Saylor.org's Comparative Politics/Types of Electoral Systems

=Overview of Electoral Systems=

What Electoral Systems Are
At the most basic level, electoral systems translate the votes cast in an election into results – the offices/seats - won by parties and candidates. The key variables are the electoral formula used (i.e. whether a plurality/majority, proportional, mixed or other system is used, and what mathematical formula is used to calculate the seat allocation), the ballot structure (i.e. whether the voter votes for a candidate or a party and whether the voter makes a single choice or expresses a series of preferences) and the district magnitude (not how many voters live in a district, but how many representatives to the legislature that district elects). It must also be stressed that, although this topic area does not focus on the administrative aspects of elections (such as the distribution of polling places, the nomination of candidates, the registration of voters, who runs the elections and so on), these issues are of critical importance, and the possible advantages of any given electoral system choice will be undermined unless due attention is paid to them. Electoral system design also affects other areas of electoral laws: the choice of electoral system has an influence on the way in which district boundaries are drawn, how voters are registered, the design of ballot papers, how votes are counted, and numerous other aspects of the electoral process.

The choice of Electoral System is one of the most important institutional decisions for any democracy. The choice of a particular electoral system has a profound effect on the future political life of the country concerned, and electoral systems, once chosen, often remain fairly constant as political interests solidify around and respond to the incentives presented by them. However, while conscious design has become far more prevalent recently, traditionally it has been rare for electoral systems to be consciously and deliberately selected. Often the choice was essentially accidental, the result of an unusual combination of circumstances, of a passing trend, or of a quirk of history, with the impact of colonialism and the effects of influential neighbours often being especially strong.

Any new democracy must choose (or inherit) an electoral system to elect its legislature. Equally, political crisis within an established democracy may lead to momentum for electoral system change, and even without political crisis, campaigners for political reform may attempt to put electoral system change onto the political agenda. Decisions to change, or indeed to keep in place, an electoral system are often affected by one of two circumstances:

■either political actors lack basic knowledge and information so that the choices and consequences of different electoral systems are not fully recognized;

■or, conversely, political actors use their knowledge of electoral systems to promote designs which they think will work to their own partisan advantage. The choices that are made may have consequences that were unforeseen when they are introduced, as well as effects which were predicted. These choices may not always be the best ones for the long-term political health of the country concerned, and at times they can have disastrous consequences for its democratic prospects.

The background to a choice of electoral system can thus be as important as the choice itself. Electoral system choice is a fundamentally political process, rather than a question to which independent technical experts can produce a single ‘correct answer’. In fact, the consideration of political advantage is almost always a factor in the choice of electoral systems—sometimes it is the only consideration—while the menu of available electoral system choices is often, in reality, a relatively constrained one. Equally, however, calculations of short-term political interest can often obscure the longer-term consequences of a particular electoral system and the interests of the wider political system. Consequently, while recognizing the practical constraints, this text attempts to approach the issue of electoral system choices in as broad and comprehensive a manner as possible.

As the crafting of political institutions is a critical task not only for new democracies but also for those established democracies that are seeking to adapt their systems to better reflect new political realities, this topic area also seeks to address the likely concerns of those persons in established democracies who may be redesigning electoral systems as well as those involved in debate on political institutions in new, fledgling and transitional democracies. Given this target audience, much of the academic literature on the subject is necessarily simplified, while at the same time this text attempts to address some of the more complex issues inherent in the area. If the text appears to be sometimes overly simplistic and at other times unduly complex, the explanation will usually lie in the attempt to balance the two objectives of clarity and comprehensiveness.

While the contexts in which emerging and established democracies make institutional choices can vary enormously, their long-term purposes are usually the same: to develop institutions which are strong enough to promote stable democracy but flexible enough to react to changing circumstances. Each type of democracy has much to learn from the experiences of the other. Institutional design is an evolving process, and this text seeks to distil the lessons learnt from the many actual examples of institutional design around the world.

Much constitutional design has taken place relatively recently: the global movement towards democratic governance in the 1980s and 1990s stimulated a new urgency in the search for enduring models of appropriate representative institutions, and a fresh evaluation of electoral systems. This process was encouraged by the realization that the choice of political institutions can have a significant impact on the wider political system. For example, it is increasingly being recognized that an electoral system can be designed both to provide local geographic representation and to promote proportionality; can promote the development of strong and viable national political parties, and ensure the representation of women and regional minorities; and can help to ‘engineer’ cooperation and accommodation in a divided society by the creative use of particular incentives and constraints. Electoral systems are today viewed as one of the most influential of all political institutions, and of crucial importance to broader issues of governance.

Guiding Principles of Electoral Systems
When an electoral system is chosen there are a number of things this system can be asked to accomplish or at least be conducive to – a stable and efficient government, coherent coalitions and strong parties are only a few. These goals – and their order of priority – are likely to differ between the different stakeholders.

In addition to this, there are general principles that can be used to guide the design of electoral system, as well as the process of choice itself. Some of the more important principles are:

Representation
The basic task for an electoral system is to translate votes into seats; to transform the expressed will of the voters into people who will represent it. There are many views of what fair representation is – geographic representation, descriptive representation, ideological or party political representation – but regardless of the view that is taken in each country, representation as a principle is a key guide when designing the most suitable electoral system.

Transparency
It is important that the mechanisms of the electoral system be as transparent as possible and known to both voters and political parties and candidates well in advance in order to avoid confusion and distrust in the results they produce at elections. In addition to this, the process through which the choice of electoral system is arrived at also benefits from transparency for the same reasons. If stakeholders’ arguments and influence over the process of review, reform or adoption are presented in an open way, the process and the electoral system arrived at will have a greater chance of being seen as legitimate.

Inclusiveness
The electoral system will have a greater chance of being accepted as fair and legitimate if it is considered to work in an inclusive manner. This means not only that the electoral law allows as many as possible citizens to vote (including inclusive suffrage, making sure that the system is easily understandable, and assuring access for all to the polling station), but also that the mechanisms of the electoral system do not overtly discriminate against any one group in society, minority or otherwise. Also, if the process through which the electoral system is arrived at is as inclusive as possible, both the process and the system may benefit as legitimacy and ownership increase, and as more stakeholders are able to bring suggestions and participate in the process of finding the most appropriate system for the society in question.

The Importance of Electoral Systems
Political institutions shape the rules of the game under which democracy is practised, and it is often argued that the easiest political institution to manipulate, for good or for bad, is the electoral system. In translating the votes cast in a general election into seats in the legislature, the choice of electoral system can effectively determine who is elected and which party gains power. While many aspects of a country’s political framework are often specified in the constitution and can thus be difficult to amend, electoral system change often only involves new legislation and can thus be subject to manipulation by unscrupulous majority.

Even with each voter casting exactly the same vote and with exactly the same number of votes for each party, one electoral system may lead to a coalition government or a minority government while another may allow a single party to assume majority control.

Electoral Systems and Party Systems
A number of other consequences of electoral systems go beyond this primary effect. Some systems encourage, or even enforce, the formation of political parties; others recognize only individual candidates. The type of party system which develops, in particular the number and the relative sizes of political parties in the legislature, is heavily influenced by the electoral system. So is the internal cohesion and discipline of parties: some systems may encourage factionalism, where different wings of one party are constantly at odds with each other, while another system might encourage parties to speak with one voice and suppress dissent. Electoral systems can also influence the way parties campaign and the way political elites behave, thus helping to determine the broader political climate; they may encourage, or retard, the forging of alliances between parties; and they can provide incentives for parties and groups to be broadly based and accommodating, or to base themselves on narrow appeals to ethnicity or kinship ties.

Electoral Systems and Conflict Management
These different impacts underline the important role that electoral systems often have in terms of conflict management. It is clear that different electoral systems can aggravate or moderate tension and conflict in a society. At one level, a tension exists between systems which put a premium on representation of minority groups and those which encourage strong single-party government. At another level, if an electoral system is not considered fair and the political framework does not allow the opposition to feel that they have the chance to win next time around, losers may feel compelled to seek power through illegal means, using non-democratic, confrontationalist and even violent tactics. And finally, because the choice of electoral system will determine the ease or complexity of the act of voting, it inevitably impacts on minorities and underprivileged groups. This is always important, but becomes particularly so in societies where there are a substantial number of inexperienced or illiterate voters.

Psychological and Mechanical Effects
Electoral systems are generally considered to have both ‘mechanical’ and ‘psychological’ effects. The mechanical impact is most apparent in the way different electoral systems tend to encourage different kinds of party system. Plurality/majority systems often tend to have a constraining effect on party numbers as only the top candidate/-s or parties in each electoral district will be elected, while proportional systems tend to be more ‘permissive’, resulting in a greater diversity of parties. The psychological impact of electoral systems reinforces this mechanical effect: under First Past The Post (FPTP) rules, voters who wish to support a minor party are often faced with a dilemma as to how best to avoid ‘wasting’ their vote, as only one candidate can be elected from any single-member district. The result of this dilemma is that many voters will not express their sincere choice but rather will vote for another candidate (usually from a major party) who they believe has a realistic chance of winning the seat. The overall effect of this is to strengthen larger parties at the expense of smaller ones. Proportional systems or systems that allow multiple ballot choices, by contrast, are more likely to facilitate the success of small parties, and hence the pressure to vote strategically is reduced.

The Importance of Context
It is important to realize that a given electoral system will not necessarily work in the same way in different countries. Although there are some common experiences in different regions of the world, the effects of a particular type of electoral system depend to a great extent on the socio-political context in which it is used. For example, while there remains general agreement that plurality/majority systems tend to restrict the range of legislative representation and Proportional Representation systems encourage it, the conventional wisdom that plurality/majority rules will produce a two-party system and PR a multiparty system is looking increasingly dated. In recent years, FPTP has not facilitated the aggregation of the party system in established democracies such as Canada and India, nor has it led to the formation of strong and lasting parties in Papua New Guinea. PR has seen the election of dominant single-party regimes in Namibia, South Africa and elsewhere. More broadly, the consequences of the choice of electoral system depend on factors such as how a society is structured in terms of ideological, religious, ethnic, racial, regional, linguistic or class divisions; whether the country is an established democracy, a transitional democracy or a new democracy; whether there is an established party system, or parties are embryonic or unformed, and how many ‘serious’ parties there are; and whether a particular party’s supporters are geographically concentrated or dispersed over a wide area.

The Broader Democratic Framework
It is also important not to see electoral systems in isolation. Their design and effects are heavily contingent upon other structures within and outside the constitution. Electoral systems are one square of an interrelated patchwork of government systems, rules and points of access to power. Successful electoral system design comes from looking at the framework of political institutions as a whole: changing one part of this framework is likely to cause adjustments in the way other institutions within it work.

For example, how does the chosen electoral system facilitate or encourage conflict resolution between party leaders and activists on the ground? How much control do party leaders have over the party’s elected representatives? Are there constitutional provisions for referendums, citizens’ initiatives or ‘direct democracy’ which may complement the institutions of representative democracy? And are the details of the electoral system specified in the constitution, as an attached schedule to the constitution, or in regular legislation? This will determine how entrenched the system is or how open it may be to change by elected majorities.

There are two issues of this kind that are worth considering in more detail. The first is the degree of centralization. Is the country federal or unitary, and, if federal, are the units symmetrical in their power or asymmetrical? The second is the choice between parliamentarism and presidentialism. Both systems have their advocates, and the traditions of different countries may influence which is chosen or even foreclose debate; but the different relationship between legislative and executive institutions has important implications for electoral system design for both. The frequent debates over the direct election of mayors and heads of the executive at local level combine both issues.

In most bicameral legislatures in federal systems of government, the two chambers are elected by different (or incongruent) methods. This makes sense for two prime reasons which have to do with the theory underpinning federalism. First, the second (or upper) house of a federal legislature is there to represent the provinces or states of the country, and each unit often receives equal representation regardless of population or territory size (e.g. the US Senate or South Africa’s National Council of Provinces).

Second, there is little point in creating a two-chamber legislature unless there is a degree of difference between the roles and possibly also of the powers of the two chambers, and using the same electoral system for both is more likely to repeat and reinforce the majority power that controls the lower chamber—particularly if the elections to both chambers are simultaneous. Upper chambers provide the opportunity for some degree of electoral innovation to include communities of interest which may not be fully represented in national elections to a lower chamber. But when elections take place at three or more levels, to the upper chamber of the legislature, the lower chamber of the legislature, and the institutions of government at regional level, it is crucial that the systems used are considered together. It may for example be possible to promote representation of minorities at regional level while discouraging or even prohibiting it at national level. Whether this is or is not desirable is a matter of political debate and choice.

Until recent years there were few examples of enduring democracies using presidential systems. However, the commitment to presidentialism in for example Latin America and parts of South-East Asia means that the question now asked is: What aspects of institutional design help make presidentialism work? There is some evidence from the Latin American experience that stability can be problematic in countries with presidential constitutions and highly fragmented party systems, and that there are tensions between divided executive and legislative are not held concurrently. However, it appears helpful branches when the presidential electoral system is over two rounds, the legislative system is List PR, and the elections to adopt an electoral system which makes it likely that the party or coalition supporting an elected president has a significant block, although not necessarily an absolute majority, of elected members of the legislature.

Plurality elections for the presidency and simultaneous presidential and legislative elections are often seen as helping to focus the party system into fewer and more viable challengers for power. However, there can be serious dangers in combining the great power that is vested in the hands of a directly elected president who is head of the executive with the use of a plurality method in a diverse or ethnically divided country where no single group has an absolute majority. The result can be devastating for legitimacy or indeed for the success of a peace process. A presidential electoral system may complement a federal system by requiring a successful candidate to achieve a winning vote not only nationwide but also a significant fraction of the vote in a minimum number of the states of the federation.

Criteria for Design
When designing an electoral system, it is best to start with a list of criteria which sum up what you want to achieve, what you want to avoid and, in a broad sense, what you want your legislature and executive government to look like. The criteria which follow cover many areas, but the list is not exhaustive, and the reader may add a host of equally valid items. It is also true that some of the criteria outlined overlap and may appear contradictory. This is because they often are contradictory: it is the nature of institutional design that trade-offs have to be made between a number of competing desires and objectives. These files focus primarily on criteria for the systems and not on the design process itself. The design process is covered in more depth in the sections Process of Change and Advice for Electoral Systems Designers.

For example, one may want to provide the opportunity for independent candidates to be elected, and at the same time to encourage the growth of strong political parties. Or electoral system designers may think it wise to craft a system which gives voters a wide degree of choice between candidates and parties, but this may make for a complicated ballot paper which causes difficulties for less-educated voters. The trick in choosing (or reforming) an electoral system is to prioritize the criteria that are most important and then assess which electoral system, or combination of systems, best maximizes the attainment of these objectives.

Providing Representation
Representation may take at least four forms.

First, geographical representation implies that each region, be it a town or a city, a province or an electoral district, has members of the legislature whom it chooses and who are ultimately accountable to their area.

Second, the ideological divisions within society may be represented in the legislature, whether through representatives from political parties or independent representatives or a combination of both.

Third, a legislature may be representative of the party-political situation that exists within the country even if political parties do not have an ideological base. If half the voters vote for one political party but that party wins no, or hardly any, seats in the legislature, then that system cannot be said to adequately represent the will of the people.

Fourth, the concept of descriptive representation considers that the legislature should be to some degree a ‘mirror of the nation’ which should look, feel, think and act in a way which reflects the people as a whole. An adequately descriptive legislature would include both men and women, the young and the old, the wealthy and the poor, and reflect the different religious affiliations, linguistic communities and ethnic groups within a society.

Making Elections Accessible and Meaningful
Elections are all well and good, but they may mean little to people if it is difficult to vote or if at the end of the day their perception that their vote makes no difference to the way the country is governed. The ease of voting is determined by factors such as how complex the ballot paper is, how easy it is for the voter to get to a polling place, how up-to-date the electoral register is, and how confident the voter is that his or her ballot will be secret.

Electoral participation—at least as a free choice—is also thought to increase when the outcome of elections, either at a national level or in the voter’s particular district, is likely to make a significant difference to the future direction of government. If you know that your preferred candidate has no chance of winning a seat in your particular district, what is the incentive to vote? In some electoral systems, the wasted votes (i.e. valid votes for losing candidates, as distinct from spoiled or invalid ballot papers, which are excluded from the count) can amount to a substantial proportion of the total national vote.

Lastly, the actual power of the body being elected helps determine whether its election has any meaning. Hollow elections in authoritarian systems which offer no genuine choice, where legislatures have little real influence on the formation of governments or on government policy, are far less important than elections to legislatures which actually have the power to determine central elements in people’s everyday lives.

Even within democratic systems, the choice of electoral system can influence the legitimacy of institutions. For example, the Australian Senate between 1919 and 1946 was elected by a highly disproportional electoral system (the Alternative Vote in multimember districts), which produced lopsided and unrepresentative results. This tended to undermine the actual legitimacy of the Senate itself in the eyes of both electors and politicians and, some observers argued, also undermined public support for the institutions of federal government in general. After the system was altered to a fairer proportional system (the Single Transferable Vote) in 1948, the Senate began to be perceived as more credible and representative, and thus respect for it and its relative importance in decision making increased.

Providing Incentives for Conciliation
Electoral systems can be seen not only as ways to constitute governing bodies but also as mechanisms for conflict management within a society. Some systems, in some circumstances, will encourage parties to make inclusive appeals for electoral support outside their own core vote base; for instance, even if a party draws its support primarily from black voters, a particular electoral system may give it the incentive to appeal also to white, or other, voters. Thus, the party’s policy platform would come less divisive and exclusionary, and more unifying and inclusive. Similar electoral system incentives might make parties less ethnically, regionally, linguistically or ideologically exclusive. Examples of how different electoral systems have worked as tools of conflict management are given throughout this text.

On the other side of the coin, electoral systems can encourage voters to look outside their own group and think of voting for parties which traditionally have represented a different group. Such voting behaviour breeds accommodation and community building. Systems which give the voter more than one vote or allow the voter to order candidates preferentially provide the space for voters to cut across preconceived social boundaries. At the 1998 Good Friday agreement election in Northern Ireland, for instance, vote transfers under the STV system benefited ‘pro-peace’ parties while still providing broadly proportional outcomes. At the 2003 election, however, a shift in first-preference votes towards hard-line parties tended to outweigh such effects.

Facilitating Stable and Efficient Government
The prospects for a stable and efficient government are not determined by the electoral system alone, but the results a system produces can contribute to stability in a number of important respects.

The key questions are

■whether voters perceive the system to be fair,

■whether government can efficiently enact legislation and govern, and

■whether the system avoids discriminating against particular parties or interest groups.

The perception of whether results are fair or not varies widely from country to country. Twice in the United Kingdom (UK) (in 1951 and 1974) the party winning the most votes in the country as a whole won fewer seats than its opponents, but this was considered more a quirk of a basically sound system than an outright unfairness which should be reversed. Conversely, similar results in New Zealand in 1978 and 1981, in which the National Party retained office despite winning fewer votes than the Labour opposition, are credited as starting the reform movement which led to a change of electoral system.

The question whether the government of the day can enact legislation efficiently is partly linked to whether it can assemble a working majority in the legislature, and this in turn is linked to the electoral system. As a general rule of thumb, plurality/majority electoral systems are more likely to produce legislatures where one party can outvote the combined opposition, while PR systems are more likely to give rise to coalition governments. Nevertheless, it has to be remembered that PR systems can also produce single-party majorities, and plurality/majority systems can leave no one party with a working majority. Much depends on the structure of the party system and the nature of the society itself.

Finally, the system should, as far as possible, act in an electorally neutral manner towards all parties and candidates; it should not openly discriminate against any political grouping. The perception that electoral politics in a democracy is an uneven playing field is a sign that the political order is weak and that instability may not be far around the corner. A dramatic example of this was the 1998 election in Lesotho, in which the Lesotho Congress for Democracy won every seat in the legislature with only 60 per cent of the votes under an FPTP system. The public unrest that followed, culminating in a request for military intervention in the country by the Southern African Development Community, demonstrated that such a result was not merely unfair but also dangerous, and the electoral system was consequently changed for future elections.

Holding the Government Accountable
Accountability is one of the bedrocks of representative government. Its absence may indeed lead to long-term instability. An accountable political system is one in which the government is responsible to the voters to the highest degree possible. Voters should be able to influence the shape of the government, either by altering the coalition of parties in power or by throwing out of office a single party which has failed to deliver. Suitably designed electoral systems facilitate this objective.

The conventional wisdom in this area may be simplistic. Traditionally, plurality/majority systems like FPTP were seen as leading to single parties taking office, while PR systems were associated with multiparty coalitions. While the broad logic of this association remains valid, there have been sufficient examples in recent years of FPTP elections leading to multiparty cabinets (e.g. in India) or of PR elections leading to the election of a strong single-party government (e.g. in South Africa) to raise doubts about the automatic assumption that one kind of electoral system will lead to particular governance outcomes. But clearly, electoral systems do have a major impact on broader issues of governance, for both presidential and parliamentary systems.

Holding Individual Representatives Accountable
Accountability at the individual level is the ability of the electorate to effectively check on those who, once elected, betray the promises they made during the campaign or demonstrate incompetence or idleness in office and ‘throw the rascals out’. Some systems emphasize the role of locally popular candidates, rather than on candidates nominated by a strong central party. Plurality/majority systems have traditionally been seen as maximizing the ability of voters to throw out unsatisfactory individual representatives. Again, this sometimes remains valid. However, the connection becomes tenuous where voters identify primarily with parties rather than candidates, as in the UK. At the same time, open and free list systems and STV are designed to allow voters to exercise candidate choice in the context of a proportional system.

Encouraging Political Parties
The weight of evidence from both established and new democracies suggests that longer-term democratic consolidation—that is, the extent to which a democratic regime is insulated from domestic challenges to the stability of the political order—requires the growth and maintenance of strong and effective political parties, and thus the electoral system should encourage this rather than promote party fragmentation.

To do this, electoral systems can be framed specifically to exclude parties with a small or minimal level of support. The development of the role of parties as a vehicle for individual political leaders is another trend which can be facilitated or retarded by electoral system design decisions. Most experts also agree that the electoral system should encourage the development of parties which are based on broad political values and ideologies as well as specific policy programmes, rather than narrow ethnic, racial or regional concerns. As well as lessening the threat of societal conflict, parties which are based on these broad ‘crosscutting cleavages’ are more likely to reflect national opinion than those which are based predominantly on sectarian or regional concerns.

Promoting Legislative Opposition and Oversight
Effective governance relies not only on those in power but, almost as much, on those who oppose and oversee them. The electoral system should help ensure the presence of a viable opposition grouping which can critically assess legislation, question the performance of the executive, safeguard minority rights, and represent its constituents effectively.

Opposition groupings should have enough representatives to be effective (assuming that their performance at the ballot box warrants it) and in a parliamentary system should be able to present a realistic alternative to the current government. Obviously the strength of the opposition depends on many other factors besides the choice of electoral system, but if the system itself makes the opposition impotent, democratic governance is inherently weakened.

A major reason for the change to an MMP electoral system in New Zealand, for example, was the systematic under-representation of smaller opposition parties under FPTP. At the same time, the electoral system should hinder the development of a ‘winner takes all’ attitude which leaves rulers blind to other views and the needs and desires of opposition voters, and sees both elections and government itself as zero-sum contests.

In a presidential system, the president needs the reliable support of a substantial group of legislators: however, the role of others in opposing and scrutinizing government legislative proposals is equally important. The separation of powers between legislature and executive effectively gives the task of executive oversight to all legislators, not only the opposition members. This makes it important to give particular thought to the elements of the electoral system which concern the relative importance of political parties and candidates, alongside the relationship between parties and their elected members.

Making the Election Process Workable and Sustainable
Elections do not take place on the pages of academic books but in the real world, and for this reason the choice of any electoral system is, to some degree, dependent on the cost and administrative capacities of the country involved. Although donor countries often provide substantial financial support for the first, and even the second, election in a country in transition to democracy, this is unlikely to be available in the long term even if it were desirable.

A sustainable political framework takes into account the resources of a country both in terms of the availability of people with the skills to be election administrators and in terms of the financial demands on the national budget.

For example, a poor country may not be able to afford the multiple elections required under a Two-Round System or be able easily to administer a complicated preferential vote count.

However, simplicity in the short term may not always make for cost effectiveness in the longer run. An electoral system may be cheap and easy to administer but it may not answer the pressing needs of a country—and when an electoral system is at odds with a country’s needs the results can be disastrous.

Alternatively, a system which appears at the outset to be a little more expensive to administer and more complex to understand may in the long run help to ensure the stability of the country and the positive direction of democratic consolidation.

Taking into Account 'International Standards'
Finally, the design of electoral systems today takes place in the context of a number of international covenants, treaties and other kinds of legal instruments affecting political issues.

While there is no single complete set of universally agreed international standards for elections, there is consensus that such standards include:

■the principles of free, fair and periodic elections that guarantee universal adult suffrage,

■the secrecy of the ballot and freedom from coercion, and

■a commitment to the principle of one person, one vote. Moreover, while there is no legal stipulation that a particular kind of electoral system is preferable to another, there is an increasing recognition of the importance of issues that are affected by electoral systems, such as the fair representation of all citizens, the equality of women and men, the rights of minorities, special considerations for the disabled, and so on.

These are formalized in international legal instruments such as the 1948 Universal Declaration of Human Rights and the 1966 International Covenant on Civil and Political Rights, and in the various conventions and commitments concerning democratic elections made by regional organizations such as the European Union (EU) and the Organization for Security and Co-operation in Europe (OSCE), Organization of American States (OAS), Council of Europe (COE) and the Commonwealth.

Criteria for Design - Conclusions
The ten outlined criteria are at times in conflict with each other or even mutually exclusive. The designers of an electoral system must therefore go through a careful process of prioritizing which criteria are most important to the particular political context before moving on to assess which system will do the best job.

A useful way forward is first to list the things which must be avoided at all costs, such as political catastrophes which could lead to the breakdown of democracy. For example, an ethnically divided country might want above all to avoid excluding minority ethnic groups from representation in order to promote the legitimacy of the electoral process and avoid the perception that the electoral system was unfair.

In contrast, while these issues might still be important to it, a fledgling democracy elsewhere might have different priorities—perhaps to ensure that a government can enact legislation efficiently without fear of gridlock, or that voters are able to remove discredited leaders if they so wish.

Establishing the priorities among such competing criteria can only be the domain of the domestic actors involved in the institutional design process.

Design Components
Once a decision has been made about the important goals to be achieved—and the important pitfalls to be avoided—in a new electoral system, there are a group of electoral system design tools which can be used to help achieve these goals. They include, among others:

■electoral system family and type

■district magnitude

■the relative role of political parties and candidates

■the form of the ballot paper

■the procedures for drawing electoral boundaries

■the electoral registration mechanisms

■the timing and synchronization of elections

■quotas and other special provisions

These tools will work differently in different combinations. Their use may depend on the level of information that is or can be available within a society, for example the numbers, diversity, and location of the population. Their effect will also depend on other institutional framework tools, such as the choice between parliamentarism and presidentialism, the requirements for registration and management of political parties, the relationship between political parties and elected members, and the role of instruments of direct democracy—referendums, citizens’ initiatives, and recall. It is worth emphasizing again that there is never a single ‘correct solution’ that can be imposed in a vacuum.

The Systems and Their Consequences
There are countless electoral system variations, but essentially they can be divided into 12 main systems, the majority of which fall into three broad families. The most common way to look at electoral systems is to group them according to how closely they translate national votes won into legislative seats won, that is, how proportional they are. To do this, one needs to look at both the votes-to-seats relationship and the level of wasted votes.

For example, South Africa used a classically proportional electoral system for its elections of 2004, and with 69.69 per cent of the popular vote the African National Congress (ANC) won 69.75 per cent of the national seats. The electoral system was highly proportional, and the number of wasted votes (i.e. those which were cast for parties which did not win seats in the Assembly) was only 0.74 per cent of the total. In direct contrast, in Mongolia in 2000, a Two-Round System only requiring a plurality of 25 per cent of the votes for candidates to be elected resulted in the Mongolian People’s Revolutionary Party (MPRP) winning 72 seats in the 76-member Parliament with around 52 per cent of the popular vote. This result was mirrored in Djibouti’s Party Block Vote election of 2003 when all 65 legislative seats were won by the Rassemblement Populaire pour le Progrès with 62.7 per cent of the vote.

However, under some circumstances, non-proportional electoral systems (such as FPTP) can give rise to relatively proportional overall results, for example, when party support is concentrated in regional fiefdoms. This was the case in another Southern African country, Malawi, in 2004. In that election, the Malawian Congress Party won 30 per cent of the seats with 25 per cent of the votes, the United Democratic Front won 27 per cent of the seats with 25 per cent of the votes, and the Alliance for Democracy won a little more than 3 per cent of the seats with just under 4 per cent of the votes.

The overall level of proportionality was high, but the clue to the fact that this was not inherently a proportional system, and so cannot be categorized as such, was that the wasted votes still amounted to almost half of all votes cast.

Equally, some design factors accentuate disproportionality. Systems with a high level of malapportionment often produce disproportional results, as do proportional systems with high thresholds—which can result in a high level of wasted votes, as in Turkey in 2002, where a 10 per cent threshold resulted in 46 per cent of votes being wasted.

Plurality/Majority Systems
The principle of plurality/majority systems is simple. After votes have been cast and totalled, those candidates or parties with the most votes are declared the winners (there may also be additional conditions). However, the way this is achieved in practice varies widely.

Five varieties of plurality/majority systems can be identified:

■First Past The Post (FPTP),

■Block Vote (BV),

■Party Block Vote(PBV),

■Alternative Vote (AV), and

■the Two-Round System (TRS).

In an FPTP system (sometimes known as a plurality single-member district system) the winner is the candidate with the most votes but not necessarily an absolute majority of the votes. When this system is used in multi-member districts, it becomes the Block Vote. Voters have as many votes as there are seats to be filled, and the highest-polling candidates fill the positions regardless of the percentage of the vote they achieve. This system—with the change that voters vote for party lists instead of individual candidates—becomes the Party Block Vote.

Majoritarian systems, such as the Alternative Vote and the Two-Round System, try to ensure that the winning candidate receives an absolute majority (i.e. over 50 per cent). Each system in essence makes use of voters’ second preferences to produce a winner with an absolute majority if one does not emerge from the first round of voting.

First Past The Post (FPTP)
The First Past The Post system is the simplest form of plurality/majority system, using single member districts and candidate-centred voting. The voter is presented with the names of the nominated candidates and votes by choosing one, and only one, of them. The winning candidate is simply the person who wins the most votes; in theory he or she could be elected with two votes, if every other candidate only secured a single vote.

Along with the UK, the cases most often analysed are Canada, India, and the United States.

Advantages and disadvantages of FPTP system
ADVANTAGES of FPTP

First Past The Post, like other plurality/majority electoral systems, is defended primarily on the grounds of simplicity and its tendency to produce winners who are representatives beholden to defined geographic areas and governability. The most often cited advantages are that:

■It provides a clear-cut choice for voters between two main parties. The inbuilt disadvantages faced by third and fragmented minority parties under FPTP in many cases cause the party system to gravitate towards a party of the ‘left’ and a party of the ‘right’, alternating in power. Third parties often wither away and almost never reach a level of popular support above which their national vote yields a comparable percentage of seats in the legislature.

■It gives rise to single-party governments. The ‘seat bonuses’ for the largest party common under FPTP (e.g. where one party wins 45 per cent of the national vote but 55 per cent of the seats) mean that coalition governments are the exception rather than the rule. This state of affairs is praised for providing cabinets which are not shackled by the restraints of having to bargain with a minority coalition partner.

■It gives rise to a coherent opposition in the legislature. In theory, the flip side of a strong single-party government is that the opposition is also given enough seats to perform a critical checking role and present itself as a realistic alternative to the government of the day. It advantages broadly-based political parties. In severely ethnically or regionally divided societies, FPTP is commended for encouraging political parties to be ‘broad churches’, encompassing many elements of society, particularly when there are only two major parties and many different societal groups. These parties can then field a diverse array of candidates for election. In Malaysia, for example, the Barisan Nasional government is made up of a broadly-based umbrella movement which fields Malay, Chinese, and Indian candidates in areas of various ethnic complexions.

■It excludes extremist parties from representation in the legislature. Unless an extremist minority party’s electoral support is geographically concentrated, it is unlikely to win any seats under FPTP. (By contrast, under a List PR system with a single national-level district and a large number of seats, a fraction of 1 per cent of the national vote can ensure representation in the legislature.)

■It promotes a link between constituents and their representatives, as it produces a legislature made up of representatives of geographical areas. Elected members represent defined areas of cities, towns, or regions rather than just party labels. Some analysts have argued that this ‘geographic accountability’ is particularly important in agrarian societies and in developing countries.

■It allows voters to choose between people rather than just between parties. Voters can assess the performance of individual candidates rather than just having to accept a list of candidates presented by a party, as can happen under some List PR electoral systems.

■It gives a chance for popular independent candidates to be elected. This may be particularly important in developing party systems, where politics still revolves more around extended ties of family, clan, or kinship and is not based on strong party political organizations.

■Finally, FPTP systems are particularly praised for being simple to use and understand. A valid vote requires only one mark beside the name or symbol of one candidate. Even if the number of candidates on the ballot paper is large, the count is easy for electoral officials to conduct.

DISADVANTAGES of FPTP

However, FPTP is frequently criticized for a number of reasons.

These include:

■It excludes smaller parties from ‘fair’ representation, in the sense that a party which wins approximately, say, 10 per cent of the votes should win approximately 10 per cent of the legislative seats. In the 1993 federal election in Canada, the Progressive Conservatives won 16 per cent of the votes but only 0.7 per cent of the seats, and in the 1998 general election in Lesotho, the Basotho National Party won 24 per cent of the votes but only 1 per cent of the seats. This is a pattern which is repeated time and time again under FPTP.

■It excludes minorities from fair representation. As a rule, under FPTP, parties put up the most broadly acceptable candidate in a particular district so as to avoid alienating the majority of electors. Thus it is rare, for example, for a black candidate to be given a major party’s nomination in a majority white district in the UK or the USA, and there is strong evidence that ethnic and racial minorities across the world are far less likely to be represented in legislatures elected by FPTP. In consequence, if voting behaviour does dovetail with ethnic divisions, then the exclusion from representation of members of ethnic minority groups can be destabilizing for the political system as a whole.

■It excludes women from the legislature. The ‘most broadly acceptable candidate’ syndrome also affects the ability of women to be elected to legislative office because they are often less likely to be selected as candidates by male-dominated party structures. Evidence across the world suggests that women are less likely to be elected to the legislature under plurality/majority systems than under PR ones.

■It can encourage the development of political parties based on clan, ethnicity or region, which may base their campaigns and policy platforms on conceptions that are attractive to the majority of people in their district or region but exclude or are hostile to others. This has been an ongoing problem in African countries like Malawi and Kenya, where large communal groups tend to be regionally concentrated. The country is thus divided into geographically separate party strongholds, with little incentive for parties to make appeals outside their home region and cultural–political base.

■It exaggerates the phenomenon of ‘regional fiefdoms’ where one party wins all the seats in a province or area. If a party has strong support in a particular part of a country, winning a plurality of votes, it will win all, or nearly all, of the seats in the legislature for that area. This both excludes minorities in that area from representation and reinforces the perception that politics is a battleground defined by who you are and where you live rather than what you believe in. This has long been put forward as an argument against FPTP in Canada.

■It leaves a large number of wasted votes which do not go towards the election of any candidate. This can be particularly dangerous if combined with regional fiefdoms, because minority party supporters in the region may begin to feel that they have no realistic hope of ever electing a candidate of their choice. It can also be dangerous where alienation from the political system increases the likelihood that extremists will be able to mobilize anti-system movements.

■It can cause vote-splitting. Where two similar parties or candidates compete under FPTP, the vote of their potential supporters is often split between them, thus allowing a less popular party or candidate to win the seat. Papua New Guinea provides a particularly clear example.

■It may be unresponsive to changes in public opinion. A pattern of geographically concentrated electoral support in a country means that one party can maintain exclusive executive control in the face of a substantial drop in overall popular support. In some democracies under FPTP, a fall from 60 per cent to 40 per cent of a party’s share of the popular vote nationally can result in a fall from 80 per cent to 60 per cent in the number of seats held, which does not affect its overall dominant position. Unless sufficient seats are highly competitive, the system can be insensitive to swings in public opinion.

■Finally, FPTP systems are dependent on the drawing of electoral boundaries. All electoral boundaries have political consequences: there is no technical process to produce a single ‘correct answer’ independently of political or other considerations. Boundary delimitation may require substantial time and resources if the results are to be accepted as legitimate. There may also be pressure to manipulate boundaries by gerrymandering or malapportionment. This was particularly apparent in the Kenyan elections of 1993 when huge disparities between the sizes of electoral districts—the largest had 23 times the number of voters the smallest had—contributed to the ruling Kenyan African National Union party’s winning a large majority in the legislature with only 30 per cent of the popular vote.

Block Vote (BV)
The Block Vote is simply the use of plurality voting in multi-member districts. Voters have as many votes as there are seats to be filled in their district, and are usually free to vote for individual candidates regardless of party affiliation. In most BV systems, they may use as many, or as few, of their votes as they wish. The system was used in Jordan in 1989, in Mongolia in 1992, and in the Philippines and Thailand until 1997, but was changed in all these countries as a result of unease with the results it produced.

Advantages and disadvantages of BV
ADVANTAGES of BV

The Block Vote is often applauded for retaining the voter’s ability to vote for individual candidates and allowing for reasonably-sized geographical districts, while at the same time increasing the role of parties compared with FPTP and strengthening those parties which demonstrate most coherence and organizational ability.

DISADVANTAGES of BV

However, the Block Vote can have unpredictable and often undesirable impacts on election outcomes. For example, when voters cast all their votes for the candidates of a single party, the system tends to exaggerate most of the disadvantages of FPTP, in particular its disproportionality. When parties nominate a candidate for each vacancy in a Block Vote system and encourage voters to support every member of their slate, this is particularly likely. In Mauritius in 1982 and 1995, for example, the party in opposition before the election won every seat in the legislature with only 64 per cent and 65 per cent of the vote, respectively. This created severe difficulties for the effective functioning of a parliamentary system based on concepts of government and opposition. The use of ‘best loser’ seats in Mauritius only partially compensates for this weakness.

In Thailand, the Block Vote was seen as having encouraged the fragmentation of the party system. Because it enables electors to vote for candidates of more than one party in the same district, members of the same party may be encouraged to compete against each other for support. The Block Vote is thus sometimes seen as being a contributor to internal party factionalism and corruption.

In recent years, a number of countries have therefore abandoned the Block Vote in favour of other systems. Thailand and the Philippines both changed from BV to a mixed system in the late 1990s. In both cases, a major justification for the change was the need to combat vote-buying and strengthen the development of political parties.

Party Block Vote (PBV)
Under Party Block Vote, unlike FPTP, there are multi-member districts. Voters have a single vote, and choose between party lists of candidates rather than between individuals. The party which wins most votes takes all the seats in the district, and its entire list of candidates is duly elected. As in FPTP, there is no requirement for the winner to have an absolute majority of the votes. As of 2004, PBV was used as the only system or the major component of the system in four countries—Cameroon, Chad, Djibouti and Singapore.

Advantages and disadvantages of PBV
ADVANTAGES of PBV

PBV is simple to use, encourages strong parties and allows for parties to put up mixed slates of candidates in order to facilitate minority representation. It can be used to help to ensure balanced ethnic representation, as it enables parties to present ethnically diverse lists of candidates for election—and may indeed be designed to require them to do so.

DISADVANTAGES of PBV

However, the Party Block Vote also suffers from most of the disadvantages of FPTP, and may indeed produce highly disproportional results where one party wins almost all of the seats with a simple majority of the votes. In Djibouti’s 1997 election, the ruling Union for the Presidential Majority coalition won every seat, leaving the two opposition parties without any representation in the legislature.

The Alternative Vote (AV)
Elections under Alternative Vote are usually held in single-member districts, like FPTP elections. However, AV gives voters considerably more options than FPTP when marking their ballot paper. Rather than simply indicating their favoured candidate, under AV electors rank the candidates in the order of their choice, by marking a ‘1’ for their favourite, ‘2’ for their second choice, ‘3’ for their third choice and so on. The system thus enables voters to express their preferences between candidates rather than simply their first choice. For this reason, it is often known as ‘preferential voting’ in the countries which use it. (The Borda Count, STV, and the Supplementary Vote are also preferential systems).

AV also differs from FPTP in the way votes are counted. Like FPTP or TRS, a candidate who has won an absolute majority of the votes (50 per cent plus one) is immediately elected. However, if no candidate has an absolute majority, under AV the candidate with the lowest number of first preferences is ‘eliminated’ from the count, and his or her ballots are examined for their second preferences. Each ballot is then transferred to whichever remaining candidate has the highest preference in the order as marked on the ballot paper. This process is repeated until one candidate has an absolute majority, and is declared duly elected. AV is thus a majoritarian system.

It is possible, but not essential, in preferential systems such as AV to require voters to number all, or most, of the candidates on the ballot paper. This avoids the possibility of votes becoming ‘wasted’ at a later stage in the count because they bear no further valid preferences. However, it can lead to an increase in the number of invalid votes, and it can sometimes give substantial importance to preferences between candidates to which the voter is indifferent or actively dislikes.

Advantages and disadvantages of AV
ADVANTAGES of AV

One advantage of transferring ballots is that it enables the votes of several candidates to accumulate, so that diverse but related interests can be combined to win representation. AV also enables supporters of candidates who have little hope of being elected to influence, via their second and later preferences, the election of a major candidate. For this reason, it is sometimes argued that AV is the best system for promoting centrist politics, as it can compel candidates to seek not only the votes of their own supporters but also the ‘second preferences’ of others. To attract these preferences, candidates must make broadly-based appeals rather than focusing on narrower issues. The experience of AV in Australia tends to support these arguments: the major parties, for example, typically try to strike bargains with minor parties for the second preferences of their supporters prior to an election—a process known as ‘preference swapping’. Furthermore, because of the majority support requirement, AV increases the consent given to elected members, and thus can enhance their perceived legitimacy.

The experience of AV in Papua New Guinea and in Australia suggests that it can provide significant incentives for accommodatory and cooperative politics. In recent years, AV, or its variant the Supplementary Vote, has also been adopted for presidential and mayoral elections in Bosnia, London, and San Francisco.

DISADVANTAGES of AV

Nevertheless, AV also has a number of disadvantages. First, it requires a reasonable degree of literacy and numeracy to be used effectively, and because it operates in single-member districts it can often produce results that are disproportional when compared to PR systems—or even in some cases compared with FPTP. Also, the potential of AV for promoting centrist outcomes is very dependent on underlying social and demographic conditions: while it successfully promoted interethnic accommodation in Papua New Guinea during the 1960s and 1970s, it has been criticized in another Pacific country, Fiji, since it was implemented there in 1997. Moreover, as its use in the Australian Senate from 1919 to 1946 noted, AV does not work well when applied to larger, multi-member districts

The Two-Round system (TRS)
The central feature of the Two-Round System is as the name suggests: it is not one election but takes place in two rounds, often a short time apart. The first round is conducted in the same way as a single-round plurality/majority election. In the most common form of TRS, this is conducted using FPTP. It is, however, also possible to conduct TRS in multi-member districts using Block Vote (as in Kiribati) or Party Block Vote (as in Mali). A candidate or party that receives a specified proportion of the vote is elected outright, with no need for a second ballot. This proportion is normally an absolute majority of valid votes cast, although several countries use a different figure when using TRS to elect a president. If no candidate or party receives an absolute majority, then a second round of voting is held and the winner of this round is declared elected.

The details of how the second round is conducted vary in practice from case to case. The most common method is for it to be a straight run-off contest between the two highest vote winners from the first round; this is called majority run-off TRS. It produces a result that is truly majoritarian in that one of the two participants will necessarily achieve an absolute majority of votes and be declared the winner. A second method, majority-plurality TRS, is used for legislative elections in France, the country most often associated with the Two-Round System. In these elections, any candidate who has received the votes of over 12.5 per cent of the registered electorate in the first round can stand in the second round. Whoever wins the highest number of votes in the second round is then declared elected, regardless of whether they have won an absolute majority or not. Unlike majority run-off, this system is not truly majoritarian, as there may be up to five or six candidates contesting the second round of elections.

Advantages and disadvantages of Two-Round System (TRS)
ADVANTAGES of TRS

■First and foremost, TRS allows voters to have a second chance to vote for their chosen candidate, or even to change their minds between the first and the second rounds. It thus shares some features in common with preferential systems like the Alternative Vote, in which voters are asked to rank-order candidates, while also enabling voters to make a completely fresh choice in the second round if they so desire.

■TRS can encourage diverse interests to coalesce behind the successful candidates from the first round in the lead-up to the second round of voting, thus encouraging bargains and trade-offs between parties and candidates. It also enables the parties and the electorate to react to changes in the political landscape that occur between the first and the second rounds of voting.

■TRS lessens the problems of ‘vote-splitting’, the common situation in many plurality/majority systems where two similar parties or candidates split their combined vote between them, thus allowing a less popular candidate to win the seat. Also, because electors do not have to rank-order candidates to express their second choice, TRS may be better suited to countries where illiteracy is widespread than systems which use preferential numbering like the Alternative Vote or the Single Transferable Vote.

DISADVANTAGES of TRS

■TRS places considerable pressure on the electoral administration by requiring it to run a second election a short time after the first, thus significantly increasing both the cost of the overall election process and the time that elapses between the holding of an election and the declaration of a result. This can lead to instability and uncertainty. TRS also places an additional burden on the voter in terms of time and effort required to cast the vote as the voter has to make it to the polling station twice, and sometimes there is a sharp decline in turnout between the first round and the second.

■TRS shares many of the disadvantages of FPTP. Research has shown that in France it produces the most disproportional results of any Western democracy, and that it tends to fragment party systems in new democracies.

■One of the most serious problems with TRS is its implications for deeply divided societies. In Angola in 1992, in what was supposed to be a peacemaking election, rebel leader Jonas Savimbi came second in the first round of a TRS presidential election to Jose dos Santos with 40 per cent of the vote as opposed to dos Santos’ 49 per cent. As it was clear that he would lose the run-off phase, he had little incentive to play the democratic opposition game and immediately restarted the civil war in Angola, which went on for another decade. In Republic of the Congo in 1993, prospects of a government landslide in the second round of a TRS election prompted the opposition to boycott the second round and take up arms. In both cases, the clear signal that one side would probably lose the election was the trigger for violence. In Algeria in 1992, the candidate of the Islamic Salvation Front (Front Islamique du Salut, FIS) led in the first round, and the military intervened to cancel the second round.

Proportional Representation (PR)
The rationale underpinning all PR systems is to consciously reduce the disparity between a party's share of the national vote and its share of the parliamentary seats; if a major party wins 40 per cent of the votes, it should win approximately 40 per cent of the seats, and a minor party with 10 per cent of the votes should also gain 10 per cent of the legislative seats. This congruity between a party’s share of the vote and its share of the seats provides an incentive for all parties to support and participate in the system.

PR requires the use of electoral districts with more than one member: it is not possible to divide a single seat elected on a single occasion proportionally. There are two major types of PR system—List PR and Single Transferable Vote (STV). Proportionality is often seen as being best achieved by the use of party lists, where political parties present lists of candidates to the voters on a national or regional basis, but preferential voting can work equally well: the Single Transferable Vote, where voters rank-order candidates in multi-member districts, is another well-established proportional system.

There are many important issues which can have a major impact on how a PR system works in practice. The greater the number of representatives to be elected from a district, the more proportional the electoral system will be. PR systems also differ in the range of choice given to the voter—whether the voter can choose between political parties, individual candidates, or both.

Advantages and disadvantages of PR systems
In many respects, the strongest arguments for PR derive from the way in which the system avoids the anomalous results of plurality/majority systems (insert link to glossary) and is better able to produce a representative legislature. For many new democracies, particularly those which face deep societal divisions, the inclusion of all significant groups in the legislature can be a near-essential condition for democratic consolidation. Failing to ensure that both minorities and majorities have a stake in developing political systems can have catastrophic consequences, such as seeking power through illegal means.

ADVANTAGES of PR systems

PR systems in general are praised for the way in which they:

■Faithfully translate votes cast into seats won, and thus avoid some of the more destabilizing and ‘unfair’ results thrown up by plurality/majority electoral systems. ‘Seat bonuses’ for the larger parties are minimized, and small parties can have their voice heard in the legislature.

■Encourage or require the formation of political parties or groups of like-minded candidates to put forward lists. This may clarify policy, ideology, or leadership differences within society, especially when, as in Timor-Leste at independence, there is no established party system.

■Give rise to very few wasted votes. When thresholds are low, almost all votes cast in PR elections go towards electing a candidate of choice. See pcb02b to read who may determine the selection process in political parties. This increases the voters’ perception that it is worth making the trip to the polling booth at election time, as they can be more confident that their vote will make a difference to the election outcome, however small.

■Facilitate minority parties’ access to representation. Unless the threshold is unduly high, or the district magnitude is unusually low, then any political party with even a small percentage of the vote can gain representation in the legislature. This fulfils the principle of inclusion, which can be crucial to stability in divided societies and has benefits for decision making in established democracies, such as achieving a more balanced representation of minorities in decision-making bodies and providing role models of minorities as elected representatives.

■Encourage parties to campaign beyond the districts in which they are strong or where the results are expected to be close. The incentive under PR systems is to maximize the overall vote regardless of where those votes might come from. Every vote, even from areas where a party is electorally weak, goes towards gaining another seat.

■Restrict the growth of ‘regional fiefdoms’. Because PR systems reward minority parties with a minority of the seats, they are less likely to lead to situations where a single party holds all the seats in a given province or district. This can be particularly important to minorities in a province which may not have significant regional concentrations or alternative points of access to power.

■Lead to greater continuity and stability of policy. The West European experience suggests that parliamentary PR systems score better with regard to governmental longevity, voter participation, and economic performance. The rationale behind this claim is that regular switches in government between two ideologically polarized parties, as can happen in FPTP systems, makes long-term economic planning more difficult, while broad PR coalition governments help engender a stability and coherence in decision making which allow for national development.

■Make power-sharing between parties and interest groups more visible. In many new democracies, power-sharing between the numerical majority of the population who hold political power and a small minority who hold economic power is an unavoidable reality. Where the numerical majority dominates the legislature and a minority sees its interests expressed in the control of the economic sphere, negotiations between different power blocks are less visible, less transparent, and less accountable (e.g. in Zimbabwe during its first 20 years of independence). It has been argued that PR, by including all interests in the legislature, offers a better hope that decisions will be taken in the public eye and by a more inclusive cross-section of the society.

DISADVANTAGES of PR systems

Most of the criticisms of PR in general are based around the tendency of PR systems to give rise to coalition governments and a fragmented party system. The arguments most often cited against PR are that it leads to:

■Coalition governments, which in turn lead to legislative gridlock and consequent inability to carry out coherent policies. There are particularly high risks during an immediate post-conflict transition period, when popular expectations of new governments are high. Quick and coherent decision making can be impeded by coalition cabinets and governments of national unity which are split by factions.

■A destabilizing fragmentation of the party system. PR can reflect and facilitate a fragmentation of the party system. It is possible that extreme pluralism can allow tiny minority parties to hold larger parties to ransom in coalition negotiations. In this respect, the inclusiveness of PR is cited as a drawback of the system. In Israel, for example, extremist religious parties are often crucial to the formation of a government, while Italy endured many years of unstable shifting coalition governments. Democratizing countries are often fearful that PR will allow personality-based and ethnic-cleavage parties to proliferate in their undeveloped party systems.

■A platform for extremist parties. In a related argument, PR systems are often criticized for giving a stage in the legislature to extremist parties of the left or the right. It has been argued that the collapse of Weimar Germany was in part due to the way in which its PR electoral system gave a toehold to extremist groups of the extreme left and right.

■Governing coalitions which have insufficient common ground in terms of either their policies or their support base. These coalitions of convenience are sometimes contrasted with coalitions of commitment produced by other systems (e.g. through the use of AV), in which parties tend to be reciprocally dependent on the votes of supporters of other parties for their election, and the coalition may thus be stronger.

■Small parties getting a disproportionately large amount of power. Large parties may be forced to form coalitions with much smaller parties, giving a party that has the support of only a small percentage of the votes the power to veto any proposal that comes from the larger parties.

■The inability of the voter to enforce accountability by throwing a party out of power or a particular candidate out of office. Under a PR system, it may be very difficult to remove a reasonably-sized centre party from power. When governments are usually coalitions, some political parties are everpresent in government, despite weak electoral performances from time to time. The Free Democratic Party (FDP) in Germany was a member of the governing coalition for all but eight of the 50 years from 1949 to 1998, although it never gained more than 12 per cent of the vote.

■Difficulties either for voters to understand or for the electoral administration to implement the sometimes complex rules of the system. Some PR systems are considered to be more difficult than non-PR systems and may require more voter education and training of poll workers to work successfully.

List PR
In its most simple form, List PR involves each party presenting a list of candidates to the electorate in each multi-member electoral district. Voters vote for a party, and parties receive seats in proportion to their overall share of the vote in the electoral district. Winning candidates are taken from the lists in order of their position on the lists. The choice of List PR does not in itself completely specify the electoral system: more details must be determined. The system used to calculate the allocation of seats after the votes have been counted can be either a Highest Average or a Largest Remainder Method. The formula chosen has a small but sometimes critical effect on the outcomes of elections under PR. In Cambodia in 1998, a change in the formula a few weeks before polling day turned out to have the effect of giving the largest party 64 seats, instead of 59, in a 121-seat National Assembly. The change had not been well publicized, and it was with difficulty that the opposition accepted the results. This example clearly demonstrates the importance for electoral system designers of apparently minor details.

There are several other important issues that need to be considered in defining precisely how a List PR system will work. A formal threshold may be required for representation in the legislature: a high threshold (for example 10 per cent, as used by Turkey) is likely to exclude smaller parties, while a low threshold (for example 1.5 per cent, as used by Israel) may promote their representation. In South Africa, there is no formal threshold, and in 2004 the African Christian Democratic Party won six seats out of 400 with only 1.6 per cent of the national vote. List PR systems also differ depending on whether and how the voter can choose between candidates as well as parties, that is, whether lists are closed, open or free (panachage). This choice has implications for the complexity of the ballot paper.

Other choices include arrangements for formal or informal ‘vote pooling’; the scope for agreements between parties, such as that provided by systems which use apparentement; and the definition of district boundaries.

Advantages and disadvantages of List PR
ADVANTAGES of List PR

■In addition to the advantages attached to PR systems generally, List PR makes it more likely that the representatives of minority cultures/groups will be elected. When, as is often the case, voting behaviour dovetails with a society’s cultural or social divisions, then List PR electoral systems can help to ensure that the legislature includes members of both majority and minority groups. This is because parties can be encouraged by the system to craft balanced candidate lists which appeal to a whole spectrum of voters’ interests. The experience of a number of new democracies (e.g. South Africa, Indonesia, and Sierra Leone) suggests that List PR gives the political space which allows parties to put up multiracial, and multi-ethnic, lists of candidates. The South African National Assembly elected in 1994 was 52 per cent black (11 per cent Zulu, the rest being of Xhosa, Sotho, Venda, Tswana, Pedi, Swazi, Shangaan and Ndebele extraction), 32 per cent white (one-third English-speaking, two-thirds Afrikaans-speaking), 7 per cent Coloured and 8 per cent Indian. The Namibian Parliament is similarly diverse, with representatives from the Ovambo, Damara, Herero, Nama, Baster and white (English and German-speaking) communities.

■List PR makes it more likely that women will be elected. PR electoral systems are almost always more friendly to the election of women than plurality/majority systems. In essence, parties are able to use the lists to promote the advancement of women politicians and allow voters the space to elect women candidates while still basing their choice on other policy concerns than gender. As noted above, in single-member districts, most parties are encouraged to put up a ‘most broadly acceptable’ candidate, and that person is seldom a woman. In all regions of the world, PR systems do better than FPTP systems in the number of women elected, and 14 of the top 20 nations when it comes to the representation of women use List PR. In 2004, the number of women representatives in legislatures elected by List PR systems was 4.3 percentage points higher than the average of 15.2 per cent for all legislatures, while that for legislatures elected by FPTP was 4.1 percentage points lower.

DISADVANTAGES of List PR

In addition to the general issues already identified relating to PR systems, the following additional disadvantages may be considered:

■Weak links between elected legislators and their constituents. When List PR is used, and particularly when seats are allocated in one single national district, as in Namibia or Israel, the system is criticized for destroying the link between voters and their representatives. Where lists are closed, voters have no opportunity to determine the identity of the persons who will represent them and no identifiable representative for their town, district or village, nor can they easily reject an individual representative if they feel that he or she has performed poorly in office or is not the kind of person they would want representing them – e.g., warlords in countries such as Bosnia or Afghanistan. Moreover, in some developing countries where the society is mainly rural, voters’ identification with their region of residence is sometimes considerably stronger than their identification with any political party or grouping. This criticism, however, may relate more to the distinction between systems in which voters vote for parties and systems in which they vote for candidates.

■Excessive entrenchment of power within party headquarters and in the hands of senior party leaderships—especially in closed-list systems. A candidate’s position on the party list, and therefore his or her likelihood of success, is dependent on currying favour with party bosses, while their relationship with the electorate is of secondary importance. In an unusual twist to the List PR system, in Guyana parties publish their list of candidates not ranked but simply ordered alphabetically. This allows party leaders even more scope to reward loyalty and punish independence because seats are only allocated to individuals once the result of the vote is known.

■The need for some kind of recognized party or political groupings to exist. This makes List PR particularly difficult to implement in those societies which do not have parties or have very embryonic and loose party structures, for example, many of the island countries of the Pacific. While technically possible to allow independent candidates to run under various forms of PR, it is difficult and introduces a number of additional complications, particularly as relates to wasted votes.

The Single Transferable Vote (STV)
STV has long been advocated by political scientists as one of the most attractive electoral systems, but its use for legislative elections has been limited to a few cases—the Republic of Ireland since 1921, Malta since 1947, and once in Estonia in 1990. It is also used for elections to the Australian Federal Senate and in several Australian states, and for European and local elections in Northern Ireland. It has been adopted for local elections in Scotland and in some authorities in New Zealand. It was also chosen as the recommendation of the British Columbia Citizens’ Assembly.

The core principles of the system were independently invented in the 19th century by Thomas Hare in Britain and Carl Andræ in Denmark. STV uses multi-member districts, and voters rank candidates in order of preference on the ballot paper in the same manner as under the Alternative Vote system. In most cases, this preference marking is optional, and voters are not required to rank-order all candidates; if they wish, they can mark only one.

After the total number of first-preference votes are tallied, the count then begins by establishing the quota of votes required for the election of a single candidate. The quota used is normally the Droop quota, calculated by the simple formula:

Quota = (votes / (seats +1)) +1

The result is determined through a series of counts. At the first count, the total number of first-preference votes for each candidate is ascertained. Any candidate who has a number of first preferences greater than or equal to the quota is immediately elected.

In second and subsequent counts, the surplus votes of elected candidates (i.e. those votes above the quota) are redistributed according to the second preferences on the ballot papers. For fairness, all the candidate’s ballot papers can be redistributed, but each at a fractional percentage of one vote, so that the total redistributed vote equals the candidate’s surplus (the Republic of Ireland uses a weighted sample instead of distributing fractions). If a candidate had 100 votes, for example, and their surplus was five votes, then each ballot paper would be redistributed according to its second preference at the value of 1/20th of a vote. After any count, if no candidate has a surplus of votes over the quota, the candidate with the lowest total of votes is eliminated. His or her votes are then redistributed in the next count to the candidates left in the race according to the second and then lower preferences shown. The process of successive counts, after each of which surplus votes are redistributed or a candidate is eliminated, continues until either all the seats for the electoral district are filled by candidates who have received the quota, or the number of candidates left in the count is only one more than the number of seats to be filled, in which case all remaining candidates bar one are elected without receiving a full quota.

Advantages and disadvantages of STV
ADVANTAGES of STV

The advantages claimed for PR generally apply to STV systems. In addition, as a mechanism for choosing representatives, STV is perhaps the most sophisticated of all electoral systems, allowing for choice between parties and between candidates within parties. The final results retain a fair degree of proportionality, and the fact that in most actual examples of STV the multi-member districts are relatively small means that a geographical link between voter and representative is retained. Furthermore, voters can influence the composition of post-election coalitions, as has been the case in the Republic of Ireland, and the system provides incentives for interparty accommodation through the reciprocal exchange of preferences between parties.

STV also provides a better chance for the election of popular independent candidates than List PR, because voters are choosing between candidates rather than between parties (although a party-list option can be added to an STV election; this is done for the Australian Senate).

DISADVANTAGES of STV

The disadvantages claimed for PR generally also apply to STV systems. In addition:

■STV is sometimes criticized on the grounds that preference voting is unfamiliar in many societies, and demands, at the very least, a degree of literacy and numeracy. ■The intricacies of an STV count are quite complex. This has been cited as one of the reasons why Estonia decided to abandon the system after its first election. STV requires continual recalculations of surplus transfer values and the like. Because of this, votes under STV need to be counted at counting centres instead of directly at the polling place. Where election integrity is a salient issue, counting in the actual polling places may be necessary to ensure legitimacy of the vote, and there will be a need to choose the electoral system accordingly. ■STV, unlike Closed List PR, can at times produce pressures for political parties to fragment internally because members of the same party are effectively competing against each other, as well as against the opposition, for votes. This could serve to promote ‘clientelistic’ politics where politicians offer electoral bribes to groups of defined voters. ■STV can lead to a party with a plurality of votes nonetheless winning fewer seats than its rivals. Malta amended its system in the mid-1980s by providing for some extra compensatory seats to be awarded to a party in the event of this happening.

Proportional representation related issues
Proportional Representation electoral systems require to a larger extent than other systems that the designer also considers a number of issues in addition to the choice of electoral system type. These issues will affect the results of the elections both mechanically and through psychological effects by changing the incentives for voters and political parties alike. Often, these effects will appear to be minor, and this may very well be true in practice. However, even minor differences in results can sometimes have serious implications on the setup of the legislature and the formation of government, and – perhaps most importantly – the perception of the legitimacy of the elections and the results. Also, even though many of these choices are likely to only affect the outcome slightly, some – like the choice of electoral district magnitude – will have considerable implications on the translation of votes into seats, and are thus likely to become a highly political issue. Therefore, a designer is advised to consider all these issues well in advance of an election and to be aware of the likely administrative as well as political implications the different options will have.

District Magnitude
There is near-universal agreement among electoral specialists that the crucial determinant of an electoral system's ability to translate votes cast into seats won proportionally is the district magnitude, which is the number of members to be elected in each electoral district.

Under a system such as FPTP, AV, or the Two-Round System, there is a district magnitude of one; voters are electing a single representative. By contrast, all PR systems, some plurality/majority systems such as Block Vote and PBV, and some other systems such as Limited Vote and SNTV, require electoral districts which elect more than one member. Under any proportional system, the number of members to be chosen in each district determines, to a significant extent, how proportional the election results will be.

The systems which achieve the greatest degree of proportionality will use very large districts, because such districts are able to ensure that even very small parties are represented in the legislature. In smaller districts, the effective threshold is higher. For example, in a district in which there are only three members to be elected, a party must gain at least 25 per cent +1 of the vote to be assured of winning a seat. A party which has the support of only 10 per cent of the electorate would be unlikely to win a seat, and the votes of this party’s supporters could therefore be said to have been wasted. In a nine-seat district, by contrast, 10 per cent +1 of the vote would guarantee that a party wins at least one seat. The problem is that as districts are made larger—both in terms of the number of seats and often, as a consequence, in terms of their geographic size as well—so the linkage between an elected member and his or her constituency grows weaker.

This can have serious consequences in societies where local factors play a strong role in politics or where voters expect their member to maintain strong links with the electorate and act as their ‘delegate’ in the legislature.

Because of this, there has been a lively debate about the best district magnitude. Most scholars agree, as a general principle, that district magnitudes of between three and seven seats tend to work quite well, and it has been suggested that odd numbers such as three, five and seven work better in practice than even numbers, particularly in a two-party system. However, this is only a rough guide, and there are many situations in which a higher number may be both desirable and necessary to ensure satisfactory representation and proportionality. In many countries, the electoral districts follow pre-existing administrative divisions, perhaps state or provincial boundaries, which means that there may be wide variations in their size. However, this approach both eliminates the need to draw additional boundaries for elections and may make it possible to relate electoral districts to existing identified and accepted communities.

Numbers at the high and low ends of the spectrum tend to deliver more extreme results. At one end of the spectrum, a whole country can form one electoral district, which normally means that the number of votes needed for election is extremely low and even very small parties can gain election. In Israel, for example, the whole country forms one district of 120 members, which means that election results are highly proportional, but also means that parties with only small shares of the vote can gain representation and that the link between an elected member and any geographical area is extremely weak.

At the other end of the spectrum, PR systems can be applied to situations in which there is a district magnitude of only two. For example, a system of List PR is applied to two-member districts in Chile. This delivers results which are quite disproportional, because no more than two parties can win representation in each district. This has tended to undermine the benefits of PR in terms of representation and legitimacy.

These examples, from the opposite ends of the spectrum, both serve to underline the crucial importance of district magnitude in any PR electoral system. It is arguably the single most important institutional choice when designing a PR system, and is also of crucial importance for a number of non-PR systems as well. The Single Non-Transferable Vote, for example tends to deliver moderately proportional results despite not being in essence a proportional formula, precisely because it is used in multi-member districts. Similarly, the Single Transferable Vote when applied to single-member districts becomes the Alternative Vote, which retains some of the advantages of STV but not its proportionality. In Party Block Vote and Block Vote systems, as district magnitude increases, proportionality is likely to decrease. To sum up, when designing an electoral system, district magnitude is in many ways the key factor in determining how the system will operate in practice, the strength of the link between voters and elected members, and the overall proportionality of election results.

On a related note, the party magnitude (the average number of successful candidates from the same party in the same electoral district) is an important factor in determining who will be elected. If only one candidate from a party is elected in a district, that candidate may well be male and a member of the majority ethnic or social groups in the district. If two or more are elected, balanced tickets may have more effect, making it likely that more women and more candidates from minorities will be successful. Larger districts (seven or more seats in size) and a relatively small number of parties will increase the party magnitude.

The Threshold
All electoral systems have thresholds of representation: that is, the minimum level of support which a party needs to gain representation. Thresholds can be legally imposed (formal thresholds) or exist as a mathematical property of the electoral system (effective or natural thresholds).

Formal thresholds are written into the constitutional or legal provisions which define the PR system. In the mixed systems of Germany, New Zealand, and Russia, for example, there is a 5 per cent threshold in the PR section: parties which fail to secure 5 per cent of the vote nationwide are ineligible to be awarded seats from the PR lists. This concept had its origins in the desire to limit the election of extremist groups in Germany, and is designed to stop very small parties from gaining representation.

However, in both Germany and New Zealand there exist ‘back-door’ routes for a party to be entitled to seats from the lists; in the case of New Zealand, a party must win at least one constituency seat, and in the case of Germany three seats, to bypass the threshold requirements. In Russia in 1995, there were no back-door routes, and almost half of the party-list votes were wasted. Elsewhere, legal thresholds range from 0.67 per cent in the Netherlands to 10 per cent in Turkey. Parties which gain less than this percentage of the vote are excluded from the count. A striking example of this was the 2002 Turkish election, in which so many parties failed to clear the 10 per cent threshold that 46 per cent of all votes were wasted. In all these cases, the existence of a formal threshold tends to increase the overall level of disproportionality, because votes for those parties which would otherwise have gained representation are wasted. In Poland in 1993, even with a comparatively low threshold of 5 per cent for parties and 8 per cent for coalitions, over 34 per cent of the votes were cast for parties and coalitions which did not surmount it.

An effective, hidden, or natural threshold is created as a mathematical by-product of features of electoral systems, of which district magnitude is the most important. For example, in a district with four seats under a PR system, just as any candidate with more than 20 per cent of the vote will be elected, any candidate with less than about 10 per cent (the exact figure will vary depending on the configuration of parties, candidates, and votes) is unlikely to be elected.

Open, Closed and Free Lists
While the List PR system is based on the principle that parties or political groupings present candidates, it is possible to give voters a degree of choice within List PR between the candidates nominated as well as between the parties. There are essentially three options that can be chosen—open, closed, and free lists.

The majority of List PR systems in the world are closed, meaning that the order of candidates elected by that list is fixed by the party itself, and voters are not able to express a preference for a particular candidate. The List PR system used in South Africa is a good example of a closed list. The ballot paper contains the party names and symbols, and a photograph of the party leader, but no names of individual candidates. Voters simply choose the party they prefer; the individual candidates elected as a result are predetermined by the parties themselves. This means that parties can include some candidates (perhaps members of minority ethnic and linguistic groups, or women) who might have difficulty getting elected otherwise. The negative aspect of closed lists is that voters have no say in determining who the representative of their party will be. Closed lists are also unresponsive to rapid changes in events. In East Germany’s pre-unification elections of 1990, the top-ranked candidate of one party was exposed as a secret-police informer only four days before the election, and immediately expelled from the party; but because lists were closed, electors had no choice but to vote for him if they wanted to support his former party.

Many List PR systems in Western Europe use open lists, in which voters can indicate not just their favoured party but their favoured candidate within that party. In most of these systems, the vote for a candidate as well as a party is optional and, because most voters mark their ballots for parties only rather than candidates, the candidate-choice option of the ballot paper often has limited effect. However, in Sweden, over 25 per cent of the voters regularly choose a candidate as well as a party, and a number of individuals are elected who would not be if the list were closed.

In Brazil and Finland, voters must vote for candidates: the number of seats received by each party is determined by the total number of votes gained by its candidates, and the order in which the party’s candidates are elected to these seats is determined by the number of individual votes they receive. While this gives voters much greater freedom over their choice of candidate, it also has some less desirable side effects. Because candidates from within the same party are effectively competing with each other for votes, this form of open list can lead to internal party conflict and fragmentation. It also means that the potential benefits to the party of having lists which feature a diverse slate of candidates can be overturned. In open-list PR elections in Sri Lanka, for example, the attempts of major Sinhalese parties to include minority Tamil candidates in winnable positions on their party lists have been rendered ineffective because many voters deliberately voted for lower-placed Sinhalese candidates instead. In Kosovo, a switch from closed to open lists actually enhanced the presence of more extremist candidates. On the same note, open lists have sometimes proved to be disadvantageous for the representation of women in highly patriarchal societies, although in Poland voters have shown themselves willing to use open list to elect more women than would have resulted from the nominations made by the parties if closed lists had been used.

Other devices are used in a small number of jurisdictions to add additional flexibility to open-list systems. In Ecuador, Luxembourg and Switzerland, electors have as many votes as there are seats to be filled and can distribute them to candidates either within a single party list or across several party lists as they see fit. The capacity to vote for more than one candidate across different party lists (known as panachage) or to cast more than one vote for a single highly favoured candidate (known as cumulation) both provide an additional measure of control to the voter and are categorized here as free list systems.

Apparentement
High effective thresholds can serve to discriminate against small parties–indeed, in some cases this is their express purpose. But in many cases, an inbuilt discrimination against smaller parties is seen as undesirable, particularly where several small parties with similar support bases ‘split’ their combined votes and consequently fall below the threshold, when one aligned grouping would have gained enough combined votes to win some seats in the legislature. To get around this problem, some countries which use List PR systems also allow small parties to group together for electoral purposes, thus forming a cartel—or apparentement or stembusaccoord—to contest the election. This means that the parties themselves remain as separate entities, and are listed separately on the ballot paper, but that the votes gained by each are counted as if they belonged to the entire cartel, thus increasing the chances that the combined vote total will be above the threshold and hence that they may be able to gain additional representation. This device is a feature of a number of List PR systems in continental Europe, in Latin America (where the umbrella parties are called lema) and in Israel. They are nevertheless a rarity within PR systems in Africa and Asia, and were abolished in Indonesia in 1999 after some small parties discovered that, although their cartel gained representation overall, they as parties actually lost seats.

Independent Candidates and PR systems
A common misconception is that independent candidates cannot run under proportional systems. This is not true, although most elections under List PR systems, will be carried out exclusively with candidates who belong to a political party. Under STV however, the very system is candidate centred and independent candidates are very common in elections in for example the Republic of Ireland.

Many times, an independent candidate will simply be treated as a one person party, presenting a list with only one name on it and will gain the seat if he or she receives enough votes in the election.

Mixed Systems
Mixed electoral systems attempt to combine the positive attributes of both plurality/majority (or other) and PR electoral systems. In a mixed system, there are two electoral systems using different formulae running alongside each other. The votes are cast by the same voters and contribute to the election of representatives under both systems. One of those systems is a plurality/majority system (or occasionally an ‘other’ system), usually a single-member district system, and the other a List PR system.

There are two forms of mixed system. When the results of the two types of election are linked, with seat allocations at the PR level being dependent on what happens in the plurality/majority (or other) district seats and compensating for any disproportionality that arises there, the system is called a Mixed Member Proportional (MMP) system. Where the two sets of elections are detached and distinct and are not dependent on each other for seat allocations, the system is called a Parallel system. While an MMP system generally results in proportional outcomes, a Parallel system is likely to give results the proportionality of which falls somewhere between that of a plurality/majority and that of a PR system.

Parallel and MMP systems have been widely adopted by new democracies in Africa and the former Soviet Union.

Mixed Member Proportional (MMP)
Under MMP systems, the PR seats are awarded to compensate for any disproportionality produced by the district seat results. For example, if one party wins 10 per cent of the vote nationally but no district seats, then it will be awarded enough seats from the PR lists to bring its representation up to 10 per cent of the seats in the legislature. Voters may get two separate choices, as in Germany and New Zealand. Alternatively, voters may make only one choice, with the party totals being derived from the totals for the individual district candidates.

The proportion of seats allocated according to the two elements of the system vary from country to country. Lesotho’s post-conflict electoral system contains 80 FPTP seats and 40 compensatory ones while Germany elects 299 candidates under each system.

Although MMP is designed to produce proportional results, it is possible that the disproportionality in the single-member district results is so great that the list seats cannot fully compensate for it. This is more likely when the PR electoral districts are defined not at national level but at regional or provincial level. A party can then win more plurality/majority seats in a region or province than its party vote in the region would entitle it to. To deal with this, proportionality can be closely approached if the size of the legislature is slightly increased: the extra seats are called overhang mandates or Überhangsmandaten. This has occurred in most elections in Germany and is also possible in New Zealand. In Lesotho, by contrast, the size of the legislature is fixed, and the results of the first MMP election in 2002 were not fully proportional.

Advantages and Disadvantages of MMP
While MMP retains the proportionality benefits of PR systems, it also ensures that elected representatives are linked to geographical districts. However, where voters have two votes—one for the party and one for their local representative—it is not always understood that the vote for the local representative is less important than the party vote in determining the overall allocation of seats in the legislature. Furthermore, MMP can create two classes of legislators—one group primarily responsible and beholden to a constituency, and another from the national party list without geographical ties and beholden to the party. This may have implications for the cohesiveness of groups of elected party representatives.

In translating votes into seats, MMP can be as proportional an electoral system as pure List PR, and therefore shares many of the previously cited advantages and disadvantages of PR. However, one reason why MMP is sometimes seen as less preferable than straight List PR is that it can give rise to what are called ‘strategic voting’ anomalies. In New Zealand in 1996, in the constituency of Wellington Central, some National Party strategists urged voters not to vote for the National Party candidate because they had calculated that under MMP his election would not give the National Party another seat but simply replace an MP who would be elected from their party list. It was therefore better for the National Party to see a candidate elected from another party, providing that candidate was in sympathy with the National Party’s ideas and ideology, than for votes to be ‘wasted’ in support of their own candidate.

Parallel Systems
Parallel systems also use both PR and plurality/majority components, but unlike MMP systems, the PR component of a parallel system does not compensate for any disproportionality within the plurality/majority districts. (It is also possible for the non-PR component of a Parallel system to come from the family of ‘other’ systems, as in Taiwan which uses SNTV.)

In a Parallel system, as in MMP, each voter may receive either one ballot paper which is used to cast a vote both for a candidate and for his or her party, as is done in South Korea (the Republic of Korea), or two separate ballot papers, one for the plurality/majority seat and one for the PR seats, as is done for example in Japan, Lithuania, and Thailand. Parallel systems have been a product of electoral system design over the last decade and a half—perhaps because they appear to combine the benefits of PR lists with those of plurality/majority (or other) representation.

Advantages and disadvantages of Parallel systems
ADVANTAGES of Parallel Systems

In terms of disproportionality, Parallel systems usually give results which fall somewhere between pure plurality/majority and pure PR systems. One advantage is that, when there are enough PR seats, small minority parties which have been unsuccessful in the plurality/majority elections can still be rewarded for their votes by winning seats in the proportional allocation. In addition, a Parallel system should, in theory, fragment the party system less than a pure PR electoral system.

DISADVANTAGES of Parallel Systems

As with MMP, it is likely that two classes of representatives will be created. Also, Parallel systems do not guarantee overall proportionality, and some parties may still be shut out of representation despite winning substantial numbers of votes. Parallel systems are also relatively complex and can leave voters confused as to the nature and operation of the electoral system.

Other Electoral Systems
Three systems do not fit neatly under any one of the above-mentioned categories. The Single Non-Transferable Vote is a multi-member-district, candidate-centred system in which voters have one vote. Limited Vote is very much like SNTV but gives voters more than one vote (however, unlike Block Vote, not as many as there are seats to be filled). Borda Count is a preferential system in single- or multi-member districts.

These systems tend to translate votes cast into seats in a way that falls somewhere between the proportionality of PR systems and the results of plurality/majority systems.

The Single Non-Transferable Vote (SNTV)
Under SNTV, each voter casts one vote for a candidate but (unlike FPTP) there is more than one seat to be filled in each electoral district. Those candidates with the highest vote totals fill these positions. SNTV can face political parties with a challenge. In, for example, a four-member district, a candidate with just over 20 per cent of the vote is guaranteed election. A party with 50 per cent of the vote could thus expect to win two seats in a four-member district. If each candidate polls 25 per cent, this will happen. If, however, one candidate polls 40 per cent and the other 10 per cent, the second candidate may not be elected. If the party puts up three candidates, the danger of ‘vote-splitting’ makes it even less likely that the party will win two seats.

Advantages and disadvantages of SNTV
ADVANTAGES of Single Non-Transferable Vote

■The most important difference between SNTV and the plurality/majority systems described earlier is that SNTV is better able to facilitate the representation of minority parties and independents. The larger the district magnitude (the number of seats in the constituency), the more proportional the system can become. In Jordan, SNTV has enabled a number of popular non-party pro-monarchist candidates to be elected, which is deemed to be an advantage within that embryonic party system.

■SNTV can encourage parties to become highly organized and instruct their voters to allocate their votes to candidates in a way which maximizes a party’s likely seat-winning potential. While SNTV gives voters a choice among a party’s list of candidates, it is also argued that the system fragments the party system less than pure PR systems do. Over 45 years of SNTV experience, Japan demonstrated quite a robust ‘one party dominant’ system.

■Independent candidates are easily accommodated.

■Finally, the system is praised for being easy to use and understand.

DISADVANTAGES of Single Non-Transferable Vote

■Parties whose votes are widely dispersed will win fewer seats than otherwise, and larger parties can receive a substantial seat bonus which turns a plurality of the vote nationally into an absolute majority in the legislature. These anomalies may lead to significant protests against the results and the system. Although the proportionality of the system can be increased by increasing the number of seats to be filled within the multi-member districts, this weakens the voter–MP relationship which is so prized by those who advocate defined geographical districts. Multi-member districts of up to 18 members in Thailand, for example, are at the very top end of what is manageable.

■As with any system where multiple candidates of the same party are competing for one vote, internal party fragmentation and discord may be accentuated. This can serve to promote clientelistic politics where politicians offer electoral bribes to groups of defined voters.

■Parties need to consider complex strategic questions of both nominations and vote management; putting up too many candidates can be as unproductive as putting up too few, and the need for a party to discipline its voters into spreading their votes equally across all a party’s candidates is paramount.

■As SNTV gives voters only one vote, the system contains few incentives for political parties to appeal to a broad spectrum of voters in an accommodatory manner. As long as they have a reasonable core vote, they can win seats without needing to appeal to ‘outsiders’. However, they could win more seats by wooing voters from other parties by putting up candidates acceptable to them.

■SNTV usually gives rise to many wasted votes, especially if nomination requirements are inclusive, enabling many candidates to put themselves forward.

The Limited Vote (LV)
Like SNTV, the Limited Vote is a plurality/majority system used in multi-member districts. Unlike SNTV, electors have more than one vote—but fewer votes than there are candidates to be elected. Counting is identical to SNTV, with the candidates with the highest vote totals winning the seats. This system is used for various local-level elections, but its application at the national level is restricted to Gibraltar and to Spain, where it has been used to elect the Spanish upper house, the Senate, since 1977. In this case, with large multi-member districts, each voter has one vote less than the number of members to be elected.

Advantages & Disadvantages of Limited Vote
Like SNTV, LV is simple for voters and relatively easy to count. However, it tends to produce less proportional results than SNTV. Many of the arguments relating to internal party competition, party management issues, and clientelistic politics apply to LV in a similar way as to SNTV.

Borda Count (BC)
A final—and unique—example of electoral system design is the modified Borda Count used in the tiny Pacific country of Nauru. The Borda Count is a preferential electoral system in which electors rank candidates as for the Alternative Vote. It can be used in both single- and multimember districts. There is only one count, there are no eliminations and preferences are simply tallied as ‘fractional votes’: in the modified Borda Count devised by Nauru, a first preference is worth one, a second preference is worth half, a third preference is worth one-third and so on. These are summed and the candidate(s) with the highest total(s) are declared the winners.

Advantages & Disadvantages of BC
The advantages and disadvantages of the Borda Count are similar to the ones of the other preferential electoral systems. Voters are able to express a detailed set of preferences, but on the other hand the system requires at least some level of numeracy to work, and it may be hard for voters to fully understand. The level of proportionality and number of wasted votes will depend largely on the size of the districts.

Electoral System Tiers and Hybrid Systems
Many electoral systems, both plurality/majority and proportional, have a single tier of representation: each voter in the country votes once and there is one set of elected representatives. In one-tier List PR systems, the lists may be at national level, as in Namibia and the Netherlands, or at regional level, as in Finland and Switzerland.

In mixed systems, there are usually two tiers of representatives, those elected under the plurality/majority system and those elected under the proportional system. In Hungary, however, there are three tiers: plurality/majority representatives of single member districts elected using TRS; and representatives at both regional and national levels elected using List PR.

It is also possible for an electoral system to have two tiers without being mixed in character. Two-tier proportional systems may have both national and regional lists (as in South Africa) or regional lists only (as in Denmark). In the two-tier plurality/majority system of the British Virgin Islands, there are representatives elected from single-member districts using FPTP and representatives elected from the Islands as a whole using Block Vote.

Electoral systems with two or more tiers need to be distinguished from hybrid systems, in which one part of a country elects its representatives using one electoral system, and another distinct part of the country elects representatives using a different system. In Panama, about two-thirds of the representatives are elected from multimember districts using List PR, while the remaining third are elected from single member districts using FPTP with no overlap of the two types of districts.

Considerations on Representation
Electoral regulations can affect a number of different areas relating to how the citizens of a country are represented. Many of these effects are direct results of the choice of electoral system itself, while others come from more explicit design of additional provisions in the constitution or electoral law.

Representation of Women
There are many ways to enhance the representation of women. Proportional systems tend to result in the election of more women, primarily by eliminating the disincentive inherent in FPTP systems of needing to present a single "most acceptable" candidate. Electoral systems which use reasonably large district magnitudes encourage parties to nominate women on the basis that balanced tickets will increase their electoral chances. Some List PR countries also require that women make up a certain proportion of the candidates nominated by each party.

In addition to the choice of electoral system, there are also a number of other strategies that can be used to increase the number of women representatives.

■First, there are reserved seats, where a certain number of seats are set aside for women in the legislature. These seats are filled either by representatives from regions or by political parties in direct proportion to their overall share of the national vote. Reserved seats typically exist in plurality/majority electoral systems, and are often entrenched in a country’s constitution. In India, seats on local authorities in some states are divided into three groups: at each election, only women may be nominated for one group of seats, thereby guaranteeing a minimum of one-third women elected.

■Second, the electoral law can require political parties to field a certain number of women candidates for election. This is most often done in PR electoral systems, for example in Namibia (30 per cent of candidates at the local level) and Peru (30 per cent of candidates). It is also required in the List PR component of Bolivia’s MMP system (30 per cent of candidates). However, the laws do not always guarantee that the target will be met unless there are strict placement mandates and enforcement mechanisms guaranteeing that women are placed in electable positions on party lists (i.e., positions on a party’s list that are likely, given the party’s expected number of votes, to be included in the party’s legislative delegation).

■Third, political parties may adopt their own internal quotas for women as legislative candidates. This is the most common mechanism used to promote the participation of women in political life, and has been used with varying degrees of success all over the world: by the ANC in South Africa, the Peronist Party (PJ) and the Radical Civic Union (UCR) in Argentina, CONDEPA (the Conscience of the Fatherland) in Bolivia, the Party of the Democratic Revolution (PRD) in Mexico, and the Labour parties in Australia and the UK, and throughout Scandinavia. The use of women-only candidate short-lists by the Labour Party at the 1997 UK elections almost doubled the number of female MPs, from 60 to 119. In 2004, 14 countries had quotas entrenched in the constitution, 32 countries had quotas provided for by legislation, and at least 125 parties in 61 countries had adopted their own voluntary party quotas. In terms of electoral system type, 17 countries with plurality/majority systems had quotas, and there were 15 in mixed electoral systems and 45 in PR systems. Two of the ‘others’—Afghanistan and Jordan—used quotas. Systems that guarantee women representation in the legislature vary where both their success and their consequences are concerned. For example, reserved seats may help guarantee that women make it into elected positions of office, but some women have argued that quotas end up being a way to appease, and ultimately sideline, women or to privilege the female relatives and friends of traditional male politicians rather than encouraging females to develop careers in politics, which can take many years. Since entry into politics is often done at the local level, even by male politicians, it may make more sense to institute quotas, at least initially, at the local rather than the national level.

Being elected to a legislature does not necessarily mean being given substantive decision-making power, and in some countries women legislators, particularly those elected from reserved or special seats, are marginalized from real decision-making responsibility. Yet in other countries, women have used the position afforded to them by quotas to make significant contributions to policy making and influence ‘traditional’ policy making.

Fur further details and data, see the IDEA/Stockholm University Global Database for Electoral Quotas for Women at www.quotaproject.org.

Representation of Minorities
There are also many ways to enhance the representation of minorities and communal groups. Again, electoral systems which use reasonably large district magnitudes encourage parties to nominate candidates from minorities on the grounds that balanced tickets will increase their electoral chances. A very low threshold, or the complete elimination of a formal threshold, in PR systems can also facilitate the representation of hitherto under-represented or unrepresented groups by encouraging the formation of parties specifically representing them. In plurality/majority systems in particular, seats are sometimes set aside in the legislature for minorities and communal groups.

Reserved seats can be used to ensure the representation of specific minority groups in the legislature. Seats are reserved for identifiable ethnic or religious minorities in countries as diverse as Colombia (‘black communities’), Croatia (the Hungarian, Italian, Czech, Slovak, Ruthenian, Ukrainian, German, and Austrian minorities), India (the scheduled tribes and castes), Jordan (Christians and Circassians), Niger (Tuareg), New Zealand (Maori), Pakistan (non-Muslim minorities), Palestine (Christians and Samaritans), Samoa (non-indigenous minorities), Slovenia (Hungarians and Italians), and Taiwan (the ‘aboriginal’ community). Representatives from these reserved seats are usually elected in much the same manner as other representatives, but are sometimes elected only by members of the particular minority community designated in the electoral law. This requires a communal roll (a roll of those voters who, by belonging to that particular community, are eligible to vote in that election). While it is often deemed to be a normative good to represent small communities of interest, it has also been argued that it is a better strategy to design structures which give rise to a representative legislature without overt manipulation of the electoral law or legal obligation, and that quota seats may breed resentment on the part of majority populations and exacerbate mistrust between various cultural groups.

Instead of formally reserved seats, regions can be over-represented to facilitate the increased representation of geographically concentrated groups. In the UK, Scotland and Wales have more MPs in the British House of Commons than they would be entitled to if population size alone were the only criterion. The same is true in the mountainous regions of Nepal.

Another possibility is the best loser system used in Mauritius, whereby some of the highest-polling losing candidates from a particular ethnic group are awarded seats in the legislature in order to balance overall ethnic representation.

Electoral boundaries can also be manipulated to promote the representation of particular groups. The Voting Rights Act in the United States has in the past allowed the government to draw weirdly shaped districts with the sole purpose of creating majority Black, Latino, or Asian-American districts; this might be called ‘affirmative gerrymandering’. However, the manipulation of any electoral system to promote or protect minority representation is rarely uncontroversial.

Communal Representation
A number of ethnically heterogeneous societies have taken the concept of reserved seats to its logical extension. Not only are seats divided on a communal basis, but the entire system of representation in the legislature is similarly based on communal considerations. There is a separate electoral register for each defined community, which elects only members of its ‘own group’ to the legislature.

In Lebanon, multi-member districts are defined, in each of which an allocation of seats between confessional groups is determined. Representatives are elected by Block Vote from communal rolls separately to the seats allocated for each confessional group. In Fiji, electors are able to vote both for their own communal candidates and for candidates in ‘open’ districts.

Most communal roll arrangements were abandoned after it became clear that communal electorates, while guaranteeing group representation, often had the perverse effect of undermining the path of accommodation between different groups, since there were no incentives for political intermixing between communities. The tasks of defining a member of a particular group and distributing seats fairly between them were also full of pitfalls. In India, for example, the separate districts which had existed under colonial rule for Muslims, Christians, Sikhs and others were abolished at independence, although some reserved seats remain in order to represent the scheduled tribes and castes. Similar communal roll systems used at various times in Pakistan, Cyprus, and Zimbabwe have also been abandoned. Despite a controversial history of use, Fiji continues to elect part of its legislature from separate communal rolls for indigenous Fijian, Indian, Rotuman, and ‘general’ electors.

While some communal roll arrangements give the task of determining who falls into which category to some form of registration body, others give this choice to the individual. The predominant example of a communal roll system still in place among contemporary democracies is the optional separate roll for Maori voters in New Zealand. Maori electors can choose to be on either the national electoral roll or a specific Maori roll, which now elects seven Maori representatives to the legislature. The results of New Zealand’s first PR elections since 1996 could, however, be said to have weakened the rationale for the communal system: twice as many Maori representatives have been elected from the general rolls as from the specific Maori roll.

The timing of Elections
Elections, whether they be for national, executive, legislative, state-wide, or local bodies, may not necessarily be held on a single day (or specific days) but can instead be staggered. The reasons for separating elections over a significant period of time can be both practical and political. Staggering of elections usually occurs when there are major logistical preparations involved (e.g. elections to the lower house of India, the Lok Sabha, take place on four days over the course of several weeks due to the need to redeploy security forces to cover all of the polling places) or when security concerns require it. Administrative and security considerations mean that it is far easier for the Indian Electoral Commission to sequence the holding of legislative votes across both time and states. The difficulties facing staggered elections include ballot security; in order for areas voting later not to be influenced by areas voting earlier, ballot papers need to be held at a secure centralized point until all voting has taken place, when all votes can be counted at once.

More common is the staggering over time of presidential, legislative, and federal state elections. There is evidence to suggest that holding presidential and legislative elections on the same day can advantage the president’s party, and can make executive-legislative fragmentation less likely and thus make government more coherent—especially in embryonic democracies. However, if there is a desire to accentuate a separation of powers or there are logistical capabilities to consider, then it may be necessary or prudent to separate presidential and legislative elections.

Remote Voting
Remote voting is used in many countries, both old and new democracies, around the world, to broaden participation. Remote voting may take place in person somewhere other than an assigned polling station or at another time, or votes may be sent by post or cast by an appointed proxy. When the requirements to qualify as a remote voter are minimal, remote voting can make up a significant proportion of the total vote. In Finland, it has been as high as 37 per cent of all votes cast, and in the 2003 legislative elections in the Marshall Islands it was 58 per cent. In Sweden, where it is commonly about 30 per cent, voters can also change their pre-cast vote if they subsequently travel to their allocated polling station on election day. However, its use may have implications for electoral system design, with issues of election integrity being salient.

Remote voting is easiest to administer under a nationwide List PR system with only one list per party, and most complicated under a system using single-member districts, which would require placing ballots from many constituencies in many different locations. Particularly if out-of-country voting is to be implemented, the practicalities of getting the right ballot paper to each elector need to be considered carefully. Requiring a country’s embassies to issue ballot papers may not sit easily with a system with a significant number of electoral districts, because of the logistic challenge of ensuring that each embassy receives the right selection of ballot papers and gives the right ballot paper to each elector. If ballot papers are to be despatched by post, there will be an impact on the election timetable in that the ballots will have to be readied farther in advance of election day.

Once cast, out-of-country votes can be included in the absentee voter’s home district (as in New Zealand); counted within single (or multiple) out-of-country districts (as in Croatia); attached to one or more particular districts (as in Indonesia); or merely added to the national vote totals when seats are allocated under a nationally based List PR system (as in The Netherlands).

Turnout Issues
There is an established relationship between the level of turnout in elections and the electoral system chosen. PR systems are in general linked with higher turnout. In plurality/majority systems, turnout tends to be higher when national election results are expected to be close than when one party looks certain to win, and also higher in individual districts where results are expected to be closer.

As a measure to improve electoral legitimacy, some countries, notably several of the post-communist former republics of the USSR, introduced mandatory minimum turnout levels: if the turnout in an electoral district did not reach, for example, 50 per cent, the election would not be considered valid. However, the use of mandatory turnout levels can create administrative nightmares if repeated elections consistently fail to achieve the required turnout levels, leaving electoral districts in limbo. Ukraine, for example, abolished mandatory turnout provisions for the 1998 elections after the experience of repeated by-elections failing to reach the required turnout in 1994.

Several countries address the issue of participation by using compulsory voting, including Australia, Belgium, Greece and many countries in Latin America. Many other countries, however, reject compulsory voting on principle. While it is probably equally compatible with any electoral system, its use can be considered simultaneously with other turnout-related issues.

For charts, tables, and additional information, please go to http://aceproject.org/ace-en/topics/es/onePage