Defying Electoral Governance: Distrust and Protests in Latin America

Tarouco, Gabriela; Tarouco, Gabriela

doi:10.26851/rucp.33.20

Servicios Personalizados

Revista

Articulo

Traducción automática

Links relacionados

Otros
Otros

Permalink

Revista Uruguaya de Ciencia Política

versión impresa ISSN 0797-9789versión On-line ISSN 1688-499X

Rev. Urug. Cienc. Polít. vol.33 Montevideo 2024 Epub 01-Dic-2024

https://doi.org/10.26851/rucp.33.20

Número temático: A 100 años de la creación de la Corte Electoral en Uruguay. Organización, administración y gobernanza electoral en América Latina y el mundo

Defying Electoral Governance: Distrust and Protests in Latin America^¹ ^²

Confrontando la gobernanza electoral: desconfianza y protestas en América Latina

Confrontando a Governança Eleitoral: Desconfiança e Protestos na América Latina

Gabriela Tarouco¹
http://orcid.org/0000-0002-7200-6095

^¹Universidad Federal de Pernambuco (Brasil). gabriela.tarouco@ufpe.br

Resumen:

¿Cuándo rechazan los partidos políticos los resultados electorales? Incluso décadas después de la transición democrática, los conflictos y protestas postelectorales siguen siendo comunes en muchos países latinoamericanos. Este artículo sostiene que la confianza en la gestión electoral puede predecir las decisiones de los competidores políticos entre dos estrategias alternativas: desafío o cumplimiento. El presente trabajo utiliza un diseño de investigación de N medio y un modelo multinivel para analizar las elecciones en 18 países latinoamericanos desde la redemocratización y encuentra apoyo al argumento de que ceteris paribus, la desconfianza en la imparcialidad de las elecciones predice las reacciones desafiantes de los competidores políticos.

Palabras clave: protestas; gobernanza electoral, partidos políticos, gestión electoral, elecciones.

Abstract:

When do political parties reject electoral results? Even decades after the democratic transition, post-electoral conflicts and protests are still common in many Latin American countries. This article argues that trust in electoral management can predict political competitorsʼ choices between two alternative strategies: defiance or compliance. The present work uses a medium-N research design and a multilevel model to analyze elections in 18 Latin American countries since redemocratization and finds support for the argument that ceteris paribus, distrust in the fairness of elections predicts the defiant reactions of political competitors.

Keywords: protests; electoral governance; political parties; election management; elections

Resumo:

Quando os partidos políticos rejeitam os resultados eleitorais? Mesmo décadas após a transição democrática, protestos pós-eleitorais ainda são comuns em muitos países latino-americanos. Este artigo argumenta que a confiança na gestão eleitoral pode predizer a escolha dos competidores entre duas estratégias alternativas em relação ao resultado: contestar ou acatar. O presente trabalho utiliza um desenho de pesquisa de N médio e um modelo multinível para analisar eleições em 18 países latino-americanos desde a redemocratização e encontra apoio para o argumento de que ceteris paribus, a desconfiança na lisura das eleições prediz a rejeição dos resultados pelos competidores.

Palavras-chave: protestos; governança eleitoral, partidos políticos, gestão eleitoral, eleições.

1. Introduction

Even decades after transitioning to democracy, in many Latin American countries, post-electoral protests are still frequent. Why do political parties sometimes choose to reject election results and, at other times, choose to accept them? This article argues that distrust in electoral governance predicts defiance from political competitors. Credibility and trust in electoral management can, in turn, affect the behavior of political competitors and their propensity to accept election results peacefully.

The relationship between lack of trust in electoral institutions and post-electoral challenges is thoroughly mentioned in the literature on electoral integrity (^{Norris et al., 2015}; ^{Zavadskaya, 2017}) but to date has not been tested in a comparative, empirical manner in Latin America. Filling this gap is justifiable given that Latin America is a region with specificities that uniquely impact a relationship already explored in other regions. The negotiated nature of the democratic transition in Latin America (^{O’Donnell and Schmitter, 1986}) has set conditions for its consolidation distinct from those in countries where transitions have proceeded through elections (^{Lindberg, 2009}; ^{Edgell et al., 2015})^³.

This research found that post-electoral protests follow concerns about electoral fairness. As previous results had shown a significant association between partisan-composed Electoral Management Bodies (EMBs) and protests (^{Tarouco, 2017}), the present study hypothesizes that distrust in electoral integrity would mediate the relationship between EMBs’ partisanship and defiance. Mediation analysis showed that the effect of partisan EMBs occurs only through the mediation of distrust.

The findings of this single-region analysis are theoretically relevant as they uncover the mechanisms of self-enforcement applicable to other new democracies and the potentially undesired consequences of the multiparty composition of the EMBs, here referred to as the partisan power-sharing model of electoral management.

The next section briefly reviews the literature on electoral governance^⁴ and trust in electoral fairness. The third section discusses post-electoral protests as strategic defiance and describes their occurrence in Latin America. The following section presents the proposed mechanism, followed by descriptions of the data and methods used. Next, the article presents the results of the statistical analysis performed. Finally, the last section offers some conclusions and points to promising future research in the area.

2. Political Parties, Electoral Management, and Trust: A Brief Review

The way political parties connect to electoral management institutions is a crucial variable in the contemporary debate on the autonomy of EMBs. Through their lawmaking role, political parties establish the institutional design of their countryʼs electoral bodies, choosing either to include their representatives in a multiparty electoral management body (appointing members directly or through the legislature) or to delegate electoral management to non-partisan actors,^⁵ such as professional managers or the judicial branch (^{Mozaffar and Schedler, 2002}).

Either of these alternatives would be better than letting the incumbent manage elections, as was the case in the classical model of electoral governance (^{Lehoucq, 2002}), with the executive branch organizing elections and parliament certifying the results. According to Lehoucq (2002), the classical model fails to produce acceptance of results when the same party controls both branches. Thus, the delegation of electoral governance to an autonomous body or an independent-model EMB (^{Catt et al., 2014}) was designed to prevent election conflicts that may lead to political instability in young democracies. Political parties would be unable to police themselves while in charge of managing electoral competition from within the executive or legislative branches (Lehoucq, 2002).

This article builds on the extant literature by adding that the autonomy of EMBs should include independence not only from the executive office and legislature but also from political parties.^⁶ This argument is met with both support and criticism in the literature and is rooted in controversial empirical research evidence. Partisan power-sharing inside electoral governance institutions is a matter of reasonable dispute in the literature.^⁷ On the optimistic side, there is the argument that partisan power-sharing in electoral management would work as a guarantee of electoral integrity because it provides transparency, avoids unequal access to information, and allows for political parties to act as each other’s watchdogs (^{Pastor, 1999}; ^{López-Pintor, 2000}). This argument finds support in research about the effects of partisan power-sharing EMBs on acceptable election results^⁸ (^{Hartlyn et al., 2008}), the confidence of parliamentarians in election processes (^{Otaola, 2017}), and the credibility of elections and accepted outcomes (^{Estevez et al., 2008}).

An alternative view stresses the risk of an EMBʼs “capture” by its stakeholders, considering that political parties stand to benefit from interfering with the process, thus compromising impartiality. Support for this argument lies in research about trust in elections (^{Molina and Hernández, 1998}), trust among political elites in low-level democracies (^{Rosas, 2010}), overall concerns with electoral fairness (^{Tarouco, 2016}), and the autonomy and impartiality of EMBs (^{Ugues Jr., 2014}).

2.1 Delegation and Partisanship in Electoral Management

Several classifications of EMBs in the literature focus on their independence from government influence. Despite their typology variation, most of these classifications oppose governmental models of electoral management to independent commissions, including those handled by partisan-appointed members (^{Norris, 2015}; ^{Pastor, 1999}; ^{López-Pintor, 2000}; ^{Schedler, 2004}).

The concept of electoral governance (^{Mozaffar and Schedler, 2002}) is unique in that it distinguishes delegation from independence, two distinct dimensions of electoral management. The present article follows this distinction, with the partisan dimension varying among non-governmental models of EMBs. This operationalization distinguishes party delegation from other sources of autonomy (mainly the government); it also considers the appointment of EMB members by the legislature as partisan (although indirectly). Even when EMB members are not affiliated with political parties, their appointment by party legislators implies a principal-agent relationship between them.^⁹

2.2 Trust in Elections

Trust in electoral fairness-or a lack thereof-appears in the literature either as a proxy for electoral integrity or as a measure of popular support for democracy and institutions, mainly as a dependent variable (^{Fortin-Rittberger et al., 2017}; ^{Maldonado and Seligson, 2014}; ^{Norris, 2022}; ^{Rosas, 2010}; ^{Zmerli and van der Meer 2017})

Trust in electoral institutions and procedures, measured using surveys, can be a good indicator of an electionʼs fairness (^{Birch, 2008}; ^{Bowler et al., 2015}). Since it is challenging to directly measure fraud and manipulation, many researchers use votersʼ evaluations as proxies, despite the potential bias of measures based on perceptions.^¹⁰

For this article, it does not matter if popular perceptions reflect the actual level of electoral integrity or not, since, in non-institutionalized democracies, claims of fraud or manipulations can be made even in free and fair elections (^{Anderson et al., 2005}; ^{Hyde and Marinov, 2014}). Instead, the perceptions themselves (their accuracy is irrelevant) are central to this discussion. From the perspective of political parties, distrust in electoral fairness is indicative of both the population’s mood and the chances that a party denouncing elections as fraudulent or manipulated will be believed. Electoral fraud and manipulation appear to increase citizens’ support for protests (^{Norris, 2014}; ^{Sedziaka and Rose, 2015}), but before citizens join a protest, some of the competitors must decide to call for one. This decision is based on calculating how responsive the public will be to the grievances.

3. Defiance

Political competitors can react to election results in two different ways: they can accept the results or challenge them. Despite some contemporary exceptions^¹¹, we would not expect to see candidates and parties refusing to accept election results in a consolidated democracy. The losing candidate’s compliance is a central aspect of any democracy, under which a losing party can act safely as political opposition while waiting for the next electoral competition. (^{Nadeau and Blais, 1993}; ^{Przeworski, 1991})

In new democracies, however, competitors are still familiarizing themselves with alternation in power. Their recent experience with an autocratic past may even feel more concrete than their belief in a democratic future. Rules of the competition are new, and confidence in fair elections depends on the expectation that, if defeated, the incumbent will respect the results-note this has yet to be tested. Losersʼ consent (^{Anderson et al., 2005}) depends on time and experience with democracy. But as ^{Lago and Martinez i Coma (2016}) find, the losing candidate is more likely to consent when and where elections are free and fair.

At the party level, the decision to call for or support a post-electoral protest has been explained through several different variables. ^{Chernykh (2015}) finds that party age, origin, and ideology-but not opposition status-may explain why some parties choose the protest strategy while others comply with results.

At the national level, there are distinct explanations for the occurrence of protests. Protests, which may depend on certain political and contextual factors (^{Norris et al., 2005}), are more likely to occur when third-party actors have some bearing on elections (^{Chernykh and Svolik, 2015}). Procedural inconsistencies (^{Schedler, 2009}) and changes in electoral institutions may also increase the probability of protests (Chernykh, 2014). The literature has characterized protesting as a permanent threat that stands to prevent abuses from incumbents (^{Fearon, 2011}) and a desirable tool in a self-enforcing democracy (^{Hyde and Marinov, 2014}).

According to ^{Beaulieu (2014}), electoral protests represent a breakdown in negotiations between the incumbent and the opposition and are caused by commitment problems, lack of information, or lack of credibility. Agreements between potential protesters and the incumbent would be difficult to reach and honor if any side were to misrepresent its intentions. From that perspective, protests can be avoided if the government and the opposition are able to coordinate and bargain.

By contrast, the present article considers post-electoral protests to be a strategy chosen by political parties according to their evaluation of contextual conditions and the anticipated probability of garnering domestic (and perhaps international) support. In non-consolidated democracies, diffuse suspicions of manipulated elections are incentive enough for losers to call for protests, as they have nothing to lose. For this reason, this article rests on the assumption that post-electoral protests are generally called for or supported by losing parties.^¹²

Rejecting election results can sometimes successfully garner domestic and international actors’ support, especially if the denouncements are credible. The credibility of complaints works as an incentive for losers to reverse electoral defeat through public outcry. This strategy has fewer costs in a young democracy than in a consolidated one because distrust in elections tends to remain widespread across society for some time after the transition to democracy from autocracy. Only after democratic consolidation does compliance become a consistently expected behavior.

In Latin America, violent uprisings followed almost 10% of all 284 elections conducted from redemocratization through 2020.^¹³ According to the NELDA (National Elections Across Democracy and Autocracy) dataset^¹⁴, there were 28 post-electoral protests among Latin American countries after the democratic transition (see Table 1).

Table 1:Post-electoral protests in Latin America since redemocratization^¹⁵

Source: NELDA6 (^{Hyde and Marinov, 2021})

Table 1 shows that most protests occurred in the Dominican Republic and Mexico, both countries with long histories of struggle for democracy. Based on the available literature, it is possible to speculate that distrust in election authorities and procedures has played a role in protests (^{Donno, 2013}; ^{Hartlyn, 1998}; ^{Eisenstadt, 2004}). The concentration of cases in a few countries posits a methodological challenge that this article addresses through a combination of distinct strategies, including a set of multilevel regressions with countries as the group level.

4. From Partisan Electoral Management to Distrust and Protests

Among several electoral governance features (^{Mozaffar and Schedler, 2002}), partisan power-sharing in electoral management has been suggested as both a problem and a solution for issues of poor electoral integrity. By sharing power in electoral management bodies, political parties may monitor each other, promoting horizontal control, transparency, and confidence. Alternatively, however, they may take advantage of the arrangement to serve their own interests in electoral procedures and disputes. Previous studies have shown that partisan electoral governance is a necessary condition for post-electoral protest from opposition parties (^{Tarouco, 2017}) and increases concerns about electoral fairness (Tarouco, 2016). The present article follows this perspective, adding a new building block to our understanding of the relationship between partisan power-sharing in electoral management and protests: distrust as the mechanism between partisanship of EMBs and defiance.

Figure 1: Distrust as the link between partisan EMBs and Defiance

Why would people distrust partisan EMBs? Why would political parties distrust institutions run by their own representatives? This article argues that trust in partisan EMBs can be undermined for two reasons. First, political parties are stakeholders with critical interests in electoral management. This condition works as an incentive for violating confidentiality and impartiality in EMB procedures. As all political parties are subject to this condition, they can expect biased behavior from all the others. Dissatisfaction can be more severe among underrepresented parties (ACE Project, 2015). Second, distrust in political parties can emerge in public opinion regardless of their effective roles in EMBs, especially in new democracies, where the rules of the competition are still building stability and confidence. The relationship between EMB partisanship and distrust has already been demonstrated empirically for the same countries and period studied here (^{Tarouco, 2016}). The losing political parties can take advantage of this overall lack of trust and garner support for their claims of manipulation by partisan EMBs, regardless of whether they are right or wrong.

That widespread distrust possibly makes politicians concerned with the legitimacy of the electoral competition itself. Perhaps because of this, the partisanship of Latin American EMBs has been the target of several institutional reforms. Many countries in the region have changed their electoral management rules during the transition process and even after redemocratization. The full or partial delegation of electoral governance to non-partisan actors was adopted by many countries in which political parties did not trust each other to conduct transitional elections. Following transitional elections, five countries made reforms toward delegation: Bolivia (1993), Costa Rica (1953), Ecuador (2009), Paraguay (1998), and Venezuela (1998). Six countries kept delegated models either fully (such as Argentina, Brazil, Chile, and Peru) or partially (such as Guatemala and Panama) since redemocratization. There was no reform toward the inclusion of political parties in EMBs (^{Tarouco, 2020}).

These reforms seem to confirm these new democracies’ main trend of denying positions on electoral management bodies to political parties. As the authors of electoral regulation (^{Clark, 2015}), political parties might have renounced control over electoral management in exchange for the legitimacy of the electoral processes on which their political fortune depends, as such legitimacy may stem from the perceived impartiality of non-partisan EMBs.

The mechanism proposed here links partisan electoral management to distrust and then to post-electoral protests, arguing that distrust encourages defiance by political competitors.

The partisan composition of EMBs increases vulnerability to electoral fraud and manipulation because political parties can manipulate electoral management to serve their own interests. Such vulnerability makes grievances regarding election results potentially more credible to public opinion. General distrust in electoral management might, in turn, increase the chances that a party’s grievances will receive widespread public support, without which it would be worthless to call for a post-electoral protest.

This article tests the hypothesis that losing parties choose to challenge election results when they think that trust in electoral management fairness is low, as parties anticipate that the public will find their grievances credible. The article argues that ceteris paribus, pre-electoral distrust in the fairness of elections predicts political actors’ post-electoral strategies.

As small-N tests have already shown, lack of delegation of electoral management to non-partisan actors is a necessary condition for opposition parties in Latin America to protest election results (^{Tarouco, 2017}). Building on these former findings, the present work advances the novel argument that lack of trust in electoral fairness is a sufficient condition for defiance, working as the mechanism of a relationship found elsewhere. Through a multilevel test, the new hypothesis tested in this article is that distrust of electoral fairness predicts defiance of electoral results.

5. Data and Methods

Table 2 summarizes the elections analyzed. The dataset combines variables from NELDA (^{Hyde and Marinov, 2012}; 2021)^¹⁶ and V-Dem (^{Coppedge, Gerring et al. 2024}) databases. The cases include elections for constituent assemblies, executive office, and legislatures from redemocratization until 2020, excluding those elections conducted under authoritarian periods, according to ^{Mainwaring and Pérez-Liñánʼs (2015}; 2013) criteria, which include the following conditions:

1) the head of government and the legislature must be chosen through open and fair competitive elections;
2) the franchise must include the great majority of the adult population;
3) political and civil rights must be protected;
4) elected authorities must exercise real governing power (without being overshadowed by non-elected actors).

Table 2: Countries and elections analyzed^¹⁷

The cases in this paper’s sample confirm the former findings (^{Tarouco, 2016}) about the association between partisan EMBs and distrust, as table 3 shows.

Table 3:Partisanship of EMBs and concerns about electoral fairness

Sources: ^{Tarouco (2016}) and ^{Hyde and Marinov (2021})Pearson chi2(2)=11.2373 Pr = 0.004

Despite a few exceptions, the association tests’ statistical results are as expected: partisanship of EMBs is significantly associated with concerns about electoral fairness.

5.1 Dependent Variables

This study mobilizes two alternative variables to operationalize the concept of defiance. One of them intends to capture the defiance itself, that is, the very occurrence of post-electoral protests. The other one reflects the counterfactual, measuring how widespread was the acceptance of the election results. The two dependent variables represent two sides of the same event: protesting election results and accepting election results. The independent variables shall have opposite effects on each of the dependent variables. Factors that increase the probability of post-electoral protests must decrease the level of acceptance of election results.

The first dependent variable is the answer to the following question in the NELDA dataset: Were there riots and protests after the election? (nelda29) It is a dummy variable that reached 9.9% affirmative answers among elections analyzed here. The second dependent variable is based on the expertsʼ responses to the following question in the V-Dem dataset: Did losing parties and candidates accept the result of this national election within three months? The answers vary from 0 to 4, where 0 = none and 4 = all, converted by the V-Dem team to an index through the IRT method (^{Coppedge, Gerring et al. 2024}). Among the elections in this study, the index varies from -2.35 to 1.95, with a mean of 1.02 and a standard deviation of 0.86. The two dependent variables relate to each other exactly as expected: the elections followed by protests also had lower levels of acceptance of results, as the test of comparison of means in Table 4 shows.

Table 4:Relationship between the two dependent variables (mean-comparison test of acceptance based on the occurrence of protest)

Source: Author's elaboration t = 10.9684; Ho: diff. = 0; degrees of freedom = 281; Pr = 0.0000

5.2 Independent variables

The main independent variable is the concern about electoral fairness, which indicates the presence of overall suspicions regarding the integrity of each election. It consists of answers to the following question in the NELDA dataset: Before elections, are there significant concerns that elections will not be free and fair? (nelda11) It is a dummy variable coded by the NELDA team and included here as a proxy for trust. The data reveal significant concerns about electoral integrity before the elections in approximately 18% of the cases examined here.

5.3 Control variables

The control variables of this study are those implemented in the literature. The type of election-executive (216 cases), legislative (61 cases), or constituent assembly (7 cases)-and an indicator of economic conditions-occurrence of an economic crisis during a given election year (about 20% of the cases)-both derived from the NELDA dataset. There is no need to control for other macro-institutions because all elections were in Latin American countries with very similar institutional designs (presidential and multiparty systems). A variable for the democracy level controls for variations among the cases in the quality and competitiveness of democratic regimes. According to ^{Mainwaring and Pérez-Liñánʼs (2013}; 2015) criteria, 63% of the elections in this sample were conducted under democratic regimes, while 37% were in semi-democratic ones. The level of democracy is a good proxy for credibility and information availability that might affect defiance probabilities (^{Beaulieu, 2014}). The sample does not include any election conducted under authoritarian rule.^²² As a last control variable, international observersʼ presence (nelda45) is included as a proxy for international visibility that might encourage protests (^{Donno, 2013}; ^{Kelley, 2012}).

6. Tests and Results

A first step toward assessing the relationship between trust and defiance is verifying their association, as in the bivariate analysis shown in table 5.

Table 5:Relationship Between Distrust and Protests

Source: ^{Hyde and Marinov (2021}) Pearson chi2 (2)=13.5504; Pr = 0.000

As expected, concerns are associated with protests. The frequency of protests is significantly greater where concerns about electoral fairness exist. Some features of the data pose some challenges to the analysis. First, the data has a cross-section-time-series format, but there is more than one election each year in each country (executive office, legislature, and runoff); thus, panel regression models do not fit. Second, there are few post-electoral protest cases, so the dependent variable varies less than what would be desirable for a regression.^²³ Because of these limitations, this article combines two different methodological strategies: 1) two sets of multilevel regressions, one for each dependent variable, with individual-level variables for each election and the countries as the group-level variable; 2) a set of logistic regressions for rare events, for the dummy dependent variable. The results of these multilevel models are presented in Tables 6-7. For the NELDA dummy dependent variable (occurrence of protest), Table 8 shows coefficients for rare events logit regression.

Table 6 Linear Regression

Source: Author's elaboration Group variable: country Number of groups: 18 Multilevel regression using MIXED in STATA 13 * p<0.05, ** p<0.01, *** p<0.001

Table 7 shows two multilevel regressions for acceptance of election results: the first has only the independent variable, and the second has several control variables added. The results indicate that concerns reduce acceptance. Besides that, election results in democratic countries are more accepted than in semi-democratic ones.

The following models in Tables 8 and 9 focus on the second dependent variable: protests’ occurrence.

Table 7:Multilevel Logistic Regression

Source: Author's elaboration Group variable: country Number of groups: 18 Multilevel regression using MELOGIT in STATA 13 * p<0.05, ** p<0.01, *** p<0.001

Table 7 shows two multilevel regressions for the occurrence of post-electoral protests: one with only the independent variable and the other with control variables added. The results support the hypothesis that distrust in electoral fairness increases the probability of protests.

The next pair of regressions in Table 8 applies an alternative technique, the rare events logistic regression model, to the same dependent variable.

Table 8: Rare events logistic regressions (pooled) Dependent variable: Were there post-electoral protests?

Source: Author's elaboration Logistic regression using RELOGIT in STATA 13 * p<0.05, ** p<0.01, *** p<0.001

Table 8 shows the results of logit coefficients and their standard errors corrected for rare events. The results are in the same expected direction as in the multilevel model.

In the set of regression tests conducted above, signs of coefficients point in the expected directions. Concerns about the fairness of elections decrease the acceptance of results and increase the probability of post-electoral protests. It is remarkable - and counter-intuitive - that democracy is statistically significant in only one of the models (Table 6), showing the very complex relationship between regime, perceptions of electoral integrity, and defiance behavior. Distrust may occur even under full democracies and protests might be just competition strategies.

The results of multilevel regressions support the hypothesis that distrust affects defiance. The results are robust to the corresponding tests using the counterfactual (compliance) as the dependent variable. The rare events logistic regressions can be read as an additional robustness check. These findings imply that without trust in electoral fairness, the losers’ consent is still threatened in Latin America.

The last test refers to distrust’s mediation role. Concerns about electoral fairness are the independent variable in this study, but the electoral management model might affect them. As shown in Table 3, partisan EMBs are associated with concerns about electoral fairness. Figure 2 illustrates the theoretical expectations about how partisan electoral management relates to defiance:

Figure 2: (c) Partisan EMBs increase defiance (T →Y); (a) Partisan EMBs increase distrust (T →M) (b) Distrust increases defiance (M →Y) Theoretical expectations of distrust as a mediator between EMBs and defiance

To test whether distrust is a mediator connecting partisan models of electoral management to defiance, Table 9 shows two models of mediation analysis, one for each operationalization of the dependent variable defiance: post-electoral protests (from Nelda) and acceptance of results (from V-Dem).

Table 9: Analysis Dependent variable: Defiance; Mediator: Distrust

Source: Author's elaboration Standard errors in parentheses. The number of observations in the data is less than the number of simulations. Mediation analysis using MEDEFF in STATA 13 * p<0.05, ** p<0.01, *** p<0.001

Results suggest that partisan EMBs have no significant direct effect on defiance^²⁵. That relationship happens only through the mediation of distrust. That result is not trivial. It reinforces the crucial relevance of trust in elections for democratic compliance. Also, it means that concerns about electoral integrity may emerge in distinct institutional designs of electoral governance. Uruguay is an emblematic example of a partisan EMB with no distrust at all.

7. Concluding remarks

Challenging electoral results may be either good or bad news for young democracies as the countries studied in this article. On the one hand, it denotes freedom of contestation and may legitimately point to flaws in electoral governance to be fixed. On the other hand, defiance may be adopted opportunistically as a strategy to change unwanted results. This article has shown that in any case, trust or distrust in electoral fairness plays a crucial role.

The defiance concept was operationalized through two variables (occurrence of post-electoral protest and acceptance of results) and was successfully predicted by distrust in electoral fairness through regression models. The statistical analyses tested both the additive effects of distrust over defiance and its mediation role, finding that it advances post-electoral protests and decreases acceptance of electoral results by losers.

The conclusion drawn from this study of Latin America is that political parties defy election results mainly when there are spread concerns-truly due or not-about how elections were conducted. They do so because elections are surrounded by distrust, motivated or not by electoral governance procedures, which encourages the strategy of challenging electoral managers over the acceptance of election results.

Interested political competitors take advantage of social context and mood in order to advance their plans. It is not trivial that distrust in electoral conduct may give rise to strategic defiance, which may, in turn, worsen the legitimacy of electoral governance.

As the sample excludes elections under authoritarian periods, we can interpret the results as indicating that democratic institutions of electoral governance in Latin America are at a non-negligible risk. Public concerns may become potentially harmful to their credibility, mainly when distrust evolves into refusing electoral results. The urgent challenge for Latin American institutions of electoral governance is to face the threat of public distrust while pursuing electoral integrity.

References

ACE Project. (2015). EMB Members: Respected Experts or Watchdogs on Each Other? [ Links ]

Anderson, C., Blais, A., Bowler, S., et al. (2005). Losers’ Consent: Elections and Democratic Legitimacy. Oxford University Press. [ Links ]

Beaulieu, E. (2014) Electoral Protest and Demoracy in the developing world. Cambridge University Press. [ Links ]

Birch, S. (2008). Electoral institutions and popular confidence in electoral processes: A cross-national analysis. Electoral Studies (27), 305-320. [ Links ]

Birch, S. (2011). Electoral Malpractice. Oxford University Press. [ Links ]

Bowler, S., Brunell, T., Donovan, T., et al. (2015). Election administration and perceptions of fair elections. Electoral Studies (38), 1-9. [ Links ]

Catt, H., Ellis, A., Maley, M., et al. (2014). Electoral management design. International IDEA. [ Links ]

Chernykh, S. (2014). When Do Political Parties Protest Election Results? Comparative Political Studies, 47(10): 1359-1383. [ Links ]

Chernykh, S. (2015). Beyond the Opposition: Who Rejects Electoral Results? Electoral Integrity Project, Working Paper Series, 4(1). [ Links ]

Chernykh, S., Svolik, M. W. (2015). Third-Party Actors and the Success of Democracy: How Electoral Commissions, Courts, and Observers Shape Incentives for Electoral Manipulation and Post-Election Protests. The Journal of Politics, 77(2), 407- 420. [ Links ]

Clark, A. (2015) The relationship between political parties and their regulators. Party Politics, 23(6), 646-656. [ Links ]

Coppedge, M., Gerring, J., Knutsen, C. H., Lindberg, S. I., Teorell, J., Altman, D., Angiolillo, F., Bernhard, M., Borella, C., Cornell, A., Fish, M. S., Fox, L., Gastaldi, L., Gjerløw, H., Glynn, A., God, A. G., Grahn, S., Hicken, A., Kinzelbach, K., … Ziblatt, D. (2024). V-Dem (Country-Year/Country-Date) Dataset v14. V. o. D. V.-D. Project. [ Links ]

Coppedge, M., Gerring, J., Lindberg, S. I., et al. (2017b). V-Dem Methodology v7.1 - Varieties of Democracy (V-Dem) Project. [ Links ]

Donno, D. (2013). Defending Democratic Norms - International actors and the politics of electoral misconduct. Oxford University Press. [ Links ]

Edgell, A., Mechkova, V., Altman, D., et al. (2015). When and Where do Elections Matter? A Global Test of the Democratization by Elections Hypothesis, 1900-2012. [ Links ]

Eisenstadt, T. (2004). Courting Democracy in Mexico: Party Strategies and Electoral Institutions. Cambridge University Press. [ Links ]

Estevez, F., Magar, E., Rosas, G. (2008). Partisanship in non-partisan electoral agencies and democratic compliance: Evidence from Mexico’s Federal Electoral Institute. Electoral Studies (27), 257-271. [ Links ]

Fearon, J. D. (2011). Self-enforcing democracy. The Quarterly Journal of Economics, 126(4), 1661-1708. [ Links ]

Fortin-Rittberger, J., Harfst, P., Dingler, S. (2017). The costs of electoral fraud: establishing the link between electoral integrity, winning an election, and satisfaction with democracy. Journal of Elections, Public Opinion and Parties, 27(3), 350-368. [ Links ]

Hartlyn, J. (1998). The Struggle for Democratic Politics in the Dominican Republic. The University of North Carolina Press. [ Links ]

Hartlyn, J., McCoy, J., Mustillo, T. M. (2008). Electoral Governance Matters: Explaining the Quality of Elections in Contemporary Latin America. Comparative Political Studies 41(1), 73-98. [ Links ]

Hernández-Huerta, V. A. (2017). Judging Presidential Elections Around the World: An Overview. Election Law Journal, 16(3), 377-396. [ Links ]

Hyde, S. D., Marinov, N. (2021). National Elections Across Democracy and Autocracy. [ Links ]

Hyde, S. D., Marinov, N. (2012). Which Elections Can Be Lost? Political Analysis, 20(2), 191-201. [ Links ]

Hyde, S. D., Marinov, N. (2014). Information and Self-Enforcing Democracy: The Role of International Election Observation. International Organization, 68(2), 329-359. [ Links ]

Kelley, J. G. (2012). Monitoring Democracy: When International Election Observation Works, and Why It Often Fails. Princeton University Press. [ Links ]

King, G., Zeng, L. (2001). Logistic Regression in Rare Events Data. Department of Government, Harvard University. http://gking.harvard.edu [ Links ]

Lago, I., Martinez i Coma, F. (2016). Challenge or Consent? Understanding Losers’ Reactions in Mass Elections. Government and Opposition. [ Links ]

Lehoucq, F. E. (2002). Can Parties Police Themselves? Electoral Governance and Democratization. International Political Science Review, 23(1), 29-46. [ Links ]

Lindberg, S. I. (2009). Democratization by elections - a new mode of transition. The John Hopkins University Press. [ Links ]

López-Pintor, R. (2000). Electoral Management Bodies as Institutions of Governance. UNDP Bureau for Development Policy. [ Links ]

Mainwaring, S., Pérez-Liñán, A. (2007). Why Regions of the World Are Important: Regional Specificities and Region-Wide Diffusion of Democracy. In G. Munck (Ed.), Regimes and Democracy in Latin America: Theories and Methods. Oxford University Press. [ Links ]

Mainwaring, S., Pérez-Liñán, A. (2013). Democracies and Dictatorships in Latin America - Emergence, Survival, and Fall. Cambridge University Press. [ Links ]

Mainwaring, S., Pérez-Liñán, A. (2015). Cross-Currents in Latin America. Journal of Democracy, 26(1), 114-127. [ Links ]

Maldonado, A., Seligson, M. (2014). Electoral trust in Latin America. In P. Norris, R. W. Frank and F. Martinez i Coma (Eds.), Advancing Electoral Integrity. Oxford University Press. [ Links ]

Molina, J., Hernández, J. (1998). La credibilidad de las elecciones latinoamericanas y sus factores. El efecto de los organismos electorales, el sistema de partidos y las actitudes políticas. XXI Congreso Internacional de la Asociación de Estudios Latinoamericanos. LASA. [ Links ]

Mozaffar, S., Schedler, A. (2002). The Comparative Study of Electoral Governance-Introduction. International Political Science Review , 23(1), 5-27. [ Links ]

Nadeau, R., Blais, A. (1993). Accepting the Election Outcome: The Effect of Participation on Losers’ Consent. British Journal of Political Science, 23(4), 553-563. [ Links ]

Norris, P. (2014). Why Electoral Integrity Matters. Cambridge University Press. [ Links ]

Norris, P. (2015). Why Elections Fail. Cambridge University Press. [ Links ]

Norris, P. (2022). In Praise of Skepticism: trust but verify. Oxford University Press. [ Links ]

Norris, P., Frank, R. W., Martinez i Coma, F. (2015). Contentious elections - from votes to violence. In P. Norris, R. W. Frank, and F. Martinez i Coma (Eds.) Contentious elections - from ballots to barricades. Routledge. [ Links ]

Norris, P., Walgrave, S., Van Aelst, P. (2005). Who demonstrates? Antistate Rebels, Conventional Participants, or Everyone? Comparative Politics, 37(2). [ Links ]

O’Donnell, G, Schmitter, P. (1986). Transitions from Authoritarian Rule. University of Notre Dame Press. [ Links ]

Otaola, M. L. (2017). To include or not to include? Party representation in electoral institutions and confidence in elections: A comparative study of Latin America. Party Politics . [ Links ]

Pastor, R. A. (1999). The Role of Electoral Administration in Democratic Transitions: implications for policy and research. Democratization, 6(4), 1-27. [ Links ]

Przeworski, A. (1991). Democracy and the market : political and economic reforms in Eastern Europe and Latin America. Cambridge University Press. [ Links ]

Rosas, G. (2010). Trust in elections and the institutional design of electoral authorities: Evidence from Latin America. Electoral Studies (29), 74-90. [ Links ]

Schedler, A. (2004). Measuring the Formal Independence of Electoral Governance. CIDE. [ Links ]

Schedler, A. (2009). Inconsistencias contaminantes: gobernacion electoral y conflicto poselectoral en las elecciones presidenciales del 2006 en México. América Latina Hoy (51), 41-59. [ Links ]

Sedziaka, A., Rose, R. (2015). Do contentious elections catalize mass protests? In P. Norris, R. W. Frank, and F. Martinez i Coma (Eds.), Contentious Elections: from ballots to barricades. Routledge. [ Links ]

Tarouco, G. (2016). The Role of Political Parties in Electoral Governance: Delegation and the Quality of Elections in Latin America. Election Law Journal: Rules, Politics, and Policy, 15(1), 83-95. [ Links ]

Tarouco, G. (2017). Electoral governance in Latin America and the behaviour of oppositions. In H. A. Garnett and M. Zavadskaya (Eds.), Electoral Integrity and Political Regimes: Actors, Strategies and Consequences. Routledge. [ Links ]

Tarouco, G. (2020). Political Parties and Electoral Governance: Institutional change in Latin America. 12.º Encontro da Associação Brasileira de Ciência Política. https://cienciapolitica.org.br/eventos/12o-encontro-abcp/anais [ Links ]

Tomz, M., King, G., Zeng, L. (1999). ReLogit: Rare Events Logistic Regression. Harvard University. http://gking.harvard.edu/ [ Links ]

Ugues Jr., A. (2014). Electoral management bodies in central America. In P. Norris, R. W., Frank, and F. Martinez i Coma (Eds.), Advancing Electoral Integrity. Oxford University Press. [ Links ]

Wasserstein, R. L., Lazar, N. A. (2016). The ASA Statement on p-Values: Context, Process, and Purpose. The American Statistician, 70(2), 129-133. [ Links ]

Zavadskaya, M. (2017). When Do Elections Subvert Authoritarianism? Electoral Malpractice and Variety of Post-Electoral Scenarios. In H. A. Garnett and M. Zavadskaya (Eds.), Electoral Integrity and Political Regimes: Actors, Strategies and Consequences . Routledge. [ Links ]

Zmerli, S., Van der Meer, T. W. G. (Eds.) (2017). Handbook on Political Trust. Edward Elgar Publishing. [ Links ]

¹A previous version of this paper was presented at the 11th ECPR General Conference in 2017. I thank the participants for their comments. I am also grateful to Toby James, Holly Garnett, Rodrigo Martins, and the RUCP reviewers for their contributions. All remaining inaccuracies are my full responsibility.

² Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

³For the importance of studying regions in comparative politics, see Mainwaring and Pérez-Liñán (2007).

⁴Besides electoral management, the concept of electoral governance, as defined by Mozaffar and Schedler (2002), includes rule making and rule adjudication levels. In this article, as the main focus is a feature of the rule application level (the delegation of electoral management to non-partisan actors), both terms (electoral management and electoral governance) are interchangeable.

⁵Mozaffar and Schedler (2002) call this model abdication. As a matter of institutional engineering, delegation of election management to non-partisan actors is usually something that happens in constitutional or transition moments, but it can also be adopted (or revoked) through legislative reforms, as has been the case in some Latin American countries.

⁶Latin American presidential countries are good examples of how political parties outside the executive branch have substantial power. Not only the government, but also opposition parties (frequently strong in legislatures), when represented in the EMBs, can be a threat to their autonomy.

⁷In this article, power sharing in electoral governance refers to multiparty composition of the Electoral Management Body. Another meaning is that provided by Norris (2015: 115): “institutional arrangements where decision-making authority over election laws and procedures is dispersed among multiple branches of government, levels of government, and specialized administrative agencies.”

⁸Hartlyn et al. (2008) classify elections as acceptable, flawed, or failed. An election is deemed acceptable “if the basic elements for procedural fairness and technical soundness are present to an important degree” (p. 77).

⁹This is the case of Mexico, where the Electoral Counselors are elected by the vote of two thirds of the members present in the Chamber of Deputies (in addition to the Legislative Branch Counselors, who have voice but no vote). So, the Mexican EMB is here coded as partisan. In this article, the term delegation corresponds to the abdication model in Mozaffar and Schedler’s (2002) typology. For an example of this kind of partisanship of EMBs appointed by legislature, see Estevez et al. (2008).

¹⁰Flawed elections can be perceived by citizens as free and fair, depending on how efficiently the fraud and manipulations are hidden.

¹¹President Trump refused to accept the 2020 election results in the US, the biggest democracy in the world. On the other hand, Bolsonaro in Brazil accepted the results, albeit belatedly. He only acknowledged the defeat after few days, after having defied electoral management integrity during the whole campaign period and despite his party filing a suit demanding electoral justice to annul some ballot boxes.

¹²This information, however, is not available in our data sources (NELDA6 dataset).

¹³The list of countries and elections can be seen in Table 3.

¹⁴Version 6 of the NELDA dataset (Hyde and Marinov, 2012; 2021).

¹⁵Excluding some elections conducted during authoritarian periods.

¹⁶The National Elections Across Democracy and Autocracy (NELDA6) dataset provides detailed information on elections around the world from 1945 to 2020. Each round of an election is coded separately, meaning that subsequent rounds of the same election are coded as separate units of observation. In this article, we use the collapsed version of the data, which aggregates elections held on the same day (for example, general elections). In these cases, the “type” variable is coded as “Executive” (Hyde and Marinov, 2021).

¹⁷Data are available online at https://pesquisapartidos.wordpress.com/dados/

²²The period analyzed is different for each country, including only elections conducted during democratic or semi-democratic periods, according to Mainwaring and Pérez-Liñán’s criteria (2013; 2015).

²³A possible third problem would be that there are fewer countries at the group level (18) than usually recommended for a multilevel analysis, but the literature on that requirement is controversial.

²⁵The total effect of partisan EMBs on defiance goes in opposite directions in each model because the dependent variables in each one are opposite ways to measure defiance. The dependent variable in the first model of Table 9 is the occurrence of post-electoral protests, while in the second model, it shows how promptly the election results were accepted.

¹⁸Democratization means change from autocracy to democratic or semidemocratic regime, according to Mainwaring and Pérez-Liñán’s criteria (2013; 2015).

¹⁹Excluding the 2009 election, which was conducted during an authoritarian period.

²⁰Excluding the authoritarian period from 1992 to 1994.

²¹Excluding elections from 2010 on, which were conducted during an authoritarian period.

²⁴Type (executive) is the dummy variable that aggregates the categories “legislative” and “constituent assembly” as 0 and “executive” as 1 from the original multinomial category “type”.

Recibido: 15 de Abril de 2024; Aprobado: 21 de Octubre de 2024

Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons