Demographic polarization and the rise of the far right: Brazil’s 2018 presidential election

Abstract

Does the recent electoral success of far-right populists represent a mere rejection of the political and economic status quo, or has it revealed deeper cultural divides? Historically, demographic cleavages have been poor predictors of vote choice and partisanship in Latin America. However, during Brazil’s 2018 presidential election campaign, right-wing candidate Jair Bolsonaro fomented conflict across lines of gender, race, and religion. We argue that his candidacy activated latent, previously unexploited grievances in the electorate. Using survey data from an original five-wave online panel conducted between July 2018 and January 2019, we examine the effect of demographic cleavages on presidential vote choice. In stark contrast to prior elections, we find clear evidence of demographic divides in 2018, partially mediated by issue positions. Bolsonaro’s campaign and subsequent election thus appear to have created new identity-based alignments in Brazil’s electorate. Our findings shed further light on the global resurgence of the far right, suggesting that far-right candidates can attract new bases of support through demographic polarization, exploiting differences in values and issue preferences by gender, race, ethnicity, and religion.

Keywords

Electoral behavior Brazil far right identity politics demographic polarization

Introduction

Even against the backdrop of surging support for far-right candidates and parties across the world, the 2018 election of Brazil’s President Jair Bolsonaro came as a shock to many observers. International media highlighted the far-rightist’s penchant for offensive statements about women and LGBT, Black, and indigenous Brazilians, as well as his nostalgia for Brazil’s military dictatorship—its “biggest mistake was to torture but not kill” (Anderson, 2019). Common wisdom held that such a candidate could not defeat moderate candidates from established parties on the center-right and center-left. Yet Bolsonaro ultimately rose from relative obscurity to win 55% of the second-round vote against Fernando Haddad from the long-dominant Workers’ Party (PT). Bolsonaro’s victory thus constituted one of the most prominent shifts to the right in a recent wave of global right-wing populist victories.

What does Bolsonaro’s election reveal about the bases of support for the global right? Did the victory merely represent the rejection of incumbent political performance amid economic and political turmoil, or did it reveal emerging cultural divides in Brazil’s electorate? In contrast to other world regions, demographic cleavages have rarely been salient—or even significantly predicted vote choice—in Latin American elections (Carlin et al., 2015). However, the recent rise of the right in some countries of the region may have changed historical patterns, as far-rightists often emphasize cleavages by gender, race, ethnicity, and religion. Bolsonaro’s campaign thus represents an opportunity to examine the demographic roots of support for the far right in a context where, traditionally, such identities only weakly correlate with partisanship and vote choice.

Recent scholarship arrives at conflicting explanations of the roles of demographics, issues, and performance in Bolsonaro’s victory. Analyzing the 2018 Brazilian Electoral Panel Study (BEPS), Rennó (2020) highlights the impact of ideological issue positions; however, the study does not control for respondent racial identification. In a cross-sectional analysis of the 2018 Brazilian Electoral Study, Amaral (2020) points to anti-partisan sentiment and ideology, and secondarily religious and gender identity. By contrast, Almeida and Guarnieri (2020) draw on Datafolha cross-sectional time series data and analyses of the 2019 AmericasBarometer to argue that it was Brazil’s economic and political crises that mattered, together with demographic identities, including race.

Our study draws on an original, five-wave online panel survey of Brazilian voters from July 2018 to January 2019.¹ We show that religious and racial cleavages defined Bolsonaro’s coalition in ways that have rarely marked political competition in Brazil’s current democratic era. Demographics independently impacted vote choice, but also shaped vote choice through issue positions. Moreover, the views of Bolsonaro’s supporters (and detractors) correlate strongly with the candidate’s stated positions, countering the notion that his election simply reflected performance-based rejection of the PT. Bolsonaro’s campaign thus created new alignments in Brazil’s electorate, producing demographic polarization by exploiting differences in values and issue preferences by gender, race, ethnicity, and religion. Our findings illuminate how far-right, populist candidates build coalitions around the world.

Demographic cleavages in comparative perspective

Scholars have long theorized a link between demographic cleavages, partisanship, and vote choice in Western democracies (Carmines and Stimson, 1989; Lipset and Rokkan, 1967), but research on identity and voting behavior has become especially prominent in response to the resurgence of the right in the past decade (Norris and Inglehart, 2019; Sides et al., 2018). In the US, race, education, gender, and religion are critical for understanding Donald Trump’s 2016 victory (e.g. Sides et al., 2018). In Europe, demographics similarly impacted the reemergence of far-right populism (Lucassen and Lubers, 2012).

Yet while demographics anchor partisanship in many older democracies, Latin American voting behavior was long unmoored from ethnicity, race, gender, and religion. In their regional overview, Carlin et al. (2015) concluded that issues outweighed demographics in Latin American voting behavior—and yet demographics became salient in polarized contexts. Only recently has ethnicity emerged as a primary cleavage in some societies with large indigenous populations, like Bolivia and Ecuador (Madrid, 2012). Afro-descendants are disadvantaged throughout Latin America, yet Black movements struggle to mobilize citizens (Telles, 2014). Similarly, despite long-standing inequities, Latin American men and women differ little in partisanship, issue attitudes, and vote choice (Htun, 2016; Morgan, 2015). Although religiosity predicts voting for the right, stable religious-partisan linkages have emerged only in a few countries, such as Chile’s Christian Democratic Party and Mexico’s National Action Party (e.g. Boas and Smith, 2015). Instead, parties with lasting social bases have largely been grounded in class, as in the cases of the Institutional Revolutionary Party (PRI) in Mexico and the Peronist Party in Argentina (Roberts and Wibbels, 1999).

The absence of long-standing demographic cleavages in Latin America makes the far right’s emergence in Brazil a puzzle. In the US, partisan “sorting” by race, gender, and religion took decades (Manza and Brooks, 1999; Tesler, 2016); Trump exacerbated divisions in an already polarized electorate. But in Brazil, no such process had occurred before 2018. Did Bolsonaro’s candidacy activate previously latent racial, religious, and gender divides?

Demographics and vote choice in Brazil

Until now, demographics have weakly rooted Brazilian partisanship and elections. Despite deep racial inequalities, race had never oriented electoral politics (Telles, 2014). Samuels (2006: 17) concluded that age, race, gender, and religion seemed “relatively unimportant” in the 2002 election. A decade later, ideological preferences were still mostly uncorrelated with ethnicity and gender (Moreno Morales, 2015; Morgan, 2015). Religion’s link to partisanship was small and inconsistent after 2000 (Smith, 2019), although evangelical voters tended to support evangelical candidates (e.g. Bohn, 2007). Moreover, the base of the country’s only party with deep social roots, the PT, shifted dramatically over time: from Catholic, labor, and leftist activists (Amaral and Meneguello, 2017; Hunter, 2010); to well-educated, urban, southern men (Hunter, 2010: 114–115); to lower-status voters benefitting from social assistance (Hunter and Power, 2007). In sum, there is little evidence that race, gender, and religion drove ideology or vote choice in Brazil before 2018.

What explained voters’ choices instead? Most scholars point to personalism and performance evaluations (Hunter and Power, 2007; Moisés, 1993). More recently, Samuels and Zucco (2018: 46) highlight a PT/anti-PT divide rooted in “attitudes about democracy and the desirability of social change.” In 2018, voters had many performance-based grievances against the PT and allies: a recession, the Lava Jato corruption scandal, and the impeachment of President Dilma Rousseff.¹ In this context, voters might reasonably opt for an outsider untainted by crisis (e.g. Hunter and Power, 2019). Yet, we have doubts about this perspective. Voters in 2018 did not reject the PT for a mainstream, reform-oriented candidate, but rather a far-right firebrand. Was this desperation from an exasperated electorate, or did Bolsonaro tap into previously unexploited divisions?

To explain voters’ shift to Bolsonaro, we foreground demographics, arguing that three identities became salient in 2018: gender, race, and religion. First, Bolsonaro championed traditional, sometimes misogynistic, views on gender. He opposed gender pay equity, and once said that a fellow member of Congress “doesn’t deserve to be raped because she’s very ugly . . . I’m not a rapist, but if I was, I wouldn’t rape her because she doesn’t deserve it” (Bolsonaro, 2008). The campaign coincided with rising women’s mobilization opposing sexual assault and domestic violence (Goñi and Watts, 2016), and Bolsonaro’s advocacy for lifting restrictions on gun ownership likely troubled many women (Datafolha, 2019; Goss, 2006). In the lead-up to the first-round election, millions of women organized the #EleNão (#NotHim) protest movement, calling voters to support anyone but Bolsonaro (Darlington, 2018).

Race provided a second fault line. Throughout his legislative career, Bolsonaro made racially offensive remarks, from opposing interracial marriage to telling Black activists to “go back to the zoo.” As a candidate, he advocated militarizing predominantly Black favelas to combat crime, claiming “the only good criminal is a dead criminal” (Waldron, 2018). Bolsonaro also scorned Brazil’s system of affirmative action, which has opened educational and employment opportunities for Afro-Brazilians since 2012 (Nugent, 2018). In short, Bolsonaro’s stances centered race in political conversations and likely alienated Afro-descendants, while courting resentful White voters.

Third, Bolsonaro likely divided Brazilians by religion. Although he continues to identify by his Catholic upbringing, Bolsonaro has long attended a Baptist church and joined the legislative evangelical caucus; hence, some voters might have perceived him as evangelical. More importantly, we suspect Bolsonaro’s opposition to LGBT groups attracted evangelicals, as sexuality and the family have become pivotal issues driving religious cleavages in Brazil (Smith, 2019).²

Thus, we argue that Bolsonaro galvanized the public along demographic lines by alienating or wooing different groups, and that issue positions mediated between demographics and vote choices. If demographic identity correlated with issue stances, which in turn influenced voting, Bolsonaro’s election cannot be viewed solely as a rejection of the status quo. Rather, Bolsonaro capitalized on latent identity politics in ways that previous candidates had not. Our analysis assesses the relative weight of demographics, issue positions, and performance evaluations, shedding light on an election that constituted a turning point in Brazilian democracy.

Data and methods

We conducted an original five-wave panel study between July 2018 and January 2019 via the Qualtrics platform.³ NetQuest, an international survey provider, recruited subjects via quota sampling, collecting 6045 total responses to obtain a sample that is nationally representative in gender and region, and that met a targeted income distribution. We examine results from the first (N = 2018, collected 6–21 July 2018), third (N = 1244, 21 September—4 October), and fourth (N = 957, 29 October—8 November) waves. By measuring attitudes in September, after demographics (July), but prior to voting (retrospective choice in the 28 October 2018 runoff election), we reduce threats to inference that could result from attitudes shaping slow-moving demographics (e.g. religion), or being influenced by vote choice.

We adjudicate between contending explanations via three models of self-reported second-round vote for Bolsonaro versus Haddad: (a) a demographics-only model, then ones adding (b) retrospective evaluations, and (c) issue attitudes.⁴ Model 1 includes Gender, Ethnicity, Religion, Level of Education, Income, and Age.⁵ Model 2 adds Sociotropic and Idiotropic Evaluations and Perceptions of Political Corruption. An index taps PT Favorability by averaging a dichotomous measure of PT partisanship and a measure of warmth toward the party’s standard bearer, Lula (scaled 0–1). If backlash against the status quo explains Bolsonaro’s election, these variables should predict voting, and may reduce coefficients for demographics. Model 3 adds issue positions that likely polarized voters, assessing whether preferences mediated demographic effects. Two items from the third wave reference policies likely to activate demographic cleavages: Right to Bear Arms and Racial Resentment, and a feeling thermometer captures approval of the LGBT community.

Results

Figure 1 estimates relationships between demographics and second-round vote for Bolsonaro (versus Haddad).⁶ All else constant, women were 6 percentage points less likely than men to choose Bolsonaro (p < .10). Brown and Black self-identification decreased Bolsonaro support by 9 and 30 percentage points, respectively; evangelicalism increased it by 36 percentage points.⁷ Hence, demographics appear more important for 2018 vote choice in Brazil than prior literature would suggest.

Figure 1.

Demographic determinants of presidential voting.

Table 1 shows results from logistic regression models regressing vote choice on (a) demographics, (b) retrospective evaluations of the PT, and (c) issue attitudes.⁸ Model 1 presents the same model as Figure 1: women, Brown, and Black respondents were less likely to vote for Bolsonaro, and evangelicals more so. Education also reduced support for Bolsonaro, and income increased it, net of other factors.

Table 1.

Logistic regression models of second round vote for Bolsonaro.

	Demographics only (Model 1)	+ PT rejection (Model 2)	+ Attitudes (Model 3)
Female	–0.343*	0.059	0.743**
	(0.197)	(0.255)	(0.338)
Brown	–0.499**	–0.247	–0.105
	(0.224)	(0.290)	(0.385)
Black	–1.718***	–1.006*	–0.440
	(0.402)	(0.530)	(0.739)
Indigenous	0.375	1.801	1.297
	(1.268)	(2.120)	(3.102)
Evangelical	2.085***	2.234***	1.281**
	(0.338)	(0.420)	(0.570)
Higher education	–0.699***	–1.191***	–1.343***
	(0.217)	(0.287)	(0.368)
Income	0.619*	0.973**	1.234**
	(0.368)	(0.475)	(0.610)
PT favorability		–5.862***	–5.210***
		(0.613)	(0.763)
Perception of corruption		–1.567*	–1.843*
		(0.806)	(1.010)
Sociotropic economic evaluation		1.508***	1.415**
		(0.452)	(0.585)
Idiotropic economic evaluation		–0.317	–1.016**
		(0.359)	(0.459)
Right to bear arms			2.736***
			(0.445)
Racial resentment			1.477***
			(0.447)
LGBT feeling thermometer			–2.232***
			(0.567)
Constant	0.451	1.904**	2.418*
	(0.469)	(0.877)	(1.278)
Observations	671	636	568

Notes: Standard errors in parentheses. Controls for Asian/other racial identification, Catholic and other religion, no and primary education, age, and region not shown. Coefficients are statistically significant at * p = .10, ** .05, *** .01. All independent variables are recoded to run from 0 to 1.

PT: Workers’ Party.

Could demographic cleavages merely reflect group differences in views of the PT’s performance? The second model shows that partisan evaluations substantially affected vote choice. Partisanship has a very large effect on voting for Bolsonaro; moving from the minimum to maximum favorability of the PT, while holding other variables at their observed values, the probability of voting for Bolsonaro drops 77 percentage points (from .81 to .04).

Contrary to common wisdom, Bolsonaro apparently attracted voters who were more satisfied with the status quo vis-à-vis corruption. Corruption perceptions have a large effect contradicting expectations: Bolsonaro’s support decreases 16 percentage points (from .78 to .62) among those saying that all (versus no) politicians are corrupt. In the Online Appendix, we confirm this unexpected finding using AmericasBarometer data. Given the pervasiveness of the Lava Jato scandal and petistas’ belief that the impeachment of Dilma Rousseff was rigged, it might make sense in retrospect that perceptions of corruption did not drive voters to Bolsonaro. In addition, over a year into the interim presidency of center-rightist Michel Temer, people who evaluated the sociotropic economy more highly were 14 percentage points more likely to vote for Bolsonaro. Performance variables attenuate the coefficients for gender and self-identification as Brown; neither variable remains statistically significant. Nonetheless, Black and evangelical self-identification remain important independent determinants of the vote.

In the third model, attitudinal variables further attenuate the impact of race, such that Black identity no longer significantly predicts the vote. However, the estimated effect of evangelical religious identification remains; evangelicalism increases Bolsonaro support by 10 percentage points. Issue positions also strongly predict support for Bolsonaro, even controlling for partisanship and performance evaluations. The strongest supporters of the right to bear arms, for example, are 27 percentage points more likely to have voted for Bolsonaro than the strongest opponents (.78 versus .51). Racial resentment increases Bolsonaro support by 12 percentage points (from .59 to .71), while warm feelings toward the LGBT community decrease it by 19 percentage points (from .75 to .56).

The attenuation of demographic effects, combined with substantial effects of related issue attitudes, suggests that voters’ stances on those issues may mediate the effects of demographics on voting for Bolsonaro. In the Online Appendix, we present the partial correlations between demographic and attitudinal variables, as well as the results of formal tests of mediation. For gender, we examine opinions about the right to bear arms as a mediator, given extensive evidence that women in Brazil and elsewhere favor tighter gun restrictions (Datafolha, 2019; Goss, 2006); for Black racial identity, we use racial resentment; and for evangelical affiliation, the LGBT feeling thermometer. After selecting these mediators on theoretical grounds, we confirmed that each demographic variable predicted its respective attitude through regression models presented in Figures A1–A3. Using causal mediation analysis, we estimate that 30.5% of the effect of evangelical religious affiliation is mediated by LGBT attitudes in the final model, and 21.5% of the effect of Black identity is mediated through racial resentment. The estimated effect for gender is also substantial (62.1% of the effect of gender is estimated to be mediated by issue attitudes, see Table A3), but imprecisely estimated. We conclude that demographic variables are largely insignificant in Model 3 because demographics shaped issue positions, which in turn affected whether voters embraced or rejected Bolsonaro.

Finally, we find that education depresses Bolsonaro support, consistent with recent work showing low support for right-wing authoritarian candidates among highly educated individuals in Latin America as a whole (Cohen and Smith, 2016). However, we view these results with some caution, as they are inconsistent with replication models using other datasets (see Tables A5–A6).

Discussion and conclusion

We analyzed the social bases of support for Brazil’s far-right president, Jair Bolsonaro, in 2018, finding that his voters were distinguished by race, gender, and religion—contrary to historical patterns. Religion shows particularly robust links to voting. Demographics mattered not only via voters’ own identities; attitudes towards groups defined by race and sexuality also influenced vote choice. Moreover, gender shaped voting through attitudes on gun possession—an issue with grave implications for women’s security.

Retrospective evaluations of the PT also mattered. Even in the final model, favorable feelings toward the PT reduced support for Bolsonaro by 58 percentage points across the independent variable’s range. Yet economic evaluations inconsistently affected vote choice, and voters who perceived Brazilian politicians as corrupt tended to oppose Bolsonaro. In sum, we find only partial support for the notion that Bolsonaro’s election represented a sweeping rejection of status quo politics in Brazil.

Scholars have historically found that race, gender, and religion are minor determinants of vote choice in Brazil and across Latin America, emphasizing instead class-based and non-programmatic politics rooted in candidate charisma or clientelism. However, Brazil’s 2018 contest was different. Identity shaped responses to Bolsonaro’s candidacy. For certain individuals, his rhetoric and policy positions regarding minority groups represented a red flag; for others, they served as a rallying point.

Questions remain regarding the origins of demographic polarization. Divides between women and men, citizens of different racial identifications, and evangelicals and the non-religious may be unusually intense in the Bolsonaro era, but it seems unlikely that they sprang suddenly to life with Bolsonaro’s presidential bid. Future work should trace how these cleavages evolved in the decade and a half of PT governments. A second lingering question relates to the impact of intensifying demographic cleavages for Brazil’s political system. The 2018 election did not sweep away all vestiges of the old politics, but it may signal the onset of a more polarized political arena in which individuals sort themselves into political factions based on identity, which is less mutable than performance evaluations. It also remains to be seen how individuals with cross-cutting identities cope with such polarization.

Beyond Brazil, these results suggest that far-right politics can unearth demographic fissures even where they were previously largely unexploited. Our results align with multiple studies of the US (e.g. Sides et al., 2018) that find demographics mattered more than economic considerations in explaining Trump’s victory. As shocking as Bolsonaro’s victory was to many domestic and international observers, our analysis here raises the possibility that his campaign succeeded not despite his divisive nature, but rather because of it.

Supplemental Material

sj-pdf-1-rap-10.1177_2053168021990204 – Supplemental material for Demographic polarization and the rise of the far right: Brazil’s 2018 presidential election

Supplemental material, sj-pdf-1-rap-10.1177_2053168021990204 for Demographic polarization and the rise of the far right: Brazil’s 2018 presidential election by Matthew L. Layton, Amy Erica Smith, Mason W. Moseley and Mollie J. Cohen in Research & Politics

Footnotes

Acknowledgements

Thanks to Lindsay Mayka and two RAP reviewers for helpful feedback on previous versions of this paper, as well as to participants in a seminar at the University of Pittsburgh. Author order was randomly assigned. We thank the LAPOP Lab and its major sponsors for access to the 2018 AmericasBarometer data.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Matthew L. Layton

Amy Erica Smith

Supplemental materials

The supplemental files are available at

The replication files are available at

Notes

Carnegie Corporation of New York Grant

This publication was made possible (in part) by a grant from the Carnegie Corporation of New York. The statements made and views expressed are solely the responsibility of the author.

References

Almeida

MHT

Guarnieri

(2020) The unlikely president: The populist captain and his voters. Revista Euro Latinoamericana de Estudios Sociales y Políticos. 1(1): 139–159. Available at: http://www.ojs.unsj.edu.ar/index.php/relasp/article/view/510 (accessed 10 February 2020).

Amaral

OE do

(2020) The victory of Jair Bolsonaro according to the Brazilian Electoral Study of 2018. Brazilian Political Science Review 14(1): e0004.

Amaral

OE do

Meneguello

(2017) The PT in power, 2003–2016. In: Kingstone

Power

(eds) Democratic Brazil Divided. Pittsburgh: University of Pittsburgh Press, 31–52.

Anderson

(2019) Jair Bolsonaro’s southern strategy. The New Yorker, 25 March.

Boas

Smith

(2015) Religion and the Latin American voter. In: Carlin

Singer

Zechmeister

(eds) The Latin American Voter: Pursuing Representation and Accountability in Challenging Contexts. Ann Arbor: University of Michigan Press, 99–121.

Bohn

(2007) Contexto político-eleitoral, minorias religiosas e voto em pleitos presidenciais (2002–2006). Opinião Pública 13(2).

Bolsonaro

(2008) BOLSONARO (PP-RJ) x Maria Do Rosário (PT-RS). 19 August. Available at: https://www.youtube.com/watch?v=atKHN_irOsQ (accessed 1 November 2019).

Carlin

Singer

(2011) Support for polyarchy in the Americas. Comparative Political Studies 44(11): 1500-1526.

Carlin

Singer

Zechmeister

(eds) (2015) The Latin American Voter: Pursuing Representation and Accountability in Challenging Contexts. Ann Arbor: University of Michigan Press.

10.

Carmines

Stimson

(1989) Issue Evolution: Race and the Transformation of American Politics. Princeton: Princeton University Press.

11.

Cohen

Smith

(2016) Do authoritarians vote for authoritarians? Evidence from Latin America. Research & Politics 3(4): 2053168016684066.

12.

Darlington

(2018) Jair Bolsonaro, candidate in Brazil, faces women’s calls: #NotHim. The New York Times, 24 September.

13.

Datafolha (2019) 66% são contra posse de armas e 70% rejeitam flexbilizar porte. Posse e porte de armas. Folha de São Paulo, 12 July.

14.

Goñi

Watts

(2016) Brazil and Argentina unite in protest against culture of sexual violence. The Guardian, 3 June.

15.

Goss

(2006) Disarmed: The Missing Movement for Gun Control. Princeton: Princeton University Press.

16.

Htun

(2016) Inclusion without Representation in Latin America: Gender Quotas and Ethnic Reservations. New York: Cambridge University Press.

17.

Hunter

(2010) The Transformation of the Workers’ Party in Brazil, 1989–2009. New York: Cambridge University Press.

18.

Hunter

Power

(2007) Rewarding Lula: Executive power, social policy, and the Brazilian elections of 2006. Latin American Politics and Society 49(1): 1–30.

19.

Hunter

Power

(2019) Bolsonaro and Brazil’s illiberal backlash. Journal of Democracy 30(1): 68-82.

20.

Lipset

Rokkan

(1967) Cleavage structures, party systems, and voter alignments: An introduction. In: Lipset

Rokkan

(eds) Party Systems and Voter Alignments: Cross-National Perspectives. New York: Free Press, 1–64.

21.

Lucassen

Lubbers

(2012) Who fears what? Explaining far-right-wing preference in Europe by distinguishing perceived cultural and economic ethnic threats. Comparative Political Studies 45(5): 547-574.

22.

Madrid

(2012) The Rise of Ethnic Politics in Latin America. New York: Cambridge University Press.

23.

Manza

Brooks

(1999) Social Cleavages and Political Change: Voter Alignments and US Party Coalitions. Oxford: Oxford University Press.

24.

Moisés

(1993) Elections, political parties and political culture in Brazil: Changes and Continuities. Journal of Latin American Studies 25(3): 575–611.

25.

Moreno Morales

(2015) Ethnicity and electoral preferences in Latin America. In: Carlin

Singer

Zechmeister

(eds) The Latin American Voter: Pursuing Representation and Accountability in Challenging Contexts. Ann Arbor: University of Michigan Press, 122–142.

26.

Morgan

(2015) Gender and the Latin American Voter. In: Carlin

Singer

Zechmeister

(eds) The Latin American Voter: Pursuing Representation and Accountability in Challenging Contexts. Ann Arbor: University of Michigan Press, 143–167.

27.

Norris

Inglehart

(2019) Cultural Backlash: Trump, Brexit, and Authoritarian Populism. New York: Cambridge University Press.

28.

Nugent

(2018) How far-right presidential candidate Jair Bolsonaro could transform Brazil. Time, 25 October.

29.

Rennó

(2020) The Bolsonaro voter: Issue positions and vote choice in the 2018 Brazilian presidential elections. Latin American Politics and Society 62(4): 1–23.

30.

Roberts

Wibbels

(1999) Party systems and electoral volatility in Latin America: A test of economic, institutional, and structural explanations. American Political Science Review 93(3): 575-590.

31.

Samuels

(2006) Sources of mass partisanship in Brazil. Latin American Politics and Society 48(2): 1–27.

32.

Samuels

Zucco

(2018) Partisans, Antipartisans, and Nonpartisans: Voting Behavior in Brazil. New York: Cambridge University Press.

33.

Sides

Tesler

Vavreck

(2018) Identity Crisis: The 2016 Presidential Campaign and the Battle for the Meaning of America. Princeton: Princeton University Press.

34.

Smith

(2019) Religion and Brazilian Democracy: Mobilizing the People of God. New York: Cambridge University Press.

35.

Telles

(2014) Pigmentocracies: Ethnicity, Race, and Color in Latin America. Chapel Hill: UNC Press Books.

36.

Tesler

(2016) Post-racial or Most-racial? Race and Politics in the Obama Era. Chicago: University of Chicago Press.

37.

Waldron

(2018) Brazil’s military takeover of security in Rio de Janeiro is a looming disaster. Huffington Post, 20 February.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.35 MB