The first thing to do when considering a posit like this is to break it down, into its constituent parts. As such, this question naturally has two main elements; does proportional representation (PR) encourage increasing numbers of political parties (as opposed to other electoral systems) and does the use of PR increase the relative instability of the state in which it is used?

As we investigate these elements we will see that there exists evidence for and against these suppositions and that the only possible conclusion of any validity is that PR does not necessarily lead to a proliferation of parties, nor does it necessarily lead to political instability. PR does, however, mean that smaller parties and coalitions are more likely to witness electoral success than in other such systems. Evidence also suggests that plurality voting does not necessarily lead to two-party systems, nor does it necessarily lead to ‘stable’ government.

What we must do now is look at the terms which we shall be using and determine their definitions. This is necessary in all investigative works as it allows for researchers and readers to coalesce on the relevant subject matter and clarify the direction of the investigation.

To that end, let us consider the first term, proportional representation (PR). PR is an electoral system that is present in about 60% of today’s democracies and is widespread throughout Europe and Latin America.1 In its absolute form, PR has the electorate cast a vote, for a party representative (or party list, in a multi-member district) which, according to the distribution rules, is then used to determine that party’s ‘proportion’ of the seats available.2

The traditional alternative to PR systems is the plurality or majority voting system, where the electorate votes for representation on a territorial basis, rather than a universal party delineation. This form of representation is thought to produce clear and stable governments and is utilised in the USA, UK and Canada.3 These systems are used as they are the simplest to understand and are often referred to as ‘first past the post’ systems; relating to the achievement of the entirety of the electoral spoils.

Some of the operational parameters of the political system of a state are determined by the electoral system employed. They fulfil a number of important functions, the primary one being to legitimise the state and allow the efficient operation of its political machinery. Electoral systems also affect the level of representation experienced by society at large, the number of political parties in government, the speed of hand-over and other such functions.

The classification of an electoral system can be done on many grounds, but this essay shall concentrate on the output method, referring to the outcome of the election; proportional or non-proportional. This is because the question is naturally loaded to this distinction and it is simple to investigate and analyse.

Plurality voting systems fall under three main categories; first past the post (FPTP); majority voting (MV) and alternative vote systems (AV). First past the post systems simply require a party, or its representative, to acquire more votes than their opposition. This is the most simple of voting procedures, but is open to criticism for its lack of representation; a candidate could win a seat with a small percentage of the overall vote.4

Majority voting requires a candidate to acquire a majority of the votes, i.e. above 50% of those cast, in order that they may be successful. The alternative vote system allows voters to give the alternative candidates a ranking in accordance with their preferences, i.e. first preference, second preference and so on – this gives a greater level of representation and is often seen as the bridge between PR and plurality voting systems.

Proportional methods can also be identified by three main categories; the party list (PL); the single transferable vote (STV) and the additional member system (AMS). The party list system sees a party ‘list’ its candidates in order of preference. These candidates are later elected according to the ‘proportion’ of votes the party receives.5

The single transferable vote system allows voters to list their candidates in order of choice, in a similarly way to the AV system, the subsequent choices are then taken into account when collating all votes.6 The additional member system ties in elements of MV and PL methods; voters have two votes, the first elects a representative for a region and the second goes on a party list, which is used to allocate additional representatives.7

If we now consider the first part of the question, PR encourages the proliferation of parties, we can find examples that both support and counter it. It was not until the beginning of the 20th Century that PR was developed and introduced and the adoption of it across Europe sparked many writings. It was Rokkan who argued that PR was adopted ‘through a convergence from below and above. The rising working class wanted to gain access to the legislatures and the most threatened of the old-fashioned parties demanded PR to protect their position against the new waves of mobilised voters created by universal suffrage.’8

The ethnic and religious divisions in Belgium and Sweden meant that they were keen to develop these ideas and late in the 19th Century they emerged as the pioneers of PR. Electoral reform was called for across many other European states, but it was Belgium and Sweden’s enthusiasm towards PR and their belief that it could adequately represent its diverse populace that most influenced its adoption.

Rokkan argued further that the existence of multi-party systems had long preceded the introduction of PR. He draws on the examples of Belgium, Denmark, Germany and Norway, suggesting that PR could have been introduced to ratify this diversity in a more easily manageable format.9

Contrary to this, Lipjhart claimed that parliaments elected under PR systems contain, on average, more parties than those elected under non-proportional representative systems.10 This is because PR is based on the premise that parties are represented according to their popularity with the electorate.

Duverger argued that single-member, simple plurality systems naturally lead to a two-party system. He claimed that the voter did not choose openly from those parties available, but picked the one that was closest fit and who had a chance of electoral success, thus ensuring that their vote amounted to ‘something’. Riker further supported this theory, suggesting that parties who obtain consistently large shares of the votes will continue to do so under this system. The electorate is, therefore, picking between two (or sometimes three) main parties, hence the term ‘two-party system’.11

However, Riker further highlighted the inconsistency with this theory, bringing up the example of India, where, until 1977, a two-party system was not evident despite having a plurality electoral system. Instead, India witnessed a dominant party system. This undermines the theory put forward by Duverger, but is explained away by Riker, who suggests that this anomaly was down to the structuralisation of voter preferences.12

It should be further noted that after 1977 the previously fragmented opposition to the dominant Congress Party formed a semi-cohesive coalition and changed the hitherto stable system into a new era. Neither of the first two largest parties was able to form a government and a coalition of 13 contrasting parties was formed. This led to a period of instability and uncertainty and, again, counters Duverger’s suppositions.

It is therefore fallacious to suggest that PR will always result in the proliferation of parties. It is, likewise, untrue to state that non-proportional systems will always result in a two-party system. There will always exist examples that counter such generalised statements, e.g. Canada, where a FPTP system is utilised, but often yields many more than two main parties.13 Similarly, the UK is often used as an example of a two-party system resulting from the FPTP voting procedure, yet it is untrue to classify as thus. The rising fortunes of the Liberal Democrats, the acceptance of Sinn Fein into Parliament and the ramifications of party sleaze14 have all contributed to a more diverse make-up of the party political landscape.

Thus, the electoral system does not necessarily influence the number of parties present; it is something that is more easily explained in terms of social, religious and ethnic divisions, also known as cleavages. These cleavages result in the emergence of political movements as opposed to impacting on the electoral system as a whole. While, in general, one may comment that multi-party systems are more likely to be evident under PR voting procedures, it is not an axiom for the political structure of the state.

Let us now consider the second half of the original posit; does PR increase instability? There exists a general consensus that this is the case; small, extremist parties are able to obtain a share of the power and as such destabilise the core through their attempts to be legislated for. This is exemplified by the decision of the Austrian People’s Party to include Jorg Haider’s far-right Freedom Party in its 2000 election campaign.

As a result of this radical example many EU member states froze diplomatic relations with Austria, demanding Haider’s resignation before they would resume such activities. In this instance, the PR system of Austria allowed Haider to make a claim to power and it is thus argued that PR makes states more unstable. However, if we consider the case of France and the success of Jean-Marie Le Pen we can see an altogether more rounded stand-point. As the leader of the far-right National Front, Le Pen won approximately 15% of the votes in France in 2000, despite the voting system being organised along majoritarian lines.

The French example is rare, but serves to note that given enough support, instability and extremism will develop, regardless of electoral structure.15 For this reason many countries have developed the idea of a voting ‘threshold’, below which none of the votes cast for you or your party will count. Extremist parties rarely receive enough votes, set by the specific state, to gain a share of power. In Denmark the threshold is set at 2%; Sweden’s 4%; Germany’s 5% and Turkey’s 10%.16

The threshold, as seen above, varies greatly from state to state and is set on either ‘effective’ or ‘formal’ grounds. The former relates to the reactionary setting of the threshold, based on the number seats to be filled, the number of parties and/or candidates and other such factors. The latter refers to states whose thresholds are set through electoral law. In most of the PR states witnessed today, the threshold has served to limit the number of parties involved in governance to a reasonable level and to deter the proliferation of extremist parties.17

By extension, we could assume that this means PR would naturally lead to instability if it were not curbed and legislated against. However, the definition of political instability is not as simple as the number of parties involved in the political arena, it may be interpreted in many different ways; the longevity of governments in office; the prevalence, or otherwise, of coalitions; the levels of civil unrest and disobedience; the size of the police force and many more.

For the purpose of this essay one should consider political stability to refer to the duration of a governmental office. At first glance this would be easy to judge, as there are countless statistics available across the globe for this material. However, it is not so. For this reason, we shall consider the duration of government to encompass the two concepts of Lijphart and Dodd – cabinet duration. In this way, we recognise that a) cabinets change only if the composition of the party changes and b) cabinets are changed when elections arise, Prime Ministers or Presidents change, etc.18

Taking this into account we can see that the type of electoral system employed will have wide ramifications on the political composition and stability of the state. The level and nature of choice for the electorate is affected, which has knock-on effects for the politicians. As such the second half of the original posit becomes more complex and it should be evident that evidence exists to both prove and disprove its validity.