Electoral rules are highly effective in determining who winners and losers are

Electoral rules are highly effective in determining who winners and losers are. As the current electoral rules are highly likely to advantage incumbent parties/politicians, they would rather favour status quo instead of a change. For this reason, considering the uncertainties and the risk of unintended consequences electoral reforms bring, political parties need strong motivation to change the voting rules from voluntary to compulsory. This study is about party determinants of introduction of compulsory voting. Adapting a qualitative comparative analysis, the effects of party positions on political spectrums on their decisions in different political contexts are investigated. The results show that both the normative and consequential considerations of parties play an important role in explaining the variation in support for compulsory voting.

Compulsory voting (CV) is an institution that obligates eligible voters to attend to the polls or be liable to sanctions. Although this is a broadly accepted definition of CV in the literature, in fact, neither a physical attendance to the polls nor sanctions for non-voters is necessarily required. For the former, vote facilitation mechanisms such as postal or other forms of remote voting renders turning out possible without actually attending to the polls. For the latter, although such an obligation requires sanctions for non-compliance by definition, there are cases where citizens are legally obliged to vote, yet there are no sanctions for the non-voters. Another point this definition brings that needs to be raised is that: who are the eligible voters? In most countries where CV rules exist, there are groups of citizens who are excluded from the compulsion. The criteria for being an eligible citizen varies among countries under CV rules. Factors such as age, literacy, marital status, gender and citizenship may be leading factors that determine the eligibility for voting.

Compulsory voting is not a new concept. Robson (1923) traces the idea of obliging electors to attend the polls back to the citizen assemblies in Switzerland in the eighteenth century: the members were required to turn up in the specified way to make collective decisions (p.570). Birch (2009) reports that compulsory voting practice was first applied in the modern elections in 17th century in American colonies (p.20). In Europe, Switzerland is the first state which introduced compulsory voting: In the canton of St. Gallen, CV rules were implemented in 1835. Liechtenstein was the first state which made voting compulsory at the national level in 1878, followed by Belgium in 1892. Except few cases as such, in all of the Western democracies with mandatory voting rules, CV was instituted in the early twentieth century. Today, CV rules are wide spread across almost all continents, with a particularly high popularity in South America. There are currently more than 20 countries with compulsory voting rules applied to general and/or local elections at the country level; Austria and Switzerland for particular cantons or regions; and France for senate elections. Moreover, there have been other countries that have experienced CV laws at one period in their history, such as Italy, the Netherlands, and Spain. 

Why does Compulsory Voting matter?

Most electoral reforms create advantages for certain political groups. Likewise, compulsory voting rules may well have consequential effects for the electoral fortunes of political parties running for office by influencing electoral outcomes as they change the size of the electorate and the characteristics of the median voter. Below, how CV leads to these changes, and what they bring along with are explained.

In fact, the change in the size of the electorate is the primary consequence of CV. Many studies have agreed on that CV boosts turnout. Different studies find high levels of increase in turnout levels after the implementation of CV. Findings show that CV boots participation rates from 7 to 16 points on average (Lijphart 1997; Powell, 1980; Jackman, 1987; Franklin, 2001). Among Latin American countries, the estimated turnout boost associated with CV is roughly 11 to 17 percentage points (Fornos, 1996). Besides this, one can easily notice the rapid change in participation levels in the elections with the implementation of CV rules in within country comparisons. (For within country comparisons, see Hill & Louth, 2004:5; Reed, 1925: 355-6). Moreover, the effects of compulsory voting laws on voter turnout is bidirectional. In other words, we do not only witness a major increase in participation levels with the implementation of CV laws. In many cases, a remarkable decrease in voter turnout is observed after the removal of legal compulsion to vote (See Irwin, 1974:292-3).

In order to unravel how CV increases turnout, it is important to understand how the rational calculus of voting is affected by CV system. When rationalist approach explains human behaviour, it moves from ‘an economic point of view’ (Downs, 1957: 14). This utilitarian explanation states that a rational citizen always prefers more than less of goods, and always makes her decisions in order to achieve her political or economic goals. In other words, this model brings the cost-return principle of economics into politics and a citizen’s decision-making procedure; ‘if the returns outweigh the costs, he votes; if not, he abstains’ says the author (p. 260). The calculus of voting equation for individuals under CV rules is as follows:

R = pB - C + D + S
R: the reward, in utiles, that an individual voter receives from her act of voting 

B: the differential benefit, in utiles, that an individual voter receives from the success of her more preferred candidate over her less preferred one 

p: the probability that the citizen will, by voting, bring about the benefit 

C: the cost to the individual of the act of voting.

D: the satisfaction from compliance with the ethic of voting/duty.

S: the sanctions for not voting.

This equation has experienced modifications throughout the time: the D element was added to the classic rational choice model by Riker and Ordeshook in 1968, after some scholars raised the paradox of voting, saying that it is not rational to vote as costs of voting exceed the returns. Apart from Riker and Ordeshook (1968), there are other descriptions of the D term in the literature (See Feddersen, 2004; Olson, 1965). Under sanctioned compulsory voting rules, a new term ‘S’ is added to the equation, which stands for the sanctions. With this updated version of the calculus of voting, the logic is simple: The cost of voting is low (if not zero) already with the help of vote facilitation mechanisms, such as automatic registration, proxy voting, holding elections on a rest day and/or more than one day, absent and/or early voting, covering travel expenses of voters. With the positive effect of fulfilment of the civic duty, and the negative effect of sanctions for abstention, the cost of not voting exceeds the cost of voting. In a way, the negative motivation sanctions create on individuals results in high participation rates in the elections.
When there is an increase in the size of the electorate, the composition of the voters also changes. This is because when voluntary, those who vote and who do not are highly influenced by the socio/economic status individuals have. More precisely, studies have shown that there is a clear inequality in political participation between different segments of the society: those with high levels of resources such as education, income etc. are more likely to participate in the political arena than others. Moreover, the declining trends in political participation levels is primarily centred on more disadvantaged groups of the society, which broadens the participation gap. As CV convinces all segments of the society to turn up, it leads to a more equal participation (Lijphart, 1997). As a result, CV eliminates the participation gap between different socio/economic groups and so that homogenize the pool of voters.
The preferences of the core-voters who are highly likely to go to the polls in most elections, and peripheral-voters who mostly abstain when voluntary are not distributed randomly. Therefore, when CV brings the natural abstainers to the polls as well as the core voters, the preferences of the median voter is expected to change. As different parties appeal to and are supported by different groups of voters; compelling the natural abstainers to the polls would not affect the vote share of the all parties evenly. Instead, parties will derive different benefits from introducing compulsory voting. Given this, political parties as the institutions that hold the power to set the rules of the game have different views on CV.

Prior research

Despite its popularity and importance in the politics, very little is know about the origins of CV. Scholars have been working on a wide range of consequences this institution, in which CV has mostly taken as an independent variable. Relatively little academic attention has been devoted to the factors that motivate political parties to implement CV.

In the earlier works, democratic considerations caused from low citizen participation were seen as the main determinants of the support for compulsory voting. In brief, because chronic low voter turnout undermines the principles of equal representation of all segments of the society, equal consideration of interests, and the legitimacy of the election results, these studies claim, incumbent parties get motivated to implement CV and guarantee higher participation rates in the elections. This approach considers the motivation behind the implementation of CV laws as the prevention of electoral abuse and strengthening democracy for a long time. Clearly, there are justifiable reasons for politicians to desire high participation levels in the elections with consideration of democratic values.  Yet, given the similar participation rates with voluntary voting rules in many countries and more importantly the deterministic effects of CV, it is not realistic to assume that CV is supported by political parties to simply increase turnout. As more contemporary research clarifies, self-interest models are efficient in explaining electoral changes; that is decision makers must find it in their interest to alter the rules they were elected under (Cox, 1997:18).

More recent studies adapt a strategic behaviour approach when interpreting the views political parties hold on CV. This approach moves from Downs (1957) and his followers who take political parties as vote maximisers that would adapt the electoral system to create electoral advantages for themselves. Under the rational choice model umbrella, there is a clear agreement in the literature about which parties benefit most from CV rules: the left wing parties (i.e. Hicks & Swank, 1992; Castles & McKinlay, 1979; Bechtel et. al., 2015; Jensen & Spoon, 2011). The reason for these claims can be explained as follows. As mentioned earlier, when voluntary, those with high level of resources are more likely to participate in the political arena than others are. Husted and Kenny (1997) show that introducing sanctions for nonvoting places a disproportionate burden on socio-economically lower-level citizens. Therefore, CV is believed to lead to a demographically increased equality in participation by compelling all eligible electorates –including the natural abstainers- to vote (for how CV leads to more equal participation, see Gellogo, 2010 - of education-; Hooghe & Pelleriaux, 1998 –of socio-economic status-; Singh, 2010 –of age-). Since these citizens with lower levels of resources are expected to support pro-redistribution left-wing political parties (Anderson & Beramendi, 2012; Pontusson & Rueda, 2010), under CV, the median voter’s preferred level of redistribution will increase. As a result, higher electoral participation favours left - wing, redistributive parties. In light of this, the literature concludes, the left wing parties are more likely to push for and ultimately introduce compulsory voting rules.

However, contrary to the conventional wisdom, in almost all the instances of supportive arguments for compulsory voting, the centrist and conservative political parties have been the driving force. Unfortunately, to my knowledge, there is no up to date study or any other source of data on which parties implemented CV in each country. Yet, the leading work by Abraham (1955) shows that in the vast majority of countries, CV laws were implemented under rightist political parties. Since most of the countries implemented CV rules in either the last quarter of the 19th century or the first half of the 20th century, this study can still be helpful. One may think that it might simply be the case because the left-wing parties did not have the power – in this context, the necessary percentage of seats in the parliament – to pass the law and introduce CV. Yet, leftist parties were opposing the introduction of CV in many cases, i.e. Greece, Argentina. Moreover, there is no consistency in political parties’ views on CV with regards to their location on the right – left scale. For instance, in France there was leftist support for the introduction of CV for the senate elections; while in Argentina, right wing government implemented CV despite the opposition from the leftist parties, and in Australia, all effective parties supported the implementation of CV at the national level and in each state but Queensland. Moreover, a change in the attitudes of the same political parties towards CV is witnessed in some cases, i.e. the centre-right New Democracy Party in Greece decided to withdraw their support from CV years after introducing it. These experiences show that the most important determinant of a party’s position on CV cannot simply be explained by its location on the right-left scale regardless of the conditions under which CV was implemented.

Recently, Helmke and Meguid (2007) raised the question ‘why conservative parties adapt CV while today’s supporters of CV are on the left?’ in their prominent work ‘Endogenous Institutions: The Origins of Compulsory Voting Laws’. The authors investigate the strategic calculations of political parties. They argue that if the party believes that those who abstain under voluntary voting are mostly their natural base of voters, and thus mobilising them to the polls would increase this party’s vote share more than the opponent parties’, then the governing party would implement CV. As so, they become the first scholars –to my knowledge- to indicate that right-wing established parties benefitted from the CV in the context of franchise expansion, in contrast with the overwhelming agreement in the existing literature since Lijphart (1997). After Helmke and Meguid (2007), few other scholars drew attention on the same issue (i.e. Pilet, 2007; Malkopoulou, 2007, 2011; Maldonado, 2015). All these studies apply a strategic behaviour model, proposing that parties implement CV only if their perceived benefit is larger than their opponent’s. These attempts are successful in investigating the topic in a certain context, which is the expansion of franchise. They all reach the conclusion that right-wing parties supported and implemented CV with consequential considerations. This is actually not necessarily the case in all instances where CV laws have been implemented. In some cases right wing parties introduce it while in others left wing parties do and we do not know why. In short, the literature misses a comprehensive work on the issue, that is because the existing studies take all cases as a whole instead of differentiating the sets of cases that would benefit certain political parties.

One can assert that parties’ benefits from electoral rules depend on a set of conditions in the country i.e. political, economic, cultural etc. For this reason, parties’ decisions to introduce CV should be investigated in consideration of the political contexts in the country. For this reason, I start with clearly investigating the instances to see whether the political parties implement CV rules in similar sets of political conditions in line with their strategical calculations.

Also, the existing studies try to explain why certain parties support the introduction of CV while others oppose with their positions on the right-left spectrum. Dimensions are best regarded as ways to summarise how actors position themselves on major issues. Left-right differences form a central reference point that has been an important indicator of party policy on various issues. This dimension is concerned with economic issues such as redistribution, welfare, and government regulation of the economy. Hence, the right-left dimension constrains party positioning on policies regarding political regulation of the market, including social policy, employment policy, and cohesion policy (Hooghe, et. al, 2002; Thomassen & Schmitt, 1997). However, as some initial observations suggest, party characteristics regarding only the right-left dimension cannot successfully explain the variation in where political parties stand on issues like CV. This is because an actor’s interpretation of the act of voting, whether it is a duty for all eligible citizens or a right that cannot be forced to be practised, is about how the actor interprets personal freedoms and social rules. Therefore, in this study, the two-dimensional political space is introduced to determine the factors that explain where parties position themselves on CV.

In my work, I borrow the libertarian – authoritarian term from Kitschelt (1994, 1995) for the second dimension I take into account. Basically, authoritarian – libertarian dimension classifies parties in terms of their views on democratic freedoms and rights. To shortly define what being at the edges of this dimension means: In his work that he bases on the works of Flanagan (1987), Flanagan & Lee (2003) and Kitschelt (1994), Stubager (2008) writes that there is the concept of hierarchy and tolerance in the core of authoritarian-libertarian values. Common features of the libertarians are listed by Flanagan (1987) as follows: an emphasis on political and personal freedom, participation (more say in government, one's community), equality, tolerance of minorities and those holding different opinions, openness to new ideas and new life styles, environmental protection and concern over quality-of-life issues, self-indulgence and self-actualisation' (p.1304). On the other hand, Inglehart (1977) indicates that those place themselves on the other side of the dimension support law and order over personal freedoms. Flanagan adds that authoritarians emphasise respect for authority, discipline and dutifulness (1987:1305).

In fact, I argue that parties positions on this dimension is helpful for answering the very current question, as how authoritarian a political party is related to whether it sees individuals’ votes as a tool for maintaining the legitimacy and order, and how strict it is when it comes to limiting personal freedoms such as decisions on whether to vote or not. In fact, arguments for and against CV are mostly concentrated on whether it is compatible with liberal personal rights or not. Furthermore, parties’ positions on the authoritarian – libertarian scale is highly related to the openness of a state, which informs the chance individuals have to influence the political decision-making procedure of the governments.

For the aforementioned reasons, I discuss that parties positions on the authoritarian – libertarian dimension inform both consequential calculations and normative considerations they hold. Their positions on the economic right – left dimension, on the other hand, is rather expected to have tight links with its consequential considerations. My argument puts that the political conditions in the country determine whether certain parties support or oppose CV, as well as what sort of considerations influence their decisions to the greatest deal.

CV is wide spread around the world and many different countries have taken it into their political agendas from time to time. For achieving the main goal of this study, an inclusive comparison of political parties with their views on CV is required. Yet, as mentioned earlier, studies on the origins of CV are quite rare. Therefore, not surprisingly, the lack of information has been the main limitation I face in this study. Moving from this, I limit the number of countries I include in the comparison I make in this study. Because the cross-national data sets I am using are relatively small in size, a qualitative comparative analysis (QCA) pioneered by Charles C. Ragin for social sciences - is hold which is suitable for small to medium N studies to determine different contexts in which a cause (or causes) influences a certain outcome; this is which I aim to do. Comparing where parties stood on CV with regards to their locations on the political dimensions allows us to have a general understanding of the tendencies different parties have to support or oppose CV in each political contexts. With this method, the causal weights of the variables are measured.

I categorise the contexts in which CV rules were implemented in different countries as follows. The first and one of the most popular contexts is the expansion of franchise. In a number of countries, CV rules were implemented along with, or just before or after a new group of people were granted the right to vote. In most instances, the expansion of franchise occurred as a result of internal and/or external demands, which were almost impossible to resist after a certain point. As the governing parties in these countries were elected by the already existing electorate and they desired the status quo, these parties presented CV rules as a solution for a possible rapid change in the electoral outcomes with the change in the size and the characteristics of the electorate.

The second context that produced the implementation of CV in many countries occurs with the first general elections after military coups. As seen in the Greek case, the fear of mass abstention –mostly as a way to protest the military intervention in politics- is the main motivation for the governments to compel all the eligible citizens to vote in such a context. This is not surprising when the possible legitimacy gap caused by a really low participation rate is considered.

The last set of conditions I group as one of the contexts is the one that we know the least about. We can name the main motivation as a legitimacy gap, which happens when there is chronic low turnout. In other words, the governments in this group did not implement CV following a certain political event etc.; therefore it is harder to address the common features of these countries.

As mentioned earlier, the sources give inconsistent information about the exact date of the implementation of CV in many of these countries. Therefore, when selecting the cases for the qualitative comparative analysis, the criterion is the availability of the necessary information. I aim to include as many countries with CV system as possible from each political context in order to investigate what accounts for the difference the parties’ stands on CV.

The number of parties in a country, and the ideological difference between them is shaped by varying elements, including both supply side (such as political and legal structure) and demand side (such as political culture) factors. As such, what right or left wing means may differ from one country to another. Likewise, the dominant issues in the political agenda varies between countries, which inform the main issues that determine party positions on the second political dimension in a country. Therefore, in order to precent miscalculations or misinterpretations, data on party position on the political spectrums relies on country specialists.

To start with, when I conceptualise countries with or without CV, I argue that the choice of terminology used in the electoral laws matters. My conceptualisation of CV is as follows: countries with a statement of ‘voting is obligatory’ in the electoral laws or constitutions are considered to have a compulsory voting system. Whereas, those which state that ‘voting is a duty’ are countries with voluntary voting (VV) systems unless they create an obligation to vote as well, same as the countries who simply clarify the act of voting as a right. Maldonado applies the same approach in his work that focuses on Latin American countries with CV (2015). Based on my conceptualisation, the list of countries which have/have had implemented CV rules is given in Table 1. As the legal structure is taken as the main criterion, the legal foundations of the voting system are also given in the table, along with the year CV was implemented in. As this study focuses on countries with democratic elections, countries like Egypt and Lebanon are not included in the list below.

This study is still work in progress. Unfortunately, the Qualitative Comparative Analysis is not completed to the date of submission. I aim to finalise and submit the analysis by the end of this month.

Because the analysis is not complete yet, the results are subject to be released soon.

Clearly, as George and Bennett put it rightly, ‘correlation does not imply causation’ (2005:21). In order to explore the causal mechanism beyond the parties’ stance on CV, case study method needs to be adapted in the following steps. With this method, a more in-depth analysis of party stands can be done which could also explore the possible causal mechanisms behind the correlations. By doing this, ‘a basis for establishing modest empirical generalisations concerning historically defined categories of social phenomena’ will be provided (Ragin, 1987:31).

Citizen participation is a crucial element for democratic governance. In order to prevent the broadening participation gap between different groups in a society, CV laws were introduced as a remedy around the world. Besides increasing participation levels in the elections, CV rules also bring indirect effects by changing the composition of the electorate. Taking this into account, political parties as vote maximisers support the introduction of CV if their perceived benefit is higher than their opponents’.

In order to investigate what sort of considerations inform the stands political parties take on CV, their positions on both right-left economic dimension, and a more broad authoritarian-libertarian dimension are analysed, as well as the set of conditions the countries had in the time CV was implemented. The initial investigations show that the weight of different considerations of the parties are determined by the context. More clearly, in certain sets of conditions, their consequential considerations - as the change in their vote share in the elections - play an important role in their decisions; while in others, their normative considerations – as their ideology – is also relatively important. Yet, the analysis is subject to be completed.


References
Anderson, C. J. & Beramendi, P. (2012). Left Parties Poor Voters and Electoral Participation in Advanced Industrial Societies. Comparative Political Studies 45(6): 714-46.
Birch, S. (2009). Full participation: A comparative study of compulsory voting. Manchester: Manchester University Press.
Bowler, S., Donovan, T. & Karp, J. A. (2006). Why Politicians Like Electoral Institutions: Self-Interest, Values, or Ideology? The Journal of Politics 68(2): 434-46.
Castles, F. & McKinlay, R. D. (1979). Does politics matter: An analysis of the public welfare commitment in advanced democratic states. European Journal of Political Research 7: 169–86
Cox, G. (1997). Making Votes Count: Strategic Coordination in the World’s Electoral Systems. New York: Cambridge University Press.
Downs, A. (1957). An Economic Theory of Democracy. Harper & Row Publishers, New York.

Feddersen, T. J. (2004) Rational Choice Theory and the Paradox of Not Voting, Journal of Economic Perspectives 18(1): 99–112

Kitschelt, H. (1994). The transformation of European Social Democracy. New York: Cambridge University Press. 

Kitschelt, H. (with McGann, Anthony J.). (1995). The radical Right in Western Europe. Ann Arbor: University of Michigan Press.
Lijphart, A. (1997). Unequal Participation: Democracy's Unresolved Dilemma. American Political Science Review 9(1): 1-14.

Maldonado, A. (2015). The Origins and Consequences of Compulsory Voting in Latin America.