This paper represents the first in a new series of publications on issues in parliamentary practice from the European Centre for Parliamentary Research and Documentation (ECPRD).

Electoral systems in Europe: An overview is an examination of the rules and mechanisms used in democratic elections across the parliaments of Europe. Its priority, as with the other studies in this series, is to provide an easy to consult and accessible introduction for both public service practitioners and members of the public on the predominant issues in parliamentary activity.

It is often the case that basic yet essential information on the workings of parliaments is diffuse and difficult to track down. This ECPRD paper brings together such information, allowing the reader to become knowledgable about European electoral systems, and elements of the political debate surrounding them.

The European Centre for Parliamentary Research and Documentation is a cooperative body under the aegis of the European Parliament and the Parliamentary Assembly of the Council of Europe. It is primarily an international network of research departments and parliamentary libraries but also involves other officials responsible for information gathering and dissemination. Its aim is to facilitate contacts and exchanges between the officials of member parliaments to the mutual benefit of all.

Created in 1977 by the Conference of Speakers of European Parliamentary Assemblies, which delegated to the Presidents of the European Parliament and the Parliamentary Assembly of the Council of Europe its formation and management, it is comprised of 41 member countries and 7 guest countries, allowing 67 parliamentary chambers to cooperate within the ECPRD.

The main activities of the ECPRD are seminars hosted by member parliaments on subjects of relevance to parliamentary officials. The two main emphases in recent years have been, broadly, how parliaments should react to the challenges of the new technology, and parliamentary management and democratisation. In addition, there are working groups which meet annually to consider the technical aspects of information and communications technology, and on macro-economic policy.

The Secretariat of the ECPRD is based in the European Parliament in Brussels. It is staffed by permanent officials of the Parliament, and acts as a clearing house for information requests for comparative research. It also manages the ECPRD website at http://www.ecprd.org.

This paper is available in English only and was edited by Simon McGee under the supervision of Adam Isaacs. The text was completed in October 2000.

The opinions expressed in this paper are the sole responsibility of the authors and do not necessarily represent the official position of the ECPRD.

Reproduction and translation for non-commercial purposes are authorised provided the source is acknowledged and the publisher is given prior notice and sent a copy.

Publisher: Dick Toornstra

ECPRD

c/o European Parliament

B - 1047 Brussels, Belgium

Preface 2

Introduction 4

PART ONE: PRINCIPAL ELECTORAL SYSTEMS EXPLAINED 6

Majoritarian systems 8

Proportional systems 9

Mixed systems 14

PART TWO: NATIONAL ELECTORAL SYSTEMS IN ACTION 16

EU Member States 18

EU Applicant Countries 30

European ECPRD countries 38

PART THREE: ELECTORAL SYSTEMS FOR THE EUROPEAN PARLIAMENT 47

Towards a uniform electoral procedure 48

The Anastassopoupos Report 50

Current arrangements for European elections 52

PART FOUR: APPENDICES 59

Glossary 60

References 63

Article 138 of the EC Treaty 66

Electoral Turnout tables 67

Elections lie at the heart of representative democracy, enacting citizens' rights to have a say in by whom they are governed. The electoral process is the ultimate symbol and act of modern democratic societies: "democracy's ceremonial; its feast, its function", H.G. Wells called it.

Particularly in the context of the ECPRD and the recent establishment of democratic systems to many of its constituent parliaments, the importance of democratic elections across Europe is paramount; a message consistently stressed by the European Union. The Copenhagen Summit in 1993, which laid out the basic conditions which any prospective European Union applicant country would need to establish to gain 'EU applicant status', represented a firm recapitulation of the EU's commitment to the principles and ideals of modern democracy. At the top of this list of conditions stood the need for former Communist countries to embrace democratic principles should they wish to join the EU: any "candidate country [must have] achieved stability of institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities". This commitment in Copenhagen was later enshrined in the Amsterdam Treaty of 1997, but was already being reflected in the institutional reform that was taking place in Central and Eastern Europe at the time.

Once these countries had committed themselves to building democratically elected representative chambers, the question of how parliamentary seats should be filled seemed to re-ignite interest among politicians, academics and others across Europe, on the subject of electoral systems. That is not to say, however, that recent scrutiny of electoral systems is solely confined to, or motivated by concerns springing from, the birth of the new democracies. On both a European Union level and on a Member State level there is, and has been for some time, much debate on electoral reform. Proposals calling for the establishment of a uniform cross-Community electoral procedure for elections to the European Parliament, which is examined here, have been numerous. Member States have also been looking towards adapting their existing rules and mechanisms. The United Kingdom's traditionally majoritarian electoral systems, for example, have been the subject of scrutiny, and in some cases change, in an effort to inject some proportionality into the system. Similarly, the Netherlands' proportional national electoral system is being looked at to see if it can accommodate elements which will enable the Dutch legislature to establish closer contacts to the electorate..

This paper serves as an introduction to the subject of electoral systems, and gives an overview of not only what the main systems are and where they are practised across Europe, but also an indication of the arguments and issues surrounding them. In particular, this paper looks at "the core debate": the question of whether electoral systems which prioritise government effectiveness and accountability should take precedence over systems which prioritise the creation of an accurately representative parliamentary body.

The paper's focus is rigidly on the crucial nature of the electoral system, which is widely defined as a body of legal rules, regulating the right to elect, the right to be elected, the general voting principles, the overall methods employed in organising and undertaking the elections and the principles of correctly establishing the results of the election process. It should be stressed that other election issues, despite being very closely connected to the functioning of elections - such as campaign funding and election advertising rules - are not featured in this paper. It is concerned with the core mechanisms, of how they collect and distribute popular support to shape the political landscape of parliaments and the nature of governments; concerns which make the issue of electoral systems one of the hottest in today's political arena.

This paper is structured in four parts. Part One gives an introduction to the main categories of electoral system – and the varieties within them – outlining how they operate and commenting on aspects of the debate surrounding them. Part Two provides basic details about the electoral systems of all European ECPRD member parliaments. These are categorised into: systems for national elections in EU Member States; systems for national elections in the EU 'Applicant Countries'; and systems for national elections in the remaining European ECPRD states. Part Three gives an outline of the systems currently used by EU Member States for elections to the European Parliament, and also addresses the pressure which has been building in favour of the establishment of a uniform electoral procedure across the EU for these elections. Part Four includes other relevant information and also seeks to aid further study of electoral systems by providing references to a variety of publications, articles and websites.

Dick TOORNSTRA

Co-Director ECPRD

No consensus exists among public servants, academics or observers as to which system of carrying-out elections can be said to lead to the most fair and most effectively shaped representative democracy. The considerable volume of work devoted to electoral systems and the existence of such a variety of popular vote-distributing mechanisms is an accurate reflection of the fact that there are so many well-constructed and well-argued systems in existence.

All electoral systems seek to take account of at least some or all of the following features. Which of these is deemed most important tends to shape people's preference of electoral system.

The aim of this section is to outline the core principles and debates surrounding the main varieties of majoritarian, proportional and mixed electoral systems. It by no means serves as an exhaustive guide, but it does provide sufficient background information to enable a good understanding of the terminology and systems featured in the electoral system tables of Parts Two and Three.

Majoritarian, or plurality, systems represent the oldest and simplest electoral system category, based on the principle that whichever candidate receives the most votes in a constituency is deemed elected. Its prime concern is with the creation of effective government. The following three varieties of majority systems operate on the basis of single-member constituencies:

Supporters of majoritarian systems of one shape or another broadly point to the following range of reasons why they are preferable to alternative systems:

- Majoritarian systems are simple. They do not require complex mathematics to produce their results and encourage transparency, since votes are easily translated into parliamentary seats. Many proportional systems, such as Single Transferable Vote, are unattractive in that they involve very lengthy and complex calculations.

- The 'winner-takes-all' philosophy behind these systems focuses on producing a clear, strong and stable body of representatives and government, not on mirroring the make-up of the general populance. Because of the way that majoritarian systems disproportionately reward large parties over small parties, the leading party is rarely required to build a coalition to create a government. A one-party government (which, one should remember, is already a complex coalition of interests) is therefore free of the problems that can emerge from being involved in a coalition which might make it weak: government is immediately broadly united; policy need not be diluted as it might in a coalition where a partner party may demand policy concessions; ministerial portfolios can be distributed without fear of party political battles or major policy cleavages (although other battles will still rage); the identity and message of a party in government need not be confused, or aggravated, by its close association with another party.

- Just as a majoritarian system tends to create strong government, it also tends to lead to a strong alternative party, such as in the UK and the USA. This creates a dynamic political arena where government needs to work hard because there is always a viable and strong alternative which the public can chose to replace them with. Philip Norton states that the British two-party system in the House of Commons has particularly strong benefits for the scrutiny of government: "opposition can be seen as having co-equal rights in terms of time during debates, and by having the opportunity to question government and participate in debate, can force the government to answer and justify its actions".

- The fact that smaller parties are not effectively represented in majoritarian systems can be seen to have some advantages. They encourage minority groups to integrate into the larger parties, which is desirable both for the minority group (which can gain the political support for some or all of its concerns) and the large group (which gains the electoral support of the minority). Different groups are brought together by political need which encourages cooperation and the creation of parties with 'broad church' appeal, rather than division by electoral systems which may only serve to highlight differences and keep minorities at a distance. It can also serve to discourage extremist minority parties from gaining parliamentary representation.

- Single-member constituencies which operate in majoritarian systems also encourage the linkage of an elected representative to a particular geographic area and the residents of that area, who become his or her constituents. This link encourages those elected to prioritise serving their constituents above satisfying the party hierarchy, since their re-election will depend entirely on those same people. This link is particularly important in the UK, where MPs have always been viewed as representatives of a specific area, not of the UK as a whole. Individual citizens feel that they are part of the democratic process through a single representative, who is there to help with individual or community casework. The election of representatives by a proportional system in a multi-member constituency (which may even be the entire country) would not determine this link.

While majoritarian systems provide the oldest model for electoral systems, proportional representation (PR) systems are currently the most widely used in Europe. Its focus is on the creation of a parliamentary chamber which accurately reflects the diverse make-up of an electorate. The two main PR system categories are Party List and Single Transferable Vote. Both of these carry a range of considerations in creating representative parliaments.

The size of the constituency (also refered to as 'district magnitude') is important since it can affect how 'accurate' the results reflect a general electorate. 'Full PR' considers the whole country as a single constituency, with seats allocated on a pro rata basis according to the number of votes cast. This is considered to yield the most 'pure' form of PR. 'Limited PR' sees elections taking place in several constituencies, and is considered to not produce as accurate a snapshot of the population, since the greater the number of constituencies (and hence the fewer the number of seats available per constituency), the harder it is to ensure complete proportionality.

As will become very obvious in Part Two, many systems used include a threshold. This aims to reduce the extent to which proportionality is taken. By putting in place a minimum level of national support required for a political party to be allowed to gain parliamentary representation, it limits the ability for very small parties to gain representation. It is understood that the electoral threshold had its origin in inter-war Germany, as a result of the Weimar Republic's concern with extremist groups. Some thresholds are nominal, others are so large that it challenges the basic idea behind using a PR system. In some countries, political parties not reaching the thresholds and getting no seats can potentially waste millions of votes. Another requirement which some countries (many in central and eastern Europe) also have in place is a minimum quorum level, which demands that for an election to be deemed valid and fully representative the turnout must meet or exceed a percentage of the electorate.

A Party List system presents multi-member constituency electorates with political parties putting forward slates or lists of candidates. It represents the principal PR system in operation, although there are many variations of it, based on constituency size (as covered above), thresholds and quorums (also examined above), rules about whether the electorate can have preferences for individuals on party lists, whether the Greatest Remainder or Highest Average systems are used and which formulas are used within them.

Before looking at the Greatest Remainder and Highest Average systems, one must cover perhaps the most important variant in the Party List system: the ability of the voter to influence the party list. The ability of the voter to influence the party lists acknowledges that voters will often have preferences for individual candidates within party lists and that they might also wish to satisfy those preferences regardless of what the party hierarchies have put in front of them. The extent to which a voter can have choice in a Party List PR system is determined by one of the following variations. Preferential Voting allows voters to decide their own order of preference, different from that indicated by the party. The voters may only, however, vote for one list. Vote-Splitting allows voters to select candidates from competing lists, and thus draw up their own list. The commonest used variation is the Closed List system, where the electorate are simply allowed to vote for one party list, not an individual. It is up to the party to decide who should get the chance to fill the seats given to them (usually determined well in advance with the candidates placed in order of priority).

The two main Party List PR seat-distribution systems are Greatest Remainder (often known as Largest Remainder), where candidate parties have to receive a quota of votes determined by a particular quota formula to gain a seat, and Highest Average, in which parties must have the highest averages after being divided by a particular formula. Although some academics have tried to show that one system is 'more proportional' than the other, Arend Lijphart quite rightly states that neither of the two systems yields more proportional systems than the other: it is merely up to which of the methods (i.e. which quota method or divisor) is chosen.

The Greatest Remainder variety of Party List PR allocates seats in two stages and uses a quota formula. The first stage establishes an 'electoral quota' to determine how many votes each party list must receive for a candidate to be allocated a seat. Each party list then receives as many seats as the number of times it satisfies the electoral quota. One of four types of quota formula systems can be used, and each can produce different results.

Seats 5

Seats+1 5+1

Since the number of seats allocated across the party lists will never represent the total number of votes obtained by a party, there are always votes and at least one seat left over. These votes are called remainders, and are the subject of a second stage which ensures that any remaining seats are distributed fairly. The quota system is abandoned, and seats are distributed in order of which party lists have the most remaining non-allocated votes.

The Highest Averages system is more commonly used than the Greatest Remainder system. It successively divides the number of votes received by a party list by a series of divisors. Seats are then distributed to the lists which secure the highest resulting quotient until all seats have been allocated. Three divisor fomulas predominate:

PARTY

A B C D

Votes (100): 50 32 13 5

Divided by 1 50 32 13 5

Divided by 2 25 16 6.5 2.5

Divided by 3 16.7 10.7 4.3 1.6

Divided by 4 12.5 8 3.3 1.3

Seats won: 3 2 0 0

PARTY

A B C D

Votes (100): 50 32 13 5

Divided by 1 50 32 13 5

Divided by 3 16.7 10.7 4.3 1.7

Divided by 5 10 6.4 2.6 1

Divided by 7 7.1 4.6 1.8 0.7

PARTY

A B C D

Votes (100): 50 32 13 5

Divided by 1.4 35.7 22.8 9.3 5

Divided by 3 16.7 10.7 4.3 1.7

Divided by 5 10 6.4 2.6 1

Divided by 7 7.1 4.6 1.8 0.7

Single Transferable Vote, or STV, is the other main PR system in use, and also represents the main other commonly-used Preferential Voting system in addition to the majoritarian Alternative Vote system, requiring constituents to express a rank order of preferences for the candidates standing in multi-member constituencies. Used in Ireland and Malta, its philosophy runs totally contrary to that of the Party List system, since it "aims to minimise the influence of party in the election of MPs". It also seeks to solve the issue of 'wasted votes', which are produced by majoritarian systems like FPTP. A candidate is elected as soon as he reaches the electoral quotient calculated by the Droop quota (or occassionally one of the another quotas used in Party List PR). The additional votes are then redistributed to the other candidates on the basis of second choices made on the ballot. The same operation is carried out in the case of the candidate who polled the fewest votes, who is eliminated. If there are still seats to be filled after the second count, the process continues.

As the most-practised electoral system category in Europe, PR has a large number of supporters and perceived advantages; the following represent the main arguments for PR:

- Results of an election conducted under a PR system more accurately reflects the way that a electorate would have voted than a majoritarian system. This means that the more anomalous and what are often considered "unfair" results that are occasionally thrown up are avoided. An example of this is highlighted by Pippa Norris in the shape of the British General Election in 1951, where the Conservative Party won the election (operating under First-Past-The-Post) with a sixteen seat parliamentary majority, having gained a 48% share of votes, despite the fact that the Labour Party amassed 48.8% of the vote.

- It is much easier for smaller political or ethnic parties to gain representation. In most cases, just over 5% of the vote is sufficient to get into the parliament. This creates a truly representative assembly, where the diversity of the country is mirrored at the highest level. This is a particularly important issue in central and eastern Europe, and some countries have even gone so far as reserving seats for national minorities; in Slovenia two parliamentary seats are reserved for Italian and Hungarian minorities.

- The number of 'wasted votes' - in other words votes which do not go towards electing any candidates - is very small. Only votes for those candidates who fail to reach the threshold can be said to be wasted. This helps to reaffirm to voters that their participation will more than likely 'make a difference'.

- PR systems do not seem to discourage voting. Despite arguments claiming that PR discourages electoral participation, due to the fact that it uses often-complex calculations to distribute seats, they do not seem to hold up if the electoral turnout of the most recent national elections in EU Member States is examined. Despite the simplicity of the UK's First-Past-The-Post system (which yielded a 71.6% turnout in the 1997 general election), many other EU countries using PR have managed to deliver higher turnouts (including Belgium, Denmark and Luxembourg which each had over 85% of the electorate participating in their most recent national elections) using variations of PR.

- The experience of PR has been that it has increased the number of women in parliaments. Some political parties have taken affirmative action and drawn up official targets for the proportion of women and those from ethnic minorities on their party lists. Defenders of such targets argue that although artificial, its aim is to temporarily speed-up change to the status quo, which can often only be done by such targets in closed PR systems.

- PR tends not to result in one party having an absolute majority of the votes, thus making coalition government the norm. This leads, in theory at least, to a more consensual and inclusive style of government.

There are a range of systems which try to strike a balance between majoritarian and proportional representation systems, and are generally (if rather untidily) categorised as Mixed Systems. These can broadly be divided into those which try to bring together elements of majoritarian and proportional systems to try and come up with a seat-distributing mechanism incorporating the best of both world (but end up tending to lean more towards one system or the other), and systems where both majoritarian and proportional mechanisms are used in different stages.

These varieties are from the former, and slant towards majoritarian systems:

This Mixed System variety leans more towards PR:

These systems, however, use separate majoritarian and proportional mechanisms for different constituencies:

For details of information sources on electoral systems, consult References in Part Four.

This section details the electoral systems and rules for the appointment of seats to every national parliament in Europe. For ease of use, the information in this section has been divided-up into three categories: the first has information on the national election systems operating in EU Member States; the second covers the national election systems of EU Applicant Countries; and the third covers national election system information for the remaining European ECPRD countries.

Each category includes a synoptic table, enabling the reader to compare elements of the different electoral systems from a one page summary, and detailed tables. The information within these detailed tables provides the following on every country: the name of chamber or chambers including corresponding website addresses (or general telephone number for parliaments without websites); the number of seats distributed; details about constituency boundaries; the length of mandate; details of the electoral system used for distributing seats; the rules for filling vacancies; rules on who is entitled to vote; rules on who is eligible to stand for election; and rules surrounding the nomination of candidates.

Details about the electoral law in each country are not included in the tables for reasons of space, but are available from the International Foundation for Electoral Systems website at http://www.ifes.org/electionlaws.htm

For ease of reference and comprehension, all principal electoral system terminology is also included in the Glossary, Part Four.

SYSTEMS FOR NATIONAL ELECTIONS IN EU MEMBER STATES

The systems used for parliamentary elections in the fifteen member states of the European Union present a varied picture, each products of different political circumstances and traditions. Bicameral, unicameral, majoritarian, proportional, and mixed systems are all well represented in the EU, displaying a diversity which is often augmented by certain aspects of the electoral law not concerning the vote-distribution mechanisms. Compulsory voting rules in Belgium and Greece, for example, sit alongside the less demanding requirements of other Member States.

Nine of the EU Member States have bicameral parliaments, while six are unicameral. In the bicameral systems, the Lower House is the most important, the result there determining the make-up of the government. A wide range of systems are in use for parliamentary elections, although proportional systems dominate. The UK uses the FPTP system in single-member constituencies. France also has single-member constituencies, but uses the two-stage Second Ballot Majority Runoff system, where an absolute majority is required to gain a seat. Spain, and Italy use the Additional Member system, while Germany uses the Mixed Member Proportional system. Finland's system is principally PR, but the representative for Åland is chosen by simple majority. The other nine member states use a variety of PR systems. The size of the constituencies may vary, and a range of different formulae for determining seat distribution are used. Some systems include a nationwide minimum popularity threshold for representation in parliament - 2% in Denmark, up to 5% in Germany and Luxembourg.

Most of the Upper Houses are chosen by appointment or indirect election, while three - Belgium, Spain and Italy - have directly elected elements. Belgium uses the same proportionality-based voting system as for the Lower House, but the representatives are chosen along the lines of the regions and language communities. Italy uses a mixed system for the Upper House, most seats being filled by simple majority voting, the rest by proportionality. Spain's Upper House has most places filled by simple majority voting, but about a sixth of seats are indirectly chosen by regional assemblies.

The indirectly appointed upper chambers have a variety of different methods for choosing members. In four of the countries - Germany, Austria, Holland and Spain - the Upper House is made up of members of regional or local assemblies. In France, the Senate is made up of people chosen by local electoral colleges. In Ireland, appointments to the Upper House are made in part by the Taoiseach (Prime Minister) and by indirect elections by some universities, county councils, panels representing different interests in society or the outgoing House. In the United Kingdom, there are four groups of members: hereditary peers (aristocracy), life peers, Law lords, and certain bishops. The government recently started a two-stage reform of the House of Lords, removing the right to sit for all but 92 hereditary peers in an "interim" house. Stage two will produce a final dispensation, following the publication of the report from a Royal Commission to consider the composition, roles and functions of the second chamber.

The majority of the electoral systems operating in the thirteen EU Applicant Countries are relatively young – in terms of how long they have been operating for – when compared to those which have been operating in most of the Member States of the European Union. However, as can be seen in the following tables, there is a large degree of similarity in the models used by the applicant countries, most of which broadly follow the EU Member State preference for proportionality.

The Lower Houses in most of these countries are elected by a form of PR, although Hungary uses the Mixed Member Proportional system and Lithuania uses the Additional Member system. Out of all the Applicant Countries, only the Czech Republic, Poland and Romania have bicameral systems, which can be traced to the need for most former communist states with new democratic systems to 'keep things simple'. In Romania, somewhat unusually, the constitution does not provide a hierarchy between the Chamber of Deputies and the directly-elected Senate. Among these, only Cyprus demands that citizens vote.

The following tables outline the details of the electoral systems used by other European ECPRD-affiliated countries which are not members of, or applicants to, the EU. These countries are notable because they include one of the biggest ECPRD members – Russia – and some of the smallest – Andorra, San Marino, Liechtenstein. Like many of the EU Applicant Countries, most of these are former Communist countries, and have – with the exception of Russia and Croatia – unicameral systems.

Despite the great diversity in the histories and locations of the countries in this section, it is interesting to see that mixed systems predominate, with only Liechtenstain, Moldova, San Marino and the Croatian Upper House using full PR systems. As seen in many other countries, some of these systems also require a threshold for representation in their parliaments – 2% in Albania through to 5% in Russia and Croatia, and even 8% in Liechtenstein. In Albania, people who had held certain state positions before March 1991 under the former Communist regime are deemed ineligible to vote.

The European Parliament is a unique legislature, composed of 626 members from fifteen different countries, who are elected by different systems, conducting elections taking place on different days, albeit within a four-day window. Many similarities exists between the systems used, but individual national circumstances currently predominate. This section outlines the moves by the European Parliament to secure a uniform electoral procedure for its elections, and also details current practice across the Community.

Different electoral arrangements have for some time been seen by the European Parliament as a whole as no longer suitable, considering the Parliament's increasing influence and relevance. Article 138 of the EC Treaty (or Article 190 of the Consolidated Treaty), the text of which is included in the Annex to this paper, calls on the Parliament to "draw up proposals for elections by direct universal suffrage in accordance with a uniform procedure in all Member States or in accordance with principles common to all Member States". Once the proposal is drawn up, the Treaty states that "The Council shall, acting unanimously after obtaining the assent of the European Parliament which shall act by a majority of its component members, lay down the appropriate provisions, which it shall recommend to Member States for adoption in accordance with their respective constitutional requirements."

Current moves towards the establishment of a uniform procedure can be seen as a continuation of the increasing establishment of the European Parliament as a fully-functioning legislature, a process which has been marked by developments such as the election of Parliament by direct universal suffrage from 1979 onwards, the granting of legislative codecision powers by the Maastricht Treaty, and granting of the right to veto the passing of the Community budget and the approval of the European Commissioners.

In light of the increasing monetary, economic and political integration of the Union, calls for the establishment of a uniform electoral procedure can be seen as an effort to create a stronger Parliament, which many hope will become the focal point of representative democracy within the EU.

It is also very likely that considering the 1999 elections witnessed the lowest average turnout across the continent in a European election since the introduction of elections to the chamber by direct universal suffrage, this desire for a uniform procedure may also be seen as a way of trying to make the European Parliament and trying to make it appeal more directly to citizens, in an attempt to reverse this downward trend. From a Community turnout highpoint of 63% in 1979, this fell to its lowest, only 49.9% of the EU electorate in 1999; particularly worrying since the influence of the European Parliament, and the entire Union, has grown tremendously since then, and a strong democratic institution can have a substantial role to play in scrutinising the rest of the European institutions.

Pressure for a uniform procedure can also be put down to the desire to bring about changes in time for the 2004 elections and the entrance of the Applicant Countries into the Union, whose accession will engulf Council's time and effort, most likely meaning that Council's consideration of uniform procedures would slip further down the agenda.

The introduction of a uniform procedure has been on the European Parliament's agenda for some time. Even before the advent of elections by direct universal suffrage, many of the calls for an elected Parliament were accompanied by the desire to start a process which would lead to the introduction of uniformity, as the Patijn report in 1975 did. The rapporteur, Mr. Schelto Patijn (NL), suggested a three-step approach: the first, to make it possible for Parliament to be elected directly on the basis of national electoral systems; the second, to put forward proposals of restricted scope in order to enable 'desired objectives' to be achieved; and the third, to propose a step-by-step approach with the ultimate aim of introducing a uniform electoral procedure.

After 1979, the Parliament drew up a proposal, the Seitlinger report, for a uniform procedure which was adopted in 1982 and was the first to outline in some detail what type of electoral system could be put in place. It suggested a system of proportional representation, with seats distributed according to the d'Hondt system, within multi-member constituencies (each with between three and 15 representatives). No electoral threshold was set. Although the Council considered this text several times, no further action was taken.

Following the 1984 elections, a new proposal intended to revive the debate within the Council was put forward. The Bocklet report, however, faced major difficulty in that the two committees responsible disagreed on whether a uniform procedure required absolutely identical electoral procedures across the board, or whether Member States should merely have concordance between the key elements of those procedures (such as the electoral system to be used, the entitlement to vote and eligibility to stand for election). Bocklet, the rapporteur, proposed that the latter was more workable. The method of voting proposed was proportional representation, using the d'Hondt method, and Member States would have the possibility of establishing a single national constituency or regional constituencies, and also of setting a threshold of not more than 5%.

Bocklet, however, did not put his report forward to be voted on in plenary. This was because he felt that it would not be adopted, since the Committee on Political Affairs had only adopted it by a small majority and the Committee on Legal Affairs opposed it, due to its flexible definition of a uniform electoral procedure. The issue was deemed to require more consideration, and resulted in the creation of an intergroup working party, which during its first few meetings in 1986 agreed to a range of basic principles:

Details of the proposal were finalised towards the end of 1986 and advocated the use of proportional representation, with preferential voting. Unfortunately for those on the working group - who had spent a considerable amount of time and effort trying to draft an adoptable and consensual document to present to the rest of Parliament - lack of agreement within political groups led to the killing of any momentum and the chance of getting Council to approve it in time for the 1989 elections.

Following these elections, Mr De Gucht was appointed rapporteur, and his final resolution was adopted in 1993. In it, De Gucht agreed with Bocklet's assessment that the concept of uniformity did not require the electoral procedure to be uniform in every respect but merely that the basic features of the procedure should be harmonized, an idea which would partly be responsible for the concept of common principles which would subsequently be incorporated into the Amsterdam Treaty.

In the report, Parliament called for the uniform electoral procedure to be based on a system of proportional representation, taking account of the votes cast throughout the territory of each Member State. Preferential voting was provided for, as was the possibility of setting a threshold of between 3% and 5% of the votes cast.

In order to take account of the question of the United Kingdom, Parliament proposed that, if a Member State used a single-member constituency system, not more than two-thirds of the seats could be distributed according to that system, with the remaining third being distributed in such a way as to ensure that the overall distribution of seats was in proportion to the total votes cast.

Following the 1994 elections the Parliament did not find much favour with the Council. Although the Council President in December 1994, Mr Von Ploetz, stated that he was minded to take the European Parliament's flexible uniform procedure further, his immediate successor, Mr Lamassoure, considered that Parliament's text was too general in its content to be considered by Council.

The Intergovernmental Conference in October 1996, however, saw increased willingness on the part of the Member States to make progress. In the lead-up to the conference, Parliament repeatedly called on the Member States to implement Article 138 of the Treaty, and although no direct action was taken on this, a consensus very quickly formed around the proposal for a reference in the Treaty to 'principles common to' all Member States.

The Treaty adopted in Amsterdam includes the following addition to Article 138: '3. The European Parliament shall draw up a proposal for elections by direct universal suffrage in accordance with a uniform procedure in all Member States or in accordance with principles common to all Member States.' This addition takes account of differences in opinion between Member States and does not require an across-the-board harmonization of the electoral procedure, which is likely to lead to a more flexibile procedure which will find favour with more participants across the institutions.

Parliament's current position is set out in a report by Mr Georgios Anastassopoulos, adopted by plenary on 15 July 1998 (by 355 votes to 146 with 39 abstentions) proposing a draft act based on common principles, but which has since not been considered by the governments of the Member States. The text of the draft act calls for the following:

Article 1: In each Member State, Members of the European Parliament shall be elected by a list system of proportional representation. Election shall be by direct universal suffrage through an equal, free and secret ballot.

Article 2: Constituencies shall be established for this election without generally violating the proportional character of the system, in order to bring the elected representatives closer to the electorate and with due regard for the specific characteristics of each Member State. This provision shall apply with effect from the elections to the European Parliament scheduled for 2004. Member States whose population does not exceed 20 million inhabitants are not required to establish constituencies.

Article 3: Where a country decides to establish several electoral constituencies on its territory, the Members of the European Parliament shall remain, in accordance with Articles 137 and 138 of the Treaty (189 and 190 of the consolidated Treaty), the representatives of the peoples of the States brought together in the Community, and not the representatives of their constituencies.

Article 4: Special arrangements may be made to take account of specific regional characteristics but they may not violate the principle of proportional representation.

Article 5: A minimum threshold for the distribution of seats may be set and at national level may not exceed 5% of the votes cast.

Article 6: Member States may permit preferential voting in accordance with procedures that they shall lay down.

Article 7: The European Parliament will consider a proposal that a certain percentage (the rapporteur proposed 10%) of the total number of seats within the European Parliament shall be filled by means of list-based proportional representation relating to a single constituency comprising the territory of the European Union Member States with effect from the European elections to be held in 2009. The implementing provisions shall be adopted by 1 January 2008 on a proposal from the European Parliament by the Council acting unanimously, and after obtaining Parliament's assent.

Article 8: The office of Member of the European Parliament shall be incompatible with the office of member of a national parliament.

Article 9: Each Member State may set a limit for candidates' expenditure linked to the conduct of the election campaign.

Article 10: Save where otherwise stipulated, the Act of 20 September 1976 concerning the election of the representatives of the European Parliament by direct universal suffrage shall remain in force.

This report is the culmination of many years of thought and debate on the issue which, when combined with the flexibility which Article 138 of the Treaty now has, allows it to appeal to more participants than ever before. Furthermore, it does not ask for any major changes in the systems of Member States. For example, a major concern with other reports was about how to create a procedure that would be acceptable to the British, who used a majoritarian system for their European elections. The establishment of a proportional representation party list system in the UK (except for Northern Ireland) before the 1999 elections now means that there is already a degree of similarity among the systems used in Member States (they are all forms of proportional representation), that would most likely make agreement between the Member States easier.

The concept of a uniform electoral procedure is perhaps in the best position that it has ever been in to be properly addressed and considered by the Council, and is likely to be resolved within the next few years, so that some or all of the proposals may very well be adopted and put in place for the 2004 elections.

The following tables give details of the electoral systems and mechanims currently used across EU Members States for elections to the European Parliament. The first is a synoptic table, giving a comparative summary of the information contained in the more detailed tables following it.

These detailed tables provide the following details for every Member State's electoral system: the website addresses of European Parliament information offices in Member States (unless the offices do not have websites, in which case the main EP website is listed); the number of seats distributed; details about constituency boundaries; the length of mandate; details of the electoral system used for distributing seats; the rules for filling vacancies; rules on who is entitled to vote; rules on who is eligible to stand for election; and rules surrounding the nomination of candidates.