Turkeys Do Not Vote for Christmas. The Brazilian Anti-Vote-Buying Law

Andreia reis do Carmo

doi:10.1515/soeu-2018-0027

Article Publicly Available

Turkeys Do Not Vote for Christmas. The Brazilian Anti-Vote-Buying Law

Andreia reis do Carmo

Published/Copyright: September 11, 2018

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Comparative Southeast European Studies Volume 66 Issue 3

Abstract

The research to date shows conflicting results about the effect of perceived corruption on political participation by the public. Corruption perception can mobilize, demobilize, or show no impact on political participation. The Brazilian Anti-Vote-Buying Law of 1999, one of the fastest laws ever to be approved in the country, is an example of how perceived corruption can increase public engagement. Using process tracing, the author tells this story identifying two sets of factors ranked according to their causal power (distal and proximal conditions) and further test the theory. The Brazilian population, who perceived corruption as high, learned about the antidemocratic aspects of vote buying and how to curb it. This ‘informed’ perception of corruption led people to politically participate by signing the popular initiative.

Introduction

In Brazil, about 6 million people (6% of voters nationally) were offered money for their votes in the 2000 municipal elections.^[1] Vote buying became the ‘top reason politicians are removed from office: 667 at the municipal level and 31 state and federal politicians’ between 2000 and 2008.^[2] These results were only possible because of a popular legal initiative. This article tells the story of the Anti-Vote-Buying Law (Lei da Compra de Votos, AVB) and analyses it using process tracing.

In 1999, Brazil was the fifth largest population (170 million people), had the eighth largest gross national product, and was one of the most unequal countries in the world^[3] (with a Gini coefficient of 0.59).^[4] Since 1977 it had been remarkably stable, ranging from 0.52 to 0.62, but in 1989 it reached 0.64.^[5] Regionally, Brazil’s life expectancy, literacy, and poverty rates vary tremendously.^[6]

Although big and diversified, Brazil mobilized nationally to enact the AVB Law. Greater popular participation was enabled by an awareness of levels of political corruption and how vote buying could be one of its causes.^[7] The deterrent effects of corruption on democracy are obvious. However, the Brazilian case is an example of how perceived corruption, if informed, can be beneficial.

The papers that study the effects of corruption perception on participation are discussed in the literature review. The methodology applied is then briefly debated. The story of the AVB Law is laid out, alternative explanations disregarded, and its distal and proximal conditions (elements that carry less or more causal power) identified. Finally, I test the theory with process-tracing tools.

Literature Review

Vote Buying

Vote buying distorts democratic procedure towards private sector interests,^[8] enabling candidates to buy poor voters.^[9] It is also harmful for institutions, which reproduce forms of political exclusion. Cyclically, the ineffective functioning of institutions caused by vote buying affects the distribution of income, increasing inequality.^[10]

Besides compulsory voting, Brazilian ‘weak parties, [and] feckless democracy’^[11] increase the incentives for vote buying due to the country’s candidate-centred electoral rules. The personalist nature of Brazilian elections is a way to overcome the lack of party identity and weak programmatic appeal. Curbing the issue can be achieved through electoral laws^[12] and increasing awareness.^[13] Nevertheless, ‘Brazil has adopted a spectrum of laws that look robust, but they have not seen a lot of change. Corruption is still stable’.^[14]

Perceiving vote buying as a wrongdoing is not as easy as one might think. ‘The democratic norm of fair elections and clean government is relatively universal [in Latin America], but so are norms about gratitude and solidarity.’^[15] Through these two conflicting norms, people understand corruption in a social-systemic or in an individual way. In the individual understanding, people look at vote buying through the ‘gratitude and solidarity norm’ lens, thus it is tolerated and seen as a common occurrence. Contrarily, those who look at vote buying through the ‘democratic norm’ lens disapprove of corruption.

Both types of understanding depend on previous experience with vote buying (not distinguished by the authors from perception or tolerance of vote buying), political sophistication (measured by education), partisan identification, and feelings of reciprocity.^[16] Thus, vote buying is less stigmatized by people who directly experience it, who are poor, who identify with parties that accept it, and by those who value reciprocity.

Corruption as a Mobilizing Force of Participation

The mobilizing power of corruption in Latin America depends on whether it is perceived, experienced, or tolerated.^[17] Citizens who have been asked for bribes (experience) and those who believe ‘it is justifiable to pay bribes sometimes’ (tolerant) are more inclined to participate in contacting public officials, protesting, and community activism. Voter turnout is an exception. The effects of perception of corruption depend on the type of participation.

Focusing on experience with corruption and participation in six Latin American countries, the positive link between the variables remains.^[18] The more individuals are asked for bribes, the more likely they are to participate politically, regardless of levels of grand corruption or the intensity of socialization of corruption acts in each country.^[19]

Experience and perception of corruption also have distinct impacts on democratic legitimacy and satisfaction with democracy. Experience with corruption is detrimental to democratic legitimacy, but corruption perception is not.^[20] Conversely, the perception of corruption is harmful to satisfaction with democracy, but experience with corruption is not.^[21]

In Eastern European new democracies, corruption perception does mobilize people to participate politically. Voters may feel a psychological need to express in some way the anger caused by greedy politicians. The expression of this anger brings emotional benefits as well. Anger can create instrumental incentives for citizens to participate, and in turn these voters develop a sense of accomplishment and fulfilment of their public duties.^[22] The meaning of corruption also matters. Citizens who understand corruption as originating from incompetence and illegitimacy tend to be motivated to remove the incumbent by participating more actively. However, if citizens see corruption as meaning they cannot trust politicians, participation is diminished.^[23]

Regarding the types of participation, results show that corruption perception ‘depresses people’s motivations to support petitions’ but stimulates ‘unofficial strikes and occupation of buildings’,^[24] and makes ‘passive voters’ (i.e. those who usually abstain) and previous incumbent voters feel more mobilized and able to vote for the opposition. Despite these results, it is recognized that corruption perception can depress voter turnout as well as motivate it, implying that there is a threshold when corruption perception inverts its effect on participation.^[25]

In Africa, perceived^[26] and experienced^[27] corruption are important factors for participation. Citizens of Senegal who believe their leaders are corrupt are more likely to participate. This result is shown in an experiment and in a national survey.^[28] The effect of corruption perception is even bigger when it comes to protesting.^[29] When experience of corruption is investigated, measured by the payment of bribes, an intriguing result appears: paying bribes motivates collective actions when the experience is relatively rare, whereas when bribe paying becomes more common, collective action is diminished.^[30]

There is a general preference among Africans to protest as a reaction to corruption allegations. More and more Africans choose to protest if the number of bribes paid also increases. Surprisingly, if the number of bribes continues to increase, it comes to a point when participation in protests falls.^[31] If corruption is too ingrained in society, its mobilizing powers are precluded.

Corruption as a Demobilizing Force of Participation

In Europe, citizens embedded in systemic societal grand corruption are more loyal to corrupt politicians, intensifying the division between those beneficiaries of the political system and those who are left aside. Outsiders express their grievances through alienation, resignation, and exit instead of indignation and voice. Loyalty, demobilization, and divisiveness explain the persistence of corrupt politicians in power. Where corruption is systemic, individuals who face the risk and costs of mobilizing towards accountability are rare.^[32]

In Brazil, corruption depresses participation through cynicism. Voters do not ‘throw the rascals out’ because they believe all candidates are the same. This belief derives from actual levels of corruption and not from an uninformed citizenry.^[33] Interestingly, several other research studies that find evidence for the demobilizing power of corruption perception explain this effect through the level of information disseminated in society.

‘Widespread suspicions about fraud and corruption in Mexico did affect electoral outcomes by making it less likely that potential opposition supporters turned out to vote.’^[34] Suspicion is caused by corruption information relayed through mass media, which is easy and ‘cheap’ for Mexicans to consume. ‘The information is typically presented as a “morality play”, [and it is] accessible to voters who rely heavily on television to obtain their information.’^[35]

Also in Mexico, information that is disseminated about corruption can have harmful results, whereas general information about citizens’ rights, government responsibility, and so on can promote accountability through transparency. When information on accountability is more visible and available through the media, government responsiveness is promoted, opportunistic behaviour is reduced, widespread theft of public resources is prevented, and parliamentary performance is improved.^[36] In Costa Rica, if the information available to voters is about how widespread corruption is in their country, they not only demobilize, but also engage in corruption.^[37]

Corruption may also have indirect impacts on participation, with ideology, retrospective voting, and economic conditions being more important variables.^[38] Whether it is positive, negative, or neither, the relation between corruption awareness and engagement in politics depends on, among other things, the type of participation (electoral or non-electoral), on the meaning of corruption, and on its salience.

Methodology

Qualitative analysis of mobilizing perceived corruption is rare and can potentially explain the cases of demobilization.^[39] The search for countries with moments of high corruption perception and intense participation led me to the Brazilian Anti-Vote-Buying Law. As a case study, the AVB Law is primarily useful for descriptive inferences but is also a powerful tool for investigating causality,^[40] which ‘more plausibly subsist[s] on the basis of a single-unit study’, therefore enabling us to ‘know more about less’ as opposed to ‘less about more’.^[41]

One tool with which to analyse case studies is process tracing: ‘The systematic examination of diagnostic evidence selected and analysed in light of research questions and hypotheses.’^[42] This method identifies the intervening causal processes, i.e. the causal chain and mechanisms in between the dependent and independent variables.^[43] Identifying causal mechanisms involves combining ‘general knowledge of the world with empirical knowledge of how X and Y interrelate’.^[44] Testing them consists of identifying the several steps composing a causal chain, specifying the empirical implications of each step, and finally looking for evidence that supports these steps at the micro-social level.^[45]

Because the effect of information on transforming ideas is an important mediating variable of the AVB Law, I use a specific type of process tracing called ideational process tracing. This analytical tool treats ideational causes differently from material causes for the simple reason that material incentives are easier to observe than ideas. Tracking ideas is useful not only to understand behaviour, but also to grasp ‘continuity and change, sequences of events, flows of information, and movements of actors across institutional settings over time’.^[46]

Although somewhat unobservable, ideas also serve an instrumentalist logic, maximizing individuals’ preferences given their causal beliefs.^[47] Ideas are visible through ‘communication and behavioural interaction among individuals and organizations’.^[48] This specific type of process tracing tracks the flow of information transforming ideas and becoming a strategic incentive to change behaviour. Ideational analysis pays ‘close attention to the ways in which the institutional and political contexts of choice generate strategic incentives’.^[49]

Ideational theory, the analytical product of this type of process tracing, is a causal explanation that links the actors’ cognitive structures to their choices.^[50] Whereas in ideational theories the variation lies in the ‘content of actors’ cognitions’,^[51] in materialist explanations it lies in an individual’s objectives. In this framework, it is assumed that the choices actors make are not fully determined by material gains, but instead by their beliefs. Testing these theories requires looking for evidence that an individual’s choices were influenced by their ideas.

With this framework in mind, this article delivers three products. After the narrative of the AVB Law, three alternative explanations are disregarded. The second product is delivered in the section entitled ‘Process Tracing Results I’, which lists the most important causes that led to the enactment of the AVB Law. These causes are organized in two groups: distal and proximal conditions. Distal conditions are events that have less causal power but were crucial for setting the background against which this process took place. Proximal conditions carry more causal power and are mostly closer in time to the outcome. I identify five distal and four proximal conditions to the enactment of the AVB Law. A visual ‘causal map’ is found in Figure 1 in the appendix.

Figure 1

Causal map. A visual representation of theory.

The third product delivered is the further testing of this theory under the section ‘Process Tracing Results II’. I carry out two tests: a double decisiveness and a hoop test. Both are useful to confirm one hypothesis and eliminate others. Double decisiveness tests provide a necessary and sufficient criterion for establishing the hypothesis, whereas hoop tests only follow the logic of necessity. Moreover, hoop tests identify evidence that is general, establishing a background with high levels of certainty.^[52]

The data were collected during fieldwork in Brasilia, Brazil, from July to October 2014. I carried out eight in-depth interviews^[53] in a snowball sampling fashion.^[54] I had access to and scanned 287 archival records from Mr Whitaker’s personal archives and from the National Confederation of the Bishops of Brazil (CNBB). For the collection of data about the AVB Law’s political process, I used the Lower Chamber and Congress Diaries, which are available online.^[55]

The Story of the Anti-Vote-Buying Law

Buying votes was considered a crime according to the Brazilian Electoral Code of 1965. Elections would be cancelled if tainted with ‘falseness, fraud, coercion, […] or by the illegal employment of advertisement procedures or the captivation of votes prohibited by the law’.^[56] The AVB Law stated that:

‘It is established as captivation of suffrage, restricted by this law, the donation, offering, promise, giving, by the candidate to the voter, in order to obtain their vote, of goods or personal advantages of any nature, including jobs or public offices, since the registering of the candidacy until the day of the election inclusive, under penalty of 1,000 to 50,000 UFIR and the abrogation of the registry or of the diploma.’^[57]

Paragraph 41-A amended the Electoral Code of 1965 and the Law of Elections (9304/1997) and instituted a punishment for vote buyers.^[58]

As a crime, buying votes did not incur punishment, for several reasons. In criminal prosecutions, both the accuser and the accused were at risk of being punished, which is an obstacle to gathering evidence.^[59] The possibility of filing appeals was also a problem. This process could go on for years, allowing politicians to finish their mandates.^[60] Moreover, the Brazilian justice system was slow and generally ineffective.^[61]

The bill was written by a working group composed of the Attorney General of Brazil at the time, Dr Aristides Junqueira Alvarenga; a former electoral judge, Dr Dyrceu Aguiar Dias Cintra Jr; and the regional public prosecutor from Ceará, Dr José Gerim Cavalcante. They understood that the ‘legislative gap’ to punish vote buying was in its interpretation. Vote buying had to be interpreted as an administrative improbity or an electoral violation, rather than a crime. The law could then be applied to the candidate and not to those who sold their vote.

As an administrative improbity, higher-ranking judges were not necessary, reducing the length of proceedings. This was the ‘masterstroke of the new legislation’.^[62] It allowed corruption to be curtailed despite the low level of judicial sanctions.^[63] Following completely different proceedings to criminal charges,^[64] the impossibility of running as a candidate, or having the diploma cancelled, was a bigger threat than the rare outcome of imprisonment.

After the law was validated in 2000, electoral courts understood that vote buying did not need to affect electoral results,^[65] and that for anyone, candidate or not, buying votes was punishable. Proof that one vote was bought was sufficient for the cancellation of the diploma or register. So where did the idea for this law come from?

The idea came from the Brazilian Roman Catholic Church, one of the biggest (73.57% of Brazilians were Catholic in 2000) and most progressive in the world. Throughout the 1980s, the Church promoted human rights and organized social classes restoring civilian rule and democratic practices.^[66] Being Catholic meant getting involved with social justice and political struggles.^[67] This changed vision, the ‘Liberation Theology’, had as its pillar the CNBB. The CNBB’s ‘values, ideals, and personnel […] penetrated political structures and social movements’.^[68]

In 1996, the Church issued a campaign called ‘Fraternity and Politics’. The Brazilian Commission for Justice and Peace (CBJP), a branch of the CNBB, focused on ‘Combating Electoral Corruption’. This side-project was motivated by the survey result from the DataBrasil Institute: 96% of participants believed that candidates bought votes.^[69] For the CBJP and its executive secretary, Francisco Whitaker, the solution to this problem had to come from a popular initiative and the bill was written.

Besides the Catholic Church, fifty-five civil society organizations supported the bill, mobilizing and gathering signatures. They all campaigned using the slogan, ‘A vote does not have a price, it has consequences’. The plan was to have a million signatures by late 1998 and to hand the bill to Congress in early 1999, validating the law for the 2000 municipal elections. However, by the beginning of 1999, the number of signatures was still far below the target. The difficulty in collecting signatures was due to the requirement that supporters provide their voter registry numbers (título eleitoral). Some citizens were also not convinced that a change in legislation would stop the practice of vote buying. Running the campaign in a national electoral period represented a third difficulty, as some thought the bill was a tool with which to attack political enemies.

There was a general social fear of repression and change. Collecting signatures represented a hard and time-consuming ‘teaching’ endeavour. Despite the obstacles, it was the CNBB’s understanding that the campaign should go on. Raising awareness about vote buying would be a victory in itself.

It was only when TV Globo National News broadcast a story about the popular initiative that the rest of the signatures were collected. In June 1999, after a three-minute broadcast, the daily average access to the CBJP website jumped from five to 5,000. The number of signatures was getting closer to the required target.^[70]

The involvement of the TV Globo Corporation was motivated by a corruption scandal known as the ‘Inspector’s Mafia’ in São Paulo. At the time (1997-2000),^[71] corrupt municipal inspectors fined construction sites if they were not bribed, delaying the corporation’s building projects.^[72] On television, these illegal actions were framed as a consequence of vote buying.

On 10 August 1999, the bill was submitted and had seven weeks to be validated for the 2000 elections. The bill did not proceed as a popular initiative but rather as a parliamentary initiative, to avoid the need to recount and check all the signatures and voter registry numbers. Eleven congressmen (one from each political party represented in Congress) subscribed the bill, which was approved with consensus within thirty-five days. It was one of the fastest laws ever to have passed in Brazilian Congress.^[73]

In the same year, the Lower House of Brazil issued the publication ‘The Combat against Electoral Corruption: The Proceedings of the First Popular Initiative Approved by the National Congress’. The publication highlights the alliance of Congress and civil society, intermediated by the Catholic Church, the moral guardian. Its somewhat romanticized narrative exalts society’s power, and proclaims how Congress, responding positively to it, accomplished its duty.^[74]

In 2002, the Movement against Electoral Corruption (Movimento Contra a Corrupção Eleitoral, MCCE) was founded. The partnership between civil society entities and the Catholic Church (under the CNBB) was now an official and permanent social movement. Their goal was to continue establishing the ‘9840 committees’, previously used to collect signatures, but since transformed into vote-buying watchdogs. In 2002 there were 130 committees in seventeen states; in 2011, 311 in all states.^[75] The committees also continued to raise awareness within communities.

Fieldwork investigation found evidence that the popular initiative was a mere constitutional tool to aggregate social support. This social support was made concrete through the population’s signatures. However, we must think of popular initiatives and social mobilization as two separate elements. Popular initiatives have less causal power than social mobilization itself. Moreover, considering this story a ‘spontaneous social and Christian manifestation’ undermines the role of key actors such as Mr Whitaker and Dr Aristides. I also argue against Nichter’s^[76] explanation, which considers the independent judiciary to be a necessary cause of the law. I deconstruct these alternative explanations below.

Alternative Explanations

The Myth of Popular Initiatives

Although documents claim that the vote-buying bill was the first popular initiative to take place, these same documents also affirm that the bill did not follow popular initiative proceedings in Congress.^[77] I believe this contradiction is motivated by the Catholic Church’s intention to make the AVB Law an example to be followed. As such, the narrative could not exclude popular initiatives because it channelled popular support. Popular participation became concrete through the carts full of pages being pushed into Congress,^[78] but it could have had any other format.

The popular initiative as a legal, technical, and constitutional mechanism did not exist. In fact, Congress did not fully appreciate the internal regulations involved.^[79] Because this new direct participatory tool was not entirely established in the constitution, it was easier to override it and turn it into an ordinary parliamentary initiative and avoid having to spend time recounting and checking signatures. Congressmen turned a blind eye to the lack of procedures for popular initiatives and facilitated the process because of social pressure.

The Myth of a Spontaneous Social and Christian Manifestation

The germ of the AVB Law can be linked to the biography of Mr Whitaker, and this is how he himself begins the story. Although Mr Whitaker, the leader of the movement,^[80] did not act alone, his previous political experience, his institutional role as executive secretary of the CBJP, and his social capital were determinant to the successful passing of the law.

In 1987, when the Brazilian constitution was being written, Mr Whitaker was one of the organizers of the ‘Pro-Popular Participation Plenary’ in the National Constituent Assembly. In 1988, he helped write São Paulo’s Organic Municipal Laws, based on the new constitution. From 1989 until 1992, Mr Whitaker was São Paulo’s city councillor twice under the Worker’s Party (Partido dos Trabalhadores). During this time, he participated in two Parliamentary Committees of Inquiry regarding corruption allegations, once as rapporteur and once as president. Being a politician had given Mr Whitaker direct experience of corruption and provided him with ideas on how to curb it.

Mr Whitaker was able to transfer his experience with direct participation, legislation, and political corruption onto the success of the AVB bill. He ended his political career and was invited by the Catholic Church in 1996 to be the executive secretary of the CBJP. His past Worker’s Party affiliation was overlooked by politicians, who thought of him as a Catholic man, and not as a politician. His religious and neutral status allowed him to access and talk to government and opposition parties equally about the bill.

Along with this curriculum vitae, Mr Whitaker had many contacts.^[81] He was a personal friend of Dr Aristides, one of the writers of the bill. They both had strong ties to the Catholic Church and Dr Aristides had been a member of the CBJP. Mr Whitaker also knew the local São Paulo TV reporter broadcasting against the ‘Inspectors’ Mafia’. This reporter linked Mr Whitaker to the director of the national TV channel, which ran the story nationally. President Cardoso’s private secretary, Mr Lucena, was also a friend, and assisted in obtaining the president’s fast sanction of the law.

As well as knowing a network of powerful people, Mr Whitaker wisely played on the Congressmen’s feeling of the need to increase popular support for Congress.^[82] Their concern for low legitimacy helped shape the perfect moment to present the bill to Congress.

The Independent Judiciary as Necessary Condition

According to Nichter,^[83] the independent character of the Brazilian judiciary led to the passing of the bill. In his explanation, prominent members of the judiciary, such as Dr Aristides who led the working group that wrote the bill, played a key role in supporting the popular initiative because of the autonomy of electoral governance. I found no evidence that the participation of these law experts was motivated by electoral courts’ independence.

Nichter^[84] misses the link between Dr Aristides and the Catholic Church. It seems more plausible that the other law experts were involved because of Dr Aristides’ past ties with the CBJP, rather than as a result of the structure and autonomy of the electoral courts. They developed a bill requested by the CNBB, not because electoral judicial institutions in Brazil allowed them to do so. More judges could have been mobilized to support the bill in that case; instead, those who were in fact involved can be traced back to the CNBB.

Process Tracing Results I. Distal and Proximal Conditions

Distal Conditions

Five main distal conditions set the scenario that led to the enactment of the AVB Law. These elements comprise a timeframe from 1987 until 1999, the year the law was approved. Although some of them are close in time to the law, they are all secondary to the outcome due to their reduced causal power compared to the proximal conditions.

1. The National Constituent Assembly, the 1988 constitution, and popular initiatives.

The transition to democracy and the 1988 constitution are landmarks in understanding the process of corruption regulation.^[85] Since the constitution of 1988, written by the National Constituent Assembly, corruption has been a salient issue in Brazil and ‘corruption scandals have become frequent in the Brazilian political scenario’.^[86]

In this constitution, popular initiatives are one of the three mechanisms that enable popular sovereignty, the other two being plebiscites and referendums. A popular initiative is a bill regarding a unique subject, signed by at least 1% of the population distributed across at least five states. Congress is not able to reject it based on legislative technicalities, format, or grammar issues.

Because no popular initiative was ever able to follow legal proceedings as such in Congress, I consider it a distal condition to the enactment of the AVB Law, contrary to some scholars and institutions.^[87] Social mobilization, support, and pressure became real with the act of subscribing the bill. Signing it somehow made people feel responsible, involved, and interested. This participatory mechanism is important for generating these feelings. However, popular initiatives do not pass laws; social mobilization does.

2. Key actor’s experience with corruption, political participation, and legislation. Before and during the National Constituent Assembly, Mr Whitaker was able to attain skills that were crucial for his leadership. His experience with corruption during his political career was one of his most important attributes.

Experience with corruption is different from its perception. To have experienced something means to have ‘knowledge or skill acquired by a period of practical experience of something, especially that gained in a particular profession’.^[88] Perception, on the other hand, is an awareness of something through the senses.^[89] This subjective feature impacts on participation differently. Mr Whitaker knew how to fight corruption using a legal mechanism of direct participation. Few scholars distinguish between experience and perception of corruption.^[90] The qualitative research enabled us to see the full extent of Mr Whitaker’s experience with corruption and how this knowledge was applied to pass the law.

3. The rise in uninformed corruption perception. The increased number of corruption scandals from 1987 to 1999

Corruption scandals covered by the media rose in number during the timeframe from 1987 to 1999. Since the return to democracy, the 1988 constitution, and the direct elections of 1989, Brazil has been plagued by corruption scandals.^[91] During president Sarney’s government from 1985 to 1990, there were three corruption scandals; during president Collor’s and Itamar Franco’s mandate from 1990 to 1995, there were five. During Fernando Henrique Cardoso’s first government from 1995 to the end of 1998, eight corruption scandals erupted, and his second mandate from 1999 to 2003 saw six corruption scandals.^[92]

Besides the increase in corruption scandals, corruption awareness also increased through the dissemination of Transparency International’s Corruption Perception Index in 1995.^[93] Since then, it has been a key topic of popular and academic debate in Brazil, although Brazil can be considered to have improved its rank. In 1996, Brazil was ranked fortieth out of fifty-four countries.^[94] A year later in 1997, Brazil was thirty-sixth out of a total of fifty-two countries; in 1998, it was forty-sixth out of eighty-five countries; and in 1999, forty-fifth out of ninety-nine countries.^[95] The salience of the issue also impacted on political trust. During Collor’s impeachment, 31% of the population trusted politicians and 26% trusted political parties; in 2005, this plummeted to 8% and 9% respectively.^[96]

I call this particular corruption perception ‘uninformed corruption perception’. It is uninformed because it is devoid of accountability information, i.e. it did not come with a solution to curb the problem. Uninformed corruption perception made people aware of the disseminated corruption. As such, it did not have a positive effect on participation, and perhaps in fact demotivated participation. However, when uninformed corruption perception was combined with accountability information, it stimulated participation.

4. International favourable scenario. Conventions

Corruption was never a threat specific to Brazil. During the 1990s, the world was discussing how to improve accountability on an international scale. The salience of the issue further pushed the domestic fight against corruption. International institutions such as the World Bank supported more than six hundred anti-corruption programs from 1996 to 2004.^[97]

In 1996, the Inter-American Corruption Convention, organized by the Organization of American States (OAS), took place. Brazil was a signatory and agreed to enact a set of laws against corruption. This convention was ‘the first multilateral legal framework established to combat public corruption in international business transactions’.^[98]

In 1997, the Organization for Economic Co-operation and Development (OECD) held an Anti-Bribery Convention with similar objectives. It criminalized the bribing of foreign officials as an attempt to gain improper advantages^[99] and regulated ‘corruption of public officials of any state, not just of the states that are parties to the Convention’.^[100] The pact is an ex-post accountability mechanism, encouraging the punishment of ‘private sector parties who engage in corrupt activities, as well as corporations and other legal persons whose employees, officers or agents participate in corruption’.^[101]

Brazil ratified both conventions in 2002^[102] and was also monitored by them. This world consensus on combating corruption internationally also caused a certain level of domestic accountability. ‘Laws against bribery abroad are effective in making investors become more sensitive to host country corruption and, as a result, further reduce their investments in corrupt countries.’^[103]

5. Congress’s concern for its low legitimacy

In 1999, corruption was also an issue in Congress’s internal debates. Members of parliament (MPs) believed that passing the law would be a way to avoid a greater erosion of Congress’s legitimacy.^[104] The low legitimacy of Congress was caused by a series of events, such as internal corruption allegations, suspicious privatizations carried out by president Fernando Henrique Cardoso, and the internal trial of a corrupt politician charged with murder, which took place on the same day the bill was introduced.^[105]

The exchange of corruption allegations between Jader Barbalho (president of Partido do Movimento Democrático Brasileiro, PMDB) and Antônio Carlos Magalhães (president of the Senate) caused an atmosphere of distrust. Jader Barbalho demanded the opening of three Parliamentary Investigation Committees (CPIs). These conflicts received media attention, further damaging Congress’s image. In plenary, MP Jorge Alberto mentions a 1999 Vox Populi research study which concluded that 38% of the population disapproved of Congress.^[106]

Privatizations being carried out by Fernando Henrique Cardoso were also suspicious. On 22 September 1999, MP Avenzoar Arruda demanded a CPI to investigate the privatizations of Cemig (a telecoms company from Minas Gerais) and those involving Banco Opportunity. It was the MP’s opinion that these scandals were causing a reaction among the people, leading them to oppose Congress as well as the president.^[107]

These events led MPs to look for a solution to their lack of acceptance within society. Without their concern about low legitimacy (a distal condition), MPs would not have been open to demands from the population. The receptiveness of the government generated by their will to increase their legitimacy led them to decide to support the bill, which is a proximal condition.

Proximal Conditions

Proximal conditions are not only closer in time to the passing of the law, but necessary to the outcome. The flow of causal power, from distal conditions, is increased. Although I list and number these causes, it is difficult to textually show how all the elements are linked. Figure 1 provides a visual map of these conditions.

1. The organization of the social movement

The organization of the social movement is intertwined with Francisco Whitaker’s past experiences. Possessed of social capital,^[108] Mr Whitaker gathered a group of people who added to that capital. He was the centre of a ‘network of strong, crosscutting personal relationships developed over time that provide[d] the basis for trust, cooperation, and collective action’.^[109] This network consisted of members of the judiciary, the media, and the Church.

Those from the judiciary collaborated with their legal expertise in writing the bill; those from the media disseminated information about the corruption scandals and the bill, and supported the national campaign, motivating society to act further. The Church made use of its far-reaching structure, exerted its powerful influence, and gave credibility to the passing of the bill.

The group became an official social movement in 2002. The Movement against Electoral Corruption (MCCE) aimed to act directly on popular information ‘about the importance of ethics in politics and the conscientious vote, as well as in the intercommunication with spaces of instituted power, along with the judiciary, executive and legislative powers’.^[110] The creation of the MCCE was also a response to the CBJP’s mission to ensure the enforcement of the law, minimizing the risk of another ‘dead letter’ law.^[111]

2. The national campaign endorsed by TV Globo

The media was already paying attention to corruption scandals and raising awareness during the AVB campaign. However, it was only when TV Globo decided to join the national campaign that the social movement gained the necessary visibility. Besides receiving information about vote buying, the Brazilian population was being alerted to the fact that there was a feasible solution.

Without the broadcast about the AVB popular initiative, the social movement would not have collected the second half of the required signatures, and awareness about vote buying would not have been as effective. Deprived of this huge learning process, the population would not have mobilized or pressured Congress, and the law would perhaps have taken longer to pass or not have passed at all.

The disseminated information caused a learning effect that not only increased corruption perception but also changed its nature. Before the national campaign, the information reaching individuals was that corruption was growing. During the campaign, people continued to see corruption as a growing problem, but they knew they could do something about it, something simple and within their reach. They only had to sign a petition and take their voter registry number when attending mass. The information that vote buying was harmful to society and that it could be stopped, pronounced by a credible institution such as the Catholic Church, led people to participate by signing the petition.

3. Social mobilization. The rise in informed corruption perception

Social mobilization only became real when corruption perception clashed with another type of information: accountability. People not only knew how widespread corruption was, but they were also aware of a possible solution. ‘Informed’ corruption perception led people to participate, whereas the uninformed perception of corruption, a distal condition, did not have that effect.

For judge Marlon Reis, ‘not only was a new electoral rule introduced. More than this, the decisive factor for applying the law was the social mobilization behind it’.^[112] Without social mobilization, the AVB Law would not have passed. I argue that social mobilization is the most important cause in this story for achieving greater accountability in Brazil.

4. Congress’s decision to support the AVB bill

At the time, supporting the bill was a reasonable solution because it could potentially produce a feeling of alignment between society and the government. Passing the law was an attempt to establish a type of complicity between society and the state, and to improve the legitimacy of Congress.

Both Michel Temer, president of the Lower House, and Antônio Carlos Magalhães, president of the Senate, decided to support and facilitate the passing of the bill. No other social demand made to Congress was this strong and visible. It may be the case that Congress underestimated society and let it pass with the intention of modifying it later. The MCCE, the law’s ‘watchdog’, prevented the rewriting of the law.

Process Tracing Results II. Testing the Theory

The likelihood that the Brazilian Congress on its own would have passed a law to stop vote buying is minimal. If MPs were confronted, they might have said that vote buying was already regulated by the three articles of the Electoral Code. In truth, these laws are dead letter laws and buying votes was still an important electoral practice for candidates. Before the popular initiative, there were no other bills proposing to solve the problem. As the popular English idiom says, ‘turkeys will not vote for Christmas’. Only an external source of change could shift this scenario.

Politicians lost the votes that were bought, reduced their chances of re-election, and thus the perks of holding an elected office became harder to achieve. Society lost whatever benefit it was receiving from the sale of votes. Popular and political support for the bill had little or no correlation with the material gains on both sides. This lack of correlation of the support of an idea vis-à-vis material gain is a ‘double decisiveness’ test confirming the hypothesis.^[113] Nobody was getting anything in exchange for the support of the vote-buying idea.

The flow of the idea to combat vote buying, from Mr Whitaker’s past political experiences to the writing of the bill, to then spread within society, is considered a hoop test.^[114] The idea can be traced; it was exogenous and available to the actors who used it. Another hoop test that the theory passes concerns the reasons for signing the petition. People’s motivation to sign did not come exclusively from the Catholic Church. If this were the case, the TV news broadcast would not have motivated further signatures in the way that it did. The influence of the Catholic Church was reduced at this point.

Social mobilization was caused by a certain level of corruption perception, which was transformed by a learning or an information effect. This mobilization exerted pressure to the point that a law could be enacted in record time. Perceived corruption among civil society was at an optimal level, a level that, through different kinds of information, allowed it to increase participation instead of apathy. In 1999, corruption in Brazil was not so ingrained as to become a self-fulfilling prophecy,^[115] or to make the population cynical,^[116] to erode collective action,^[117] to stop people from being angry and voicing their discontent in the ballot polls,^[118] to corrode the belief that one can influence politics,^[119] or to make social reciprocity norms overcome social democratic ones.^[120]

There are several mechanisms through which corruption perception might increase or decrease participation. In the case of Brazil and the AVB Law, if we magnify the link between awareness of corruption and the social mobilization that occurred in 1999, we see a spread of new information, in two different moments, which directed the popular reaction towards a feasible solution. The Brazilian population learned about the social and political impacts of vote buying.

Information fed corruption perception. Without the national campaign circulating the harmful aspects of vote buying and how it could be deterred, corruption perception would not have provided the muscle for a participatory reaction. The population learned not only how corrupt the state was, but also how it affected their lives and what they could do about it. This nourishing of corruption perception is only possible at rare critical junctures.

All the mechanisms above could explain this change, but there is only evidence for the ‘information effect’. This story shows how vote buying was the norm in a world where reciprocity was the rule; when there is rule of law, vote buying becomes a violation. Learning about vote buying and how to stop it also made people feel that others would participate too, motivating them further and making them feel empowered.

Final Remarks

The Brazilian AVB Law is a case in which corruption perception impacted positively on political participation during a specific critical juncture, of which information formed a part. Two consecutive types of perceived corruption—uninformed and informed—stimulated people to sign the popular initiative. Brazil in this period (1987-1999) is one more example of the mobilizing power of perceived corruption.

In order to buy votes, elections are necessary. After democratization and until 1999, there had been only three national presidential elections (1989, 1994, and 1998) and two municipal elections (1992 and 1996). In these five elections, buying votes was not a legal issue and was a rather new phenomenon. The state of perceived corruption was such that the spread of information could cause a popular reaction. As a democracy gets older and perceived corruption more normalized, the chances of an information effect are reduced.

‘Uninformed’ corruption perception played a role as a distal condition because it was a recent phenomenon happening during the democratization of Brazil. Although uninformed, there had not been enough time for it to become ingrained and normalized by society. The longer ‘uninformed’ corruption perception lasts, the less likely society is to mobilize in the event of ‘informed’ corruption perception. I believe that in cases in which ‘uninformed’ corruption perception is systemic, electoral accountability is corroded. Whether or not corruption is systemic is also crucial to understand the effects of grand corruption on European accountability,^[121] and the effects of bribes in African accountability.^[122] The findings of this research include Brazil in these results.

Only corruption perception in a certain state can stimulate participation among the mass population. Aside from the nature of the corruption perception, not all types of participation will be affected. The popular initiative was a low-cost form of mobilization facilitated by the social movement. Moreover, not all forms of information are able to elevate corruption perception into mobilization. Information about how corrupt a country is might push citizens away from politics, but it is still necessary to make the issue salient. After corruption became a relevant issue in this case, accountability information drove people to voice their complaints.

The mobilizing power of corruption perception was seen in the signing of the popular initiative. It was triggered by the spread of information on corruption scandals, which was then transformed by accountability information into social action. Types of participation and information are important variables in assessing the power of mobilization of corruption perception.

Experience with corruption also played an important role in stimulating participation, although it was restricted to the leading actors of the social movement. The experienced elite was able to put in motion a national awareness campaign, lowering the cost of participation and using the right political tools. Perception and experience with corruption differ not only in nature, but also in their effects on social groups.

This study still leaves questions unanswered. When does uninformed corruption perception irretrievably lose its mobilization power? How much social mobilization does electoral accountability need in order to be successful? Are there other mechanisms besides the information effect that explain the mobilizing power of corruption perception? Other case studies from different moments in time are necessary, as well as cross-national comparisons and quantitative designs.

Andreia Reis do Carmo is a PhD candidate in the King’s Brazil Institute at King’s College London, UK.

Published Online: 2018-09-11

Published in Print: 2018-09-25

Articles in the same Issue

https://doi.org/10.1515/soeu-2018-0027