Money and success in the 2008, 2012, and 2016 local elections in Brazil

Abstract

The close relationship between money and votes has been exposed in contemporary democracies through the influence of the former in election results and in the creation of public policies. Studies have pointed out that the progressive inequity of financial resources available to the candidates running for elections unbalances democracy and promotes plutocracy, increasing the oligarchical ruling over the political processes by the wealthiest citizens and groups. This work contributes to this debate by analyzing the relationship between fundraising and electoral performance and the success of 317,107 candidates running in local legislative elections in 2008, 2012, and 2016 in 441 Brazilian municipalities with over 50,000 inhabitants. This is an unprecedented study regarding its period, filling a gap in electoral studies on sub-national elections, particularly regarding elections for the legislative branch. Data was collected from the Brazilian Superior Electoral Court, from candidate lists, results, and accounting. The tests and data analysis and information employed correlation and regression statistical resources, along with central measures and odds ratio. The results confirmed a strong association between money, vote, and electoral success, including in the 2016 election when company donations had been prohibited. The maintenance of a pattern, regarding both time and geography, indicates the existence of structural characteristics of a plutocratic character in the election funding of those running for the council in the municipalities studied and, consequently, in the creation of public policies in Brazil.

Keywords:
election funding; municipal election; democracy; plutocracy; city councilors

Resumo

A intimidade entre dinheiro e voto tem sido exposta nas democracias contemporâneas por meio de sua influência nos resultados eleitorais e na produção de políticas públicas. Estudos apontam que a progressiva iniquidade entre os recursos financeiros dos candidatos nas eleições abala a democracia e promove a plutocracia, ou seja, amplia o domínio oligárquico dos processos políticos pelos cidadãos e pelos grupos mais ricos. Contribuindo para esse debate, o presente trabalho analisa a relação entre receitas de financiamento e o desempenho e o sucesso eleitoral de 317.107 candidaturas a vereador nas eleições de 2008, 2012 e 2016, em 441 municípios brasileiros com mais de 50 mil eleitores. Trata-se de um estudo inédito em sua abrangência temporal, que permite suprir uma carência de exames sobre eleições subnacionais, em especial nas disputas ao Legislativo municipal. Os dados utilizados têm origem no Tribunal Superior Eleitoral, nas tabelas de candidaturas, nos resultados e na prestação de contas. Os testes e as análises de dados e informações utilizam recursos estatísticos de correlação e regressão, além de medidas centrais e razão de chances. Os resultados permitem confirmar uma forte associação entre dinheiro, voto e sucesso eleitoral, inclusive no pleito de 2016, quando estiveram proibidas as doações empresariais. A manutenção de um padrão, tanto temporal quanto geográfico, indica a existência de características estruturais de caráter plutocrático no financiamento eleitoral dos candidatos a vereador nas cidades pesquisadas e, consequentemente, sobre a produção de políticas públicas no Brasil.

Palavras-chave:
financiamento eleitoral; eleição municipal; democracia; plutocracia; vereadores

Resumen

La intimidad entre dinero y voto ha sido expuesta en las democracias contemporáneas a través de su influencia en los resultados electorales y en la producción de políticas públicas. Los estudios muestran que la desigualdad progresiva entre los recursos financieros de los candidatos en las elecciones debilita la democracia y promueve la plutocracia, es decir, aumenta el dominio oligárquico en los procesos políticos por parte de los ciudadanos y grupos más ricos. Contribuyendo a este debate, el presente trabajo analiza la relación entre los ingresos por financiamiento y el desempeño y el éxito electoral de 317.107 candidatos a concejal en las elecciones de 2008, 2012 y 2016 en 441 municipios brasileños con más de 50 mil votantes. Es un estudio sin precedentes en su alcance temporal que permite suplir la falta de estudios sobre elecciones subnacionales, especialmente en disputas relacionadas con el Poder Legislativo Municipal. Los datos utilizados provienen del Tribunal Superior Electoral, de los cuadros de candidaturas, resultados y rendición de cuentas. Las pruebas y análisis de datos e información utilizan recursos estadísticos de correlación y regresión, además de medidas centrales y razón de probabilidades. Los resultados confirman una fuerte asociación entre dinero, votación y éxito electoral, incluso en las elecciones de 2016, cuando se prohibieron las donaciones corporativas. El mantenimiento de un patrón tanto temporal como geográfico indica la existencia de características estructurales de carácter plutocrático en el financiamiento electoral de los candidatos a concejal en las ciudades encuestadas y, en consecuencia, en la producción de políticas públicas en Brasil.

Palabras clave:
financiamiento electoral; elección municipal; democracia; plutocracia; concejales

1. INTRODUCTION

In an attempt to contribute to the debate on the limits and challenges of political campaign funding in Brazil, this article analyzes the influence of plutocratic relationships on elections. We investigated 317,107 candidates for city council in the 2008, 2012, and 2016 elections in 441 municipalities with over 50 thousand voters. The longitudinal analysis sets our work apart from the Brazilian literature concerned with the role that money plays in municipal elections (Sampaio & Figueiredo, 2019). We underline the importance of our work due to our large-scale analysis of an elective office that has historically been the gateway for legislators and administrators into public life.

Legislative efforts have strived to reduce or limit the influence of money in elections - some have been successful; others, innocuous or counterproductive. Between 1993 and 2014, under Law 8.713/1993 (and its amendments), legal entities could donate to candidates, political parties, and electoral committees. In the 2014 elections, circa R$ 4.2 billion in donations was declared to the electoral court, with just over R$ 3 billion originating from corporate sources. Prior elections - national, state, or municipal - registered equally high billion-dollar figures from corporate investments.

Law 13,165/2015 banned corporate campaign donations in subsequent elections in an attempt to reduce the impact of financial resources in the electoral process. This alteration originated from the Direct Unconstitutionality Action (ADI) 4.650/11, proposed by the Order of Attorneys of Brazil (OAB) to the Supreme Federal Court (STF), in September 2011.

In the ADI, the OAB claimed that electoral success is largely contingent upon on costly campaigns, making politics overtly dependent on economic power. The action argued that this dynamic would be detrimental to democracy, engendering grave distortions. First, as it promotes political inequality by increasing the influence of the wealthiest over electoral results and, consequently, over government actions. Second, as it reduces the odds of candidates without substantial financial resources or close ties to holders of economic power. Third, as it produces nefarious associations between donors and politicians, who ultimately become compelled to return the favor after the elections (Supremo Tribunal Federal, 2011). In 2015, during the STF session which decided to ban corporate donations, Minister Luiz Fux declared that “we have reached an absolute chaotic scenario in which economic power illegally seizes political power” (Supremo Tribunal Federal, 2016, p. 276).

The first test run for the effects of Law 13,165/ 015 was the 2016 municipal elections, a singular research opportunity. The media’s expectation was that the new rules would limit the influence of money on electoral performance and success in Brazil, promoting greater equity among competitors. Did this effectively happen? In search of an answer, we guided our research according to the following questions: did greater campaign financial revenues result in more votes, i.e., better electoral performance in 2016? Did it increase the odds of electoral success, i.e., of the candidates being elected?

Our hypothesis was that the 2016 elections, under the ban on corporate financing, preserved the pattern of the 2008 and 2012 elections with a sharp correlation relationship between financial resources, electoral performance, and electoral success of candidates for city council in the municipalities surveyed.

The article is divided into 4 parts, in addition to this Introduction. The first section presents a brief discussion of our research problem as well as a theoretical background on electoral financing and its impact on Brazilian elections and democracy. We then describe the materials and methods, i.e., the procedures used for collecting and organizing data, the municipalities in our research object, and the tools, especially statistics, used in the analyses.

In the empirical section, we present and discuss the main findings of our work, first by way of correlation tests and linear regression, in order to verify the relationship between revenue and electoral performance of the candidates for city council; next, by calculating the average revenue and distribution of candidates by quartiles of campaign funding, in addition to probability ratio tests, to establish the relationship between revenues and electoral success. In the final remarks, we summarize our main findings and propose some perspectives for further research on plutocratic dynamics and its implications for the Brazilian democratic regime.

2. PLUTODEMOCRACY: MONEY, VOTE AND PUBLIC POWER

Voting is a prerequisite of democracy, as elections allow citizens to evaluate government officials, preserving or changing the political agents responsible for coordinating decision-making processes and implementing public policies. One of the effects of universal suffrage has been the increased cost of political competition and the growing demand for substantial financial resources to accommodate the advertising and publicity methods imposed by electoral mediatization.

Fundraising capacity has assumed a predominant role in the electoral process, to the point that most candidates who raise little or almost nothing find it unfeasible to manage central campaign activities, such as virtually interacting with the mass of voters. Without resources, the chances of electoral success are drastically reduced.

The volume of funds raised has been a highly predictive factor for election results, making economic power a vital element for the success of a competitor or for an incumbent to remain in the political arena (Cervi, 2010; Dias, Nossa & Monte-Mor, 2018; Figueiredo, 2009; Heiler, 2011; Junckes, Horochovski, Camargo, Silva & Chimin, 2019; Lemos, Marcelino & Pederiva, 2010; Mancuso & Speck, 2015; Peixoto, 2010; Reis & Eduardo, 2019; Sacchet & Speck, 2012a, 2012b).

The high commitment of some companies to politics fed the perception that donations result in favors, contributing to an understanding of Brazilian democracy as a plutocracy (Speck, 2016). The concept stems from a prior analysis of the political role of oligarchies, whereby money imposes itself, even in formally democratic regimes. Bianchi and Aliaga (2011), as well as Couto (2012), argued that the recovery and problematization of this Greek origin term were, in modern times, introduced by the so-called classic elite theory of the early 20^th century, whose main exponents include Michels (1982), Mosca (1939), and Pareto (1966). For them, the prominence of money would thwart the very premises of democracy, reducing it to an elite-dominated regime, in a plutodemocracy (Finer, 1968).

The uncritical acceptance of a perspective akin to the elitist risks burying democracy in the common grave of other regimes, denying improvements to the political system in the prospect of an expanded and less asymmetric participation of all citizens. Nonetheless, in recent years several authors have called attention to the limits of representative liberal democracy, asserting that the regime is experiencing difficulties with the upsurge of inequalities and authoritarianism on a global scale (Levitsky & Ziblatt, 2018; Miguel, 2014). According to Milanovic (2016), we are seeing the emergence of a global plutocracy, a phenomenon concurrent with the rise of populism in the early 21^st century, reinstating the notion of demagogic plutocracy as the fate of liberal democracy (Pareto, 1966).

Given the need to define the terms of the debate in this work, we consider plutocracy to be the political system in which the possession of economic wealth equates to the materialization of preferences and interests by securing political offices and influencing the structure of the State in decision-making processes and the production of public policies. By alternating or combining democracy and plutocracy, candidates whose agendas align with major economic interests are more likely to be elected. In other words, “representation would become plutocratic, and not of the polis” (Speck, 2005, p. 141).

Without considering that money itself may be “the root of all evil”, as Scarrow (2007) warned, the emerging question is when and how much money becomes a problem when influencing, or even determining, electoral results and public governance.

Dawood (2015) indicated that there is limited consensus on topics such as democratic representation, corruption, political influence, and governance. Although the empirical evidence of the “sale” of political influence remains under explored within the American literature, evidence points to an association between the flow of legally donated money and the production of public policies. Similar conclusions and limitations were reported by Santos (2016) in his doctoral thesis on the relationship between electoral financing and legislative production in Brazil, with the research findings published in Dinheiro, eleições e poder (Carazza, 2018).

In his studies, Santos (2016) supported the hypothesis that corporate donations affect electoral results and are motivated by the prospect of preferential treatment during the term of office. In his argument, he mentions the increase of resources in political campaigns, the dominance of millionaire corporate donations over the minimal portion of the electorate that becomes donors, the expressive participation of self-financing, the increasing centrality of political parties, and the concentration of corporate resources in candidates with higher odds of winning. The author concludes that as the Executive and Legislative branches perform increasingly specialized legislative tasks, more opportunities emerge for interest groups to engage in rent seeking.

Santos, Silva, Figueiredo and Rocha (2015) showed that the political alignment of Brazilian federal deputies between 1999 and 2007 in Congress was often controlled by the proportion of corporate resources that they received when candidates, at least within the scope of the legislative interests of the National Industry Confederation (CNI in the Portuguese Acronym). An analysis of the 2011-2014 legislature found a strong positive relationship between the volume of donations from the industrial sector and the number of proposals converging to the interests of the sector (Mancuso, Santos, Resende & Barboza, 2020). The relationship between electoral financing and legislative production also includes the asymmetrical presence of better-funded parliamentarians inside committees that best correspond to the interests of the economic sectors and major donor groups (Geara, Junckes, Horochovski & Clemente, 2018; Horochovski, Junckes & Serafim, 2020).

Listing a series of influencing factors exerted by elected officials and with the use of discontinuous regression analysis, Araújo (2012) inferred that donors for federal deputies elected in 2006 increased the number of public contracts between 2008 and 2010, in contrast to donors for unelected candidates. Using the same methodology and analyzing data on contracts made between 2004 and 2010, Boas, Hidalgo, and Richardson (2014) found that companies operating public works projects could obtain gains in their contracts of up to 14 times the value of donations made to a candidate belonging to the party of the Executive. Arvate, Barbosa and Fuzitani (2013) arrived at a similar conclusion when investigating the dynamics of state elections. Companies seek to influence their contracting relationship with the public sector through political campaign donations, and therefore the interrelationship between corporate actors may reveal more about their influence strategies than the analysis of their particular actions (Chagas, Haddad & Chagas, 2015).

Donations, or electoral investments (Mancuso, 2015), are mainly affected by political determinants - political capital, party size and ideology, among others - as well as the candidates’ individual characteristics - gender, education, and occupation. According to Silva and Cervi (2017), campaign donations are unequally distributed among candidates, benefiting certain political and social profiles. Thus, one of the most conspicuous impacts of money is the imbalance it poses on electoral processes as candidates with more resources are more likely to win.

In the 2014 elections for federal deputy, private companies and party leaders showed a propensity to maintain the status quo, directing financial resources to ensure incumbency. The allocation of corporate donations, whether directly or indirectly (via parties), did not favor important democratic values; on the contrary, it contributed to maintaining a political elite connected to traditionally privileged social structures (Mancuso, Horochovski & Camargo, 2018). In other words, Brazil perpetuates what Weber called the “plutocratic recruitment” of the political class (Codato, 2014), in which electoral financing stands as one of the leading explanatory factors for preserving this logic (Cervi, Costa, Codato & Perissinotto, 2015).

Studies on the determinants of investments and the relationship with electoral results have benefited from the widely available electoral data on candidates and the financial accounting reports in public repositories, especially those released by the Supreme Electoral Court (TSE). The vast majority of these researches stem from the inquiry by Samuels (2001) regarding the source and concentration level of resources to measure the impacts of money over competitive democracy in Brazil.

By organizing large databases and using methods ranging from simple statistical resources to complex tests and network metrics, studies in general have found that incumbency, district size, party size, gender, race, education, and belonging to the government base are variables that significantly influence a candidate’s revenue. As a result, there is a strong association between revenue, performance, and electoral success.

Mancuso and Speck (2014) described the three most relevant sources of funding, in descending order: corporate donations, donations from individuals, and the candidates’ own resources. Furthermore, as their argument goes, when we include direct funding to candidates as well as (indirect) funding through parties by the private sector, elections for senator, governor, and president are almost entirely funded by corporate donations. Upon analyzing the 2010 and 2014 campaigns for the Chamber of Deputies, Silva and Cervi (2017) showed that elected and re-elected candidates received the highest percentage of donations from legal entities.

In the 2010 elections, electoral revenues amounted close to 3 billion reais. Circa 75% of this amount refers to donations from 19 thousand private companies, whereby 50% of the amount donated by legal entities came from only 70 companies or groups. In the 2014 elections, approximately 4.2 billion reais were raised, 73% of which, or just over 3 billion reais, came from companies (Mancuso et al., 2018).

Horochovski, Junckes, Silva, Silva and Camargo (2016) emphasized the existence of a power structure through the relationships between donors and recipients in the 2010 elections. The authors identified that the greatest odds for electoral success, as well as the assertiveness of donors when distributing their resources to victorious candidates, pertain to the occupation of central and strategic positions in the campaign financing network. The greater or lesser centrality of a candidacy within the financing network implies a difference of up to 147 times in the odds of electoral success.

Junckes et al. (2019) showed that no more than 5.6% of natural persons and 21.9% of legal entities decisively influenced the election of almost all candidates in 2014. Candidates and donors which had only one relationship on the network, especially natural persons, were defined as “extras”. The authors identified a core comprised of 0.16% of donors (239 companies) responsible for funding 81% of the elected candidates. Among the candidates funded by these corporate super donors are 26 of the 27 elected senators, 464 of the 513 federal deputies, 25 of the 27 governors, and 79% of state and district deputies, in addition to the president. It is a highly connected network, comprised of the main business groups financing party agents and candidates.

Given the impacts on democracy and influence on public policies, the study of electoral financing has become one of the greatest challenges for scholars dedicated to Brazilian politics. While there has been an increasing number of studies on the matter, the literature has commonly addressed general elections. Few studies have analyzed municipal elections (Sampaio & Figueiredo, 2019). Our investigation addresses this shortcoming with a contribution to the debate on the plutocratic dimensions of electoral financing. The next section describes our process.

3. MATERIALS AND METHODS

We gathered the data for the 2008, 2012, and 2016 municipal elections from the Territorial Studies Group (Gete) of the State University of Ponta Grossa (UEPG), which we then debugged, tabulated, and organized in electronic spreadsheets. A CSV file (codific. 1252) with the data used in our study is available at https://tinyurl.com/rtco7ed and can be downloaded for reference or retesting. Our selection consists of Brazilian municipalities with over 50 thousand voters in 2016, as suggested by Speck and Cervi (2016) and Sampaio and Figueiredo (2019). For the list of municipalities, see http://www.tse.jus.br/eleicoes/estatisticas/estatisticas-eleitorais. Figure 1 shows the distribution of the 441 municipalities included in our survey.

Figure 1
Geographic location of municipalities with more than 50 thousand voters

In 2016, these municipalities concentrated 61% (123,927,389) of the estimated resident population in the country, according to https://sidra.ibge.gov.br/pesquisa/estimapop/tabelas, and 59.3% (85,387,248) of voters. In the 2008, 2012, and 2016 elections the surveyed municipalities represented, respectively, 58%, 57%, and 50% of the total campaign revenues, and 24%, 28% and 29% of the total candidates for city council. Table 1 shows the distribution of the 317,107 eligible candidates and the electoral funding amount of R$ 2,866,057,116 in the three elections, with values updated by the Broad Consumer Price Index (IPCA), provided by the Brazilian Institute of Geography and Statistics (IBGE), for October 2016.

Table 1
Eligible candidates for city council and campaign revenues (municipalities with more than 50 thousand voters, by election)

Year	Candidates	Campaign Revenue
	N	R$
2008	79,420	871,500,638
2012	113,488	1,280,946,80
2016	124,199	713,610,098

• Source: Elaborated by the authors based on data from Gete-UEPG

As previously mentioned, our hypothesis is that the 2016 elections, under the ban on corporate financing, followed the pattern of the 2008 and 2012 elections with a strong correlation between financial resources, electoral performance, and electoral success of candidates for city council in the municipalities surveyed.

Our predictor variable is, therefore, the total amount of donations received and declared in the campaign, and the dependent variables are the candidates’ performance, measured by the total votes obtained, and electoral success, measured by the binary elected vs. unelected. The methodology is quantitative with statistical analysis, with the use of correlation and linear regression tests, between campaign revenues and votes, comparisons between average revenues of elected and unelected candidates, distribution of these candidates by funding contribution quartiles, and odds ratio of electoral success. The calculations and tests were performed with Jamovi R software.

In 2016, the number of voters ranged from 50,192 to 8,886,324, and the values per municipality from R$ 149,280 to R$ 42,996,783. Since the districts in our survey contain such variations, we normalized the data by establishing relative indexes of the “candidate’s share of the total revenue from all candidates in the district/municipality” (PCRM) and “candidate’s share of the total votes in the district/municipality” (PCVM). We applied the following formulas to obtain this data:

$PCRM=\frac{TRC}{TRCM}$

PCRM is the candidate’s share of the total revenue from all candidates in the district/municipality, TRC is the total revenue of the candidate, and TRCM is the total revenue of the candidates in the district/municipality.

$PCVM=\frac{TVC}{TVM}$

PCVM is the candidate’s share of the total votes in the district/municipality, TVC is the candidate’s total votes, and TVM is the total number of votes in the district/municipality.

In the following section, we present and discuss our findings from these methodological procedures.

4. DEMONSTRATION AND DISCUSSION OF THE RESULTS

4.1 The Relationship between Revenue and Electoral Performance

To answer the question “does higher campaign revenue result in more votes?” we correlated revenues declared by candidates and their vote tallies, using both the absolute data for the revenues and votes of the 317,107 candidates and the relative data of the PCRM and PCVM.

4.1.1 Absolute data

Graphs 1, 2, and 3 ¹ are dispersion diagrams showing the shape and direction of the relationship between the variables votes (dependent) and revenues (explanatory) in the 2008, 2012, and 2016 elections.

Graph 1
Dispersion diagram between campaign revenues and votes in the 2008 elections

Graph 2
Dispersion diagram between campaign revenues and votes in the 2012 elections

Graph 3
Dispersion diagram between campaign revenues and votes in the 2016 elections

Once we verified the data distribution, we performed correlation and linear regression tests. Table 2 shows the main results, specifically the Pearson correlation coefficient (r), the coefficient of determination (r²), and the P-value (p).

Table 2
Correlation coefficients between campaign revenues and votes, by election

Year	r	r2	p
2008	0.7416	0.5500	<0.0001 2
2012	0.7312	0.5347	<0.0001
2016	0.6192	0.3835	<0.0001

• Source: Elaborated by the authors based on data from Gete-UEPG.

The three elections showed a strong correlation between campaign revenues and votes obtained.³ Nonetheless, in the 2016 elections a specific candidate, Eduardo Suplicy from São Paulo, obtained 301,446 votes with a total revenue of R$ 207,190, while the runner-up amassed 107,957 votes with a revenue of R$ 2,396,056. To verify the influence of this event on the correlation, we performed a test excluding this candidate, in which we obtained a higher correlation coefficient: 0.7073 (r²=0.5003 and p<0.0001).

The direction and strength of the correlation allowed us to find the equations of regression lines that best describe the relationship between the variables (Graph 4).

Graph 4
Linear regression lines between campaign revenues and votes in the 2008, 2012, and 2016 elections

The model shows that, without financial resources, the number of votes decreases with every election. The equation for the 2008 elections indicates that a candidate without revenue would obtain circa 326 votes - intercept of the axes or the value “a” in the regression equation. In the 2012 elections this number would drop to 245 votes. And in the 2016 elections the candidate would receive only 149 votes in the lack of any funds.

Adding one thousand reais in electoral revenues - a relatively plausible value for most candidates under fair conditions- we would have 355 votes in 2008, 266 in 2012, and 201 in 2016. With 10 thousand reais, a much less equitable amount, the results would be 615 votes in 2008, 456 in 2012, and 668 in 2016. If we add 100 thousand reais, by no means an equitable amount, the results would be 3,216 votes in 2008, 2,355 in 2012, and 5,339 in 2016. In other words, according to the model, in 2016 the more disparate the conditions for electoral investments, the more propitious the plutocratic logic becomes. Graph 4 illustrates, with other examples, the prediction of the number of votes for other values.

We also projected the number of votes per amount of money. We adopted the thousand votes mark, the “dream” of any candidate in municipalities with more than 50 thousand voters. The model allowed us to estimate that, in the 2008 elections, R$ 23,321 was needed to “produce” a thousand votes; in 2012, R$ 35,781; and, in 2016, only R$ 16,396. An impulsive analyst could very well celebrate this cost reduction in 2016. While the vote in 2016 was undeniably “cheaper”, from the data presented thus far we may also infer that cheaper votes were acquired, in greater volume, by those with more money. In short, the influence of money on number of votes persisted over the course of the analyzed cycle, growing in 2016, when it became less expensive to influence results.

In an attempt to better detail the correlation between revenues and number of votes, we calculated the correlation coefficient per municipality. In the 441 cities surveyed, we observed that only 1 in 2008 as well as only 1 in 2012 had a weak correlation. In 2008, 318 municipalities (72.1%) had a strong correlation. In 2012, we found 343 (77.8%), and in 2016, 371 (84.1%). In tandem with the previous data, we can say that the strength of the correlation between money and votes in the 3 elections grows considerably in the municipalities studied.

4.1.2 Relative data

Considering the magnitude of the electoral districts analyzed, we correlated the PCRM and the PCVM. Table 3 presents our main findings.

Table 3
Correlation coefficients between PCRM and PCVM, by election

Year	R	r2	p
2008	0.6766	0.4578	<0.0001
2012	0.6661	0.4437	<0.0001
2016	0.6978	0.4869	<0.0001

• Source: Elaborated by the authors based on data from Gete-UEPG.

Comparing the values of r in Table 2 - correlation between the absolute data of revenues and votes - we confirmed a strong and positive correlation between the variables analyzed in the 3 elections, with similar linear equations.

As the relative data allowed us to draft a revenue ranking in each electoral district, we grouped the candidates into quartiles of share of the total revenue in the municipality and share of the total votes in the municipality. A similar procedure was performed by Sacchet and Speck (2012b) by gathering candidates together into successful quintiles of revenue and votes. Table 4 combines both shares, forming 16 subgroups per year.

Table 4
Proportion of candidates (by quartiles) according to the combination between PCRM and PCVM, by election

Proportion of candidates in each subgroup/quartile* (%)			Candidate’s share of the total votes in the municipality
			4		3	2	1	Total
A	2008	Candidate’s share of the total revenue in the municipality	4	17.30	5.76	1.64	0.29	25.00
			3	5.19	9.90	7.40	2.51	25.00
			2	1.19	5.42	9.53	8.86	25.00
			1	1.32	3.92	6.43	3.33	25.00
			Total	25.00	25.00	25.00	25.00	100.00
	2012	Candidate’s share of the total revenue in the municipality	4	17.32	5.60	1.74	0.33	25.00
			3	5.23	9.82	7.38	2.57	25.00
			2	1.43	6.18	9.75	7.64	25.00
			1	1.02	3.40	6.12	4.46	25.00
			Total	25.00	25.00	25.00	25.00	100.00
	2016	Candidate’s share of the total revenue in the municipality	4	17.32	5.33	1.82	0.53	25.00
			3	5.18	9.60	7.07	3.14	25.00
			2	1.62	6.60	9.18	7.60	25.00
			1	0.88	3.48	6.92	3.72	25.00
			Total	25.00	25.00	25.00	25.00	100.00

• * Subgroup/quartile 4 is the upper quartile, i.e., containing candidates with the highest revenue or most votes; subgroup/quartile 1 is the lowest.
• Source: Elaborated by the authors based on data from Gete-UEPG.

The concentration of subgroups with the highest proportions of candidates on the diagonal line, highlighted in gray, shows the relationship between higher revenue and number of votes, being therefore “further evidence of the importance of this resource for electoral performance”, as expressed by Sacchet and Speck (2012b, p. 191).

We also calculated the Pearson’s correlation coefficient by municipality using the proportion data. Of the 441 municipalities in our study, only 1 in 2008, and 1 in 2012 showed weak correlations. Over 70% of the municipalities had strong correlations. In 2008, this correlation intensity was found in 318 municipalities (72.1%). In 2012, we found 343 (77.8%), and in 2016, 371 (84.1%), the highest number in the 3 elections. Although with some variations, we found similar results in the previous section and in this section. That is, regardless of district size the coefficients almost always indicate a strong correlation between revenues and votes.

In this first part of the results, we sought to verify if higher campaign revenues result in a higher vote tally. In the 3 elections for city council, the correlation is statistically strong and positive between the variables analyzed. Corroborating the literature on electoral financing, we answered the question and showed that money makes all the difference when it comes to earning votes in Brazilian municipal elections, and was even intensified in 2016.

4.2 The Relationship between Revenue and Electoral Success

We sought to identify and analyze the relationship between campaign revenues and electoral results, investigating whether elected candidates received more money than unelected candidates, thus assessing whether higher revenues increase the odds of electoral success. We performed a twofold analysis: first, with absolute data, investigating the differences in revenue between elected and unelected candidates; second, with relative data, regarding the candidate’s share of the total revenue from all candidates in the district/municipality (PCRM), we sought to verify, in addition to placing elected and unelected candidates by revenue quartiles, the odds ratios of success by virtue of such placements.

4.2.1 Absolute data

In this section we analyze the differences between the amount of resources received by elected and unelected candidates, calculating and comparing the average revenues in each group. We began with a general analysis of the 3 elections, whose results are summarized in Table 5.

Table 5
Candidates, campaign revenue, and average revenue by election and situation (elected and unelected)

Year	Situation	Candidates		Campaign revenue		Average revenue	Difference Elected/Unelected
		N	%	R$	%	R$	%
2008	Elected	5,668	7.1	312,962,705	35.9	55,216	629.1
	Unelected	73,752	92.9	558,537,933	64.1	7,573
	Total	79,420	00.0	871,500,638	100.0	0,973
2012	Elected	7,424	6.5	454,260,714	35.5	61,188	685.1
	Unelected	106,064	93.5	826,685,666	64.5	7,794
	Total	113,488	00.0	1,280,946,380	100.0	1,287
2016	Elected	7,565	6.1	236,466,776	33.1	31,258	664.1
	Unelected	116,634	93.9	477,143,322	66.9	4,091
	Total	124,199	00.0	713,610,098	100.0	5,746

• Source: Elaborated by the authors based on data from Gete-UEPG.

In the 2008, 2012, and 2016 elections, the elected city councilors corresponded to 7.1%, 6.5%, and 6.1% of the candidates, respectively, but concentrated 35.9%, 35.5% and 33.1% of the revenue. In other words, the average revenue per elected candidate is much higher than an unelected one, with discrepancies above 600% in the three years, reaching 685% in 2012.

Only 2 municipalities in 2008, 3 in 2012, and 5 in 2016 registered differences between 0% and 100%. We found 36 municipalities with differences greater than 1,000% in 2008, 60 in 2012, and 45 in 2016. The greatest disparity was in the city of Rio de Janeiro in 2012, when the average revenue of elected candidates exceeded non-elected candidates by 2,517%. In the other municipalities, the differences varied between 101% and 1,000%.

In short, the average revenue of elected candidates is always higher when compared to unelected candidates across all cities and in every election. In approximately 99% of the 441 municipalities analyzed, each elected councilor had, on average, at least twice the resources of defeated candidates. Therefore, the highly asymmetric conditions for participating in the democratic process and the influence of money are blatant when it comes to electoral success for city councilor.

4.2.2 Relative data

We used the relative data for the elected and unelected candidates’ share of the total revenue from all candidates in the district/municipality (PCRM) to analyze their standing in the revenue ranking, classifying elected and unelected candidates by revenue quartiles and calculating the odds of electoral victory (RC) according to our classification. The results are in Table 6.

Table 6
Elected and unelected candidates classified in quartiles according to PCRM and odds ratio, by election

Year	Quartiles*	Elected		Unelected		RC - 1st and 4th Quartile
		N	%	N	%
2008	4	5,064	89.3	14,791	20.1	251.4	Candidates in the 4th quartile are 251.4 times more likely to be elected.
	3	509	9.0	19,346	26.2
	2	68	1.2	19,787	26.8
	1	27	0.5	19,828	26.9
	Total	5,668	100.0	73,752	100.0
2012	4	6,806	91.7	21,566	20.3	470.9	Candidates in the 4th quartile are 470.9 times more likely to be elected.
	3	517	7.0	27,855	26.3
	2	82	1.1	28,290	26.7
	1	19	0.3	28,353	26.7
	Total	7,424	100.0	106,064	100.0
2016	4	6,998	92.5	24,051	20.6	29.9	Candidates in the 4th quartile are 429.9 times more likely to be elected.
	3	481	6.4	30,569	26.2
	2	65	0.9	30,985	26.6
	1	21	0.3	31,029	26.6
	Total	7,565	100.0	116,634	100.0

• * Subgroup/quartile 4 is the upper quartile, i.e., containing candidates with the highest revenue or most votes; subgroup/quartile 1 is the lowest.
• Source: Elaborated by the authors based on data from Gete-UEPG.

In all elections, the vast majority of elected city councilors are in the top quartile, which concentrates the 25% of candidates with the highest revenues. In 2008, 89.3% of elected candidates were in this quartile. In 2012, they accounted for 91.7%, and in 2016, 92.5%. In the lower quartile, there are few victors: only 0.5% in 2008 and 0.3% in 2012 and 2016. Comparing elected and unelected candidates from the upper and lower quartiles, the former holds a vast advantage. Being in the top quartile, i.e., having raised more funds, increased the odds of electoral success in 251 times in 2008, 471 times in 2012, and 430 times in 2016.

All elected candidates were in the upper quartile in 124 municipalities in 2008, 133 in 2012, and 135 in 2016, and comprised over 80% of elected candidates in most municipalities - 322 in 2008, 354 in 2012, and 370 in 2016. In only 3 municipalities, in 2008 and 2016, those elected in the upper quartile did not exceed 50% of the total. In the remaining 438 municipalities (and every municipality in 2012), at least 51% of elected candidates belonged to the highest revenue quartile. We were even unable to calculate the odds ratios in 93% of the municipalities in 2008 and 2016, and in 94% of the municipalities in 2012, as not a single elected candidate belonged to the lower revenue quartile. In other words, in 93% of the municipalities in our study, a postulant for city councilor within the 25% lowest revenue quartile would have zero chance of being elected.

Both the absolute and relative data presented here converge with the literature, and, borrowing the words of Sacchet and Speck (2012b, p. 183), “we can conclude that campaign financing has a close relationship with electoral success”.

5. CLOSING REMARKS

The results of this research substantially confirm our hypothesis that the ban on corporate campaign financing, starting from the 2016 elections, was insufficient to alter the prevailing pattern found in previous municipal elections for city councilor, which revealed a strong relationship between financial resources, electoral performance, and electoral success in the municipalities surveyed.

The correlations and regressions showed that, in the last election, the association between money and votes was tighter than in 2008 and 2012, hence the greater inequity between candidates. The analysis of average revenues, classification by quartiles, and odds ratio suggest that, in order to increase the odds of electoral success, a candidate should ideally be in the upper tier of electoral funding. In 93% of Brazilian municipalities, not a single elected candidate belonged to the lowest revenue quartile.

The endurance of a temporal and geographical pattern suggests the existence of more or less structural characteristics in the financing of candidates for city councilor in the municipalities surveyed, which is in tandem with researches focused on other elective offices. We conclude that the clearest impacts of money on elections are an imbalance in competition between candidates and the emergence of a plutocratic dynamic that corrupts democracy in the country. These impacts were in no way affected with the end of corporate financing.

This situation reinforces the need, reiterated by several authors, for further research on the plutocratic dynamics and its consequences for Brazil’s democratic regime. One of the possibilities in this direction is to research the profile of donors emerging as the main financers in the absence of companies. Wealthy candidates funding their own campaigns? A business elite and capital owners? Party leaderships deciding the allocation of growing public resources? These are all hypotheses to be tested.

ACKNOWLEDGMENTS

We express our gratitude to the National Council for Scientific and Technological Development (CNPq) for their funding and support for Project 8629 - Impacts of the ban on corporate financing in the 2016 elections in Brazil. We also thank the anonymous reviewers for their valuable contributions which enabled significant improvements to the article.

REFERÊNCIAS

• Araújo, G. B. (2012). Contribuições de campanha influenciam decisões públicas? O caso dos contratos públicos federais e das emendas ao orçamento no Brasil (Tese de Doutorado). Universidade de São Paulo, São Paulo, SP.
• Arvate, P. R; Barbosa, K; & Fuzitani, E. (2013). Campaign donation and government contracts in Brazilian states(Working Paper n. 336). São Paulo, SP: Escola de Economia de São Paulo da Fundação Getulio Vargas.
• Bianchi, A; & Aliaga, L. (2011). Força e consenso como fundamentos do Estado: Pareto e Gramsci. Revista Brasileira de Ciência Política, 5, 17-36. Recuperado de https://doi.org/10.1590/S0103-33522011000100002
  » https://doi.org/10.1590/S0103-33522011000100002
• Boas, T. C; Hidalgo, F. D; & Richardson, N. P. (2014). The spoils of victory: campaign donations and government contracts in Brazil. The Journal of Politics, 76(2), 415-429. Recuperado dehttps://doi.org/10.1017/S002238161300145X
  » https://doi.org/10.1017/S002238161300145X
• Callegari-Jacques, S. M. (2003). Bioestatística: princípios e aplicações Porto Alegre, RS: Artmed.
• Carazza, B. (2018). Dinheiro, eleições e poder: as engrenagens do sistema político brasileiro São Paulo, SP: Companhia das Letras.
• Cervi, E. U. (2010). Financiamento de campanhas e desempenho eleitoral no Brasil: análise das contribuições de pessoas físicas, jurídicas e partidos políticos às eleições de 2008 nas capitais de Estado. Revista Brasileira de Ciência Política, 4, 135-167. Recuperado de https://periodicos.unb.br/index.php/rbcp/article/view/1712/1508
  » https://periodicos.unb.br/index.php/rbcp/article/view/1712/1508
• Cervi, E. U; Costa, L. D; Codato, A; & Perissinotto, R. (2015). Dinheiro, profissão e partido: a vitória na eleição para deputado federal no Brasil em 2010. Revista Sociedade e Estado, 30(1), 189-205. Recuperado dehttp://www.scielo.br/scielo.php?script=sci_arttext&pid=S0102-69922015000100189&lng=en&nrm=iso
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0102-69922015000100189&lng=en&nrm=iso
• Chagas, L. S; Haddad, E. A; & Chagas, A. L. S. (2015). Uma análise da interação entre empresas nas contribuições de campanhas para obtenção de contratos com o Governo Federal(Working Paper TD Nereus 07-2015). São Paulo, SP: Núcleo de Economia Regional e Urbana da FEA/USP.
• Codato, A. (2014). Classe política e regime autoritário: os advogados do Estado Novo em São Paulo. Revista Brasileira de Ciências Sociais, 29(84), 145-163. Recuperado dehttps://www.scielo.br/scielo.php?pid=S0102-69092014000100010&script=sci_arttext
  » https://www.scielo.br/scielo.php?pid=S0102-69092014000100010&script=sci_arttext
• Couto, C. G. (2012). Oligarquia e processos de oligarquização: o aporte de Michels à análise política contemporânea. Revista de Sociologia Política, 20(44), 47-62. Recuperado de http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-44782012000400004&lng=en&nrm=iso
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-44782012000400004&lng=en&nrm=iso
• Dawood, Y. (2015). Campaign Finance and American Democracy. Annual Review of Political Science, 18(1), 329-348. Recuperado de https://doi.org/10.1146/annurev-polisci-010814-104523
  » https://doi.org/10.1146/annurev-polisci-010814-104523
• Dias, B. P; Nossa, V; & Monte-Mor, D. S. (2018). O investimento público influencia na reeleição? Um estudo empírico nos municípios do estado do Espírito Santo. Revista de Administração Pública, 52(5), 880-898. Recuperado dehttp://bibliotecadigital.fgv.br/ojs/index.php/rap/article/view/76973/73753
  » http://bibliotecadigital.fgv.br/ojs/index.php/rap/article/view/76973/73753
• Figueiredo, D. B Filho. (2009). O elo corporativo? Grupos de interesse, financiamento de campanha e regulação eleitoral (Dissertação de Mestrado). Universidade Federal de Pernambuco, Recife, PE.
• Finer, S. E. (1968). Pareto and pluto-democracy: The retreat to Galapagos. The American Political Science Review, 62(2), 440-450. Recuperado de https://doi.org/10.2307/1952939
  » https://doi.org/10.2307/1952939
• Geara, L. E; Junckes, I. J; Horochovski, R. R; & Clemente, A. J. (2018). O financiamento eleitoral nas comissões permanentes da Câmara dos Deputados: uma análise do metarrelacionamento entre empresas e parlamentares. E-Legis-Revista Eletrônica do Programa de Pós-Graduação da Câmara dos Deputados, 11(26), 60-81. Recuperado de http://e-legis.camara.leg.br/cefor/index.php/e-legis/article/view/409
  » http://e-legis.camara.leg.br/cefor/index.php/e-legis/article/view/409
• Heiler, J. G. (2011). Democracia: o jogo das incertezas x financiamento de campanhas. Uma análise das prestações de contas das campanhas de vereadores de SC (Dissertação de Mestrado). Universidade Federal de Santa Catarina, Florianópolis, SC.
• Horochovski, R. R; Junckes, I. J; & Serafim, T. R. (2020). A quem interessam as políticas ambientais? Redes de financiamento eleitoral empresarial na Comissão de Meio Ambiente e Desenvolvimento Sustentável (Cmads) da Câmara dos Deputados. E-Legis-Revista Eletrônica do Programa de Pós-Graduação da Câmara dos Deputados, 13(31), 32-60. Recuperado dehttp://e-legis.camara.leg.br/cefor/index.php/e-legis/article/view/533
  » http://e-legis.camara.leg.br/cefor/index.php/e-legis/article/view/533
• Horochovski, R. R; Junckes, I. J; Silva, E. A; Silva, J. M; & Camargo, N. F. (2016). Estruturas de poder nas redes de financiamento político nas eleições de 2010 no Brasil. Opinião Pública, 22(1), 28-55. Recuperado dehttp://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-62762016000100028
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-62762016000100028
• Junckes, I. J; Horochovski, R. R; Camargo, N. F; Silva, E. A; & Chimin, A. B Jr;. (2019). Poder e democracia: uma análise da rede de financiamento eleitoral em 2014 no Brasil. Revista Brasileira de Ciências Sociais, 34(100), 1-22. Recuperado de https://doi.org/10.1590/3410006/2019
  » https://doi.org/10.1590/3410006/2019
• Lemos, L. B; Marcelino, D; & Pederiva, J. H. (2010). Porque dinheiro importa: a dinâmica das contribuições eleitorais para o Congresso Nacional em 2002 e 2006. Opinião pública, 16(2), 366-393. Recuperado de http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-62762010000200004
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-62762010000200004
• Levitsky, S; & Ziblatt, D. (2018). Como as democracias morrem São Paulo, SP: Zahar.
• Mancuso, W. P. (2015). Investimento eleitoral no Brasil: balanço da literatura (2001-2012) e agenda de pesquisa. Revista de Sociologia e Política, 23(54), 155-183. Recuperado de http://www.scielo.br/pdf/rsocp/v23n54/0104-4478-rsocp-23-54-0155.pdf
  » http://www.scielo.br/pdf/rsocp/v23n54/0104-4478-rsocp-23-54-0155.pdf
• Mancuso, W. P; Horochovski, R. R; & Camargo, N. F. (2018). Financiamento eleitoral empresarial direto e indireto nas eleições nacionais de 2014. Revista Brasileira de Ciência Política, 27, 9-36. Recuperado de http://www.scielo.br/pdf/rbcpol/n27/2178-4884-rbcpol-27-9.pdf
  » http://www.scielo.br/pdf/rbcpol/n27/2178-4884-rbcpol-27-9.pdf
• Mancuso, W. P; Santos, M. L; Resende, C. A. D. S; & Barboza, D. P. (2020). Financiamento eleitoral e comportamento parlamentar: a relação entre doações da indústria e proposição de leis (Texto para discussão, 2578). Rio de Janeiro, RJ: Instituto de Pesquisa Econômica Aplicada. Recuperado de https://www.ipea.gov.br/portal/images/stories/PDFs/TDs/td_2578.pdf
• Mancuso, W. P; & Speck, B. W. (2014). Financiamento de campanhas e prestação de contas. Cadernos Adenauer, 15(1), 135-150. Recuperado dehttp://www.kas.de/wf/doc/13776-1442-5-30.pdf
  » http://www.kas.de/wf/doc/13776-1442-5-30.pdf
• Mancuso, W. P; & Speck, B. W. (2015). Financiamento empresarial na eleição para deputado federal (2002-2010): determinantes e consequências. Teoria & Sociedade, 23(2), 103-125. Recuperado dehttp://bibliotecadigital.tse.jus.br/xmlui/handle/bdtse/4566
  » http://bibliotecadigital.tse.jus.br/xmlui/handle/bdtse/4566
• Michels, R. (1982). Sociologia dos partidos políticos Brasília, DF: UnB.
• Miguel, L. F. (2014). Democracia e representação: territórios em disputa São Paulo, SP: Unesp.
• Milanovic, B. (2016). Global inequality: a new approach for the age of globalization Cambridge, MA: Harvard University Press.
• Mosca, G. (1939). The Ruling Class New York, NY: McGraw Hill Book Company.
• Pareto, V. (1966). As elites e o uso da força na sociedade. In A. Souza(Org.), Sociologia política Rio de Janeiro, RJ: Zahar.
• Peixoto, V. (2010). Eleições e financiamento de campanhas no Brasil (Tese de Doutorado). Instituto Universitário de Pesquisas, Rio de Janeiro, RJ.
• Reis, B. P. W; & Eduardo, F. L. (2019). A distância entre intenção e gesto: consequências da proibição de doações eleitorais por pessoas jurídicas nas eleições municipais de 2016. Boletim de Análise Político-Institucional, 21, 67-76. Recuperado dehttps://www.ipea.gov.br/portal/images/stories/PDFs/boletim_analise_politico/191021_bapi_21_artigo_6.pdf
  » https://www.ipea.gov.br/portal/images/stories/PDFs/boletim_analise_politico/191021_bapi_21_artigo_6.pdf
• Sacchet, T; & Speck, B. W. (2012a). Dinheiro e sexo na política brasileira: financiamento de campanha e desempenho eleitoral em cargos legislativos. In: J. E. D. Alves, C. R. J. Pinto, & F. Jordão (Orgs.), Mulheres nas eleições 2010 São Paulo, SP: ABCP/Secretaria de Políticas para as Mulheres.
• Sacchet, T; & Speck, B. W. (2012b). Financiamento eleitoral, representação política e gênero: uma análise das eleições de 2006. Opinião pública, 18(1), 177-197. Recuperado de http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-62762012000100009
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-62762012000100009
• Sampaio, D; & Figueiredo, D. B ; Filho. (2019). Como o dinheiro influencia as eleições municipais no Brasil: uma revisão sistemática. Revista Brasileira de Informação Bibliográfica em Ciências Sociais, 88, 1-25. Recuperado dehttp://anpocs.com/images/BIB/n88/1017666bib88052019.pdf
  » http://anpocs.com/images/BIB/n88/1017666bib88052019.pdf
• Samuels, D. (2001). Money, elections, and democracy in Brazil. Latin American Politics and Society, 43(2), 27-48. Recuperado dehttps://doi.org/10.1111/j.1548-2456.2001.tb00398.x
  » https://doi.org/10.1111/j.1548-2456.2001.tb00398.x
• Santos, B. C. (2016). Interesses econômicos, representação política e produção legislativa no Brasil sob a ótica do financiamento de campanhas eleitorais (Tese de Doutorado). Universidade Federal de Minas Gerais, Belo Horizonte, MG.
• Santos, M. L; Silva, M. B; Figueiredo, D. B Filho & Rocha, E. C. (2015). Financiamento de campanha e apoio parlamentar à Agenda Legislativa da Indústria na Câmara dos Deputados. Opinião Pública, 21(1), 33-59. Recuperado de https://doi.org/10.1590/1807-019121133
  » https://doi.org/10.1590/1807-019121133
• Scarrow, S. E. (2007). Political finance in comparative perspective. Annu. Rev. Polit. Sci; 10, 193-210. Recuperado de https://www.annualreviews.org/doi/abs/10.1146/annurev.polisci.10.080505.100115
  » https://www.annualreviews.org/doi/abs/10.1146/annurev.polisci.10.080505.100115
• Silva, B. F; & Cervi, E. U. (2017). Padrões de financiamento eleitoral no Brasil: as receitas de postulantes à Câmara dos Deputados em 2010 e 2014. Revista Brasileira de Ciência Política, 23, 75-110. Recuperado dehttp://www.scielo.br/scielo.php?script=sci_arttext&pid=S0103-33522017000200075
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0103-33522017000200075
• Speck, B. W. (2005). Reagir a escândalos ou perseguir ideais? A regulação do financiamento político no Brasil. Cadernos Adenauer, 6(2), 123-159. Recuperado dehttps://www.kas.de/c/document_library/get_file?uuid=80ada830-0761-8346-58c6-2fc8deb12bb0&groupId=265553
  » https://www.kas.de/c/document_library/get_file?uuid=80ada830-0761-8346-58c6-2fc8deb12bb0&groupId=265553
• Speck, B. W. (2016). Game over: duas décadas de financiamento de campanhas com doações de empresas no Brasil. Revista de Estudios Brasileños, 3(4), 125-135. Recuperado dehttp://bibliotecadigital.tse.jus.br/xmlui/handle/bdtse/4404
  » http://bibliotecadigital.tse.jus.br/xmlui/handle/bdtse/4404
• Speck, B. W; & Cervi, E. U. (2016). Dinheiro, tempo e memória eleitoral: os mecanismos que levam ao voto nas eleições para prefeito em 2012. Dados, 59(1), 53-90. Recuperado dehttp://www.scielo.br/scielo.php?script=sci_arttext&pid=S0011-52582016000100053
  » http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0011-52582016000100053
• Supremo Tribunal Federal. (2011). Ação Direta de Inconstitucionalidade nº 4.650 - Distrito Federal. Petição inicial Relator: Ministro Luiz Fux. Brasília, DF: Diário da Justiça Eletrônico. Recuperado dehttp://redir.stf.jus.br/paginadorpub/paginador.jsp?docTP=TP&docID=1432694&prcI D=4136819#
  » http://redir.stf.jus.br/paginadorpub/paginador.jsp?docTP=TP&docID=1432694&prcI D=4136819#
• Supremo Tribunal Federal. (2016). Ação Direta de Inconstitucionalidade nº 4.650 - Distrito Federal. Inteiro Teor do Acórdão Relator: Ministro Luiz Fux. Brasília, DF: Diário da Justiça Eletrônico. Recuperado dehttp://portal.stf.jus.br/processos/downloadPeca.asp?id=308746530&ext=.pdf
  » http://portal.stf.jus.br/processos/downloadPeca.asp?id=308746530&ext=.pdf

• 1
  For better visualization, the graphs are presented, on both axes (x and y), on a base-10 logarithmic scale. As the scale starts at 1 (one), values equal to 0 (zero), whether for revenue or votes, were not plotted. This issue, however, does not interfere with our purposes to verify the nature and direction of the relationship between the variables.
• 2
  Since p<0,05, there is statistical evidence that the variables are related.
• 3
  If 0.00 <r <0.30, the correlation is weak; if 0.30≤r <0.60, the correlation is moderate; if 0.60≤r <0.90, the correlation is strong; if 0.90≤r <1.00, the correlation is very strong (Callegari-Jacques, 2003). We shall use this classification in this article for analytical and comparative purposes.

• [Translated version] Note: All quotes in English translated by this article’s translator.

Publication Dates

• Publication in this collection
  09 July 2021
• Date of issue
  May-Jun 2021

History

• Received
  26 Mar 2020
• Accepted
  17 Sept 2020