國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Therefore, the invalid voting in Latin America poses an empirical puzzle. To address this puzzle, this thesis aims to systematically examine the factors that impact voters’ propensity to cast invalid votes. Specifically, I will test hypotheses regarding compulsory voting system, concurrent elections, level of economic performance, level of education, level of democracy, and corruption on invalid voting rates. Based on data of presidential elections and lower chamber elections in 18 Latin American countries from 1916 to 2018, I conduct mixed-method analyses to analyze the impacts of different factors. I conduct quantitative analyses to show the relative importance of each variable on invalid vote. I also conduct qualitative analyses to illustrate how invalid voting campaigns work in Latin America. In the aggregate, invalid voting responds predictably to certain features of institutional, socioeconomic, and political such as compulsory voting system, concurrent elections, economic development, and corruption, in a different way, for the presidential and lower chamber election. And while I find evidence that a country with an enforced compulsory voting system and a high level of corruption affect more the increasing of invalid vote rates for lower chamber elections than presidential elections like it happened in Argentina 2001 and Ecuador 2006 elections. Concurrent elections tend to increase invalid vote only for presidential elections; and, a higher level of economic development tends to increase invalid vote just for lower chamber elections. Overall, this study aims to fulfill the gap in the literature and provide policy implications for voting behavior and democratization.
1.2 Why Study Invalid votes?
Generally speaking, there are two types of invalid vote. The first type is a ballot left in unmarked, which is often called “blank votes.” The second type is a ballot that is marked in a way that makes it impossible to identify the voter’s choice, or a ballot that has been torn or defaced, which is often called “null votes” or “spoiled votes.” Following previous literature (Cohen 2018;
Lysek et al. 2019; Power and Garand 2007; Uggla 2008), this thesis considers blank votes, null votes, and spoiled votes as invalid votes, and this thesis does not distinguish whether an invalid vote is cast intentionally or by error.
Voting is arguable the most remarkable feature of a democratic regime by which voters' political preferences manifest in a single act. Hence, understanding the factors that affect the
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
voter's choice to cast invalid votes is of crucial importance in the country's political life. Whether citizens cast invalid votes intentionally or accidentally, an extraordinarily high percentage of invalid votes might gravely undermine the well-functioning of the electoral system (Lysek et al.
2019) as well as democratic legitimacy (Lysek ep.al 2019; Power and Grand 2007). This issue is particularly crucial for new democracies as Latin America (Kouba and Lysek 2016).
Moreover, although invalid votes do not count in favor of a particular political party or candidate, they are not innocuous in tight electoral contests. For instance, in a highly competitive election, a small portion of null or blank votes could make a big difference in the electoral results, such as enhance the probability of holding a run-off election (Cohen 2018).
Another implication is related to the government performance. High levels of invalid voting rates might represent that voters are generally disenchanted with the candidates, political parties, or the incumbent government, which could lead to political instability. In this sense, invalid voting might not only serve as a mean of protest, but also a way to change the configuration of the electoral scenario. Therefore, high levels of invalid voting rates could push the ruling party to coordinate with other political actors and improve the government performance (Cohen 2016). In short, it is crucial to understand why some countries have higher invalid voting than others.
Why is it important to study invalid vote in Latin America? First, all Latin American countries regulate the electoral process through a special law which involves an explicit section to those categories of votes often be regarded as invalid. In Argentina, it is regulated by National Electoral Code - Decree N° 2135 (article 101), Bolivia adopted the Electoral System Law- Law N°
026 (article 161) while Brazil enacted the Electoral Code- Law N° 4.737 (article 224). Chile regulates the electoral process through the Law of Voting and Popular indexes- Law N° 18.700 (article 71). In Colombia, the Electoral Code - Decree N° 2241 (article 137) besides, in Costa Rica, the Electoral Code - Law N° 8765 (article 194), Dominican Republic, the Electoral Law- Law N°
275-97 (article 129), Ecuador is the Electoral Law- Law N° 59-86 (article 59) and, in El Salvador is the Electoral Code - Decree N° 413 (article 207). In Guatemala, it is ruled by the Electoral and
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Political Parties Law-Decree 1-85 (article 237), in Honduras, the Elections and Political Organizations Act - Decree N° 44-2004 (article 207).
For Mexico, the General Law of Electoral Institutions and Procedures published was on January 27, 2017 (article 288), while in Nicaragua, it is regulated by the Electoral Law- Law N° 331 (article 125). Likewise, Panama has the Electoral Code (article 268), Paraguay, the Electoral Code Nº 834/96 (articles 225 and 226), Peru is the Electoral Law- Law Nº 26859 (article 286), Uruguay with the Electoral Law- Law Nº 7812 (article 106) and Venezuela adopted the Organic Law for Electoral Processes in 2012 (article 106).
Second, as shown in the previous section, invalid voting is much higher in Latin America than in other regions of the world. However, compared to studies of invalid votes in other countries or regions (Aldashev and Mastrobuoni 2019; Hill and Young 2007; Martins 2017;
McAllister and Makkai 1993; Zulfikarpasic 2001), the issue of invalid votes in Latin America has received considerably less attention. While there are a number of single-country studies of invalid votes for Latin America (Arbache et al. 2015; Cisneros 2012; Driscoll and Nelson 2014; Power and Roberts 1995; Sanchez 2016; Sanchez Berzain 2019), there are few comparative analyses across Latin American countries (e.g., Cohen 2018; Power and Garand 2007). Unlike those studies mentioned above, this study is unique because it uses comprehensive electoral data in Latin America to examine determinants of invalid votes, and thus this thesis might facilitate a better understanding of what explains different levels of invalid votes across time and space.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Chapter Two: Theoretical Perspective
To explain why some countries, have a higher level of invalid vote while others do not, previous studies have focused on three approaches: institutional framework, socioeconomic characteristics, and political context (Martinez I Coma and Werner 2018; Power and Garand 2007). Each approach focuses on a number of variables, which will be discussed as follows.