Introduction

Brazil is the country in the Americas with the least number of women in legislative positions in the Lower House of the legislature (or Chamber of Deputies, as it is known in Brazil). Despite enacting a 30% gender quotas policy two decades ago, the number of women occupying these positions in the country has remained below 10%. Brazil was one of the first countries in the Americas to implement such a policy, in 1997. However, while the region, at that time, had an average percentage of women in these positions of 13.3%, today this is 28.2% (the second highest worldwide), by contrast, in Brazil it has only risen from 6.6% to 9.5% (Inter-parliamentary Union 2017a).¹

The considerable regional increase seen over this 20-year period is the result of a steady, though uneven growth in the number of women elected in different countries, mainly propelled by the implementation of gender quotas. Argentina and Costa Rica became the most well-known cases of gender quotas success when, at the beginning of the 1990s and in the 2000s, respectively, they become the first countries in Latin America to reach the 30% threshold of women in national parliamentary positions. This threshold is deemed important in the specialized literature, in that it creates a critical mass that should encourage coalition building among women (^{Dahlerup 1988}; ^{Kanter 1977}). Argentina took the lead in 1991 by enacting a 30% quota for candidacies for its Lower House. Costa Rica then followed suit in 1996 passing a 40% quota. Subsequently, many other countries, including Brazil, established similar policies, with some even moving towards gender parity. Today all countries in Latin America, except for Guatemala and Venezuela, apply gender quotas to national legislative elections. Bolivia, with a 50/50 approach, is the regional leader, with 53% of its parliament made up of women, followed by Mexico, with 43%, where a gender parity system also applies. At the lower end stands Brazil, where, despite having a quota of 30% has less than 10% of women in its Chamber of Deputies.

For readers who are familiar with the role quotas have had in increasing the presence of women in decision-making positions in different countries, yet unaware of the “peculiarities” of the Brazilian political culture and system, it may seem puzzling that Brazil has less than 10% of women in its national legislature, and stands at the lower end of the spectrum in Latin America in that respect, given that it has had a gender quotas policy for over 20 years. This article will seek to explain this paradox in two ways: first, by discussing the implementation of that policy, the aspects which are needed to make it work, in the opinion of studies of the subject, and what Brazil lacks in that regard. It will become clear that up to a very recent time, Brazil had a very weak quota policy. In fact, Brazil only implemented the quotas for its national and sub-national chambers in the elections of 2014. However, despite a considerable increase in the number of women candidates, the percentage who were elected remained the same. This suggests that other factors are more crucial than the percentage of female candidates in determining the number of women elected in the law-making bodies of Brazil. Second, this article will explain how the political system works in Brazil, emphasizing singularities of its electoral system and in, particular, how the laws on campaign finance and the behavior of donors have limited the possibilities of women candidates.

This article will argue that the political engineering of Brazil, in particular its electoral system and rules on campaign finance is key in explaining why the presence of women in political decision-making positions is lower even than that of countries without quotas. Proportional Representation (PR) with preferential voting, together with unrestrained campaign finance have resulted in candidates running in elections with significantly unequal amounts of financial resources, negatively affecting the electoral performance of women.

Data analyzed in the article come mainly from national and international databases of the Brazilian Superior Electoral Tribunal (STF), International Parliamentary Union (IPU), and the International Institute for Democracy and Electoral Assistance (International IDEA), and are complemented by secondary sources, like articles and books.

1. Quotas and Women’s Participation in Representative Positions

One of the greatest defects of democracies and representative systems around the world has to do with the participation of women. Although the discrimination against women is no longer explained (at least not openly) as a natural phenomenon, justified by an innate inability for rational thinking that makes them unfit for public life, as used to be the case among traditional political thinkers,² the small percentage of women in government today is an indication of intentional or involuntary obstacles that require deliberate action if they are to be overcome. The realization that this imbalance in political power is not natural or “inborn” has led to a search for ways to strengthen women’s participation in legislatures. Among these, gender quotas have been the mechanism that is most widely used.

On a world level, the average percentage of women members of parliaments (both houses combined) is 23.6%. There are no significant differences between percentages in Lower and Upper Houses -23.6% and 23% respectively (Inter-parliamentary Union 2017a).³ Although small, this represents a significant increase in comparison to twenty years ago when it was only 11.7%. While in December 1997 only 5 countries (4 of them in Scandinavia) had percentages above 30% (Inter-parliamentary Union 2017b),⁴ today there are 50, and 12 exceed the 40% benchmark (Inter-parliamentary Union 2017a). This significant increase has not been due to natural evolution, but is largely the consequence of specific electoral reforms and, in particular, the implementation of gender quotas for legislative positions which, in accordance with a country´s law or constitution, either stipulate a minimum percentage of women in the legislature, or, more commonly, a minimum and a maximum percentage of men and women on a party’s list of candidates.

In recent decades, such quotas have gained legitimacy and become a viable electoral strategy around the world, particularly after the Fourth World Women’s Conference in Beijing, in 1995. The Beijing Platform for Action recommended that governments, political parties and multilateral organizations take affirmative actions to ensure a fairer balance between genders and give a stronger voice to women in politics, which, in turn, encouraged women activists to pressure the political parties and parliaments of their respective countries to adopt that policy. Since then, the number of countries adopting this measure has significantly grown and today there are 128 countries that implement some sort of gender quotas in legislative elections (International IDEA 2017).⁵

In Latin America, Argentina took the lead in 1991 by adopting a 30% quota for its Chamber of Deputies.⁶ Costa Rica followed suit in 1996 by adopting a 40% quota (which was later increased to 50%). Afterwards, Bolivia, Brazil, the Dominican Republic, Ecuador, Panama, Peru and Venezuela⁷ approved quotas in 1997, followed by Honduras and Mexico in 2000 (Schwindt-^{Bayer 2010}). Today, only Guatemala and Venezuela do not use such measures. The percentage range from 20% in Paraguay to 50% in Bolivia, Costa Rica, Ecuador, Mexico, Nicaragua and Panama. While the norm was 30% during the 1990s and 2000s, the latest tendency has been gender parity.

Quotas have been a key to the significant rise in the numbers of women elected to the national parliaments of Latin America recently. At 28.6% (data from December 2017), today’s regional average is still far from the target of parity. Yet, comparatively, it represents an important step forward compared to the situation 20 years ago, when fewer countries had such quotas and the average was only 12.9% (Inter-parliamentary Union 2017b). They have not been uniform across the region and some countries, like Brazil, continue to lag behind, while others have steadily progressed and even stand out on an international level: Bolivia, for example, where 53.1% of the members of its national parliament are women, is the second in the world to have achieved gender parity. In others, such as Nicaragua and Mexico, the figure is more than 40%, while in Argentina, Costa Rica, and Ecuador, it is more than 30% (Inter-parliamentary Union 2017a). See table 3.1 for further details.

Common sense may led us to assume a rough correlation between women’s candidacies and the number of women in a parliament. In other words, the higher the quota, the more the women who are elected. However, this is not necessarily the case, as a number of other factors besides the size of quotas influence electoral outcomes. The specialized literature considers that besides its size, other factors directly related to this policy, and that interfere with its outcomes are enforcement mechanisms and placement mandates (^{Dahlerup and Freidenvall 2005}; ^{Htun and Jones 2002}; Jones 2004; Krook 2009; ^{Matland 2006}; Schwindt-Bayer 2015, 2010).

The size of the quota is obviously important, since, by forcing political parties to increase the number of women’s candidates, the aim is to ensure that more women are elected. However, there are different strategies which parties can use to avoid compliance with quotas that, if successful, will jeopardize its main purpose. In that regard, legal safeguards are needed to secure that objective. Another is a placement mandate, which requires parties to rank male and female candidates on their list in accordance with the quotas. What is the point of a high quota percentage if parties then nominate women for seats they are unlikely to win? Without the mandate, women are likely to be lower down on the lists with a smaller chance of being elected, as often happened in Latin America.

Brazil is a good example of how a weak regulation of political parties may make the quota ineffective. Since 1997, Brazilian law stipulates that the lists of parties and coalitions should have a minimum of 30% and a maximum of 70% of either men or women⁸ for elections on both a local and national level. However, the parties did not start to follow this rule in local elections until 2012 and in national ones until 2014. Loopholes in the legislation and the absence of meaningful sanctions for its violation led to widespread non-compliance.

Quotas were first implemented in Brazil under Law 9.100/95, in 1995. This was a provisional measure, only valid for candidacies for the municipal council elections of 1996 and it stipulated that at least 20% of the candidates should be women, but all parties ignored it and did not suffer any sanctions. In 1997, Law 9.504/97 increased the quotas and extended them to state and federal elections, in which each party or coalition was legally obliged to “reserve” a minimum of 30% and a maximum of 70% for the candidacies of each sex. Once again, in the following elections (1998 and onwards), the parties ignored the spirit of the law and were not sanctioned. It is only recently that they have been forced to comply with the law.

Previously, the word “reserve”, instead of “fill” in the law gave them a loophole, because even though men were excluded from 30% of their candidacies, they were not obliged to “fill” those candidacies with women. The law also allowed for an increase in the total number up to twice the number of seats in the legislature.

In 2009, Law 12.034 modified Law 9.504/97. The third paragraph of article 10 stated that: "of the number of vacancies resulting from the rules set out in this law, each party or coalition shall fill their lists with a minimum of 30% and a maximum of 70% of candidacies of each sex.”⁹ It also stipulated that a failure to comply with this rule would lead to a ban on the registration of their slates. Thus, the parties had no option but to implement the quotas. As well as making it impossible to avoid the percentage of women specified by the law, changing “reserve” to “fill” strengthened the quotas of women and thus, this policy became effective for the first time in the municipal elections of 2012 and, later, in the 2014 state and national elections, that is, 15 and 17 years, respectively, after the adoption of gender quotas. The table below provides data on four national and state elections (two of them prior to changes in the quota law).

The data in Table 1 show that compliance with the quotas in 2014 led to a significant rise in the percentage of women chosen as candidates for state and federal deputies. However, that did not result in more women being elected, because, of the 30% of women candidates for state deputy, only 11.3% were elected and of the 29.4% for federal deputy, only 9.9% were.

In contrast with others countries in the region, quotas have not been an effective mechanism for getting more women elected to representative bodies in Brazil. In fact, if we compare the percentage of women candidates elected in the last elections with the percentage of previous ones, we cannot help concluding that things have got worse over time. When we look at the odds ratio -a statistical measure that compares the number of candidates to the numbers who are elected- we see that, for state deputies, it fell from 0.84 in 2002 to 0.82 in 2006, to 0.62 in 2010, and to 0.38 in 2014. For federal deputies, it fell from 0.71 in 2002 to 0.68 in 2006 to 0.46 in 2010 and to a meager 0.34 in 2014. This means that in the last elections the women candidates for federal deputies had 1/3 of a chance of being elected, compared to the men. In that regard, a parity between women and men with regard to the chances of election (odds ratio equal or superior to 1) was only achieved in the period before the adoption of quotas, when the percentage of women in parliament was only 6.6% (^{Sacchet 2011}).

The steady fall in the odds ratio between 2006 and 2014 indicates that the political representation of women has worsened over time. Thus, contrary to what common sense may tell us and international and regional trends, the chances of women being elected Brazil for women are shrinking rather than expanding with the quotas. Although they are important for other purposes (^{Sacchet 2008}), the quotas in Brazil have not help increase the participation of women in parliamentary politics, which suggests that the real problem lies elsewhere. As studies of the subject have indicated, when assessing the effectiveness of such quotas we have to take into account not only the three key aspects of the quotas themselves -size (the stipulated percentage), a placement mandate (or lack of one) and a weak or strong enforcement of the law- but also the institutional framework in which they are implemented.

That is, the success of quotas for women in Latin America has varied in accordance with institutional factors which are not always related to the laws on quotas themselves. One of them is the electoral system (^{Norris 2006}; ^{Rule 1987}; Schwindt-Bayer 2015; among many others) and another, which is less discussed, the rules for financing campaigns. A candidate needs money to win an election, but such funding is not always available to women, at least in amounts that would level the playing field.

2. The Electoral System and Women’s Representation

The electoral system is a key variable when analyzing electoral processes. Norris claims that it “represents, perhaps, the most powerful instrument available for institutional engineering, with far-reaching consequences for party systems, the composition of legislatures, and the durability of democratic arrangements” (2004, 209). Among other functions, the electoral system sets the rules that organize elections and defines how the votes won by parties, coalitions and candidates will be converted into seats. It thus strongly influences the behavior of parties and electoral outcomes.

Majority, proportional or mixed electoral systems have specificities which, in different ways, influence the electoral arrangements and strategies followed by political parties, and shaping of political representation. The Proportional Representation System (PR) is able to secure a more pluralistic composition of political legislatures, because it roughly turns winning votes won into parliamentary seats. This is in contrast to the majority system, in which the winner takes all seats. PR, therefore, allows for a better distribution of parliamentary seats among different political groups, parties or coalitions.

In terms of strengthening the representation of women and minority groups, PR is considered preferable because it creates incentives for parties to present a more balanced ticket. PR involves larger district magnitudes, which lead to higher party magnitudes, or a greater number of seats available for parties in a parliament, which, in turn, affects their strategy for choosing candidates (^{Norris 2006}; Matland 1998; ^{Rule 1987}).¹⁰ More seats weaken the competition for candidacies and increases the chance of women.

In addition, it is in the interest of the parties’ democratic credentials to put forward a more diverse electoral list, composed of candidates from different backgrounds, since it will potentially attract votes from wider sectors of the population. In the single member districts used in majority systems, the parties tend to go for “safer” candidates, who are usually neither women nor members of minority groups. Therefore, in theory, the countries whose elections follow the majority system should have fewer women in their legislatures. Since the mixed system combines elements of both systems, the countries that use it should have fewer women than the ones which use PR, but more than those with majority systems.

Academic studies have confirmed a propensity for PR to return more seats for women and minority groups than majority or mixed systems (Lijphart 1994; Mansbridge 1999; Matland 1998; Matland and Studlar 1996; ^{Norris 2006}, 2004; ^{Reynolds 1999}; ^{Moser 2001}). Norris’ study (2005) showed that women had almost double the chances of winning seats under a PR rather than a majority system: while they occupied 10.5% of seats in the parliaments of countries with majority systems, they had 19.6% in PR systems. In countries with mixed systems, women accounted for 13.6% of the members of parliament (Norris 2006).

Recent data I gathered from the Inter-parliamentary Union database, shown in graph 1 below, indicate that the differences remain the same today, even if the extent of the difference is no longer so marked. Countries with PR have an average of 26%, followed by those with mixed systems, 24%, and lastly by those with majority systems, 16%.¹¹

Source: Author’s analysis based on the Inter-parliamentary Union database, 2017.

Graph 1 Women’s Membership of National Parliaments around the World by Electoral System (%) 

Studies of Latin America have confirmed that the electoral system plays a central role in the successful candidacies of women and, in turn, is in agreement with studies on a world level. However, they have found that not all PR systems are equally women-friendly. In terms of electing women, PR works best when it is linked to closed-candidate lists (^{Baldez 2004}; ^{Htun and Jones 2002}; Htun and Piscopo 2010; Jones 2009). Jones (2009) reported a considerable variation in the electoral results of women in Latin American countries which used PR. In countries that used quotas and had a closed list PR system, women won 30% of the seats. In countries with quotas but an open-list PR system, the figure was 22%. In countries without quotas and an open-list PR system, it was 13%; and lastly, in countries without quotas and a closed-list PR system, it was 10%. This suggests that in order to understand the electoral performance of women it is necessary to look beyond the type of electoral system, and examine the model of electoral lists.

At the current time (see table 3), 14 out 18 countries in Latin America use PR, and of these only one does not employ gender quotas, yet there are considerable differences in the percentage of women in their parliaments. Therefore, they provide good case studies of the way in which the electoral system affects the election of women to parliaments.

3. Women in Latin American Parliaments: Quotas and the Electoral System

Table 2 below provides an up to date snapshot of countries in Latin America, in relation to the key variables (mentioned above) for the electoral performance of women. The data include: the number of women in legislatures (of the lower house, or unicameral system only); the type of electoral system and type of lists it operates; the percentage of quotas (if any); whether or not there is a placement mandate for women and men; and finally, the existence of a weak or strong enforcement of quotas. Because the cases are relatively few (only 18), the statistical analysis is limited to establishing which of these variables are the most influential. However, the data lends itself to a useful descriptive comparative analysis of the current state of affairs.

Of the 18 countries in Table 2, nine employ a closed-list PR, five use an open-list PR, and four have mixed electoral systems. Only two -Guatemala and Venezuela- do not employ gender quotas. A straightforward comparative analysis on the percentages of women in Lower House positions in these countries -again, within the abovementioned statistical limits- allows us to make some inferences concerning the role of quotas and the electoral system. First, the central role played by quotas is evident. As seen in graph 2, the average percentage of female representatives in lower houses in countries that employ gender quotas is 30%, while in those without it, it is 17%. Second, of the countries that employ quotas, the highest average percentage, 38%, is found in those with mixed electoral systems, followed by countries with closed list PR, with 30%, and last those with open list PR, with 23%, as shown in graph 3.

Source: Database: Inter-parliamentary Union (2017a). Data analyzed and organized in author’s graph.

Graph 2 Countries with and without quotas in Latin America (%)¹²  

Source: Database: Inter-parliamentary Union data (2017a). Data analyzed and organized in graph by the author

Graph 3 . Countries with quotas in Latin America and Electoral System (%)¹³  

The Countries with mixed electoral systems stand out as they have the highest average percentage of women in elected positions. Indeed, Bolivia, with 53%, has the highest percentage in Latin America and the second highest in the world. The leading position of countries with mixed electoral systems counters findings from previous studies, which showed that the closed list PR system was more likely to elect more women. However, a closer look at individual cases may help to clarify the contradiction, for while Bolivia and Mexico stand out, at 53% and 43% respectively, the figure for Panama is only 18%. Interestingly, all of these countries have recently adopted a gender parity system for legislative candidates (Bolivia and Panama in 2014 and Mexico in 2017). Yet the results in Panama have been far less satisfactory than in the other two.

As seen earlier, the mixed electoral system combines the features of both the PR and Majority systems; thus, it uses through both multi-member and single-member districts. Conventional understanding suggests that fewer electoral positions should favor male candidates. Therefore, it is reasonable to expect that countries with mixed electoral systems will elect less women than countries with PR, where all the districts are multi-member. Following the same rationale, gender quotas should work less effectively in countries with mixed systems, for it is unlikely that political parties will apply this policy in single member districts, given their expected preference for male candidates. In this case, compliance with quotas is only likely to succeed if the electoral rules ensure that the quotas will also apply to single member districts. This is what happens in Bolivia. Not only does Bolivia have a very strong quota legislation, that is, one backed up by the alternation of men and women in parties’ slates and by strong enforcement mechanisms, but it also applies gender parity to single-member districts. In fact, parity in participation between women and men is secured by the constitution in relation to all positions of authority be it of the legislative, executive or judiciary power (^{Venturini and Villela 2016}). On the other hand, the quota regulations and enforcement mechanisms in Panama are weak. Furthermore, the law only covers the nomination lists in the primaries of parties (International IDEA 2017), which means that a balanced ticket is uncertain even in multi-member districts, and the policy is far less effective in Panama than in Bolivia.¹⁴

Differences in the electoral results of countries that apply quotas in closed list PR systems are also common. Nicaragua stands out, with 46% of parliamentarians being women and a gender parity system for selecting candidates (50/50); in Argentina 39% of the members of parliament are women and the quota is 30%; Costa Rica follows, with 35% of its parliament being women and a quota of 50%; for El Salvador, 32% of the members are women and the quota is 30%; the Dominican Republic and Uruguay both have quotas of 33%, with 27% and 20% of legislators being women, respectively; in Honduras the percentage is 26% and the quota is 40%; and finally, Paraguay comes last, with 14% and a 20% quota.

What these percentages show is that an analysis of women’s performance in close-list PR systems is not as straightforward as might be expected, for although results in some countries confirm the findings of most studies of the subject so far, a significant number do not. Nicaragua, for example, which has the highest percentage of women in parliament among these countries, does not have any mechanisms for enforcing quotas (International IDEA 2017), so compliance with this policy is dependent on the goodwill of its political parties (which, of course, is influenced by the bargaining power the women in them have [^{Sacchet 2008}]). El Salvador is another interesting case, with a quota of 30% and 33% of elected positions held by women, despite the fact that its electoral laws do not stipulate a placement mandate or impose strong sanctions on non-compliance (International IDEA 2017). Surprisingly, therefore, the willingness of the parties to comply with such quotas in those countries is higher than the average, which indicates that other factors may explain their behavior and require further analysis.¹⁵

By contrast, there are countries that have a high quota percentage and strong enforcement mechanisms, but do not have such satisfactory results. This is the case of Costa Rica, which has a strong and long-standing gender quota, which is now 50%, but the percentage of women in parliament is 35%. The same applies to Uruguay, which only has 20% of women in its lower chamber, despite having a strong quota of 33%. Given their electoral system and the strength of their quota legislation, a better electoral performance by women in these countries would be expected. Although previous studies have claimed that a system which combines a closed-list PR, a high quota percentage, a placement mandate and strong enforcement mechanisms is friendlier to women, the country cases analyzed in this article point to a more ambiguous situation.

On average, the countries with open-list PR have the lowest percentages of women in their legislatures, (23%), and except for Ecuador, they all have percentages smaller than 30%. At 38% Ecuador is one of the seven countries in the region with percentages above 30%, despite the fact that its electoral system is regarded as less women-friendly. Without Ecuador, the average percentage for countries with open-list PR would drop to less than 20%. Peru is another relatively successful case in this type of electoral system, while it has a comparatively low quota (30%), it has nearly the same average percentage (28%) of women in its unicameral legislature. Brazil, on the other hand, with a similar electoral system, stands out as the country with the lowest percentage. A closer look at the specific features of the electoral system and voter’s behavior in these countries will help us to understand these differences.

The National Assembly of Ecuador, its unicameral parliament, has 137 seats which are assigned in accordance with an open-list PR electoral system. Voters in Ecuador have two options in elections: a) to vote for one or many candidates -who can be from the same or from different political parties- in numbers proportional to seats available in the electoral district; b) to vote for the list of the party of their choice, instead of candidates. Since electoral lists in Ecuador follow a gender parity rule, and rank male and female candidates in alternate positions (^{Archenti and Tulla 2013}), 50% of the candidates chosen by voters who opt for the second alternative will be women. Since the second option is generally preferred by Ecuadorian voters, its electoral system works more like a closed-list PR one.

In Peru, the open-list PR electoral system has a different structure. It requires the electors to vote first for a list and then, if they wish, they can vote for up to two candidates on the same list (^{Schmidt 2006}). This has made its elections less focused on the candidates than Brazil’s. Schmidt argues that the option of voting for individual candidates in Peru is usually chosen by better-educated and more progressive voters, who have deliberately voted for women. The opportunity to vote for two candidates instead of one has also helped increase the number of women elected as, following campaigns led by women’s rights activists there, it has led voters to cast a vote for a man and another for a woman.

In Brazil, electors can vote either for a party or for a candidate. However, in contrast with Ecuador and Peru, they often prefer the latter option. Respecting the proportionality of votes won by parties and coalitions, candidates who individually manage to win the largest number of votes are those who will fill the seats in parliament. This mechanism for filling legislative seats, independent of any prior ranking order, has led to a high level of independency of candidates from their political parties and favored those who are well-known, like incumbents, or well financed (the two advantages are usually related). Women are in disadvantage concerning both these aspects (electoral finance and incumbency). They are also in disadvantage because the mechanism by which coalitions can run twice as much the number of candidates as the number of seats available weakens the impact of quotas. In effect, this has made women to be added to the lists as marginal candidates.

Therefore, the differences in the open-list PR systems of Ecuador, Peru and Brazil may explain the discrepancies between the electoral success of women in each country. Even excluding Ecuador and Peru, however, other countries whose systems are similar to that of Brazil and also have quotas, such as Chile (22.5%) and Colombia (19%), have significantly higher percentages of women in parliament, while Venezuela (22%), which lacks quotas, has more than double. A clearer understanding of this situation would require a more exhaustive analysis and data which is not always available. Although that is beyond the scope of this article, I nevertheless maintain that Brazil’s policy on campaign finance is a good starting point.

4. The Electoral System and Campaign Finance in Brazil: How Have They Affected Women’s Representation and What to Expect from Recent Changes in the Law?

Money is essential in today’s electoral campaigns. It allows candidates to make contact with the public so that they can explain their platforms and likewise allows the voters to make an informed decision about who they will vote for, an option which both sides have a right to in a democracy. In this sense, electoral funding is a central element for political equality, both from the perspective of the rights of voters and representatives.

A number of studies have shown a positive correlation between the money spent in campaigns and the number of votes won by candidates in Brazil (^{Britto 2009}; ^{Cervi 2010}; ^{Silva 2010}; ^{Mancuso 2012}; ^{Peixoto 2009}; ^{Samuels 2001}; ^{Sacchet and Speck 2012}a, 2012b; Speck 2005). This link between electoral spending and votes is obvious to former female candidates I interviewed, and who were very aware that the reason for this problem of uneven playing field due to differences in electoral spending. In a more general sense, the disparities in campaign funding are harmful for equality and democracy.

The costs of running a campaign in Brazil have risen significantly from one election to the next (Mancuso 2015; ^{Marenco 2008}; ^{Sacchet 2011}; Sacchet and speck 2012; ^{Santos 2011}), making Brazil amongst countries with the highest electoral spending (^{Samuels 2001}). The database of the Superior Electoral Tribunal shows that the total spending on the elections for federal deputy more than tripled between 2002 and 2014. In 2002 it was R$ 189,363,870.09 (Brazilian Reais), which (taking deflation into account) jumped to R$ 584,500,704.70 in 2014. In the elections for state deputies, the rise was similar. Meanwhile, the funding gap between male and female candidates has widened, as shown in table 3 and table 4 below.

Data in Table 3 and (table 4) show that the average amounts of funds raised by female candidates in elections from 2002 to 2014 was significantly inferior to that raised by male candidates, for both State and Federal Deputy Positions. It also evinces that the rate of difference has increased from one election to the next. While in 2002 the average percentage of funds raised by female candidates running for positions of federal deputy was equivalent to 51% of the amount raised by men in 2014 it represented only 23%. For the position of State deputy, the figures were equally dismaying. In 2002 funds raised by women represented 96% of funds raised by men, but in 2014 this difference increased significantly and women managed to raise only 40% of the amount raised by men.

Imbalance in terms of campaign resources has been propelled by both an electoral system that favors the individualization of campaigns, distancing candidates from parties, as considered earlier in this article, and by an over permissive legislation on campaign finance that incentivizes over-spending. National and state elections up to now have been regulated by a legislation that allowed for total freedom on campaign spending and the predominance of corporations among campaign donor. Recently, there were some changes in the law (that I will discuss below), that will apply to the October elections of 2018, which is expected to help curb overspending and the influence of corporations over elections.

Academic studies of the subject have long warned of the danger of parties turning into cartels with a uniform ideology, corporate-type structure and a growing distance from their bases as they obtain an easier access to state funds (^{Katz and Mair 1995}). Although it is premature to speak of the cartelization predicted by Katz and Mair (^{Ribeiro 2013}), the parties have indeed distanced themselves from their ideals and supporters, and are increasingly focused on political patronage. However, despite long-standing availability of public resources to parties by means of access to both official funds and free airtime for political advertisement in televisions and radios, the “cartelization” of parties in Brazil, if anything, has not derived from their access to public resources, but rather from their transactions with businesses.

Until 2015, businesses could donate up to 2% of the gross income earned in the year prior to the election, which, in the case of big corporations, is a considerable amount of money. In 2010, considering donations for all electoral positions, contributions from businesses represented 74% of the total (Mancuso 2015). This percentage is likely to have been higher in more recent elections. The predominance of corporations among campaign donors made support from businesses almost vital for electoral success and favored, in particular, well-connected incumbents. In a context of growing campaign expenditure encouraged by generous “donations” from the private sector, businesses have on the one hand increased their influence over electoral and political matters,¹⁶and on the other, been central in defining who wins and who loses as candidate

The second largest source of campaign funding are individual donations (Mancuso 2015). By stipulating that donations could not exceed 10% of an individual’s gross annual income, the law once again favored wealthy individuals, since 10% of a millionaire’s income is vastly more than that of one who earns the minimum wage. Thus, the regulations we just mentioned have favored the predominance of individual or corporate donors, and in turn, the election of representatives who advance their particular interests. Unsurprisingly, the reform of campaign finance has been a major issue in Brazilian politics.

They also mean that women candidates are underfunded, as shown in tables 4.1 and 4.2 above and the correlation tables below. A number of studies have shown that that equally applies to local elections (^{Sacchet 2013}, 2011; Sacchet and Speck 2012a, 2012b). The largest gap in campaign revenue is found in donation from the businesses source (Sacchet 2011; Sacchet and Speck 2012). Given the upper hand of corporations over electoral proceeds, this creates much disadvantage for women in terms of winning parliamentary seats, particularly if they are not incumbents. The unfavorable positioning of women as far as support from businesses is concerned is then exacerbated by the fact that they also get considerably less financial backing from individual donors, political parties and tend to invest far less of their own income in their campaigns (Sacchet 2011).

Women are also at a disadvantage because campaign finance is an even more important electoral resource for them. Research highlight that on average the amount of money needed by women to be elected is higher than that needed by men ^{(Sacchet and Speck 2012}a; Speck and Mancuso 2014). This suggests that money may compensate for both voters’ prejudice and disparity in political capital -given that, at a rate of 90% to 10% in parliament, men are the majority of political incumbents.

The correlation tables above corroborate my findings about the crucial role campaign finance plays in the lower chances women have of being elected in Brazil and the link between funding and male candidates. For male candidates for federal deputy, the correlation with campaign revenue of 12% is statistically significant at a level of 95%. It increases to 45% when campaign revenue is correlated to men who are elected -also at a significance level of 95%. For state deputy, the correlation is similar, although the coefficients of correlation are smaller.

Table 5 and (table 6) also highlight a positive correlation between the election of 2014 and female candidates both for federal and state deputies, which reflects a considerable rise in the number of female candidates that year (as shown in table 1.1). In the elections for federal deputy, there was a positive correlation of 14% between 2014 and women’s candidacies. However, the correlation also shows that despite the significant rise in the number of female candidates, there was a negative correlation, though not statistically significant, between the number of women elected and the year of the elections. The same applies to state deputies, but with a higher coefficient of female candidates (17%) and a negative and statistically significant correlation with the number of women elected that year.

Given the strong preference for voting for individual candidates, the high correlation between campaign finance and electoral success and the fact that female candidates raise significantly less money yet are more dependent on funding to win, it is not surprising that presence of women in the Brazilian parliament has been less than 10% for decades, despite significant increases in the number of female candidates in recent years. The individualization of campaigns, typical of an open-list PR system with the characteristics of Brazil’s, coupled with over-permissive rules on campaign financing have negatively affected the chances of women wining parliamentary seats.

Other countries in Latin America with open-list PR electoral systems seem to be less permissive about campaign spending and violations of the rules on campaign finance. The latest available data (from 2015) indicates that only in Chile were individual candidates allowed to receive funding from the private sector, and in all of them, including Chile, there were statutory ceilings on the money spent by candidates and political parties (International IDEA 2015).¹⁷ These differences in mechanisms used to limit the influence of financial capital on the electoral process may help explain why Brazil has the lowest percentage of women in legislative positions in the region, and even trails countries without gender quotas.

5. Recent Changes in Electoral Legislation and Prospects for the Future

In September 2015, the Federal Supreme Court (STF), the highest court of justice in Brazil, ruled that donations from corporations to both parties and candidates are unconstitutional. This ban led to a significant reduction in spending on the local elections of 2016. Then, in 2017, scandals about deals between businessmen and politicians in which there was a tradeoff between large donations and “benefits” from the government, like tax breaks and laws favoring private interests, led the public to pressure for further reforms, which, in turn, resulted in the congressional approval of a modest reform in 2017. After much negotiation within and between the Chamber of Deputies and the Senate and the final vetoes by the President, electoral legislation was amended. These changes will come into force during the 2018 elections.

Two of these changes directly concern the issues discussed in this article: ceilings on campaign expenditure and the establishment of a public campaign fund. The amount of the ceiling varies in accordance with the legislative or executive posts the candidates run for. The ceilings (including money from public funds) are as follows:

At first sight, the ceilings seem like a good way to make elections more democratic, since they seem to somewhat even electoral spending and curb the power of wealthy donors. In fact, they are far too high to fulfil that purpose. This is evident when we consider data from recent elections. Data from the last parliamentary elections, of 2014, analyzed for this article, show that of 5,869 candidates who run for the position of federal deputies, only 96 raised electoral funds in amounts superior to that of the ceiling set up recently for this position. From these candidates only seven, that is 7%, where female candidates. In relation to the position of state deputy from 15,254 candidates running, only 232 reached the ceiling of 1 million. Of these candidates just 28, or 12%, were women. Therefore, these ceiling are limited to sort out effectively the issue of inequality in disputes resulting from the imbalance in campaign’s revenues. .

The public campaign fund aims to fill the vacuum left by the ban on donations from corporations ordered by the STF in 2015. It amounts to R$ 1.7 billion and will be distributed according to electoral and political specificities. Given that money is a central constraining factor in elections for candidates from specific backgrounds -such as women, members of minorities and those running for smaller more ideological political parties¹⁸-, the creation of a public fund has the potential to provide an incentive for new actors to enter and eventually succeed in electoral disputes. Besides, the ban on donations from corporations might further help boost the election of candidates with these profiles, insofar as they are less likely to receive contribution from businesses.

For public funding to have a democratizing function, parties need to act more deliberately when distributing their resources among their candidates. As far as the low presence of women in legislative positions are concerned, parties could act more consciously in order to change todays state of affairs, allocating higher percentages of funds to women’s electoral campaigns. Otherwise, they should at least secure that funds allocated to women were similar to their percentage running as candidate. Since there is no proper mechanism in the legislation to force or provide incentives for parties to invest in women’s campaigns, as has recently been seen in other countries in the region such as in Chile (^{Casas-Zamora and Falguera 2016}), changes in these directions would have to rely on the goodwill of parties. However, given parties long-standing preferences for male candidates, it is unlikely they will freely adopt a perspective of equality in the distribution of their resources between male and female candidates, let alone to follow affirmative action whereby higher amounts of funds would be destined to women.

The congress has not reached an agreement on changes to the electoral system. For a number of years now, the parliament has shown an increasing interest in a reform of the current open-list PR system, in the form of a majority electoral system based on smaller electoral districts and with the aim of bringing representatives closer to the people they represent. Nevertheless, if approved, that reform is unlikely to benefit women. Given that the parties have failed to implement quotas for almost two decades in an electoral system based on multi-member districts, they are unlikely to support quotas in single member districts, where the selection of women candidates would exclude male ones.

Conclusions

This article has analyzed the political representation of women in Latin America, in order to explain why Brazil, despite implementing a 30% gender quota for nearly two decades, has less than 10% of women in its Chamber of Deputies. Following conventional knowledge, based on studies in the field, the analysis highlighted that quotas do not necessarily work as expected to increase women’s presence in representative positions. To a certain extent, it confirms the findings of studies which suggests that the degree to which this policy works depends on the one hand on factors related to the quotas policy itself -such as size, the existence or not of a placement mandate whether or not they are strongly enforced- and on the other, on the type of electoral system in place. Closed-list PR systems are considered to be more favorable to the election of women. However, this analysis reveals a murkier picture in Latin America today, since the percentage of women elected to the parliaments of some countries does not closely correspond to those findings, which indicates that other factors have to be taken into. I suggest that electoral finance is one of them.

Focusing on Brazil, it suggests that a combination of two factors is crucial for explaining the handicaps female candidates face. These are: the open-list PR electoral system and over-permissive rules on campaign spending. The open-list PR system may seem to make elections more democratic, since, ultimately, the voters decide whom they elect. However, by encouraging voting for individual candidates, it strengthens the influence of money on elections insofar as the candidates with more money have the advantage over those with less, especially women candidates, whose funding is significantly lower. Thus, even though the number of women candidates has steadily increased, reaching the 30% legal quota threshold in recent years, their chances of being elected have shrunk, as my analysis in this article has shown.

Electoral funding is vital for the running of campaigns and giving voters a knowledge of the positions of the candidates. Without it, candidates are less able to explain their stands and voters, less likely to hear them. Thus, in itself campaign funding is a democratic measure. However, freedom to spend on campaigns and large donations by individuals and, particularly, corporations, as it has been the case in Brazil, not only turn electoral disputes uneven, but makes it likely that some issues and interests get more publicity and are able to influence the political process more than others. In this sense, balance in electoral spending is a prerequisite for fairer campaigns and more democratic policy development. The imbalance in campaign finance particularly affects the representation of women and members of minority groups, as this article shows.

The recent legal reforms of campaign funding in Brazil are far from ideal. While the new ceilings may cut the usual spending of some candidates significantly, they are irrelevant for the overwhelming majority whose funding is far below the limits. However, so long as the new rules were strictly enforced, we were likely to see some changes in the composition, and supposedly the agendas, of the legislative branch. A curb on the political power of big corporations not only helps to make elections fairer, but reduces their influence on policy decisions. Since a willing compliance with electoral rules has not been a characteristic of Brazil’s parties and politicians, the success of this new legislation will largely depend on the State´s enforcement of the limits on campaign spending, especially in the form of secret slush funds which the private sector provides to a certain parties and candidates.

Will this reform be an incentive for getting more women elected? This is difficult to predict, since it will depend on the political actors’ willingness to fully comply with the new rules and, above all, on whether or not increasing the number of women elected becomes an issue of the interest of the parties and the legislature. Given specific constraints to female candidates in electoral processes highlighted in this article, increasing women’s participation in representative positions require purposive action. This means that the low presence of women in political positions must become a political issue in itself for parties and members of congress while advancing the political reform in course. Otherwise, the imbalance in numbers and political influence between men and women is likely to continue in the foreseeable future.