

0.23

0.36

2

0.06

0.04

3

0.03

0.04

4

0.02

0.03

5

0.05

0.09

6

0.04

0.04

7

0.07

0.04

8

0.06

0.04

9

0.06

0.07

10

0.39

0.25

Total

1.00

1.00

Source: LAPOP 2019

 

 

On the other hand, there are no significant differences between bolsonaristas and other voters regarding abortion, when the mother risks dying. Nearly 80% of respondents agree that mothers can have an abortion if childbirth poses a risk to their health (Table 9).

There are also no significant differences in the approval of government support for immigrants, even if Venezuelans (LAPOP has introduced some experiments to see the reaction to immigration when it comes to Venezuelan). 65% of respondents are in favor of governmental aid to immigrants, regardless of their vote preference.

 

 

Table 9. Position regarding abortion in case of the risk of death of the mother among those who voted and did not vote for Bolsonaro

 

Abortion is justified

in case of risk to mother health

Others

Bolsonaro

Disagree

0.23

0.26

Agree

0.77

0.74

Total

1.00

1.00

Source: LAPOP 2019

 

 

If dissatisfaction with the economic situation made the European voters turn towards populism, the opposite seems to have happened here. Table 10 shows that when asked about the country’s situation in the last two months, 32% of bolsonaristas think it has improved against 15% of those who voted for other candidates.

 

Table 10. Perception regarding country’s situation in the last twelve months among those who voted and did not vote for Bolsonaro

 

Economic situation of the country

in the last 2 months

Others

Bolsonaro

Better

0.15

0.32

Same

0.48

0.43

Worst

0.38

0.26

Total

1.00

1.00

Source: LAPOP 2019

 

 

Only 30% of respondents say they have any or too much interest in politics. 23% said they sympathized with a party, 45% of those with the PT, 7% with the PSDB, and the same proportion with the MDB. These numbers are in line with research on party identification in Brazil. The surprise is that 20% of those who said to sympathize with any party has pointed out to the PSL, Bolsonaro’s party, until recently an unknown and insignificant political organization.

Among those who voted for Bolsonaro, anti-PT attitudes are more robust, as expected. On a scale ranging from 1 to 10 -where 1 means does not like the party at all- the average PT score is 4 for the whole sample. Among the bolsonaristas, the score drops to 2,4. In comparison, the PSDB’s average score was 3,6 and MDB’s was 3,3.

 

 

 

Preliminary

conclusions

In brief, Bolsonaro seems to have conquered a niche of conservative supporters regarding moral values, mobilizing people that identify themselves with the political right, support law and order policies, are critical to the PT but are not anti-systemic or specially disaffected towards democracy. Voters that, under different circumstances, could have chosen other conservative candidates not necessarily from the extreme-right.

In October 2018, the newspaper Folha de S.Paulo, using survey data from Datafolha, arrived at a similar conclusion:

 

Despite having chosen Jair Bolsonaro (PSL) as president on Sunday (28), the Brazilian population is opposed to the retired Army captain on central issues of his governmental proposals or his rhetoric: freedom for arms possession by civilians; criticism of homosexuals; and the defense of the military dictatorship. (Seto, 2018)

 

How to reconcile those findings with the image of a polarized political scenario? How can we explain that supporters of the current extreme-right populist president value democracy, trust institutions and identify themselves with parties as much as other voters? How can we understand that such voters have chosen a president that routinely displays anti-democratic behavior and rhetorics and jeopardizes fundamental rights?

We think that the literature on populism or on the rise of the extreme right in Europe helps us little.

Conventional wisdom has pointed out to the importance of anti-system feelings, to the emotional rejection of politicians and parties leading to the choice of the candidate that presented himself as an outsider.

On the other hand, in the academic field, there are some alternative theories. One is antipetism (Samuels & Zucco, 2018): people voted against the party that ruled Brazil during 14 years. Another one rests on the literature on voting behavior and may give us some interesting hints. Both the directional theory of voting (Rabonowitz & Macdonald, 1989) and the discount theory (Grofman, 1985; Adam, Merril and Grofman, 2001; Tomz and Van Howeling, 2008) argue that, under certain circumstances, it is rational for relatively moderate voters to support more radical candidates. In presidential regimes, in times of great conflict between the executive and the legislature, or when exogenous restrictions, such as economic crises or pressure from external agents, limit the government’s range of maneuver, voters discount the Executive’s ability to implement its proposals. Therefore, they will accept a radical agenda aware that it is probably unfeasible. Empirical evidence shows that this occurred in France in 1988 and Norway in 1989 (Adam, Merril, Grofman, 2001) and, more recently, experiments confirmed that moderate or poorly politicized voters tend to discount the candidate’s radical promises (Tomz & Van Howeling, 2008) .

In Brazil, between the mid-1990’s and the mid-2010’s, the executive and legislative cooperated to solve what was then seen as the country’s most pressing problems: inflation and poverty. Therefore, the voters behaved according to the Downsian model, choosing the proposals closest to their ideal point. After the 2013 events, political polarization has increased, cooperation between the executive and the legislature deteriorated, and there has been little progress in solving problems that came to be considered severe, such as unemployment and corruption. The average voter may have started to discount candidates’ ability to implement solutions and support more extreme proposals. In Bolsonaro’s election, many of his voters could have put his most extreme statements into perspective.

In this article we cannot test the hypothesis suggested by those theories. However, we can ask ourselves about the consequences of this type of electoral behavior in comparison to the theory of populism. Only time will tell which of these models better explain what has happened.

 

Bibliographic

references

Abramowitz, A. (2010) The disappearing center - Engaged citizens, polarization and American Democracy, New Haven, Yale University Press.

Donahue, K. & Heck, P. (2019) Cycles of hatred and rage - What Right-wing Extremists in Europe and their parties tell us about the US, Switzerland, Palgrave - Macmillan

Dornbusch, R. & Edwards, S. (1991) “The Macroeconomics of Populism”. In Dornbusch, R. & Edwards, S. (ed.) The Macroeconomics of Populism in Latin America, Chicago, University of Chicago Press.

Eichengreen, B. (2018) The populist temptation - Economic grievance and political reaction in the modern Era, Oxford, Oxford University Press.

Finchelstein, F. (2019) From fascism to populism in history, Oakland, California University Press.

Frank. T. (2016) Listen liberal: what happened to the party of the people?, Australia, Scribe Publishing.

Ginsburg, T. & Huq, A.Z. (2018) How to save a constitutional democracy, Chicago, University of Chicago Press.

Green, J. & McElwee, S. (2019) “The differential effects of economic conditions and racial attitudes in the election of Donald Trump”, Perspectives in Politcs, Nº 17, pp. 358-379.

Hochschild, A.R. (2016) Strangers in their own land: anger and mourning on the American Right, New York, The New Press.

Judis, J. (2016) The populist explosion: how the great recession transformed American and European politics, New York, Columbia Global Reports.

Krastev, I. (2013) Democracy disrupted, Pennsylvania, Penn University Press.

Laclau, E. (2005) La Razón Populista, Fondo de Cultura Económica, Buenos Aires.

Levitsky, S. & Ziblatt, D. (2018) How democracies die, New York, Crown Publishing.

Lilla, M. (2017) O progresssista de ontem e o do amanhã, São Paulo, Cia. das Letras.

Major, B., Blodorn, A., & Blascovich, G. (2018). “The threat of increasing diversity: Why many White Americans support Trump in the 2016 presidential election”, Group Processes & Intergroup Relations, Nº 21, pp. 931-940.

Mann, T. E. & Ornstein, N.J. (2018) It’s even worse than it looks was: How the American constitutional system collided with the new politics of extremism, New York, Basic Books.

Norris, P. & Inglehart, R. (2019) Cultural Backlash - Trump, Brexit and Authoritarian Populism, Cambridge, Cambridge University Press.

Milanovic, B. (2010) The Haves and the Have-Nots: A Brief and Idiosyncratic History of Global InequalityBasic Books, New York.

Moffit, B. (2016) The global rise of populism: performance, political style and representation, Stanford, Stanford University Press

Mouffe, C. (2019) Por un populismo de izquierda, Buenos Aires, Siglo XXI.

Mounk, Y. (2018) The people vs democracy, Cambridge University Press.

Mudde, C. & Kaltvasser, C.R. (2017) Populism: a very short introduction, Oxford University Press, New York.

Müller, J.W, (2017) What is Populism?, Penguin Books/ Random House.

Rodrik, D. (2018) Populism and the Economics of Globalization, Copy at http://j.mp/2sowhXj.

Samuels, David J. Zucco, C. (2018) Partisans, antipartisans, and nonpartisans: Voting behavior in Brazil. Cambridge University Press.

Seto, G. (2020) “Eleitorado discorda de Bolsonaro sobre armas, ditadura e críticas a homossexuais, diz Datafolha”. Folha de São Paulo, 2018. Available in https://www1.folha.uol.com.br/poder/2018/10/eleitorado-discorda-de-bolsonaro-sobre-armas-e-criticas-a-homossexuais-diz-datafolha.shtml. Acessed march 31st, 2020.

Traverso, E. (2017) Las nuevas caras de la derecha, Argentina: Siglo XXI.

Vance, J. D. (2018). Hillbilly elegy: a memoir of Family and Culture in crisis, New York, Harper.

 

Anexo 1

Vote for Bolsonaro

according to sociodemographic and economic characteristics

 

 

Vote for Bolosnaro

Age

0.013 (0.016)

Sex (Woman)

-0.668 (0.448)

Income 2 a 3 SM

0.309 (0.662)

Income 3 a 4 SM

0.189 (0.901)

Income 4 a 5 SM

-0.114 (0.851)

Income > 5 SM

0.227 (0.948)

Indigenous

-1.057 (2.025)

Black

-0.511 (0.664)

Brown

-0.459 (0.564)

Other

0.201 (2.162)

Yellow

-0.709 (1.073)

Northeast

-0.934 (0.894)

Middle-West

-0.075 (1.112)

Southeast

-0.089 (0.858)

South

0.085 (1.032)

Protestant, Traditional Protestant

or Non-Evangelical Protestant

0.469 (0.741)

Non-Christian Eastern Religions

1.039 (4.089)

None (Believe in a Higher Self

but do not belong to any religion)

-0.393 (0.755)

Evangelical and Pentecostal

1.025* (0.617)

Traditional Religions

-0.929 (1.735)

Agnostic or atheist

(does not believe in God)

0.447 (1.628)

Other

-0.309 (1.757)

Kardecist Spiritist

-0.590 (1.393)

Prty Id

0.022 (0.513)

Constant

0.338 (1.206)

Observations

945

Note: *p<0.1; **p<0.05; ***p<0.01

Anexo 2

 

Ideological self-positioning

according to socioeconomic characteristics and vote

 

 

Position on the left-right scale

Age

0.013 (0.020)

Woman

-0.529 (0.554)

Income 2 a 3 SM

0.451 (0.810)

Income 3 a 4 SM

0.722 (1.086)

Income 4 a 5 SM

0.656 (1.083)

Income > 5 SM

0.733 (1.132)

Indigenous

0.556 (2.582)

Black

-0.599 (0.822)

Brown

-0.305 (0.700)

Other

0.184 (2.574)

Yellow

-0.669 (1.351)

Northeast

-0.240 (1.121)

Middle-West

0.037 (1.396)

Southeast

0.0001 (1.072)

South

-0.532 (1.270)

Protestant, Traditional Protestant

or Non-Evangelical Protestant

0.453 (0.948)

Non-Christian Eastern Religions

2.043 (5.232)

None (Believe in a Higher Self

but do not belong to any religion)

-0.138 (0.941)

Evangelical and Pentecostal

0.461 (0.728)

Traditional Religions

-0.061 (2.029)

Agnostic or atheist

(does not believe in God)

0.496 (1.967)

Other

1.423 (2.182)

Kardecist Spiritist

0.816 (1.847)

Vote for Bolsonaro

1.491** (0.582)

Constant

4.891*** (1.552)

Observations

894

Log Likelihood

-2,180.597

Akaike Inf. Crit.

4,411.194

Note: *p<0.1; **p<0.05; ***p<0.01