2. Classifying Electoral Systems

Before we can examine the impact of rules on voting behavior and political representation, I first need to outline a typology of the main families of electoral systems and classify their subtypes. The most important institutions influencing electoral rules can be divided into three nested components, ranging from the most diffuse to the most specific levels.

 

● The constitutional structure represents the broadest institutional context, denoted, most importantly, by whether the executive is presidential or parliamentary, whether the national legislature is bicameral or unicameral, and whether power is centralized in unitary government or more widely dispersed through federal arrangements.

 

● The electoral system concerns multiple aspects of electoral law, and the most basic features involve the ballot structure, determining how voters can express their choices, the electoral threshold, or the minimum votes needed by a party to secure representation, the electoral formula, determining how votes are counted to allocate seats, and the district magnitude, referring to the number of seats per district. Electoral systems are categorized in this study into three primary families: majoritarian, combined, and proportional, each with many subsidiary types.

   

● Last, electoral procedures concern more detailed rules, codes of conduct, and official guidelines, including practical and technical issues that can also prove important to the outcome, such as the distribution of polling places, rules governing the nomination procedure for candidates, the qualifications for citizenship, facilities for voter registration and for casting a ballot, the design of the ballot paper, procedures for scrutiny of the election results, the use of compulsory voting, the process of boundary revisions, and regulations governing campaign finance and election broadcasting.

   

The constitutional structure is important obviously because it sets the institutional context for many aspects of political behavior, but systematic comparison of all these features is also well beyond the scope of this limited study.1 I focus instead upon classifying electoral systems used in all independent nation-states around the globe, to examine their distribution worldwide. In subsequent chapters I consider specific electoral procedures and legal rules in more detail, such as the use of statutory gender quotas on women’s representation or the impact of voting facilities on turnout. The way that electoral rules work is illustrated by examples from the countries under comparison in the CSES study. Electoral systems can be compared at every level of office – presidential, parliamentary, supranational and subnational – but to compare like with like, I focus here on national elections, including systems used for parliamentary elections for the lower house and for presidential contests.

 

 The Classification of Electoral Systems

 

Ever since the seminal work of Maurice Duverger (1954) and Douglas Rae (1967), a flourishing literature has classified the main types of electoral systems and sought to analyze their consequences.2 Any classification needs to strike a difficult balance, being detailed enough to reflect subtle and nuanced differences between systems, which can be almost infinitely varied, while also being sufficiently parsimonious and clear so as to distinguish the major types that are actually used around the globe. Worldwide, excluding dependent territories, we can compare the electoral systems for the lower house of parliament in 191 independent nation-states. Of these nations, seven authoritarian regimes currently lack a working, directly elected parliament, including Saudi Arabia, Brunei, and Libya. Electoral systems in the remaining countries are classified into three major families (see Figure 2.1), each including a number of subcategories: majoritarian formulae (including FirstPast-the-Post, Second Ballot, the Bloc Vote, Single Non-Transferable Vote, and the Alternative Vote systems);3 combined systems (incorporating both majoritarian and proportional formulae); and proportional formulae (including the Party Lists system as well as the Single Transferable Vote system).

 

 

The comparison in Figure 2.1 shows that in elections to the lower house, about half of all countries worldwide use majoritarian formulae, whereas one-third use proportional formulae and the remainder employ combined systems. As discussed earlier, electoral systems vary according to a number of key dimensions; the most important concern the electoral formula, ballot structure, effective threshold, district magnitude, malapportionment, assembly size, and the use of open/closed lists. Within the family of proportional systems, for example, in Israel the combination of a single national constituency and a low minimum vote threshold allows the election of far more parties than in Poland, which has a 7% threshold and small electoral districts. Moreover, electoral laws and administrative procedures, broadly defined, regulate campaigns in numerous ways beyond the basic electoral formulae, from the administration of voting facilities to the provision of political broadcasts, the rules of campaign funding, the drawing of constituency boundaries, the citizenship qualifications for the franchise, and the legal requirements for candidate nomination.

figure 2.1. Electoral systems used worldwide for the lower house of parliament, 1997. Notes: FPTP = First-Past-the-Post; 2nd Ballot; Bloc Vote; AV = Alternative Vote; SNTV = Single Non-Transferable Vote; STV = Single Transferable Vote. Systems were classified in May 1997 based on Appendix A in Andrew Reynolds and Ben Reilly. Eds. The International IDEA Handbook of Electoral System Design. Stockholm: International IDEA. For more details see http://www.aceproject.org/.

 

Majoritarian Formulae

 

Worldwide, in total, 91 out of 191 countries use majoritarian formulae in national election to the lower house of parliament. The aim of majoritarian electoral systems is to create a “natural” or a “manufactured” majority, that is, to produce an effective one-party government with a working parliamentary majority while simultaneously penalizing minor parties, especially those with spatially dispersed support. In “winner-takes-all” elections, the leading party boosts its legislative base, while the trailing parties get meager rewards. The design aims to concentrate legislative power in the hands of a single-party government, not to generate parliamentary representation of all minority views. This category of electoral systems can be subdivided into those where the winner needs to achieve a simple plurality of votes or those where they need to gain an absolute majority of votes (50%+).

 

 Plurality Elections

 

The system of First-Past-the-Post or single-member plurality elections is used for election to the lower chamber in fifty-four countries worldwide, including the United Kingdom, Canada, India, the United States, and many Commonwealth states. This is the oldest electoral system, dating back at least to the twelfth century, and it is also the simplest. Plurality electoral systems can also use multimember constituencies; for example, some dual-member seats persisted in Britain until 1948. As discussed later, the Bloc Vote continues to be employed in nine nations, such as Bermuda and Laos, using multimember districts with plurality thresholds. But today First-Past-the-Post elections for the lower house at Westminster are all based on single-member districts with candidate-ballots. The basic system of how FPTP works in parliamentary general elections is widely familiar: countries are divided into territorial single-member constituencies; voters within each constituency cast a single ballot (marked by an “X”) for one candidate (see Figure 2.2); the candidate with the largest share of the vote in each seat is elected; and, in turn, the party with the largest number of parliamentary seats forms the government. Under First-Past-the-Post candidates usually do not need to pass a minimum threshold of votes to be elected,4 nor do they require an absolute majority of votes to be elected. Instead, all they need is a simple plurality, that is, one more vote than their closest rivals. Hence, in seats where the vote splits almost equally three ways, the winning candidate may have only 35% of the vote, while the other contestants fail with 34% and 32%, respectively. Although two-thirds of all voters supported other candidates, the plurality of votes is decisive.

figure 2.2. An example of the First-Past-the-Post (FPTP) ballot in the U.K. general election

 

Under this system, the party share of parliamentary seats, not their share of the popular vote, counts for the formation of government. Government may also be elected without a plurality of votes, so long as it has a parliamentary majority. In 1951, for instance, the British Conservative Party was returned to power with a 16-seat majority in parliament based on 48.0% of the popular vote, although Labour won slightly more (48.8%) of the vote. In February 1974 the reverse pattern occurred: the Conservatives gained a slightly higher share of the national vote, but Labour won more seats and formed the government. Another example is the 2000 U.S. presidential contest, where, across the whole country, out of more than 100 million votes cast, the result gave Gore a lead of 357,852 in the popular vote, or 0.4%, but Bush beat Gore by 271 to 267 votes in the Electoral College. Moreover, under FPTP, governments are commonly returned without a majority of votes. No governing party in the United Kingdom has won as much as half the popular vote since 1935. For instance, in 1983 Mrs. Thatcher was returned with a landslide of seats, producing a substantial parliamentary majority of 144, yet with the support of less than a third of the total electorate (30.8%).

 

One of the best-known features of winner-takes-all elections is that they create high thresholds for minor parties with support that is spatially dispersed across many constituencies. In single-member seats, if the candidates standing for the minor parties frequently place second, third, or fourth, then even though these parties may obtain substantial support across the whole country, nevertheless, they will fail to win a share of seats that in any way reflects their share of the national vote. This characteristic is the basis of Maurice Duverger’s well-known assertion that the “simple-majority single ballot system favors the two party system” whereas “both the simplemajority system with second ballot and proportional representation favor multi-partyism.”5 As discussed fully in the next chapter, the accuracy of these claims has attracted much debate in the literature.6 One important qualification to these generalizations is the recognition that FPTP is based on territorial constituencies, and the geographical distribution of votes is critical to the outcome for minor parties and for minority social groups.7 Green parties, for example, which usually have shallow support spread evenly across multiple constituencies, do far less well under FPTP than do nationalist parties with support concentrated in a few areas. Hence, for example, in the 1993 Canadian elections the Progressive Conservatives won 16.1% of the vote but suffered a chronic meltdown that reduced their parliamentary representation to only 2 MPs. In contrast, the Bloc Québécois, concentrated in one region, won 18.1% of the vote but returned a solid phalanx of fiftyfour MPs. In the same election, the New Democratic Party won even fewer votes (6.6%), but they emerged with nine MPs, far more than the Progressive Conservatives.8 In a similar way, in the United States, ethnic groups with concentrated support, such as African-American or Latino voters in inner-city urban areas, can get more representatives into the U.S. Congress than do groups such as Korean-Americans, who are widely dispersed across multiple legislative districts.9

 

Malapportionment (producing constituencies containing differently sized electorates) and gerrymandering (the intentional drawing of electoral boundaries for partisan advantage) can both exacerbate partisan biases in constituency boundaries, but electoral geography is also a large part of the cause. Single-member constituencies usually contain roughly equal numbers of the electorate; for example, the United States is divided into 435 congressional districts, each including roughly equal populations, with one House representative per district. Boundaries are reviewed at periodic intervals, based on the census, to equalize the electorate. Yet the number of electors per constituency can vary substantially within nations, where boundary commissions take account of “natural” communities, where census information is incomplete or flawed, or where periodic boundary reviews fail to keep up with periods of rapid migration. There are also substantial differences crossnationally: India, for example, has 545 representatives for a population of 898 million, so that each MP serves about 1.6 million people. By contrast, Ireland has 166 members in the Dial for a population of 3.5 million, or one seat per 21,000 people. The geographic size of constituencies also varies a great deal within countries, from small, densely packed inner-city seats to sprawling and more remote rural areas.

 

The way that FPTP systems work in practice can be understood most clearly with illustrations from the elections compared in the CSES surveys, including the 1997 British general election, the 1997 Canadian election, and the 1996 United States presidential and congressional elections.  Although all Anglo-American democracies, important differences  in how these systems operate include variations in the number of parties contesting elections, the size of the legislatures, the number of electors per district, the dominant types of social cleavages in the electorate, the geographic distribution of voters, the regulations governing campaign finance and party election broadcasts, and the maximum number of years between elections.

 

The system of FPTP used for Westminster elections to the British House of Commons generally produces a manufactured “winner’s bonus,” exaggerating the proportion of seats won by the party in first place compared with their proportion of votes. For proponents of plurality elections, this bias is a virtue because it can guarantee a decisive outcome at Westminster, and a workable parliamentary majority, even in a close contest in the electorate.11

 

One simple and intuitive way to capture the size of the winner’s bonus produced by any electoral system is to divide the proportion of votes into the proportion of seats. A ratio of 1:1 would suggest no bias at all. But in contrast, the size of the bias in the winner’s bonus at Westminster has fluctuated over time but has also gradually risen since the 1950s until the 1997 election, when the winner’s bonus was the second highest ever recorded in the postwar era (only surpassed by the 2001 election). This phenomenon is the product of three factors: the geographical spread of party support in Britain, the effects of anti-Conservative tactical voting, and disparities in the size of constituency electorates.12 The 1997 British general election witnessed one of the most dramatic results in British postwar history, where eighteen years of Conservative government under Margaret Thatcher and then John Major were replaced by the Labour landslide of seats under the prime ministerial leadership of Tony Blair. The United Kingdom is divided into 659 single-member parliamentary constituencies where voters cast a single ballot and MPs are elected on a simple plurality of votes. At Westminster, the party share of parliamentary seats, not their share of the popular vote, counts for the formation of the government. Under FPTP, British governments are commonly elected with less than a majority of votes; in 1997 Tony Blair was returned with almost two-thirds of the House of Commons, and a massive parliamentary majority of 179 out of 659 seats, based on 43.3% of the U.K. vote. As the party in first place, Labour enjoyed a seats-to-votes ratio of 1.47 whereas in contrast, with 30.7% of the vote, the Conservatives gained only 25% of all seats, producing a seats-to-votes ratio of 0.81.

 

The U.S. system is also based on FPTP in single-member districts for multiple offices including congressional races for the House and Senate, and the system of the Electoral College used for presidential contests. The ballot paper presents the voter with more complex choices than in Britain due to multi-level elections, as shown by Figure 2.3, as well as due to the use of referenda and initiatives in many states and the sheer frequency of primary, congressional, and presidential elections. The winner’s bonus under majoritarian systems is also exemplified by the outcome of the 1996 U.S. presidential election pitting the incumbent, President Bill Clinton, against the Republican nominee, Senator Bob Dole; in this contest President Clinton was returned with 70.4% of the Electoral College vote, mainly by winning the largest states, but this substantial lead was based on only 50.1% of the popular vote across the whole country. In 1996 the congressional results for the 435-seat House of Representatives was highly proportional, however, because FPTP leads to proportional results in two-party systems when the vote totals of the two parties are fairly close. Roughly in accordance with the “cube” law, disproportionality increases as the vote totals diverge.

 

The 1997 Canadian federal election saw at least a partial consolidation of the multiparty system that had developed so dramatically with the emergence of two new parties, the Bloc Québécois and Reform, during the 1993 contest. The result of the 1997 Canadian election saw the return of the Liberals under the leadership of Jean Chrétien, although with a sharply reduced majority of only four seats and with 38% of the popular vote.13 The Bloc Québécois lost its status as the official opposition, dropping from fifty-four to forty-four seats after a sharp decline in support. By contrast, the Reform party moved into second place in the House of Commons, with sixty seats, although with its strongest base in the West. Both the Progressive Conservatives and the New Democratic Party improved their positions after their disastrous results in 1993. The level of proportionality in the Canadian system was similar to that found in the British general election, with the Liberal Party and the Bloc Québécois enjoying the highest votes-to-seats bonus, and both countries had far lower proportionality than the United States. The existence of a multiparty system within plurality elections could be expected to lead to stronger calls for electoral reform by moving toward a proportional or combined formula, but the regional basis of party competition allows minor parties to be elected to parliament despite the hurdles created by the Canadian electoral system.14

figure 2.3. An example of First-Past-the-Post (FPTP) ballot with multiple offices for the U.S. General Election: Sonoma County, California

 

The Single Non-Transferable Vote, the Cumulative Vote, the Limited Vote, and the Bloc Vote

 

Many other variants on the majoritarian formula are available. From 1948 to 1993, Japanese voters used the Single Non-Transferable Vote for the lower house of the Diet, where each citizen casts a single vote in small multimember districts. Multiple candidates from the same party compete with each other for support within each district. Those candidates with the highest vote totals (a simple plurality) are elected. Under these rules, parties need to consider how many candidates to nominate strategically in each district, and how to make sure that their supporters spread their votes across all their candidates. The system has been classified as “semi-proportional” (Reynolds and Reilly), or even “proportional” (Sartori), but it seems preferable to regard this as a variation of the majoritarian family because candidates need a simple plurality of votes in their districts to be elected, and there is no quota or requirement for proportionality across districts. The system continues to be employed for parliamentary elections in Jordan and Vanuatu, as well as for two-thirds of the legislators in the Taiwanese elections under comparison (see the “Combined Systems” section in this chapter).15 Other alternatives that fall within the majoritarian category, although not employed at national level for the lower house, include the Cumulative Vote, where citizens are given as many votes as representatives and where votes can be cumulated on a single candidate (used in dual-member seats in nineteenth-century Britain, where voters could “plump” both votes for one candidate, and in the state of Illinois in the United States until 1980). The Limited Vote system is similar, but citizens are given fewer votes than the number of members to be elected (used in elections to the Spanish Senate). The Bloc Vote system is similar to FPTP but with multimember districts. Each elector is given as many votes as there are seats to be filled, and they are usually free to vote for individual candidates regardless of party. The candidates winning a simple plurality of votes in each constituency win office. This system has been used for national parliamentary elections in nine countries, including Laos, Thailand, and Mauritius. Such contests allow citizens to prioritize particular candidates within parties, as well as maintain the link between representatives and local communities. On the other hand, where electors cast all their votes for a single party rather than distinguishing among candidates for different parties, this can exaggerate the disproportionality of the results and give an overwhelming parliamentary majority to the leading party.

 

Second Ballot Elections

   

Other systems use alternative mechanisms to ensure that the winning candidate gets an overall majority of votes. Second Ballot systems (also known as runoff elections) are used in two-dozen nations worldwide for election to the lower house. In these, any candidate obtaining an absolute majority of votes (50% or more) in the first round is declared elected. If no candidate reaches a majority in this stage of the process, a second round of elections is held between the two candidates with the highest share of the vote. The traditional way that this process is understood is that the first vote is regarded as largely expressive or sincere (voting with the heart), whereas the second is regarded as the more decisive ballot between the major contenders, where strategic considerations and alliances among left and right party blocs come into stronger play (voting “with the head”). In the countries under comparison, the Second Ballot system was employed for two-thirds of the seats in the Lithuanian combined system, as well as in seven of the presidential elections. Runoff elections are most common in presidential elections, but they are also used for elections to the lower house in France, in eleven ex-French colonies (including Chad, Haiti, Mali, and Gabon), in seven authoritarian ex-Soviet Eastern European states (such as Belarus, Kyrgyzstan, Moldova, Uzbekistan, Kazakhstan), and in some unreconstructed Communist states (Cuba and North Korea), as well as in the U.S. state of Louisiana. This system can be seen as encouraging centrist party competition, as well as bolstering the legitimacy of the eventual winner by ensuring that they receive the support of at least half the public. On the other hand, the rules harshly penalize minor parties, and the need for citizens to go to the polls on at least two occasions in rapid succession can induce voter fatigue, thereby depressing turnout. This phenomenon was exemplified by the May–June 2002 French elections where voters were called to the polls four times following nonconcurrent presidential and parliamentary elections.

   

Alternative Vote

 

figure 2.4. An example of the Alternative Vote ballot for the Australian House of Representatives

figure 2.4. An example of the Alternative Vote ballot for the Australian House of
Representatives

 

The Alternative Vote, used in elections to the Australian House of Representatives and in Ireland for presidential elections and by-elections, is also majoritarian. This system, or “preferential voting” as it is commonly known in Australia, was introduced for Australian federal elections in 1919 and is now employed in all states except Tasmania, which uses STV.16 Australia is divided into 148 single-member constituencies. Instead of a simple “X” on the ballot paper, voters rank their preferences among candidate (1,2,3 . . .) (see Figure 2.4). To win, candidates need an absolute majority of votes. Where no one candidate wins more than 50% after first preferences are counted, then the candidate with the least votes is eliminated, and his or her votes are redistributed among the other candidates. The process continues until an absolute majority is secured. In the 1996 Australian federal elections under comparison, for example, the victory of the conservative Liberal–National coalition ended the longest period of Labour Party government in Australia’s history. The contest saw an extremely close call on the first preferences, with the Australian Labour Party (ALP) and the Liberal Party receiving identical shares of the vote (38.7%). In the final preferences, however, the ALP won 46.4% compared with 53.6% for non-ALP candidates. As a result, the Liberal–National government won ninety-three seats, and a substantial majority, while Labour won only forty-nine.17 This process worked as intended by translating an extremely close result in the first preference vote into a decisive majority of parliamentary seats for the leading party elected to government. This process systematically discriminates against those parties and candidates at the bottom of the poll to promote single-party government for the winner. The Alternative Vote functions similarly in many ways as the Second Ballot system, with the important distinction being that there is no opportunity for citizens to re-vote or for parties to create new alliances, in the light of the outcome of the first preference ballots. The balloting and counting process is also more efficient, avoiding repeated trips to the polling station and possible reductions in turnout due to voter fatigue.

   

Proportional Representation Formulae

   

Adversarial democracies and majoritarian electoral systems emphasize popular control by the party in government. By contrast, consensus democracies and PR electoral systems focus on the inclusion of all voices, emphasizing the need for bargaining and compromise within parliament, government, and the policymaking process. The basic principle of PR is that parliamentary seats are allocated according to the proportion of votes cast for each party. The main variations concern the use of open or closed lists of candidates, the formula for translating votes into seats, the level of the electoral threshold, and the size of the district magnitude. The Party Lists system exemplifies the proportional formula but the Single Transferable Vote system should also be included in this category because it allocates seats based on quotas.

   

Party Lists Systems

   

Proportional electoral systems based on party lists in multimember constituencies are widespread throughout Western Europe. Worldwide, 62 out of 191 countries use party list PR (see Figure 2.1). Party lists may be open as in Norway, Finland, the Netherlands, and Italy, in which case voters can express preferences for particular candidates within the list. Or they may be closed, as in Israel, Portugal, Spain, and Germany, in which case voters can only select which party to support, and each party decides the ranking of their candidates on the list. The rank order of candidates on the party list determines who is elected to parliament. In Israel, all the country is one constituency divided into 120 seats; but, often, lists are regional, as in the Czech Republic, where 200 total members are elected from eight regional lists. Proportional Party Lists systems are used in fifteen of the countries under comparison in the CSES dataset.18 A typical party list ballot paper from South Africa is illustrated in Figure 2.5.

figure 2.5. An example of a Party List ballot for the South African Parliament

 

The electoral formula for the lower-house legislative elections varies among proportional systems (see Table 2.1). Votes can be allocated to seats based on the highest averages method. This requires the number of votes for each party to be divided successively by a series of divisors, and seats are allocated to parties that secure the highest resulting quotient, up to the total number of seats available. The most widely used is the d’Hondt formula, using divisors (such as 1,2,3, etc.), employed in Poland, Romania, Spain, and Israel. The “pure” Sainte-Laguë method, used in New Zealand, divides the votes with odd numbers (1,3,5,7, etc.). The “modified” Sainte-Laguë method replaces the first divisor by 1.4 but is otherwise identical to the pure version. An alternative is the largest remainder method, which uses a minimum quota, which can be calculated in a number of ways. In the simplest, the Hare quota, used in Denmark and Costa Rica, and, for the list constituencies in Taiwan, Ukraine, and Lithuania, the total number of valid votes in each constituency is divided by the total number of seats to be allocated. The Droop quota, used in South Africa, the Czech Republic, and Greece, raises the divisor by the number of seats plus one, producing a slightly less proportional result.

<Notes>: PR, Proportional Representation; FPTP, First-Past-the-Post; AV, Alternative Vote; SMD, Single-member Districts; List, Party List; ENPP, Effective Number of Parliamentary Parties; Prop., Proportionality. ENPP is calculated following the method of Laakso and Taagepera (1979). The Index of Proportionality is calculatedas the difference between a party’s share of the vote and its share of the total seats in parliament, summed, divided by two, and subtracted from 100. Theoretically,it can range from 0 to 100. This is a standardized version of the Loosemore-Hanby Index. For details see Rose, Munro, and Mackie (1998). The formal vote threshold is theminimum share of the vote (in the district or nation) required by law to qualify for a seat, and this is distinct from the informal threshold or the actual minimum shareof the vote required to win a seat. Note that the classification distinguishes between combined dependent systems, where the outcome depends upon the proportion of votescast in the party lists, and independent combined systems used in Japan, Russia, and Korea, where the single-member districts and party lists operate in parallel. It shouldbe noted that Belgium subsequently introduced a 5% formal vote threshold for the May 2003 general elections.

<Sources>: Voting Age Population: IDEA Voter Turnout from 1945 to 1997. Available online at www.idea.int; Successive volumes of Electoral Studies; Richard Rose, Neil Munro, and Tom Mackie. 1998. Elections in Central and Eastern Europe Since 1990. Strathclyde, U.K.: Center for the Study of Public Policy; Richard Rose. Ed. International Encyclopedia of Elections. Washington, D.C.: Congressional Quarterly Press 2000; http://www.aceproject.org/; Lawrence LeDuc, Richard G. Niemi, and PippaNorris. Eds. 2002. Comparing Democracies 2: New Challenges in the Study of Elections and Voting. London: Sage; Table 1.2; CSES Macro-Level Dataset.

  

Other important differences in countries under comparison within the PR category include the formal threshold that parties must pass to qualify for seats. It should be noted that the formal threshold set by statute or specified in constitutional requirements is distinct from the effective vote threshold, which is the actual minimum share of the vote that leads to gaining at least one seat. The formal threshold ranges from the lowest level of 0.67% of the national vote, used in the Netherlands, to up to 7% of the vote, used in Poland. Worldwide, one of the highest vote thresholds is in Turkey, with a 10% hurdle, whereas there is no formal threshold in some countries such as South Africa, where less than 0.25% of the national vote is necessary for election. The formal threshold can have an important impact upon proportionality and the opportunities for minor parties. District magnitude, or the mean number of seats per constituency, also varies substantially. In Israel, for example, all 120 members of the Knesset run in one nationwide constituency. By contrast, in Spain, the 350 members are elected in fifty list districts, each district electing 7 members on average. Generally, under PR systems, the larger the district magnitude the more proportional the outcome, and the lower the hurdles facing smaller parties.

 

Single Transferable Vote

 

The other alternative system in the proportional category is the STV, currently employed in legislative elections in Ireland, Malta, and for the Australian Senate.19 The system can be classified as proportional because of the use of the quota for election. Under this system, each country is divided into multimember constituencies that each have about four or five representatives. Parties put forward as many candidates as they think could win in each constituency. Voters rank their preferences among candidates in an ordinal fashion (1st, 2nd, 3rd, . . .). The total number of votes is counted, and then the number of seats divides this vote total in the constituency to produce a quota. To be elected, candidates must reach the minimum quota. When the first preferences are counted, if no candidates reach the quota, then the candidate with the least votes is eliminated, and his or her votes are redistributed according to second preferences. This process continues until all seats are filled. Proponents argue that by allowing citizens to identify a rank order for their preferences within parties, or by ballot-splitting their votes across different parties, STV provides greater freedom of choice than do other systems.20 Moreover, by retaining proportionality, these rules also generate a fair outcome in terms of the votes-to-seats ratio.

 

Combined Systems

   

An increasing number of countries, including Italy, New Zealand, and Russia, use combined systems, employing different electoral formulae in the same contest, although with a variety of alternative designs. In this regard, we follow Massicotte and Blais in classifying combined systems (otherwise known as mixed, hybrid, or side-by-side systems) according to their mechanics, not by their outcome.21 If we followed the latter strategy, such as defining or labeling electoral systems based on their levels of proportionality, then this approach could create circular arguments. There is an important distinction within this category, which is overlooked in some discussions, between combined-dependent systems, where both parts are interrelated, and combined-independent systems, where two electoral formulae operate in parallel toward each other.

   

Combined-Dependent Systems

   

Combined-dependent systems, exemplified by the German and New Zealand parliamentary elections, include both single-member and party-list constituencies, but the distribution of seats is proportional to the share of the vote cast in the party list. As a result, the outcome of combined-dependent systems is closer to the proportional than the majoritarian end of the spectrum, although the logic of voter choice in these systems means that they still remain different from pure PR. The best-known application is in Germany, where electors can each cast two votes (see Figure 2.6). Half the members of the Bundestag (328) are elected in single-member constituencies based on a simple plurality of votes. The remaining MPs are elected from closed party lists in each region (Land). Parties that receive less than a specified minimum threshold of list votes (5%) are not be entitled to any seats. The total number of seats that a party receives in Germany is based on the Niemeyer method, which ensures that seats are proportional to second votes cast for party lists. Smaller parties that received, say, 10% of the list vote, but that did not win any single-member seats outright, are topped up until they have 10% of all the seats in parliament. It is possible for a party to be allocated “surplus” seats when it wins more district seats in the single-member district vote than it is entitled to under the result of the list vote.

figure 2.6. An example of a Combined or Mixed-Member ballot used for the German Bundestag

 

New Zealand is also classified as a combined-dependent system because the outcome is proportional to the party-list share of the vote. The Mixed Member Proportion (MMP) system (as it is known in New Zealand) gives each elector two votes, one for the district candidate in single-member seats and one for the party list.22 As in Germany, the list PR seats compensate for any disproportionality produced by the single-member districts. In total, 65 of the 120 members of the House of Representatives are elected in singlemember constituencies based on a simple plurality of votes in single-member districts. The remainder is elected from closed national party lists. Parties receiving less than 5% of list votes fall below the minimal threshold to qualify for any seats. All other parties are allocated seats based on the Sainte-Laguë method, which ensures that the total allocation of seats is highly proportional to the share of votes cast for party lists. Smaller parties that received, say, 10% of the list vote but that did not win any single-member seats outright are topped up until they have 10% of all the seats in the House of Representatives. The 1996 New Zealand election saw the entry of six parties into parliament and produced a National–New Zealand First coalition government.

 

Combined-Independent Systems

 

Other electoral systems under comparison can be classified as combinedindependent systems, following the Massicotte and Blais distinction, with two electoral systems used in parallel, exemplified by the Ukraine and Taiwan.23 In these systems, the votes are counted separately in both types of seat so that the share of the vote cast for each party on the party lists is unrelated to the distribution of seats in the single-member districts. As a result, combined-independent systems are closer to the majoritarian than to the proportional end of the spectrum.

   

The March 1996 elections to the National Assembly in Taiwan exemplify this system. The Taiwanese National Assembly is composed of 334 seats, of which 234 are filled by the single non-transferable vote. Voters cast a single vote in one of fifty-eight multimember districts, each having 5 to 10 seats. The votes of all candidates belonging to the same party in all districts are aggregated into party votes, and the list PR seats are allocated among those parties meeting the 5% threshold. There are 80 PR list seats on a nationwide constituency and 20 PR list seats reserved for the overseas Chinese community. Taiwan has a three-party system, with the Nationalist Party (KMT) being dominant since 1945; the Democratic Progressive Party, founded in 1986, providing the main opposition; and the New Party, founded in 1993, having the smallest support. The major cleavage in Taiwanese party politics is the issue of national identity, dividing those who identify themselves as mainlanders who favor re-unification with China and those native Taiwanese who favor independence. The New Party is commonly considered most pro-unification and the Democratic Progressive Party the most pro-independence.24

 

The Ukrainian elections also illustrate how combined-independent systems work. The March 29, 1998 parliamentary contests were the second elections held since Ukrainian independence. Ukrainian voters could each cast two ballots. Half the deputies were elected by First-Past-the-Post in single-member districts, and others were elected from nationwide party lists, with a 4% threshold. Unlike the system in New Zealand and Germany, the two systems operated separately so that many smaller parties were elected from the single-member districts. The 1998 elections were contested by thirty parties and party blocs, although only ten of these groups could be said to have a clear programmatic profile and organizational base.25 The Ukrainian result produced both an extremely fragmented and unstable party system: eight parties were elected via party lists and seventeen won seats via the single-member districts, along with 116 Independents. The election produced the highest Effective Number of Parliamentary Parties (ENPP) (5.98) in the countries under comparison, and it also generated fairly disproportional votes-to-seats ratios that benefited the larger parties. Ethnicity was reflected in the appeal of particular parties, including the Russophile Social Liberal Union, the Party of Regional Revival, and the Soyuz (Union) Party, and also in the way that ethnic Russians were twice as likely to support the Communist Party as were ethnic Ukrainians.26

   

For the comparison of the consequences of electoral systems, such as the link between different types of formula and patterns of party competition or electoral turnout, in this study I compare the broadest range of countries worldwide that is possible from available sources of international data. For the survey analysis, however, I compare a more limited range of legislative and presidential elections. For parliamentary elections for the lower house, in the countries under comparison in Module I of the CSES dataset, fifteen elections were held from 1996 to 2002 using proportional electoral systems. Ten nations held parliamentary elections using combined electoral systems, including independent and dependent subtypes. Last, four countries held parliamentary elections for the lower house under majoritarian rules. There are also many important differences in electoral systems within each category, summarized in Table 2.1, for example, in the ballot structure of FPTP in the United Kingdom and the Alternative Vote in Australia, in the proportion of members elected in single-member and proportional districts in combined systems, as well as in the level of electoral thresholds facing minor parties.

 

The distribution of electoral systems around the world, illustrated in Table 2.2 and in Figure 2.7, confirms the regional patterns and the residual legacy stamped upon constitutions by their colonial histories. Three-quarters of the former British colonies continue to use a majoritarian electoral system today, as do two-thirds of the ex-French colonies. By contrast, threequarters of the former Portuguese colonies, two-thirds of the ex-Spanish colonies, and all the former Dutch colonies use proportional electoral systems today. The post-Communist states freed from rule by the Soviet Union divided almost evenly among the three major electoral families, although slightly more countries (37%) have adopted proportional systems. Although Eastern Europe leans toward majoritarian arrangements, Central Europe adopted more proportional systems.

Note: The percentage of each colonial group using different types of electoral systems. Countries were classified by electoral system using the typology and sources in Figure 2.1 and by their predominant colonial history from the CIA World Fact Book, 2002. Available online at www.cia.org. The comparison covers 191 nation-states worldwide, excluding the seven states without direct elections during this period.

   

Presidential Electoral Systems

   

The countries under comparison in Module I of the CSES dataset also allow comparison of ten presidential elections (illustrated in Table 2.3), all held under majoritarian or plurality rules.27 The simple plurality FPTP was used in Mexico and Taiwan. The Second Ballot “majority-runoff” system (also known as the double ballot) is used worldwide in fifteen of the twenty-five countries with direct presidential elections, including in Austria, Colombia, Finland, France, Belarus, and Russia, and in seven of the nations under comparison in the CSES dataset. In these elections, if no candidate gets at least 50% of the vote in the first round, then the top two candidates face each other in a second round to insure a majority of votes for the leading candidate. This system is exemplified by the 1996 Russian presidential election, where seventy-eight candidates registered to run for election, of which seventeen qualified for nomination. In the first round, Boris Yeltsin won 35.3% of the vote, with Gennadii Zyuganov, the Communist candidate, close behind with 32% and Alexander Lebed third, with 14.5% of the vote. After the other candidates dropped out, and Lebed swung his supporters behind Yeltsin, the final result of the second election was a decisive 53.8% of the vote for Yeltsin against 40.3% for Zyuganov.28 Runoff elections aim to consolidate support behind the major contenders and to encourage broad cross-party coalition building in the final stages of the campaign.

 

The United States uses the unique device of the Electoral College. The president is not decided directly by popular vote, instead popular votes are collected within each state and, since 1964, the District of Columbia. Each state casts all of its electoral votes for the candidate receiving a plurality of votes within each state (the unit rule). Each state is allowed as many electoral votes as it has senators and representatives in Congress. This means that even sparsely populated states such as Alaska have at least three electoral votes. Nevertheless, the most populous states each cast by far the greatest number of electoral votes, and, therefore, presidential contenders devote most attention and strategic resources (spending, political ads, and visits) during the campaign to states such as New York, California, New Jersey, Pennsylvania, Ohio, Florida, Illinois, Michigan, and Texas, especially when polls suggest that the race is close in these areas. The importance of these rules is exemplified by the outcome of the 2000 election, where Republican George W. Bush won a 271–266 majority in the electoral college despite the fact that his opponent, Al Gore, won about half a million more popular votes. The results called attention to the need to alter the Electoral College, which has not experienced major reform since 1804, despite the fact that critics have regarded the system as archaic, outmoded, and essentially flawed.29 Other important variations among the presidential electoral systems under comparison include length of time in office (ranging from four to six years) and whether presidential elections are held in conjunction with legislative contests (which could be expected to strengthen the party coattails of presidential candidates and, therefore, create stronger legislative–executive links) or whether they are held separately, which reinforces the separation of powers.

Note: This does not count other electoral options on the ballot, such as for local, regional, state-level, upper house/senate, European, or other elected office, or any referenda issues. (i) Note Israel includes direct elections for the prime minister, not president. ✕ = Not applicable in the election under comparison. &radic; = Applicable.

 

The consequences of different arrangements also generate different electoral decisions by citizens, including how often they are called to the ballot box and what choices they face. Table 2.4 illustrates the major variations in the countries and national elections under comparison. The least demands are in parliamentary democracies such as Australia and Britain where citizens only cast one ballot at national level, although there are many other types of contest such as Australian state and local elections, and British elections to the European parliament, as well as to the Scottish Parliament and Welsh Assembly and to local government. By contrast, at national level Russian citizens are called to the polls twice for the second ballot presidential elections, as well as to cast two votes for the Duma. The other nations present different demands upon citizens ranging between these extremes. Obviously, greater options for voting provide citizens with more opportunity for political expression, for example, with split-ticket voting between levels, but at the same time, frequent demands from successive elections at multiple levels of office carry the danger of voter fatigue.

 

Conclusion: The Consequences of Electoral Systems

   

Often the choice of electoral system seems mechanistic, abstract, and highly technical, with constitutional engineering designed to bring about certain objectives. But the issue of how the electoral system should function reflects essentially contested normative concepts of representative government. For advocates of adversarial democracy, the most important considerations for electoral systems are that the votes cast in elections (not the subsequent process of coalition building) should determine the party or parties in government. The government should be empowered to implement its program during its full term of office, without depending upon the support of minority parties. The government should remain accountable for its actions to parliament, and ultimately to the public. And, at periodic intervals, the electorate should be allowed to judge the government’s record, evaluate prospective policy platforms offered by the opposition parties, and cast their votes accordingly. Minor parties in third or fourth place are discriminated against by majoritarian elections for the sake of governability. From this perspective, proportional elections are ineffective because they can produce indecisive outcomes, unstable regimes, disproportionate power for minor parties in “kingmaker” roles, and a lack of clear-cut accountability and transparency in decision-making.

 

By contrast, proponents of consensual democracy argue that majoritarian systems place too much faith in the winning party, especially in plural societies divided by ethnic conflict, with too few constraints on government during its term of office. For the vision of consensual democracy, the electoral system should promote a process of conciliation, consultation, and coalition-building within parliaments. Parties above a minimum threshold should be included in the legislature in rough proportion to their level of electoral support. The party or parties in government should craft policies based on a consensus among their coalition partners. Moreover, the composition of parliament should reflect the main divisions in society and the electorate, so that all citizens have spokespersons articulating their interests, perspectives, and concerns in national debates. In this view, majoritarian systems over-reward the winner, producing “an elected dictatorship” where a government based on a plurality can steamroller its policies and implement its programs, without the need for consultation and compromise with other parties in parliament or other groups in society. The unfairness and disproportionate results of plurality electoral systems, outside of two-party contests, means that some voices in the electorate are systematically excluded from public debate.

 

We can conclude, agnostically, that there is no single best electoral system: the central arguments between adversarial and consensual democratic theorists represent irresolvable value conflicts. For societies that are divided by deep-rooted ethnic or religious conflict, such as Rwanda, Bosnia, or Israel, proportional electoral systems may prove more inclusive, as Lijphart argues. But, as others warn, PR elections may also reinforce, rather than ameliorate, such cleavages.30 For states that are highly centralized, such as Britain or New Zealand, majoritarian systems can insulate the government from the need for broader consultation and for democratic checks and balances. In constitutional design it appears that, despite the widespread appeal of the rhetoric of “electoral engineering” for optimal decision-making, in practice there are no easy choices. A wide range of alternative rules can potentially influence the impact of these electoral systems on both patterns of voting behavior and political representation. The mechanical effects of electoral rules are easier to predict than the psychological ones, and, in both cases, many effects are highly contingent because they are embedded within many other institutional, political, cultural, and social contexts. In the next chapters I discuss the normative debates about electoral systems in more detail and then consider some of the most important consequences of electoral rules for voting behavior, including party competition, the strength of social cleavages and partisan identification, and patterns of electoral turnout.