Tolerancia Com a Corrupção Ou Cegueira Politica

8/18/2019 Tolerancia Com a Corrupção Ou Cegueira Politica

http://slidepdf.com/reader/full/tolerancia-com-a-corrupcao-ou-cegueira-politica 1/33

1

Tolerance of Corruption or Ideological Blindness?*

Carlos Pereira, Professor, Getulio Vargas Foundation (FGV)

Lucia S. G. Barros, Ph.D. Candidate, Getulio Vargas Foundation (FGV)

Very Preliminary Draft Please do not quote or cite without authorization!!!

Abstract

It is quite puzzling that corruption persists in democratic countries given that voters have the

electoral chance to choose honest politicians for political positions and punish those ones that

misbehave. The purpose of this paper is twofold: (1) to investigate the impact of type of

misconduct and ideological matching on voters’ perception of corruption; and (2) whether the

perception of corruption influences voters’ choice. After critically dialoguing the relevant

literature, our investigation is carried out through an experimental approach conducted by

Facebook users. Our preliminary results indicate that when voters and the candidate share

ideological preferences, they are less likely to consider the candidate’s misbehavior as corrupt

than when they have different ideological preferences. It also indicates that ideology, both in

economic and in social dimensions, not only affects voters’ perception of corruption but also

it affects voters’ choice, regardless of the type of corruption (nepotism versus money).

*

Very first draft prepared to be presented at the Conference on Accountability and Public Goods Provision that

will take place at Brown University, Providence, on December 5-6, 2014.



2

Introduction

Corruption is a recurrent topic in the literature (Ewoh, Matei, and Matei, 2013), as it is

a pervasive phenomenon, costing to our society about $2.6 trillion dollars, what represents 5

per cent of global GDP (Graycar and Sidebottom, 2012). Even though democracy is

sometimes negatively related to corruption (Tiwari, 2012), this relation is not clear, with

findings pointing that corruption and democracy are not related (Lambsdorff, 2006; Treisman,

2007) or even positively correlated (Keefer and Vlaicu, 2008).

It should be surprising, however, that corruption persists in democratic countries,

where voters would have the chance to choose honest people for political positions. Pereira

and Melo (2016) argue that even though voters hold politicians accountable for corruption,

they do not do so to the point of preventing misbehavior, especially when politicians are able

to reward voters with public goods in exchange for political support. Their findings suggest

that public spending moderates the negative impact of corruption on the probability of

reelection, and this occurs even when voters are informed about a mayor’s involvement in

corruption. In particular, they claim that while corruption decreases the probability of

incumbent re-election, the negative effect of corruption diminishes as public expenditure

rises. In a similar direction, Manzetti and Wilson (2007) claim that, especially in poorer

countries, corrupt incumbents obtain votes by delivering basic goods and patronage jobs.

They suggest that voters are mobilized by particularistic benefits and investigate the extent to

which political support varies conditional on government effectiveness, suggesting that those

who perceive high levels of corruption are significantly less likely to sanction the government

in countries where government effectiveness is low, than in countries where government

effectiveness is high. Golden (2009) also claims that even informed voters may vote for

allegedly corrupt incumbents if they expect to receive material benefits that other parties or

candidates cannot guarantee.

Alternatively, Winters and Weitz-Shapiro (2013) tests the implicit trade off in rouba

mas faz (steals, but get things done) argument in the Brazilian case. They explicitly presented

their hypothesis as alternatives: the informed rouba mas faz trade-off versus an information

constraint hypothesis that voters, once exposed to information about politicians’ past

behavior, will punish corrupt incumbents. The authors report that respondents strongly reject

the rouba mas faz trade-off, supporting the alternative “informational” hypothesis. The

authors utilize a nationwide survey experiment in Brazil, in which respondents were randomlyexposed to vignettes containing information about the politicians’ involvement (or lack



3

thereof) in corruption and party affiliation while being asked about the likelihood of support

by a hypothetical voter. Surprisingly, the authors report that respondents rejected the rouba

mas faz trade-off, displaying greater sensitivity to information on political misconduct.

The literature has also suggested other reasons why people would choose to vote for a

dishonest person. One widespread line of inquiry focuses on informational failures to explain

the seeming paradox of corrupt yet popular politicians. Voters, the argument goes, value

rectitude in office. When they acquire information about corruption they will sanction

wrongdoing (Ferraz and Finan 2008, Alesina and Tabellini 2006; Klasnja 2011). Uninformed

voters may reward corrupt incumbents primarily because of their inability to distinguish clean

politicians from dirty ones. This is exacerbated in contexts plagued by high corruption, where

mutual accusations and scandals are common. It follows that support for corrupt politicians is,

by and large, attributable to the existence of information asymmetries, as the inability of

voters to observe a politician’s true type generates moral hazards. By increasing the level of

information available to voters, corrupt incumbents are more likely to be sanctioned for

deviant or dishonest behavior. The information hypothesis suggests that voters’ support to

corrupt politicians takes place when they lack information about candidates’ involvement in

corruption upon which they then could act in the voting booth (Winters and Shapiro, 2013).

The authors found that this phenomenon happens in Brazil, where people voted for corrupt

politicians, even expressing a strong preference for punishing them. Thus, judicial

investigations by themselves do not necessarily lead to electoral punishment, unless such

episodes are widely reported by the media (Chang, Golden and Hill 2010; Ferraz and Finan

2008).

The alternative explanation to the “informational” argument for the phenomenon of

corrupt behavior being tolerated by the society is the “trading” or “trade off hypothesis,”

which suggests that corruption is one facet of a multidimensional voting space (Persson and

Tabellini, 2003; Rundquist, Strom and Peters 1977).

Our argument is that voting for a clean contestant proves costly if the contestant holds

preferences that deviate from the voter’s ideal point—be these ideological, religious or

kinship related. In the context of small towns, for instance, where voters are split into

ideologically defined constituencies, support for corrupt incumbents is driven by the

knowledge that a rival’s preferences lay far from the voter’s ideal point. In such cases, voters

prefer a “dirty” candidate who shares their ideological preferences or enjoys the family

support to a clean contestant from a different ideological camp or family background.Myerson (1993), for instance, explained that parties or candidates are different in honesty and



4

ideology. When there are candidates available for all ideological positions - what happens for

proportional representation and multimember districts – there is less room for dishonest

candidates, since voters have more options. However, in single-member districts, where only

one candidate can win the election, voters may tolerate a dishonest candidate if he or she is

ideologically preferred. That is, citizens weight corruption charges along with ideological

considerations (Golden 2009).

In line with this argument, Chang and Kerr (2007) distinguish between perceptions of

and tolerance for corruption, suggesting that voters who belong to the patronage network of

the incumbent are more tolerant of corruption and perceive corruption at a higher level

compared to patronage outsiders. The same finding applies to partisanship and ethnic

affiliation.

Considering that ideology may predict voting for a corrupt candidate, it is not clear

what psychological processes are behind this choice. On one hand, standard economic model

of rational behavior explains that a person consciously acts in a dishonest manner when the

benefits are higher than its costs (Allingham and Sandmo, 1972). If the voting behavior

follows the same pattern, one should choose a corrupt candidate when the benefits of electing

him or her are higher than the costs of his or her corruption.

On the other hand, a psychological perspective introduces the idea that cheating may

hurt one’s self-concept of being honest (Mazar, Amir, and Ariely, 2008). Thus, even though

there are situations in which the benefits of cheating may compensate its costs, one may

decide not to cheat to avoid the guilt of being dishonest. According to the authors, an

alternative is to behave dishonestly enough to profit but honestly enough to delude themselves

of their own integrity.

Considering this perspective, it is possible that some kinds of corruption may be seen

as ‘not that serious’, which allows voters to see corrupt candidates as honest. Mazar, and

Ariely (2006) supports this view, introducing the idea of self-deception, whereby individuals

reframe an act to themselves in a way that makes the act not be perceived as dishonest. A

variable that is particularly sensitive to reframing is the individual’s motivation to reappraise

the context, which may be higher or lower depending on extent to which the misconduct

threatens the self.

Because people tend to reappraise dishonest information more often when the person

who commits it is oneself instead of others (Chimonas, Brennan and Rothman, 2007), and

ideology is part of one’s identity (Iborra, 2005; Teles, 2008; Pimentel, 2008), it is possible



5

that voters reappraise more corruption information when the candidate shares their own

ideological positions.

So, the purpose of this paper is: (1) to investigate the extent to which different

dimensions of ideological preferences (economic and social) match on voters’ perception of

corruption; and (2) whether ideological preferences and the perception of corruption

influences voting choice. This investigation was carried out through an experimental

approach. After discussing our theoretical approach and hypothesis, we present the results,

discussion, limitations and suggestions for future investigations.

Theory and Hypothesis

Corruption and Ideology

When facing the dilemma of choosing a candidate, voters are concerned about the

policies that will be implemented, which involve trade-offs such as more or less government

spending (Ansolabehere, 2006). Because each voter has spatial preferences over the issue, he

or she chooses the candidate closest to an ideal policy, which means that vote follows a

certain ideology.

The term ideology was originally coined by De Tracy (1817) to connote a science of

ideas and it still can be understood as someone’s value or belief system which is accepted as

fact or truth (Singleton and Honeycutt, 2012). There are some systems that have been

identified by the literature, such as liberal, conservative, socialist, feminist, fascist, or other

(Festenstein and Kenny, 2005).

People may identify themselves through one of these ideological traditions, which, in

turn, will influence choice in an election. Literature suggests that ideological self-

identification is one of the strongest and most consistent predictors of politicians’ evaluation

and choice in a number of countries, such as the United States (Mann, 1980; Conover and

Feldman, 1981; Scotto, Stephenson, and Kornberg, 2004; Jost, 2006; Devine, 2012) and in

Brazil (Singer, 1998; Carreirão, 2002).

One common simplification of the concept of ideology is as a position on a liberal-

conservative continuum (Erikson, Wright, and McIver 1993, Berry, Ringquist, Fording, and

Russell, 1998), where voters positions themselves according to their attitudes toward the size

and role of government (Devine, 2012) or toward the willingness to change the status quo

(Conover and Feldman, 1981). In two-candidate elections, they vote for the candidate whoseideology is closest to their own (Berry et al. 1998).



6

Singer (1998) explains that even when people are not aware about the meaning of

ideological positions (e.g. the differences between left and right orientation), an intuitive

knowledge and feeling about the preferences of a candidate is sufficient for allowing them to

vote coherently with their own ideas. Conover and Feldman (1981) explain that ideological

identifications have largely symbolic meanings.

However, what would happen when the ideologically preferred candidate is corrupt? It

is known that people generally prefer to vote for an honest candidate (Myerson, 1993, Person

and Tabellini, 2004, Winters, and Shapiro, 2010, de Figueiredo, Hidalgo, and Kashara, 2011),

thus, this situation creates a problem for the voter, who has to choose between honesty and

ideology.

This situation happens in particular when there are only two options for the voter,

what happens when there are only two established parties, what is the case of the North

American electoral system (Myerson, 1993), or when there are only two competing

candidates, which is the case, for instance, of a runoff presidential election in Brazil

(Albuquerque, 1992).

Addressing this puzzle, we argue that voters might tolerate a dishonest candidate if he

or she is ideologically preferred, because switching to an honest candidate risks giving the

victory to a candidate of the opposite ideology. This idea is in accordance with Rundquist,

Strom and Peters (1977), Myerson (1993), and Golden (2009), who discuss the possibilities

that voters might support a corrupt politician in case he mirrors their ideological preferences.

Because ideology is a multi-dimensional concept, voters can access candidates’ values

and behaviors in terms of some policy dimensions, such as economic, foreign, and social

policies (Devine, 2012). Thus, if voters have to choose between a corrupt candidate and a

candidate of the opposite ideology, their dilemma may be different when the corrupt

candidate is similar to their own ideology in all dimensions, or in just in one of them.

In order to simplify ideology’s assessment in this paper, we measured it in two

dimensions: economic and social. Its measurement is better explained at the method section.

This assessment is consistent with a number of scholars, who explain that a position on a left-

right continuum remains the most meaningful indicator of ideological positioning in

contemporary democracies (Badescu and Sum 2005, Power and Zucco, 2009, Zucco, 2009,

Zucco and Lauderdale, 2011).

Bringing these two dimensions to our context, we suggest that voters are more willing

to tolerate a dishonest candidate when he or she is ideologically preferred in both economic



7

and social dimensions than when he or she is ideologically preferred in only one of these

dimensions. In other words, we hypothesize:

H1: Voters are more likely to choose a corrupt candidate when they share the same

ideology:

H1a: when voters and the candidate share the same ideology in the economic

dimension, voters are more likely to choose the corrupt candidate than when

they mismatch.

H1b: when voters and the candidate share the same ideology in the social

dimension, voters are more likely to choose the corrupt candidate than when

they mismatch.

H1c: when voters and the candidate share the same ideology in both

dimensions, voters are more likely to choose the corrupt candidate than when

they match in only one dimension.

If these hypotheses hold true, the next step is to investigate why these effects happen.

We propose two explanations: First, people may consciously choose the corrupt candidate

because they believe that it is better to tolerate a corrupt politician than to tolerate a person

whose ideology is contrary to their own. Second, people may ideologically identify

themselves with the corrupt candidate in such a way that they may not see the candidate’s

corruption as something very wrong.

According to the standard economic model of rational behavior, one would act in a

dishonest manner when the benefits are higher than its costs, which is a cognitive process

(Allingham, and Sandmo, 1972; Becker and Stigler, 1974). Considering this trade-off, in an

election context, it is possible that voters consciously tolerate corrupt candidates when their

benefits are believed to compensate their costs.

This cognitive trade-off has been found in studies in which a not “totally honest”, but

efficient person is preferred if compared to an honest, but “not that efficient” person

(Figueiredo, 2004, Almeida, 2006, 2007). There is even a literature suggesting that there are

circumstances in which corruption is not that harmful, but instead, it has some social,

economic and institutional roles in transition societies (Sousa, 2008). For this reason, we

propose that:



8

H2a: Voters are more likely to perceive that a corrupt candidate who shares the same

ideology is corrupt, but ideological matching makes the cost-benefit tradeoff of voting in a

corrupt candidate is worthwhile.

Literature suggests that people tend to reappraise dishonest information more often

when the person who commits it is oneself instead of others (Chimonas, Brennan and

Rothman, 2007). For instance, physicians reported that receiving gifts from the

pharmaceutical industry is generally wrong, but when they faced this situation themselves,

their evaluation changed, especially when they remembered how much sacrifice they made

during medical training (Sah and Loewenstein, 2010).

Because ideology is part of one’s identity (Iborra, 2005; Teles, 2008; Pimentel, 2008),

it is possible that ideology will contribute to corruption’s reappraisal. For instance, Van

Kenhove, Vermeir and Verniers (2001), explains that an extreme-left view that considers that

the rich are responsible for social inequality makes people reappraise theft in terms that it

becomes not considered to be a crime if the theft is from the rich. Therefore, thinking about

an electoral context, we suggest that voters reappraise more corruption information when the

candidate shares their own ideological positions. For this reason, we propose that:

H2b: When voters and the corrupt candidate’s ideology match, voters are less likely to

perceive him or her to be corrupt.

Corruption’s Reappraisal

Evidence from psychology suggests that people have internal reward mechanisms that

exert influence on individuals’ decisions (Mazar and Ariely, 2006). When the temptation for

dishonesty arises, people’s moral standards are also taken into account (Mazar et al., 2008).

The authors explain that even if misbehaving seems to be worthwhile, the fact that this

behavior is perceived as being wrong seems to be sufficient to prevent one from doing so.

Taking this fact into account, when developing a general model of dishonest behavior,

Mazar and Ariely (2006) proposed that the decision for dishonesty includes both internal

psychological reward mechanisms for honesty and dishonesty. Mazar et al. (2008) explained

that people often feel the conflict between benefiting from doing something wrong and acting

according to their belief of being honest. As a result, they tend to cheat, but ‘just a little’.

In other words, the authors explain that people tend to commit small dishonest acts,which allow them to have some benefit, but do not affect an honest self-image. In fact, under



9

some circumstances, the perceived “wrongness” of a dishonest behavior influences one’s

decision to commit it has already been found in the literature (Newman, 1979, Erffmeyer,

1984, Goldstone and Chin, 1993).

For example, Farrignton and Kidd (1977) found that more people claimed to own a

coin, which did not belong to them, when its value was low compared to high. Gino and

Pierce (2010) found that people tend to discount the wrongness of crossing ethical boundaries

to hurt or help others when the action restores equity; Mazar, Amir, and Ariely (2005) also

found that students cheated only 20 percent of the possible average magnitude, even when

they had no chance to be caught.

Considering these ideas, it is possible that our evaluation of other people’s acts, such

as politicians’, might follow the same pattern. Even though it has not been explained by

literature yet, it seems reasonable that doing something ‘just a little’ wrong may sound

acceptable because it would not hurt honesty, but if wrongness becomes larger than this

‘little’, it may be perceived as too much and the politician would turn to be considered

corrupt.

In the political context, the literature provides evidence of the relativism of what is

considered corruption and what is not (Schwenke, 2000, Le Billon, 2003), as ethical norms

are ubiquitous (Resnik, 2011). Corruption is commonly defined as the misuse of public office

for private gain (Le Billon, 2003) and it is not clear what actions are considered a ‘misuse’.

This idea suggests that there are modes of corruption that are considered ‘more wrong’ than

others.

Johnston (1986) explains that the kinds of behaviors that will be socially defined as

corrupt may vary as a function of social attachments and customs, such as kinship, political

culture and popular attachment to government (or lack of it). It may also vary as a function of

attributes of the political process, such as its speed, patterns of access and exclusion; and

economic characteristics, such as relative size of the public sector.

Therefore, it is possible that the same behavior will be considered corruption for a

group of people and not for others. For example, Filgueiras (2009) found that people who

understand public interest as everybody’s responsibility perceive that an act committed by

anyone, which affects the government, is corruption. On the contrary, people who understand

public interest as a government’s responsibility perceive that an act is corruption only when a

politician or a public bureaucrat commits it.

Mazar et al. (2008) explained that the fact that something is wrong seems to besufficient to prevent one from doing so, because people want to have an honest self-image.



10

Going back to an election context, imagine that a person has to make a choice between a

corrupt and an honest candidate. Even if the corrupt candidate seems to be a better choice

(e.g. more experienced), the fact that he is perceived as corrupt (e.g. what he or she did is

perceived to be ‘very wrong’) should be sufficient to prevent this person from voting for this

candidate.

This idea would rule out the hypothesis that the choice for a corrupt candidate is based

on a cognitive cost-benefit analysis (Allingham and Sandmo, 1972). In other words, the

choice for a corrupt candidate would not depend on a perception that he or she is in fact

corrupt but voting for him or her is worthwhile, but on a perception that what the candidate

did is not ‘wrong enough’.

This idea would also rule out the information hypothesis that voters’ support to corrupt

politicians happens when they lack information about candidates’ involvement in corruption

(Winters, and Shapiro, 2010). Instead of not having enough information, we propose that

voters’ support to corrupt politicians happens when the available information does not make

the candidate seems to be ‘corrupt enough’.

In other words, ideology may create a state of blindness, in which voters see the

candidate’s misconduct as not ‘wrong enough’. If corruption is not ‘very wrong’, voters may

believe that it is not a problem for them to vote for this candidate. Thus, we propose that:

H3: The way corruption is perceived (how wrong it is) affects choice.

Political Context

It is important to discuss the political and electoral context in which the experiment

was applied. The survey was posted on the Facebook platform between the first round and the

runoff presidential election in Brazil; precisely, two weeks prior the runoff election, which

took place on October 26 2014. This political environment became an excellent opportunity to

test the impact of voters’ ideological attachment on their tolerance to corruption.

Despite the high number of presidential candidates running for the presidency and the

great degree of uncertainty concerning who would be able to win the electoral contest, the

extremely competitive first round of the electoral race in 2014 consolidated as winners the

two main political and opposing ideological blocks that have polarized Brazil since 1994. On

the one hand, the Workers’ Party (PT), having President Dilma Rousseff running forreelection, and on the other hand, the Brazilian Social Democratic Party (PSDB) with his



11

candidate and former governor of the state of Minas Gerais, Aécio Neves. As none of the

candidates obtained the simple majority in the first round, the two first runners had to dispute

a runoff election three weeks later.

During the electoral process, a major corruption scandal was disclosure in the largest

state-oil company, Petrobras, in kickbacks financed by cash creamed from inflated contracts.

The scale of the wrongdoing could dwarf the other major scandal of recent years, known as

Mensalão – monthly bribes to legislators - and further undermine not only the reputation of

the a company that has long been seen as a national treasure, but also the political reputation

of the incumbent president running for reelection. Petrobras is seen as so important that both

the lower house and the Senate have launched inquiry committees to investigate the scandal.

The country spited in two opposite sides. On the one hand, PT’s supporters alleged

that corruption has been a pathology attributable to the Brazilian political system as a whole

and not necessarily would involve or be restricted to PT administrations. It was also argued

that during the former Cardoso’s administration and in states governed by the opposition,

many corruption scandals have also emerged, raising thus doubts about who would be in fact

corrupt or not. Although President Rousseff was chairperson of Petrobras’ board from 2003 to

2010, she has denied any knowledge of graft.

On the other hand, the opposition has blamed the Rousseff’s government to be

responsible for the largest corruption scandal ever in the Brazilian history, and their political

and economic consequences remain unpredictable. The opposition has also accused that PT

and coalition allies from blocking further investigation by the committees in Congress.

Although by very close electoral margin, President Rousseff has managed to be reelected.

Despite the hypothetical scenario of the experiment, the voting choice situation in

which voters had to access information about corruption and their ideological preferences

were present during the two weeks in which they answered the survey experiment.

Method

To test all hypotheses, we ran a web-based survey experiment. We used a 2 (mode of

corruption: money vs. nepotism) X 2 (economic ideology: matching vs. mismatching) X 2

(behavior ideology: matching vs. mismatching) between subjects design. We used two modes

of corruption to check if our results hold for both of them.

Our sample comprises 1.045 Brazilian subjects who were randomly assigned to one ofthe experimental conditions. We recruited these participants by placing an advertisement on



12

Facebook platform. This recruitment method is effective because it allows greater

demographic diversity and representativeness than student samples (Samuels & Zucco, 2012).

Interested participants answered an online survey. The procedure was the following: when

people agreed to participate, they received instructions asking them to imagine a second term

presidential election in which they would have to choose between two opposing candidates to

vote.

The participants (1) read the profile of both candidates; (2) made the decision in which

candidate they would vote for; (3) answered a questionnaire measuring their own ideology;

(4) read the manipulation; (5) made a second decision in whom to vote for; (6) answered a

questionnaire measuring the type of processing (whether the corrupt candidate was seen as a

cost-benefit choice – he steals, but he gets things done – or if he was seen as not corrupt); and

(6) answered some socio-economic questions.

The first option was the Candidate A, who had large political experience. He had been

a municipal mayor twice, and a former state-governor, and he had 70% of citizen’s approval,

and was considered very efficient by the population. The second option was the Candidate B,

who lacks administrative experience. He had been a municipal mayor just once, and he had

30% of citizen’s approval only. When he was a mayor, he did very little to improve people’s

quality of life. In this campaign, his platform focused on honesty in politics and public

service. After reading these political profiles, subjects chose for whom they would vote.

The questionnaire measuring type of ideology of the respondents contained the

indicators displayed on Figure 1. This measure is an adaptation of the scale used by the Pew

Research Institute. The Brazilian Institute Datafolha also used similar scale. We chose this

scale because it captures both dimensions of ideology (economic and social) and it has been

used in a Brazilian representative sample, which eventually allow us to compare our findings

with the Brazilian population.

As said before, the survey used two dimension of ideology, measured in terms of the

role of the government in the economy, and in social values related to religion, gay marriage

and legalization of marijuana. These dimensions were validated by exploratory and

confirmatory factorial analysis. All the indicators had two opposite sentences and subjects

should agree with one of them or say they do not know.

Based on the traditional left-right ideological continuum, we called economic-

conservative the voters leaning toward state intervention in the economy and economic-

liberals voters supporting competition and less regulation. Concerning social values, we



13

considered as social-liberals voters in favor of civil liberties such as gay marriage and social-

conservatives those voters supporting restrictions on individual behavior like the use of drugs.

[Figure 1 about here]

The manipulations came next. Subjects read a vignette showing again the profile of

both candidates and saying that they found more information about Candidate A. In the

money condition, Candidate A was accused of causing a BR$ 4 million loss in public

accounts for stealing money (vs. appointing a relative to a public position). In the economic

liberal condition, Candidate A was in favor of privatizations and less intervention on economy

(vs. creation of state-owned companies and more intervention/regulation on economy), and in

the social liberal condition, he was in favor (or against) legalizing the consumption of

marijuana and of homosexual marriage.

Based on their score differences, subjects were classified into matching or

mismatching (whether his or her ideological view matches or not with Candidate A’s that was

displayed to him or her) for both economic and social views. After that, subjects answered to

a questionnaire measuring type of processing, which contained the indicators displayed on

Figure 2. Because corruption can be chosen either due to a cognitive tradeoff (Figueiredo,

2004, Almeida, 2006, 2007) or to the perception that it was not ‘wrong enough’ (Mazar et al.

2008), this scale contains two dimensions, which was confirmed by exploratory and

confirmatory factorial analysis. Subjects rated the statements using a Likert-rating, ranging

from 1 (‘I totally disagree’) to 5 (‘I totally agree’).


Results

The characteristics of the sample are summarized in Table 1. As expected, there are no

significant differences across conditions, which indicates that subjects were randomly

assigned to the conditions and therefore our results are not driven by any demographic

characteristics.

[Table 1 about here]



14

First, we checked if the voting choice depended on type of corruption. We did not find

any difference between the types of corruption on voting choice (! 2 (2) = 3.13; p = .21). Table

2 shows the results divided by type of corruption. Given that we did not find differences in

our replicates with regard to the type of corruption on voting choice, we decided to analyze

these two groups together.


Then, we examined the extent to which voting choice depends on ideological

preferences. As we can see in Tables 3 and 4, there exist significant differences between

subjects in which ideology matched compared to when they did not match, both in economic

(! 2 (2) = 40.43; p < .001) and social (! 2 (2) = 48.47; p < .001) ideological dimensions. In other

words, when voters and the candidate shared the same ideological preferences, voters are

more inclined to vote for the corrupt candidate than when they have distinct ideological

preferences.

[Tables 3 and 4 about here]

The next step was to investigate if there is any difference among all possible

combinations of matching and mismatching in both economic and social dimensions of

ideology. Table 5 shows that when both ideological dimensions matches, Candidate A

receives a higher proportion of votes compared with the situation in which only the economic

dimension matched (! 2 (2) = 37.37; p < .001) as well as when only social dimension of

ideology matched (! 2 (2) = 31.38; p < .001). Consistently, when both ideological dimensions

mismatched, Candidate A received a lower proportion of the votes compared when only

economic matched (! 2 (2) = 10.18; p < .01), and when only behavior matched (! 2 (2) = 12.15;

p < .01). However, there was no statistically significant difference when only economic

ideology matched and when only social ideology matched (! 2 (2) =.16; p = .92). This result

confirms thus our hypotheses 1a, 1b, and 1c.


After that, we explored if any specific ideology drove this effect. Table 6 shows howsubjects in our sample made their voting choices based in each combination of our ideological



15

dimensions for each of the four conditions of matching and mismatching. We categorized

subject’s ideology into conservative and liberal. As mentioned earlier, for economic ideology,

conservative voters are in favor of a higher degree of interventionism, while liberals favor

economic competition and less regulation. For social ideology, liberals are in favor of gay

marriage and liberalization of drugs, while conservative voters believe in god as reference of

values.


Given that only 5.2% of our sample presented conservative social ideology, the

number of subjects was too small to split into eight conditions. However, among the subjects

that had liberal social ideology, we found some interesting differences.

Among economic conservatives, when both ideological dimensions matched,

Candidate A received a higher proportion of the votes compared when only social dimension

of ideology matched (! 2 (2) = 20.41; p < .001), and when only economic dimension of

ideology matched (! 2 (2) = 17.43; p < .001). However, when both ideological dimensions

mismatched, Candidate A did not receive a significant lower proportion of the votes

compared when only economic matched (! 2 (2) =.94; p = .624), and when only social matched

(! 2 (2) = 3.41; p = .18). Moreover, we did not find a significant difference when only

economic matched and when only social matched (! 2 (2) = 4.44; p < .108). It means that for

this group, the corrupt candidate should mirror their both ideological dimensions to get their

votes.

Among economic liberals, when both ideological dimensions matched, Candidate A

surprisingly received a lower proportion of the votes compared to when only economic

dimension matched (! 2 (2) = 41.42; p < .001), but, as we expected, much higher than when

only social dimension matched (! 2 (2) = 19.12; p < .001). When both ideological dimensions

mismatched, Candidate A received a marginal higher proportion of the votes compared with

when only economic dimension matched (! 2 (2) = 5.66; p = .059), and but much lower than

when only social dimension matched (! 2 (2) = 15.35; p < .001). It means that for economic

liberals, economy is more important than social issues to drive their vote decision. Another

indication that confirmed that there is a significant difference between when only economic

matched and when only social matched (! 2 (2) = 24.08; p < .001).

Psychological Mechanisms



16

The next step is to investigate the psychological mechanisms behind these voting

choices. Do people consciously evaluate the cost-benefit tradeoff of choosing a corrupt

candidate? Or do they unconsciously reappraise the corrupt behavior and consider it as “not

corrupt”? To investigate it, we run two factorial ANOVAs, first, where cost-benefit

evaluation serves as the dependent variable and then, where reappraisal serves as the

dependent variable. Our independent variables are economic and social matching or

mismatching, and subjects’ social and economic ideological dimensions.

For the cost-benefit tradeoff as the dependent variable, when the economic ideology of

the respondent matches with the candidate’s ideology, corruption is easier to be accepted as a

cost-benefit trade-off (Mmat = 2.14 vs. Mmis = 1.85; F = 7.89; p < .01). No other independent

variable predicts significant differences on cost-benefit tradeoff. Therefore, we find partial

support for H2a, because corruption is only accepted as worthwhile for the economic

dimension of ideology.

For reappraisal as our dependent variable, we find a direct effect of respondent’s social

ideology on reappraisal, so that when people have conservative social ideology, they

reappraise more than when they have liberal social ideology (Mcon = 2.01 vs. Mlib = 1.65; F =

5.28; p < .05). Because most of our sample has liberal social ideology, this finding may not

explain voting choice in this experiment.

More interestingly, however, is when both economic and social ideology matches.

Voters reappraise corruption information significantly more often (F = 5.97; p < .05) than

when only economic or social ideologies matches. It means that when the candidate shares the

same economic and social ideologies as the respondents, they think the candidate is less

corrupt, supporting H2b Figure 3 shows the different means and illustrates this finding.


Moreover, we found a significant four-way interaction of economic matching, social

matching, economic ideology, and social ideology on reappraisal (F = 7.61; p < .01). Figure 4

shows the different means and illustrates this finding. For social liberals, we found that the

effect when the candidate shared the same economic and social ideologies as the respondent,

he or she thinks the candidate is less corrupt, is stronger for economic conservatives than for

economic liberals.




17

Finally, we ran a logistic regression of the modes of processing (reappraisal and cost-

benefit tradeoff) and their interaction on choice (transformed into a dummy variable, where 1

equals voting for Candidate A, and 0, otherwise). Reappraisal, cost-benefit tradeoff, and their

interaction were significant predictors of choice (" = .994, SE = .212, p < .001; " = 1.629, SE

= .162, p < .001; " = -.370, SE = .075, p < .001, respectively).

It means that Candidate A was more likely to be voted when the voter saw him as less

corrupt, and when the candidate’s behavior is seen to be wrong, but worthwhile. When high

levels of both processes happened at the same time, their effects are reduced, but still

significant. We found support for H3, since the level of corruption’s reappraisal predicts

choice. However, the conscious cost-benefit tradeoff is also important to explain choice. It

means that the answer for voting for a corrupt candidate can be both tolerance of corruption

and ideological blindness.

Discussion

Even though, in democratic countries, voters have the electoral chance to choose

honest candidates for political positions and punish those ones that misbehave, there are

sometimes conditions for them not to do so. By using an experimental approach, we show that

ideology is one of these reasons. Specifically, our experimental analysis demonstrates that

voters are more likely to choose a dishonest candidate when they share the same ideology,

even acknowledging that this candidate may be corrupt. We additionally show that this effect

is even stronger when both economic and social ideological dimensions match.

We go a step further and investigate the psychological mechanisms behind this effect.

Interestingly, when people read information that an ideologically preferred candidate is

corrupt, they are more likely to reappraise it than when the candidate has the opposed

ideology. In other words, ideological matching facilitates the belief that the candidate’s

misconduct is not that wrong.

In addition, our findings suggest that both tolerance of corruption and ideological

blindness influence voting choice, in such a way that when an ideologically preferred

candidate is corrupt, voters see their corruption as mild, but even when they acknowledge that

the corruption exists, choosing the corrupt candidate is still worthwhile.

Even though our findings contribute to our understanding of why people choosecorrupt candidates, they meet some limitations. For instance, in the experimental procedure,



18

people first had to choose one candidate and, afterwards, fill in the type of processing form.

This may have made respondents rationalize their choice, what may have affected their

answers. Other experimental designs may replicate this study using a different order.

Another limitation is that the manipulation said that the candidate was suspected from

being corrupt, but there was no evidence to prove it. If he or she was already condemned, the

answers’ pattern may have been different. This modification in the experimental design can

also be tried in future research.

Finally, future research can investigate whether voters can differently reappraise other

modes of corruption. For example, if instead of nepotism, the non-monetary type of

corruption were patronage or public contracts, would the reappraisal process be different?

This kind of investigation could shed light on what kinds of behaviors are considered to be

corruption by the Brazilian (or other nationalities) voters.



19

References

Albuquerque, J. A. (1992). Identidade, oposição e pragmatismo: uma teoria política do voto.

Lua Nova: Revista de Cultura e Política, (26), 53-79.

Allingham, M. G., & Sandmo, A. (1972). Income Tax Evasion: A Theoretical Analysis.

Journal of Public Economics, 1, 323–38.

Almeida, A. C. (2006). Por Que Lula? Rio de Janeiro: Record.

Almeida, A. C. (2007). A Cabeça do Brasileiro. Rio de Janeiro: Record.

Ansolabehere, S. (2006). Voters, Candidates, and Parties. In: R. E. Goodin (Eds.). The Oxford

Handbooks of Political Science (pp. 29-49). New York: Oxford University Press Inc.

Asquer, R. (2012) “Why do Citizens of Democracies Tolerate Corruption? Preliminary

Evidence from the World Value Survey” (Unpublished manuscript).Becker, G. S., & Stigler, G. J. (1974). Law enforcement, malfeasance, and compensation of

enforces. Journal of Legal Studies, 3, 1.

Badescu, G., & Sum, P. E. (2005, April). The Importance of Left-Right Orientations in the

New Democracies. International Conference on Elections and Democratic

Governance, Taipei, 10–11.

Berry, W. D., Ringquist, E. J., Fording, R. C., & Russell, L. Measuring Citizen and

Government Ideology in the American States, 1960-93. American Journal of Political

Science, 42(1), 327-348.

Borba, J. (2005). Cultura política, ideologia e comportamento eleitoral: alguns apontamentos

teóricos sobre o caso brasileiro. Opinião Pública, 11(1), 147-168.

Borges, A. (2010). Burocracia, Política de Patronagem e Gasto Educacional nos Estados

Brasileiros: O Caso da Secretaria de Educação do Estado da Bahia. Estudos de

Sociologia, 15(29).

Carreirão, Y. S. (2000). A decisão do voto nas eleições presidenciais no Brasil (1989 a 1998):

a importância do voto por avaliação de desempenho. Doctoral Dissertation, USP, São

Paulo, São Paulo, Brazil.

Carreirão, Y. S. (2002). Identificação ideológica e voto para presidente. Opinião Pública, 1,

54-79.

Chabal P, Daloz, J. P. (1999). Africa Works: Disorder as Political Instrument. James Currey:

Oxford.



20

Chandra, K. (2003). Why Ethnic Parties Succeed: Patronage and Ethnic Headcounts in India.

UC Los Angeles: Department of Sociology, UCLA. Retrieved from:

http://escholarship.org/uc/item/0vb620b2

Chimonas, S., Brennan, T. A., & Rothman, D. J. (2007). Physicians and drug representatives:

exploring the dynamics of the relationship. Journal of general internal medicine,

22(2), 184-190.

Conover, P. J., & Feldman, S. (1981). The Origins and Meaning of Liberal/Conservative Self-

Identifications. American Journal of Political Science, 25(4), 617-645.

De Tracy, A. D. (1817). A Treatise on Political Economy. Georgetown: Joseph Milligan.

Devine, C. J. (2012). Social Issues, Authoritarianism, and Ideological Conceptualization:

How Policy Dimensions and Psychological Factors Influence Ideological Labeling.

Political Psychology, 33(4), 531-552.

Erikson, R. S., Wright Jr., G. C., & McIver, J. (1993). Statehouse Democracy. New York:

Cambridge University Press.

Erffmeyer, E. S. (1984). Rule-violation behavior on the golf course. Perceptual & Motor

Skills, 59, 591-596.

Ewoh, A. I. E. Matei, A. & Matei, L. (2013). Corruption, public integrity, and globalization in

South-Eastern European states. A comparative analysis. Theoretical and Applied

Economics, 1(578), 7-34.

Farrington, D. P., & Kidd, R. F. (1977). Is financial dishonesty a rational decision? British

Journal of Social and Clinical Psychology, 16, 139–146.

Filgueiras, F. (2009). A tolerância à corrupção no Brasil: uma antinomia entre normas morais

e prática social. Opinião Pública, 15(2), 386-421.

Festenstein, M., & Kenny, M. (2005). Political Ideologies: a Reader and Guide. Oxford:

Oxford UP.

Figueiredo, M. (2004). Convite ao ‘Rouba, mas faz,’ Insight Inteligência, 24-7.

Figueiredo, M. F. de, Hidalgo, F. D., & Kasahara, Y. (2011). When do voters punish corrupt

politicians? Experimental evidence from Brazil. Unpublished Manuscript, University

of California Berkeley.

Fjeldstad, O. H., Kolstad, I., & Lange, S. (2003). Autonomy, incentives and patronage. A

study of corruption in the Tanzania and Uganda revenue authorities. Michelsen

Institute.

Gino, F., & Pierce, L. (2010). Lying to Level the Playing Field: Why People May DishonestlyHelp or Hurt Others to Create Equity. Journal of Business Ethics, 95: 89–103.



21

Golden, M. (2009) “The Electoral Underpinnings of Corruption in Rich and Poor Democratic

Polities” (Unpublished manuscript)

Goldstone, R. L., & Chin, C. (1993). Dishonesty in Self-Report of Copies Made: Moral

Relativity and the Copy Machine. Basic and Applied Social Psychology, 14(1), 19–32.

Graycar, A., & Sidebottom, A. (2012). Corruption and control: a corruption reduction

approach. Journal of Financial Crime, 19(4), 384-399.

Green, E. (2010). Patronage, District Creation, and Reform in Uganda. Studies In

Comparative International Development , 45(1), 83-103.

Hsee, C. K., Yu, F., Zang, J., & Zhang, Y. (2003). Medium Maximization. Journal of

Consumer Research, 30(1), 1-14.

Iborra, A. (2005). Beyond Identity and Ideology: Processes of Transcendence Associated with

the Experience of Conversion. Culture & Psychology, 11(1), 89-100.

Johnston, M. (1986). The political consequences of corruption: a reassessment. Comparative

Politics, 18(4), 459–477.

Jost, J. T. (2006). The end of the end of ideology. American Psychologist , 61, 651–670.

Keefer, P., & Vlaicu, R. (2008). Credibility, Clientelism and Democracy. Journal of Law,

Economics and Organization, 24(2), 1-36.

Lambsdorff, J. G. (1996). Institutions and Economic Performance: Cross Country Tests Using

Alternative Institutional Measures. Economics and Politics, 7, 207-27.

Le Billon, P. (2003). Buying Peace or Fuelling War: The Role of Corruption in Armed

Conflicts. Journal of International Development , 15, 413–426.

Mann, T. (1980). Unsafe at Any Margin: Interpreting Congressional Elections. Washington,

DC: American Enterprise Institute.

Mazar, N., Amir, O. & Ariely, D. (2008). The Dishonesty of Honest People: A Theory of

Self-Concept Maintenance. Journal of Marketing Research, 155, 633–644.

Mazar, N. & Ariely, D. (2006). Dishonesty in everyday life and its policy implications,

Working paper series // Federal Reserve Bank of Boston, No. 06-3,

http://hdl.handle.net/10419/55638

Melo, M., Pereira, C. & Figueiredo, C. (2009). Political and Institutional Checks on

Corruption: Explaining the Performance of Brazilian Audit Institutions. Comparative

Political Studies, 42(9): 1217-1244.

Myerson, R. (1993). Effectiveness of electoral systems for reducing government corruption: a

game theoretic analysis. Games and Economic Behavior , 5, 118–32.





23

Singer, A. (1998). Identificação ideológica e voto no Brasil: o caso das eleições presidenciais

de 1989 e 1994. Doctoral Dissertation, FFLCH/USP, São Paulo, SP, Brazil.

Singleton, S. V., & Honeycutt, A. (2012). Utilizing Political Ideologies to Market a Political

Candidate. Journal Of Business & Economics Research, 10(1), 19-23.

Smith, G. (2001). The 2001 general election: Factors influencing the brand image of political

parties and their leaders. Journal of Marketing Management , 17(9-10), 989-1006.

Snyder Jr., J. M., & Ting, M. M. (2002). An informational rationale for political parties.

American Journal of Political Science, 90-110.

Sousa, L. (2008). As agências anticorrupção como peças centrais de um sistema de

integridade. Revista da CGU , 4(1), 20-45.

Stimson, J. A., MacKuen, M. B., & Erikson, R. S. (1995). Dynamic representation. American

Political Science Review, 89, 543–565.

Teles, C. D. P. (2009). Linguagem Escolar e a Construção da Identidade e Consciência Racial

da Criança Negra na Educação Infantil. Anagrama: Revista Científica Interdisciplinar

da Graduação, 1(4).

Tiwari, A. K. (2012). Corruption, democracy and bureaucracy. Theoretical and Applied

Economics, 9(574), 17-28.

Treisman, D. (1998). The Causes of Corruption: A Cross National Study, University of

California, Los Angeles, April, available at

http://www.isr.umich.edu/cps/pewpa/archive/archive_98/19980019.pdf .

Van Kenhove, P., Vermeir, I., & Verniers, S. (2001). An empirical investigation of the

relationships between ethical beliefs, ethical ideology, political preference and need

for closure. Journal of Business Ethics, 32(4), 347-361.

Winters, Matthew S. and Rebecca Weitz-Shapiro. 2013. “Lacking Information or Condoning

Corruption? When do Voters Support Corrupt Politicians.” Journal of Comparative

Politics 45(4): 418-436.

Zucco C., Jr., (2009). Ideology or What? Legislative Behavior in Multiparty Presidential

Settings. The Journal of Politics, 71(3), 1076–1092.

Zucco, C., Jr, & Lauderdale, B. E. (2011). Distinguishing between influences on Brazilian

legislative behavior. Legislative Studies Quarterly, 36(3), 363-396.



24

Figure 1: Indicators of the Economic and Social Ideology Scale

Role of the government in the economy – Cronbach alpha = .6

Econ1 Income should be distributed in an equal way.Income differences serve as an incentive to

individual effort.

Econ2

Competition is good because it stimulates

people to work hard and develop new ideas.

Competition is bad because it brings up the worst

part of people.

Econ3

Government regulation of business is

necessary to protect the public interest.

Government regulation of business usually does

more harm than good.

Econ4

It is better to pay more taxes and have

education and health free of charge.

It is better to pay fewer taxes and hire private

education and health.

Social values – Cronbach alpha = .6

Social1

It is necessary to believe in God in order to be

moral and have good values.

It is not necessary to believe in God in order to be

moral and have good values.

Social2

Drug use should not be forbidden, because it

is the user who suffers the consequences.

Drug use should be forbidden, because all society

suffers its consequences.

Social3 Homosexuality should be accepted by society. Homosexuality should be discouraged by society.



25

Figure 2: Indicators of the Type of Processing Scale

Cost-Benefit 1 It is worthwhile voting for Candidate A, even though he is corrupt.

Cost-Benefit 2 The constituents’ benefits generated by Candidate A made his misbehavior worthwhile.

Cost-Benefit 3 I am aware that Candidate A is corrupt. But at least he gets things done.

Cronbach's Alpha = .8

Reappraisal 1

Candidate A was not corrupt. He just relied on a necessary political mechanism in order

to get things done.

Reappraisal 2 There was no dishonest behavior of Candidate A. He did what everybody does.

Reappraisal 3

Considering the political context and national conditions, Candidate A’s behavior is

appropriate.

Cronbach's Alpha = .8



26

Table 1: Sample Summary Statistics

Condition

Manipulation 1 2 3 4 5 6 7 8

Total

Economy -Liberal

No No No No Yes Yes Yes Yes

Social -

LiberalYes Yes No No Yes Yes No No

Corruption -

Money No Yes No Yes No Yes No Yes

N 131 127 132 121 127 120 142 145 1.045

First Choice -

Candidate A90,8% 94,5% 93,2% 87,6% 91,3% 89,2% 91,5% 91,0% 91,2%

Ideology -

Economic

Score

1,86 1,88 1,94 1,90 2,01 1,88 1,82 1,92 1,90

Ideology -

Social Score1,50 1,49 1,52 1,52 1,54 1,58 1,47 1,54 1,52

Gender - Male 62,0% 55,2% 64,9% 65,5% 57,9% 61,0% 50,0% 65,3% 60,2%

Mean Age 32,9 31,4 30,1 31,9 32,2 31,1 30,3 29,9 31,2

Marital Status

- Single54,3% 68,8% 67,9% 59,7% 61,1% 65,3% 63,6% 62,5% 62,9%

Education -

Bachelor

degree

16,3% 19,2% 18,3% 15,1% 23,8% 20,3% 22,1% 22,2% 19,8%

State - São

Paulo41,1% 44,0% 35,9% 45,8% 42,1% 49,2% 40,7% 47,6% 43,2%

Monthly

income - above

US$2.800

48,8% 50,0% 51,2% 55,0% 41,7% 48,7% 54,4% 50,0% 50,0%



27

Table 2: Cross tabulation - New Choice vs. Type of Corruption

New Choice

Candidate A Candidate B Not Vote

% Nepotism 33,3% 35,1% 31,6%

% Money 38,3% 31,2% 30,5%

% Total 35,9% 33,1% 31,0%



28

Table 3: Cross tabulation - New Choice vs. Economic Matching

New Choice


% Within

Mismatching

26,7% 39,2% 34,1%

% Within

Matching

45,3% 26,8% 27,8%

% Total 35,9% 33,1% 31,0%



29

Table 4: Cross tabulation - New Choice vs. Social Matching

New Choice


% Within

Mismatching

26,2% 39,3% 34,5%

% Within

Matching

46,7% 26,3% 27,1%

% Total 35,9% 33,1% 31,0%



30

Table 5: Cross tabulation – New Choice vs. Economic Matching vs. Social Matching

Candidate A Candidate B Not Vote N

Both matched 58,8% 20,2% 21,0% 257

Economic matched / social mismatched 31,9% 33,5% 34,6% 257

Economic mismatched / social matched 33,6% 32,8% 33,6% 238

Both mismatched 21,2% 44,4% 34,5% 293

Total 35,9% 33,1% 31,0% 1045



31

Table 6: Cross tabulation - New Choice vs. All Ideological Conditions

Subjects’ Ideology

New Choice N


Social -

Conservative

Economic -

Conservative

BothMismatched

12,5% 62,5% 25,0% 8

Social

Matched33,3% 44,4% 22,2% 9

EconomyMatched

20,0% 60,0% 20,0% 20

BothMatched

30,0% 20,0% 50,0% 10

Total 23,4% 48,9% 27,7% 47

Social -Conservative

Economic -Liberal

BothMismatched

0,0% 0,0% 100,0% 1

SocialMatched 0,0% 0,0% 100,0% 1

EconomyMatched

33,3% 66,7% 0,0% 3

BothMatched

100,0% 0,0% 0,0% 2

Total 42,9% 28,6% 28,6% 7

Social - LiberalEconomic -

Conservative

Both

Mismatched20,6% 44,4% 35,1% 248

SocialMatched

36,6% 31,7% 31,7% 202

EconomyMatched

27,3% 34,0% 38,8% 209

BothMatched

58,0% 22,2% 19,8% 212

Total 35,0% 33,5% 31,5% 871

Social - LiberalEconomic -

Liberal

BothMismatched

27,8% 41,7% 30,6% 36

Social

Matched11,5% 38,5% 50,0% 26

EconomyMatched

80,0% 4,0% 16,0% 25

BothMatched

69,7% 9,1% 21,2% 33

Total 46,7% 24,2% 29,2% 120

Total N 1045



32

Figure 3: Effect of Economic and Social Matching on Reappraisal



Figure 4: Four-way Interaction of Economic Matching, Social Matching, Economic

Ideology, and Social Ideology on Reappraisal

2.27

1.78

2.25

1.38

1.84

1.66 1.651.53

2.50

1.00

1.33

4.00

1.73

1.35

1.76

1.51

!"#$#%&" (

)*+",-.

!"#$#%&" (

)&/%*+",-.

!"#$#%&" (

)*+",-.

!"#$#%&" (

)&/%*+",-.

!"#$#%&" (

)*+",-.

!"#$#%&" (

)&/%*+",-.

!"#$#%&" (

)*+",-.

!"#$#%&" (

)&/%*+",-.

0#"&*1 ( )*+",-. 0#"&*1 ( )&/%*+",-. 0#"&*1 ( )*+",-. 0#"&*1 ( )&/%*+",-.

0#"&*1 ( 2#$/-34*54- 0#"&*1 ( 6&7-3*1

!"#$#%&" ( 2#$/-34*54-

!"#$#%&" ( 6&7-3*1

Documents

Tolerancia Com a Corrupção Ou Cegueira Politica