Electoral Governance in Brazil^*

Abstract

Electoral governance has increasingly more frequently been the object of study of the comparative politics literature. This article examines the electoral governance institutional model adopted in Brazil and its consequences for political/electoral competition. It is argued herein that Brazil's Electoral Justice System, motivated by the institutional design, has ended up becoming one of the main actors of the country's recent democratic consolidation, being decisive not only with regard to rule adjudication and application, but also to rulemaking. With the purpose of assessing this governance model in action, three important recent rulings by Brazil's Electoral Justice System are analysed here: verticalization of the coalitions, reduction in the number of councillors, and party loyalty.

Keywords:
Brazil; Electoral governance; Electoral Justice System; Judiciary branch; Higher Electoral Court; Judicialization of politics

Introduction

The rising role of the Brazilian Electoral Justice System in the consolidation of Brazil's democratic regime has focused widespread attention on an institution that hitherto had been very little known and debated. It is true that research on electoral governance in the comparative political literature is recent. The subject gained importance due to a concern for the credibility of the electoral results of democracies born during the third democratic wave (Huntington 1993). The key concern in these new regimes was to guarantee that results coming from the polls were fair, transparent and, above all, accepted by political competitors.

This recent literature argues that electoral governance – construed as a set of rules and institutions that frame the electoral contest – has been overlooked as a variable in studies on democratic transition and consolidation, as most studies focus on normative issues such as government systems and the electoral formulas adopted (Elklit and Reynolds 2000; Mozaffar and Schedler 2002).

For Pastor (1999), the political literature on electoral systems is often driven by the “4-Ps: politics, parties, polling, and proportionality”. The author further states that, even in broader comparative studies on the profile of democratic regimes, the question of electoral governance has been overlooked, as in the case of the study conducted by Lijphart (1999).

Recently, the International Political Science Review dedicated an entire issue to electoral governance, a concept thus defined in an article by Mozaffar and Schedler in the same periodical:

Electoral governance is the wider set of activities that creates and maintains the broad institutional framework in which voting and electoral competition take place. It operates on three levels: rulemaking, rule application, and rule adjudication. Rulemaking involves designing the basic rules of the electoral game; rule application involves implementing these rules to organize the electoral game; rule adjudication involves resolving disputes arising within the game (Mozaffar and Schedler 2002, 7).

Rulemaking is the choice and framing of the basic rules of the electoral game. At this level of electoral governance the following are established, for example: the electoral formula, electoral districts, the magnitude of the elections, dates on which they will be held and other legal matters designed to ensure a level playing field for the contestants. Here, some rules given little attention by the political literature, such as the rules of (in) eligibility and organization of the bodies responsible for the administration of the elections are also defined.

The rule application level deals with the implementation and management of the electoral game, such as the registration of parties, candidates and voters, distribution of ballot boxes, procedures to be adopted on election day and other rules designed to ensure transparency, efficiency, and neutrality in managing the game. Basically, this level addresses issues concerning the administration of the electoral game.

Lastly, rule adjudication is the resolution of potential disputes involving competitors – the electoral bones of contention. Apart from clarifying and resolving controversies arising from the electoral race, this level also carries out the vote tallying and the announcement of final electoral results.

These three different levels of electoral governance are not the responsibility of one single body. For instance, rulemaking is almost always defined in the constitutional text and the electoral code. Thus, it is basically a Legislative Branch duty. The other two activities, however, are usually the responsibility of specific bodies – that we may generically call Electoral Management Bodies (EMBs).

Part of the effort of this recent literature is to create criteria to assess these bodies’ institutional design and its impact on the electoral game, a design which itself becomes a variable in the analyses on the consolidation of democratic regimes.

Effective electoral governance alone does not guarantee good elections, of course, because a complex variety of social, economic and political variables affect the process, integrity, and outcome of democratic elections. But good elections are impossible without effective electoral governance (Mozaffar and Schedler 2002, 6).

These authors argue that, in recent democracies, a regime's greater or lesser stability is a function of the electoral governance model adopted. Good electoral governance conducted by an adequate EMB may give credibility to the electoral results by stabilizing and resolving disputes for political power (Pastor 1999; Schedler 2002; Hartlyn et al. 2008).

Even when building on minimalist definitions of democracy, the research agenda proves to be relevant. For institutions regulating political competition, the greater good is to guarantee the fairness of the electoral process. The legitimacy of poll results is the sine qua non for the major political forces to accept the electoral results. It is worth pointing out that a large part of electoral governance activities focus on ensuring opportunities for distinct political groups to win elections and, especially, to have their victory recognized and accepted. The race can only exist insofar as political groups are apt and confident to run in the elections.

Accordingly, this article seeks to contribute to fill this gap. To this end, we intend to present an interpretation of the institutional design of Brazil's electoral governance and its consequences for partisan political competition. The key argument will be that although the Brazilian model has favoured trust in the “electoral truth”, it has also been responsible for the strong presence of the Judiciary in all electoral governance activities, including rulemaking.

To reinforce this argument we will analyse three recent rulings by the Electoral Justice System that will show the impacts of the model of electoral competition adopted – verticalization, number of councillors and party loyalty.

The Brazilian Governance Mode²

A series of political episodes in recent Brazilian history showcase the importance of the Electoral Justice System in ensuring significant standards of electoral fairness, thus revealing the strong institutional stability it has reached in the Brazilian political system. Evidence of this is the structure set in place in 1932, after the debacle of the Old Republic, which endured the many political changes that characterized a significant part of Brazilian history in the 20^th century.

Moreover, some posit a key role played by Brazil's Electoral Justice System in the long, recent democratic transition. “Without an institution of that kind, there would hardly have been trust in the competition; even more so if we take into account the political and legal restrictions of that time” (Sadek 1995, 41).

The comparative literature argues that one of the conditions for the establishment of fairness in the electoral process is the independence of the regulatory body from partisan interests (Hartlyn et al. 2008). The argument is that decision-making body independence protects the definition of the rules of the game from partisan interests. The argument is likened to that used in the debate on public policies for the relationship between bureaucracy and politics.

Analyses of the Brazilian case reflect the findings of comparative studies. In writing the history of the Electoral Justice System in Brazil, Fleisher and Barreto (2009) and Sadek (1995) showed that EMB independence from the holders of political power allows the Electoral Justice System to act independently from majority interests.

Actually, the neutrality acquired by the Brazilian model of electoral governance in relation to political interests was decisive for the paths taken in the transition. Nonetheless, here we argue that the persistence of the model throughout the democratic consolidation has yielded two other outcomes.

The first one is the insulation of the Electoral Management Body (EMB). This is not merely a model that protects governance from more immediate partisan interests but, rather and more importantly, a model that may have weakened EMB accountability.

The second effect is that it created favourable institutional conditions for the EMB – especially for its highest body, the Higher Electoral Court (Tribunal Superior Eleitoral (TSE)) – to advance over rulemaking as well. Given the characteristics of Brazil's governance, there have been recent TSE decisions that have deeply altered the electoral rules, including with creative interpretations of the constitutional text.

But in order to deepen the argument we need to know this institutional design better.

Very little has changed in the Brazilian Electoral Justice System since it was created in 1932. The 1988 Constitution states that the Electoral Justice System bodies are: 1) the Higher Electoral Court, 2) Regional Electoral Courts, 3) Electoral Judges and 4) Electoral Boards.

The Higher Electoral Court is the highest body for decisions regarding the management and implementation of the electoral process and the highest court when it comes to rule application activities in Brazilian electoral governance. In addition, it is the court of last resort in cases of electoral controversy (rule adjudication).

Chart 1
Brazil's Electoral Management Bodies (EMBs)

The TSE consists of seven members: three justices of the Supreme Federal Court (Supremo Tribunal Federal (STF)), two from the Higher Court of Justice (Superior Tribunal de Justiça (STJ)) and two citizens with remarkable legal knowledge and good moral reputation nominated by the STF and chosen by the President. These external members are usually lawyer working in the electoral field.

Regional Electoral Courts (Tribunais Regionais Eleitorais (TREs)) are based in every state capital. Aside from taking part in the management and implementation of the electoral process, a TRE is the second highest court for rule adjudication matters. Each TRE consists of seven members: two appellate judges from the state-level Court of Justice, two state lower court judges chosen by the Court of Justice, one federal lower court judge and two citizens of remarkable legal knowledge and good moral reputation nominated by the Court of Justice and chosen by the President.

The electoral judge is chosen by the TRE from among a state's judges. Their jurisdiction is the electoral zone. They also take part in the management and execution of the electoral process and act as the first court for rule adjudication.

The Electoral Boards are ad hoc bodies, set in place only for the execution of the electoral process. Sixty days prior to the elections, the TRE selects a judge and two to four citizens of remarkable legal knowledge and good moral reputation. In this way, the Board assists the Electoral Judge to carry out the duties required by the electoral process in a given electoral zone.

Brazil's Electoral Management Bodies have no tenured members of their own. Although the TSE, the TREs and the Electoral Judges’ Offices (where the electoral judges work) are permanent and therefore rely on their own tenured staff, the justices who become members of the Electoral Justice System are not obliged to stop their activities in other areas of the Judiciary. Not even the lawyers chosen are obliged to interrupt their professional activities.⁴

Such a profile can even make it difficult for the Electoral Justice System to meet all the demands it is mandated to by law. Examples of this are the audits that the Electoral Justice staff must conduct into campaign expenditures – a potential structural flaw that may compromise control quality (Taylor 2008).

Moreover, every sphere is guided by the “rule of intersection” in composing its board, in line with the Judiciary's structure. The highest electoral body intersects with the highest judicial sphere, as the lowest electoral body intersects with the lowest judicial bodies. Indeed, we may say that the highest electoral sphere intersects not only with the highest judicial spheres but also, and more importantly, it strongly intersects with the Constitutional Court (STF), a feature that will prove critical in underpinning our arguments.

Before going further with this point, it is worth deepening the analysis of the TSE's institutional functioning.

It is possible to state that the TSE is an STF body for electoral matters – not de jure, but de facto. As we saw earlier, the TSE consists of seven members. Three of them are chosen from among the justices of the Supreme Federal Court, while two others are lawyers who are nominated by the same STF, which, arguably, exerts strong influence over these members. It could be said that, at least, these lawyers are nominated in accordance with a projected and expected profile by the Justices of the Constitutional Court. The two other justices come from the STJ. However, we must also consider that these STJ justices exert less influence on the overall profile of the TSE than STF justices do. What allows us to make this statement is the length of their tenures.

All the members of the Electoral Justice System hold a two-year term of office, which may be extended for another two years. An STJ internal rule, however, bans its justices from holding a TSE office for more than one term, while for justices coming from the STF and independent lawyers the four-year term of office has become common practice.

The reason for the STJ Justices to remain with the TSE for less time is the higher number of Justices composing that court, and the need to comply with a rotation scheme to be met by its members. The STJ is composed of 33 Justices; so, in order for every STJ justice to have a chance to become a TSE member, an informal rule was set establishing that none of them shall hold office for a second two-year term.

STF members total 11 and there is no internal rule imposing on its justices the need to become a TSE member. The procedures regarding justices chosen are quite informal and depend mostly on the disposition manifested by a given justice to hold such office.

Thus, from the first semester of 1989 until the end of the second semester of 2007, 21 STJ Justices were titular members of the TSE while only 16 different STF Justices had held that same office.

Another important point is that no STJ justice who has held a TSE office for a two-year term has ever returned to the post. Yet, among the STF Justices it is not uncommon for a justice to have held office for two consecutive two-year terms and return years later for another two terms. From 1989 to late 2007, these were the cases of Justices Néri da Silveira, Carlos Velloso, Sepúlveda Pertence and Marco Aurélio Mello.

Undeniably, the time a justice spends at the TSE ensures his/her experience in the matter, even more so when the court in question does not have an exclusive body of justices. Hence, we can say that the capacity to influence the profile to be adopted by the TSE in adjudicating conflicts and in administrative decisions regarding the electoral process is stronger in STF justices than STJ justices.

We must consider that 13 of the 21 STJ justices that have held office in the TSE had worked in Electoral Justice before as regional electoral justices (in TREs) or as lower-sphere electoral judges. Such experience notwithstanding, the shorter period STJ justices spend at the TSE compared to STF justices reduces these justices’ power of influence on the institution's profile.

In addition to this predominance of the STF among the members that make up the TSE, the body's presidency and vice-presidency are exclusive offices of the justices of the Constitutional Court. Thus, the agenda and guidelines for the administrative functioning of the body itself are influenced by the profile of the STF justice occupying the presidency.

That is why we can affirm that the TSE is an STF body for electoral matters. After all, the power to influence the TSE by STF justices is undeniably greater when compared to its other members. One possible implication of this is that the rules of the competitive game can be submitted to the interpretation of a broader set of norms, including constitutional norms.

At times, when reviewing electoral legislation, the TSE makes a decision by resorting to an interpretation of the constitutional text. At other times, as it interprets the constitutional text, the STF rules for changing the rules of the electoral game. And there is no sign of any interpretive conflict between these institutions.

On the contrary, there is no record of any TSE ruling that once submitted to the STF was then reversed. In most cases, the STF does not even accept reviewing the appeal on the understanding that the TSE's interpretation is the last word in terms of electoral matters. And, when the STF hands down a decision on a theme about which it had already received an opinion by the TSE, there is no divergence in their rulings. They end up reinforcing each other.⁵

A number of countries adopt electoral governance models that set forth the powers of the Judiciary in the electoral contest. In Costa Rica, for example, the responsibility for the nomination and appointment of EMB members is exclusively of the Supreme Court of Justice. The Higher Electoral Court consists of three members, chosen for an eight-year term. The minimum requirements for selection of members are a bachelor's degree in Law and prior professional experience.

Two aspects must be underscored in comparing Costa Rica and Brazil. The first one is that in Costa Rica there is no “rule of intersection”, that is, EMB justices are elected by the Higher Court of Justice, but are not members thereof. Another feature concerns the profile of the Supreme Court itself: all of its justices are elected by parliament for an eight-year term, contrary to the classic republican principle of not setting terms-of-office for members of constitutional courts. This characteristic sets other accountability mechanisms for the Judiciary, which, in the long run, indirectly affects the EMBs’ accountability.

In Latin America, participation by the Judiciary in the electoral governance process also occurs in countries like Mexico, Peru and Chile.⁶ In these countries, however, governance activities are not concentrated in a single EMB. In cases like these, the Judiciary ultimately plays a decisive role in rule adjudication matters. Rule application, in turn, is the power of a different body. Thus, electoral governance is the responsibility of two distinct and independent bodies.

In short, Brazilian electoral governance can be defined by a combination of five characteristics: 1) concentration of governance activities in a single EMB (rule application and rule adjudication), 2) full independence from political parties and the Legislative Branch – given that nomination and appointment of its members are processes conducted by the Judiciary, 3) EMBs with non-exclusive members – given that members hold offices in electoral bodies but do not give up their activities in their bodies of origin, 4) “rule of intersection”, in that the major part of the justices’ offices is held by members of the Judiciary and 5) predominance of the Constitutional Court over the highest EMB.

Thus it is the combination of these characteristics that creates a favourable institutional environment for the insulation of the Electoral Management Body and its active participation in rulemaking. As the partisan contest consolidates itself and some themes of this game gain greater relevance, the electoral governance model adopted ultimately produces more outcomes with greater frequency.

With the aim of illustrating these outcomes, we will conduct a brief analysis of three recent milestone rulings that have radically altered the rules of the competitive game verticalization, number of councillors, and party loyalty. The main goal here is to observe the Brazilian electoral governance model in motion.

Verticalization⁷

In 2002, a year when elections for president, governors, senators and state and federal representatives took place, the TSE (court of last resort of the Electoral Management Body) handed down a ruling regarding the rules that guided partisan coalitions quite different from all former practices and legal interpretations.

Until that year, political parties were allowed to make alliances for state-level majoritarian elections (governors) regardless of the coalitions they came to make for the purpose of the national majoritarian election (presidential). In other words, there was no legal obligation to keep state and national coalitions symmetrical. So much so, that it was common for parties that were allies in gubernatorial elections to be adversaries in the presidential elections.

The year before the election, however, the Democratic Labour Party (PDT) consulted⁸ the TSE with the following question: “can a political party (Party A) enter into a coalition for the presidential election with other parties (Parties B, C and D) and, at the same time, enter into a coalition with third parties (E, F and G, who also have their own presidential candidate) for the purpose of the election for state governor?” In other words, they wanted to know if it was possible to establish a state-level coalition with parties that were adversaries in the presidential election.⁹

In the answer to the consultation, the EMB, acting as a court of last resort, decided against the procedure: political parties could not make “incongruent” alliances. Thus, coalitions for presidential elections should impose certain limits on state coalitions.

In the two previous elections, 1994 and 1998,¹⁰ this practice had been common. The interplay between state and national levels constituted one of the most important pieces on the electoral strategies game board (Braga 2006; Limongi and Cortez 2010). In ruling on the matter, with a view to reforming the political system, the Electoral Justice System imposed an unprecedented and unexpected constraint on political parties at the time.

Each dot on the graph 1 represents a different political party. The closer to 100 along the axis, the more vertical (congruent) were the party's alliances in the majoritarian elections in which it took part. On the other hand, the more distant from 100, the smaller the percentage of vertical alliances entered into by the party in state-level elections.

In other words, the party at the 100 point on the axis did not ally with opposing parties in the presidential elections in all states where it ran in gubernatorial elections.

In contrast, the party located furthest from 100 on the “verticalization” axis allied with adversaries in the presidential elections in a greater number of states where it ran in majoritarian elections.

Thus, on the right side of the axis are the parties that would have been closest in 1994 and 1998 to the obligation laid down from 2002. On the opposite side, the left side, are the parties that would have been the farthest from the verticalization practice, as designed by the TSE for the 2002 elections.

Graph 1
Level of verticalization between State and national alliances – 1994 and 1998

Building on data from these two elections, we can therefore say that the decision handed down by the Electoral Justice System went counter to partisan strategies and altered the electoral competition. The objective here is not to debate if the practice then in force was positive or negative for the Brazilian partisan system, but to show how the highest EMB found the practice to be negative and imposed a new standard that, in its view, was healthier for the electoral contest and the political system.

In justifying their opinions, some Justices argued that the coalition patterns, as they existed in Brazil, impeded the appearance of strong national parties. Hence, the most relevant national issues were being conditioned by less comprehensive regional interests.

With the opinion in their hands, the TSE Justices decided to impose the rule of verticalization as from 2002 to reform the partisan system and contribute to democratic consolidation.

The legal argument that conducted the decision drew on an interpretation of Article 17 of the Constitution that sets forth that political parties must have a “national character”. Since, according to the opinion, by establishing asymmetrical coalitions political parties would be submitting the national character to regional particularities, the practice was found to be unconstitutional.

Consequently, a new norm was introduced into the Brazilian normative framework guiding the formation of partisan coalitions. It is important to stress that the new rule came into effect without having been submitted to traditional political procedures. It was not the result of a bill or an amendment to the Constitution, nor was it sanctioned by parliamentarian majorities. It was simply the result of an innovative interpretation of the law.

The reaction of the political parties to the decision was obvious, and the developments of this reaction were a rather clear revelation of the modus operandi of our electoral governance.

Claiming that the TSE ruling had failed to comply with the constitutional text in two aspects – first, because it restrained partisan autonomy and second, because it would be changing electoral rules less than a year before the elections – many political parties filed injunctions so as not to comply with the new rule. None of these arguments, however, were backed by the Constitutional Court and the decision by the TSE prevailed.

Two elections were held under the rule of verticalization (2002 and 2006) until Parliament managed to overcome the rule by amending the constitutional text in order to once again allow political parties to engage in electoral alliances for the 2010 elections.

In the case of verticalization, therefore, there was an interpretation of the constitutional text developed by the highest EMB that altered the practice in force regarding political coalitions altogether. Thus, the status quo of political competition was changed by a new rule without the new rule having been debated, voted or sanctioned by the traditional political representative bodies

Number of Councillors

The issue of the number of councillors shows an interaction that is a little different from the previous case, yet is also revealing of the effects that the electoral governance model adopted in Brazil can yield.

In 2004, a year of municipal elections (mayors and councillors), the STF decided that a part of the municipal organic law of Mira Estrela, a small northern municipality in the state of São Paulo, regarding the city's definition of its number of councillors was unconstitutional.

The ruling was based on the interpretation of a word of the constitutional text. In line with the widespread redemocratization spirit, the Constitution granted autonomy to the municipalities to define the size of their legislatures within limits set forth therein. Municipal organic laws could define the number of councillors based on three tiers linked to a demographic standard of proportionality. Thus, small, medium and large municipalities would have autonomy within certain minimum and maximum limits.

The argument of the STF's winning thesis was that, despite its having set three classification tiers, the constitutional text made reference to the need to respect proportionality in the number of councillors and the size of the population. Even though the municipality in question was within the limits established by the constitutional text, the STF Justices understood that the city had failed to comply with the proportionality principle, as other and bigger municipalities had fewer councillors.

To support the thesis, the case's rapporteur recommended criteria to establish the principle of proportionality. This principle prompted the creation of 36 classification tiers, rather than the three constitutionally established tiers.

In principle the STF ruling should only be applicable inter partes, that is, it should affect only the municipality in question. For the decision to be extended to other municipalities another petition must be filed with the STF, which would review the concrete case and hand down another ruling or uphold the decision made before.¹¹

The outcomes were, however, different. Soon after the STF ruled the unconstitutionality of the municipal organic law, the TSE published a resolution extending the same reasoning developed at the STF to all other Brazilian municipalities. Instead of the three constitutional tiers, municipalities had to adapt to the 36 new tiers defined by the criteria of the Electoral Justice System.

This resolution by the TSE was even harder-hitting because mayoral elections were being held that year. The decision, therefore, altered the magnitude of proportional elections in several municipalities.

In order to better measure the impact of that decision, when we compare the total number of councillors in Brazil in 2000 with the number running in the 2004 elections, we realize that the resolution by the highest EMB reduced by 14% the number of seats in the municipal legislative bodies.

The singularity of this case in relation to the case of verticalization is that the electoral rule was altered by the TSE after a decision by the Constitutional Court. In the previous case the decision by the TSE had been backed by the STF, whereas in this case the STF ruling was magnified by the TSE. Although the interactive pattern is not exactly the same, the closeness between the EMB and the Constitutional Court is worth pointing out. And it is this closeness that may have contributed to the advancement of electoral governance on rulemaking.

Table 1
Difference in number of councillors from 2000 to 2004

Region	Councillors 2000	Councillors 2004	Absolute variations	Variation (%)
BR	60,317	51,819	8,475	-14%
N	4,506	4,186	320	-7%
NE	19,625	16,539	3,064	-16%
SE	19,389	15,842	3,547	-18%
S	12,127	10,967	1, 160	-10%
CO	4,669	4,285	384	-8%
BR - Brazil / N - North / S - South / CO - Central-West / NE - Northeast / SE - Southeast

Source

What is interesting is that both at the STF and the TSE the prevailing opinion was that municipal autonomy had created a promiscuous and unhealthy mechanism with regard to the use of public funds. According to this opinion, by allowing local politicians to decide autonomously on the size of their legislative bodies the political system had created incentives for this power to become a bargaining chip in political negotiations or, worse, to meet the interests of groups.

The concrete data used by the Justices to back their opinions was municipal growth after the 1988 Constitution.

Table 2
Evolution of the number of municipalities in Brazil, 1980-2004

Region	1980	1991	1993	1997	2001	2004	Growth over 1980-2004 period
BR	3,991	4,491	4,974	5,507	5,561	5,562	1,571	39.4%
N	205	298	398	449	449	449	244	119%
S	719	873	1,058	1,159	1,189	1,188	469	65.2%
CO	282	379	427	446	463	465	183	64.9%
NE	1,375	1,509	1,558	1,787	1,792	1,792	417	30.3%
SE	1, 410	1,432	1,533	1,666	1,668	1,668	258	18.3%
BR - Brazil / N - North / S - South / CO - Central-West / NE - Northeast / SE - Southeast

Source

Thus, even though the number of municipalities rose by 39% over this period, the number of councillors was reduced by 14% as a result of the criterion applied by the EMB.

No sooner had the TSE published the resolution than the political parties reacted leading a Constitutional Court claiming that: 1) the resolution published in April 2004 was against the constitutional principle that bans electoral laws from being altered less than a year before elections (principle of annuality) and 2) it also hindered municipal autonomy, altering the federative pact safeguarded in entrenched constitutional clauses and, further, that the TSE had infringed the principle of separation of powers, making laws in the Legislative Branch's place.

In a quasi-unanimous decision, with the single dissenting vote of Justice Marco Aurélio, the STF basically decided that the resolution was based on an interpretation of the constitutional text and could therefore not be considered a law. Such an interpretation would simultaneously rule out the theses of infringement of separation of powers, of the principle of annuality in electoral matters — i.e., electoral rules cannot enter into force the same year as they are issued — and amendment of an entrenchment clause of the Constitution.

Once again the thesis that resolutions issued by the TSE cannot be disputed in their (un)constitutionality, as they cannot be characterized as norms, was present. And this argument was upheld both in the case of verticalization and of the councillors, even though the resolutions in both cases clearly brought new rules for partisan competition.

In addition to those two Direct Unconstitutionality Actions, several petitions were filed with the STF seeking to reverse the ruling. None of these suits was successful and, unsurprisingly, the STF confirmed the number of councillors for all municipalities as established by the TSE resolution because, after all, that decision had been based on the opinion formulated by the STF itself.

Two elections for municipal legislatures were held under the criteria applied by the EMB, 2004 and 2008. For the upcoming municipal elections (2012) the federal lawmaker amended the constitutional text to reverse the court's decision, thus recovering the nearly 8,000 seats suppressed in 2004.

In the case studied, the electoral governance model adopted made it possible for an opinion released in an electoral year by the STF to quickly become effective for all municipalities. A decision by the Supreme Court with effect inter partes ended up producing effects erga omnes by a decision of the TSE. Had it not been for the intersection between the TSE and the STF, the matter most likely would not have had the trajectory it had, or at least not as easily.

Party Loyalty

In the first half of 2007, when the TSE answered a consultation filed, it decided that the terms of office of parliamentarians elected in 2006 who had just been sworn in belonged to the parties they had been elected by, not to the parliamentarians. Consequently, they could be disqualified from their offices if they changed parties during the Legislature.

The consultation to the TSE was proposed by political party Democrats (DEM) in the following terms: “[do] parties and coalitions have the right to preserve a seat obtained through the proportional electoral system when there is a request for cancellation of an affiliation or a transfer of a candidate elected by one party to another?”.

In the consultation, the party made its case as to what it expected the TSE justices to consider in their answer. The parties requested that the following situations be included: 1) the election of candidates to proportional offices is the result of the electoral quotient as achieved by the various parties and coalitions, 2) party affiliation is a constitutional condition for eligibility that serves to indicate to the voter the link between the candidate and the party's ideology and 3) the calculation of averages is a function of the valid votes given to parties and coalitions.

The Brazilian proportional, open-list electoral system is also called the single transferable vote system. That is, by voting for a candidate of a given party and/or coalition the voter transfers votes to the remaining candidates, who will benefit from the party's total votes.

The consultation was therefore guided by the following rationale: as party/coalition elected candidates benefit from the nominal votes given to all the candidates on the list and from the votes given to the party, the lawmaker's office, rather than belonging to the candidate, belonged to the party. And the TSE's answer confirmed that.

This debate is quite old in Brazilian politics. In Congress there are several bills addressing the issue, while the academic literature already features very robust studies focusing on the phenomenon (Melo 2000; Desposato 2006).

As part of the redemocratization process in Brazil, in 1985 Congress approved a Constitutional Amendment (EC 25) that withdrew from the text a penalty of disqualification from office for those lawmakers who switched parties. It was then that what the literature has come to call the first phase of partisan switching began.

This phase was characterized by the accommodation of political parties as a result of the recently regained political liberties. With that, new political parties were founded and a new competitive environment was established, with the progressive adoption of direct elections at all levels of the federation.

Almost at the same time, petitions began to be filed, with the TSE and the STF seeking the annulment of the migrants’ tenures to secure seats won for the party. In March 1989, the TSE received the following consultation: “will a councillor elected by a given political party, upon joining another party, keep the office he/she was elected for?” The court's opinion was pronounced without the records of the plenary debates or any other manifestation that would allow us to conclude there was any evidence of doubt regarding the matter: “In our legal order there is no annulment of tenure for party disloyalty” (CTA Nr. 9.948/89).

The second phase of the party switching pattern started soon after this initial accommodation. As from the legislature that began in 1991 in the National Congress, there began a new party switching pattern, for “the movement of the deputies started to reflect a rationale internal to the partisan political competition” (Melo 2000, 6).

In other words, the party switching pattern as of 1991 reflected the way the partisan game has unfolded in Brazil under coalition presidentialism. Driven by the need to build their political bases, political parties that supported the Executive stimulated party switching, while parliamentarians, seeking their own immediate career interests, started shifting across the various partisan options. Table 3 shows partisan switching figures in Brazil.

Table 3
Partisan switching from 1983 to 2007

Legislature	Total representatives	Total migrants	% of migrants
1983 to 1987 (47th)	528	165	31.3
1987 to 1991 (48th)	560	154	27.5
1991 to 1995 (49th)	620	200	32.3
1995 to1999 (50th)	621	167	26.9
1999 to 2003 (51st)	644	171	26.5
2003 to 2007 (52nd)	618	197	31.8
Total	3,591	1,054	29.3

SourceMelo (2000)

When the TSE received the consultation on party switching in 2007, a strategy in the legal argument proved decisive for the ruling handed down. It is worth noting that in the terms of the question formulated by the DEM there was no express intention to disqualify a parliamentarian from office, only the intention of defining the titularity of the parliamentarian's office: did it belong to the person elected or to the party?

As the Constitution is quite precise in Article 55 as to the roster of factors that can lead to a parliamentarian's disqualification from office – and changing political parties is not on that list –, the decision was not based on that norm. The decision ultimately invoked other constitutional provisions.

The argument that gained centrality in the discussions was the express condition for eligibility for partisan affiliation. If the existence of candidacies independent from political parties is not allowed in the country, the right to the office won belongs to the party and not to the candidate. The reasoning is inverted.

The other argument involves the open list proportional representation. After all, in this system, votes are distributed according to partisan quotients, thus creating a bond between the parliamentarian elected and the party for which he/she ran in the elections.

At any rate, even though the legal argument built on distinct issues, upon deciding that the titularity of the tenure belongs to the party, in practice, the Court re-established the penalty of disqualification from office for those parliamentarians who switch parties during their tenures. After all, if the office does not belong to the parliamentarians, when migrating from one party to another, representatives cannot take their offices with them.

This practical outcome of the decision was made clearer after the publication by the EMB of Resolution 22,610/07 establishing the criteria governing the rights of parties over the offices won and the judgment of those who leave the parties by which they were elected. Besides, in answering another consultation, it included those elected through the majoritarian system in the rule; thus, any politician elected in Brazil after 2007 who migrates to another party is subject to disqualification from office.

There was still some political resistance to the application of the new rule. Even as recently as 2007, parties that had lost the most parliamentarians were requesting to the directing board of the Chamber of Deputies the right to occupy the seats that belonged to those deputies who, having been elected by the party, had left it immediately afterwards.¹²

The directing board of the Chamber rejected the request prompting parties to seek the STF in search of what they deemed to be a right recognized by the EMB. Each of them filed an injunction. In October 2007 came the decision by the STF, ratifying the thesis by the TSE that the office belonged to the parties and not the candidates, obliging the directing board to accept the decision.

The 53^rd legislature of the Chamber of Deputies had 23 of its lawmakers’ offices challenged at the TSE as a result of party switching. Of the 23, only one parliamentarian was found guilty. That is, only one of the party-switching cases was ruled as “without just cause” by the TSE.

Beside the cases under its original jurisdiction, the TSE received more than 2,000 appeals to review decisions made by lower EMB spheres (TREs). By the end of 2008, for instance, some 1,308 councillors had already been disqualified from their offices in the Municipal Chambers (close to 2.5% of all the country's councillors) due to party switching.

As in the cases we saw earlier, in Congress there are Constitutional Amendment Bills (Propostas de Emendas Constitucionais (PECs)) with the aim of overhauling judicial rulings. According to one of the PECs, there must be a 30-day “moratorium” prior to the end of the electoral registrations, dubbed by the public opinion the “infidelity window”, during which party transfers may take place without any sanction. As yet, however, it is the TSE's resolution that prevails.

Yet again, we have the EMB interpreting the constitutional text through a consultation and deeply altering a practice current in the political game. Furthermore, once again we have interactive cooperation between the Constitutional Court and the EMB. When the STF was asked to give an opinion, it adopted the same opinion on constitutional norms as had been handed down by the TSE.

What we are arguing here is that had it not been for the electoral governance model adopted, the EMB would not have advanced over the constitutional text and the attunement with the interpretation given by the Constitutional Court would not have been as fine.

This case is particularly emblematic as to allow us to state that the highest EMB sphere (the TSE) is indeed an STF body for electoral matters. In addition to being able to advance over the constitutional text, its electoral governance powers have enabled it to incorporate judicial interpretations in the resolutions it drafts to regulate the partisan political game. Thus, besides being the rule adjudicator and rule applicator, the TSE once again acted as rulemaker.

Conclusion

The key argument in this paper was that the electoral governance model adopted in Brazil combines elements conducive to an environment of judicialization of political competition, namely: the concentration of governance activities in a single EMB, exclusion of any political actor from the nomination and/or selection of members of the Electoral Body, the “rule of intersection” with the Judiciary, and control of the TSE – the highest EMB sphere – by the Constitutional Court.

At the same time as they removed partisan political interests from electoral governance, these elements, combined, transformed the TSE into an STF body for electoral matters, vesting the former in such institutional power as to even allow interpretations regarding the political game to include constitutional norms.

We also argue that this model has engendered two main outcomes: 1) the insulation of the EMB and 2) the conditions for the EMB to become a rule maker as well.

As for the first outcome, we can say that as the TSE merged with the Judiciary Branch, it developed a shield against the interference of any representative institution, whether political, social or partisan interest groups. If this granted it independence to act, it may also have reduced opportunities for accountability.

Surely we could argue that the Electoral Justice System's accountability has not been affected; after all, two of the rulings analysed here were later reversed by Parliament (verticalization and number of councillors), and the same is bound to occur with the other ruling (party loyalty).¹³

However, one must bear in mind the high costs incurred by Parliament to reverse these rulings. Firstly, because such rulings set conditions for the electoral contest, for at least two consecutive terms, that were far from the curve of preference of the political actors.

Secondly because, in order to reverse the rulings, Parliament had to amend the constitutional text, hence prompting two outcomes (costs): 1) the need to achieve qualified majorities – a task which, in a markedly consensus-dependent institutional environment as the Brazilian one, is even more difficult, and 2) the entrenchment of further details in a Constitution already criticized for the number of relationships it seeks to regulate (Couto and Arantes 2008).

We argue that the Brazilian electoral governance model's accountability flaws stem not only from the costs entailed in reversing such rulings, but first and foremost, from a lack of institutional mechanisms designed to avert such costly decisions before they come to affect the status quo.

At this point it is important to remember that the most creative TSE decisions have been made in response to “consultations” (see note 6). This situation is a function of the level of insulation attained by Brazil's Electoral Justice System and of how this feature reduced opportunities for accountability. After all, let's face it, deeply altering electoral rules by means of “consultations” does not favour debate, especially not the checks and balances system.

As for the second outcome we can say that by merging the EMB's highest sphere with the Constitutional Court, the model adopted created institutional conditions such that rulings on electoral competition may be based on innovative interpretations of the constitutional text. This allowed the EMB to make decisions that ultimately prompted the overhaul of competitive rules, thus altering the normative framework without mobilizing traditional lawmaking spheres.

For in the cases of the verticalization of coalitions and partisan loyalty, as it answered consultations, the TSE promoted original interpretations of the constitutional text which, challenged at the STF, were upheld by the Constitutional Court. Indeed, in the case of party loyalty, it was precisely an interpretation by the TSE of the constitutional text that informed the majority vote at the STF. In the case of the number of councillors, there was an inversion in the interaction. As it interpreted constitutional norms, the STF altered the rules of the competitive game, magnifying the outcomes of its decisions through decisions made by the TSE.

Importantly, the debate on the need for comprehensive reform in the Brazilian political system is as old as it is controversial. We cannot say that there is consensus on the need for it – so little compared to what kind of reform needs to be done (Rennó 2007). The fact to note here is that the most striking changes on the rules of political parties did not come from the classical political arena. Political reform in Brazil in recent decades has advanced more by judicial decisions than by policies.

There is still much to be researched in Electoral Governance in general and the Brazilian model in particular. This paper has sought to contribute to fill this gap by shedding light on an institution that is as important as it has been neglected by our literature, and also by defining some core characteristics of the institutional design of Brazil's electoral governance model and its effects on the country's partisan political competition.

• 1
  Some of the arguments developed here are in Marchetti (2008).
• 4
  Such permission is based on a ruling by the STF in Direct Unconstitutionality Action 1,127/94 interpreting the statutes of the practice of law and of the Brazilian Bar Association Ordem dos Advogados do Brasil (OAB) (Law 8,906/94). This law sets forth an impediment to exercise law for “members of bodies of the Judiciary Branch, public attorney's offices, audit courts and boards, ad hoc courts, peace justices, class judges, as well as all those holding judging offices in collective deliberation bodies of the direct and indirect public administration”. The STF opinion opened an exception in the case of Electoral Justice System.
• 5
  On very few occasions has the STF reviewed appeals involving electoral matters. An important precedent was set with the ruling in Direct Unconstitutionality Action 4,018 of February 2008, when the STF ruled for the unconstitutionality of a decision of the Goiás State TRE. A recent decision on the application of the Clean Slate Law in the 2010 elections was quite significant in this respect. After a drawn ruling, the STF decided that, instead of ruling on the matter, what had to prevail was the constitutional interpretation developed in the TSE ruling. The Constitutional Court stated that the last word in interpreting the constitutional text should be the TSE's (on the decision, see Marchetti (2011)).
• 6
  I am grateful to Julio Rios-Figueroa for this remark.
• 7
  A more detailed analysis of this issue can be found in Marchetti (2010).
• 8
  The consultation is an administrative instrument that aims to solve doubts regarding electoral matters. Thus, for a consultation to be submitted and accepted by the TSE there is no need for litigation to exist or for different parties to be involved. It only requires that there be doubt regarding the legislation and that it be formulated to the TSE by an authority with national jurisdiction or by a political party's national body.
• 9
  There is no clear and definitive information regarding the party's motivation for such consultation. After all, as we will demonstrate further ahead, the practice had been present in the two previous elections, thus allowing for very few doubts. However, we will not venture into speculating about motives as our interest is to know how the Electoral Justice made its decision.
• 10
  It was only in the 1994 elections that electoral races for different levels started to take place simultaneously, as the presidential elections began to be held on the same date as the gubernatorial elections.
• 11
  To better understand the hybrid characteristics of the Brazilian constitutional control (diffuse and concentrated) see Arantes (2005).
• 12
  In proportional terms, the Socialist People's Party (PPS) lost the most candidates, a third of the 21 elected. The Brazilian Social Democracy Party (PSDB) lost 13.6%, and the DEM (Democrats), 12.3% of the seats won in the elections.
• 13
  Again, I am grateful to Julio-Rios Figueroa for this critical remark.
• Revised by Robert Stuart

Bibliographical References

• Arantes, Rogério. 2005. “Constitutionalism, the expansion of justice and the judicialization of politics in Brazil”. In. The judicialization of politics in Latin America, ed. R. Sieder, L. Schjolden and A. Angell. New York: Palgrave.
• Braga, Maria do Socorro. 2006. O processo partidário-eleitoral brasileiro: padrões de competição política (1982-2002) São Paulo: Humanitas.
• Couto, Cláudio, and Rogério Arantes,. 2008. Constitution, government and democracy in Brazil. World Political Science Review 4 (2): 1-33.
• Desposato, Scott. 2006. Parties for rent? Ambition, ideology and party switching in Brazil's chamber of deputies. American Journal of Political Science 50 (1): 62-80.
• Elklit, Jorgen, and Andrew Reynolds. 2000. The impact of election administration on the legitimacy of emerging democracies: A new research agenda. 96^th Annual Meeting of the American Political Science Association (APSA), Washington DC.
• Fleisher, David, and Barreto, Leonardo. 2009. El impacto de la Justicia Electoral sobre el sistema político brasileño. America Latina Hoy, no. 51:117-38.
• Hartlyn, Jonathan, Jennifer McCoy, and Thomas Mustillo. 2008. Electoral governance matters: Explaining the quality of elections in contemporary Latin America. Comparative Political Studies 41 (1): 73-98.
• Huntington, Samuel. 1993. The third wave: Democratization in the 20th Century Norman: University of Oklahoma Press.
• Lijphart, Arend. 1999. Patterns of Democracy: Government forms & performance in thirty-six countries. New Haven: Yale University Press.
• Limongi, Fernando, and Rafael CORTEZ. 2010. As eleições de 2010 e o quadro partidário. Novos Estudos, no. 88: 21-37.
• Marchetti, Vitor. 2008. Governança eleitoral: o modelo brasileiro de Justiça Eleitoral. Dados 51 (4): 865-893.
• Marchetti, Vitor. 2010. “O TSE (Tribunal Superior Eleitoral) e a “verticalização” das coligações”. In Coligações partidárias na nova democracia brasileira: perfis e tendências, ed. S. Krause, H. Dantas and L. F. Miguel. São Paulo: Konrad Adenauer/ Ed. Unesp.
• Marchetti, Vitor. 2011. A Ficha Limpa no contexto da governança eleitoral brasileira. Interesse Nacional, no. 12:33-42.
• Melo, Carlos Ranulfo. (2000). Partidos e migração partidária na câmara dos deputados. Dados 43 (2).
• Mozaffar, Shaheen, and Andreas Schedler. 2002. The comparative study of electoral governance – Introduction. International Political Science Review 23: 5-27.
• Pastor, Robert A. 1999. The role of electoral administration in democratic transitions: Implications for policy and research. Democratization 6 (4): 1-27.
• Rennó, Lucio. 2007. “Reforma política: consensos necessários e improváveis”. In A democracia brasileira: balanço e perspectivas para o século 21, org. Carlos Ranulfo Melo and Manuel Alcántara. Belo Horizonte: UFMG.
• Sadek, Maria Tereza. 1995. A Justiça eleitoral e a consolidação da democracia no Brasil. São Paulo: Konrad Adenauer.
• Schedler, Andreas. 2002. The nested game of democratization by elections International Political Science Review 23: 103-122.
• Taylor, Matthew. 2008. A blind eye and bound hands: Electoral courts, the federal Judiciary and political accountability. Paper prepared for the Accountability Institutions and Political Corruption in Brazil Workshop, in Oxford, London.

Publication Dates

• Publication in this collection
  27 Oct 2023
• Date of issue
  2012

History

• Received
  Dec 2011
• Accepted
  Apr 2012