HIDDEN POLITICAL OBJECTIVES OF ELECTORAL SYSTEMS – Editorial by John Szemerey FCIJ
 John Szemerey FCIJ
John is a political journalist, retired in summer 2005 after working over 30 years in the information services of the European Commission, Brussels. |
Most people do not know and do not care how electoral systems work. As far as they are concerned, they have done their duty by marking their vote on a ballot paper, or by pressing buttons on an electoral machine. This is unfortunate, as the electoral system used to elect legislatures is directly responsible for the parliament and government that the system elects. Every electoral system – and there are several hundred of them – has a political objective, quite apart from its basic function of electing members to a parliament, council, committee or local legislature. Different electoral systems produce totally different end results. It is therefore important to know the results of each electoral system before discarding an existing system and adopting another one. While the man or woman on the street thinks that electoral systems exist to elect MPs and councillors, politicians tend to judge electoral systems by how good they are at electing members of their political parties to the legislature. In other words, politicians evaluate electoral systems by how effective they are at electing the candidates of their party, and by how good they are at giving power to their party. However, electoral systems have many more objectives than to help or prevent a political party come to power. Objectives range from producing a legislature that can create a stable government or unstable governments with MPs elected from several parties but with none having enough MPs to have a majority in parliament to form a government. (Yes, some systems deliberately produce unstable governments in countries that do not want to have a strong central government.)
Then there are electoral systems that give an advantage to parties of the political centre. Others make it very difficult if not impossible for independent MPs or small political parties to be elected. Normally the parties in control of a legislature do not look kindly at proposals to change the electoral system, unless the suggested changes to the system are likely to give them more elected MPs or unless they make it more difficult for candidates of rival parties to be elected. For some time people have been calling for India to change its electoral system. However, no-one has specified exactly which electoral system should be used, and what are the political objectives of that system. Before the public agrees, it would be wise to discover exactly what electoral system is being proposed, and what are the real political objectives of that system. An exercise that proponents of change should try is to take the voting figures of the last two or three general elections and allocate them as if they had been cast using the system that they propose. This can produce interesting and perhaps surprising results. First, however, they should decide what is the purpose of elections. There are two main viewpoints, which are contradictory. One is that parliament should represent all parts of the country and all residents in the country. The other point of view is that parliament should reflect the views and opinions of the whole country. In theory, elected MPs should represent the balance of opinions in the country. In fact, most MPs are free-thinking individuals. It is highly unlikely, if not impossible, for all the candidates on the list of a political party to have the same opinion about everything. It is even less likely that elected MPs – of different parties – from constituencies should fairly represent the balance of opinions in the country. However, the view of those elected makes a big difference to the views and decisions of parliament. So it is the party officials who decide whom to put on the party list, and what place they have on the list – if they are sufficiently high on the list to be elected – that have enormous power in setting up the legislature in countries with PR electoral systems. The only way it would be possible to have a parliament with members representing all viewpoints within the country would be to have an independent committee or organisation nominate candidates to every constituency, and for these nominated candidates to be declared elected. The committee could ensure that the nominated and elected MPs represent a fair balance of the views held in the country. But clearly this is not a free and fair way of electing parliament. This system would be totally incompatible with democracy.
Electoral systems
There are three major families of electoral system in use in democratic countries. These are Direct Representation (DR) (often referred to as First Past the Post), Proportional Representation (PR) and Preferential Representation (PrefR). Direct Representation was introduced before political parties existed. It ensures that everyone in a country has his or her elected MP, to whom constituents can go for help and advice. The country is divided up into many similar sized constituencies, each of which elects one MP to parliament. The duty of elected MPs is to represent in parliament the constituency for which they were elected. DR also ensures that all parts of the country are represented in parliament.
I must point out, however, that the anti-DR propaganda, that votes for a candidate who is not elected are wasted, is nonsense. The only votes that are wasted are those that are not cast. But all votes cast play an important role in elections – in the same way that a football team or a cricket team that loses a game did not waste their time playing. PR comes in many shapes and sizes, but its basic objective is to ensure that the political balance of a legislature is in proportion to how the electors voted. For this system to work, either the whole country is taken as one single constituency, or in larger countries the country is sub-divided into electoral regions. PR is meant to ensure that the political balance of MPs elected nationally or from each electoral region reflects the balance of votes for the different parties. Voting in PR systems is either by voting for a party list, which accepts the order into which the candidates have been put, or by voting for one or several candidates on the party list. Such a vote does two things: it is counted as a vote for the party list, and it gives preference to the candidate or candidates for whom the elector voted.
Pref R has many of the advantages of DR, in that constituencies either have one or a few elected MPs, who actually represent their constituency in parliament. Voters have more of a say as to which candidate(s) is elected than is the case with PR. They vote by putting numbers against the name of candidates, to indicate their preferences. So their first preference is given a 1, their second preference a 2, and so on. When first votes are counted, the candidate with fewest votes is removed from the list, and the votes he or she received are reallocated to the candidates who were his/her voters’ second preference. As many counts are held as are needed for the right number of candidates to be elected for each constituency or region. At the end of each count the candidate with the fewest votes is removed from the list, and his/her votes are reallocated to these voters’ next preference. There are two principal PrefR electoral systems: the Alternative Vote (AV) and the Single Transferable Vote (STV). AV elects one MP from each constituency, while STV elects several from each constituency.
DR “unfair” claim Liberals
A clear example of a political party trying to have the electoral system changed for its own benefit is the continuing campaign of the Liberal Party, now the Liberal-Democrat Party, in the UK. The Liberals have for years been complaining that fewer of their parliamentary candidates are elected than should be elected, in light of the total number of votes that the party receives nationally.
What is ironic is that the Liberals have been clamouring for proportional representation, whereas PR is the last thing they want. They really want the Single Transferable Vote (STV), a preferential electoral system that is designed to help parties of the centre. STV constituencies elect from two to 30 or more MPs per constituency.
The system works best when constituencies elect between three and five MPs. This is because if the constituency only elects two MPs, preferential votes are not very effective, as a maximum of only two parties or viewpoints can be represented in parliament. However, if constituencies elect 15 – 30 MPs, they will elect several independent or minority party candidates – whose presence in parliament will make it difficult for a stable government to be elected, and it will weaken the position of the bigger parties. How does STV help parties of the political centre?
If there are three or four MPs to be elected from each constituency, voters will give their first preference votes to the party they usually support, usually a party of the Left or of the Right. If that party has a second candidate, he or she will receive voters’ second preference votes. When there are no more candidates of voters’ preferred political party, the next – second, third or fourth – preference vote will not be cast for an opposing party (a voter of the Left is highly unlikely to give his/her second preference vote to a candidate of the Right, and a voter of the Right will be highly unlikely to vote Left). The next preference vote will be cast for a party of the Centre, such as the Liberals or the Greens. This is so with Left voters and Right voters. They both tend to give their next preference vote to a party of the Centre.
This results in candidates of the centre getting the second or third preference votes of voters who would normally not vote for them but who find them more acceptable, or less unacceptable, than a party whose policies are directly opposite to those of the party they support.
With PrefR, Liberal or Green candidates benefit from receiving the second or third preference votes of voters of the Left and of the Right, which in many cases give them enough votes for them to be elected as the third or fourth MP in a multimember STV constituency.
The end result is that with a preferential election system the Liberals and the Greens tend to win considerably more seats in parliament than if the election were held by direct representation or proportional representation. PR is a group of several hundred electoral systems that tend to be undemocratic.
In an ideal PR system, each party has a national list of candidates. In most cases, voters cannot choose between candidates but they vote for a party list. Candidates are selected and put on the party list by a senior committee within each major party. In advance, the percentage of the votes that will go to each major party is known. So if a party can expect to get 25% of the vote, the first 25% of candidates on the party list will be elected. If another party expects to get 10% of the vote, the first 10% of candidates on its list will be elected.
It is therefore the senior party apparatchiks, who decide who will be on the party list and where that candidate will be on the list, who in fact decide who will be elected. Voters can only have a marginal influence on the result by voting differently from the expected percentages of support for each party. So one or two more candidates of that party will be elected if the party receives 27% of the vote instead of 25%, and other parties that receive slightly fewer votes than expected will have one or two candidates fewer elected.
French example
In 1979 I did research into how France elected its 81 members of the European Parliament, as the French had decided to elect their MEPs by PR with national lists. There were four major political parties standing in the election and several small parties.
First, not one candidate of a small party was elected under the PR system with national lists. This had largely been known in advance, so what was of interest was how many candidates from each of the four main party lists would be elected. The percentage of the vote that the four main parties were expected to receive was known in advance, so it was each party’s election committee that decided which candidates would be on the party’s list, and where on the list, that in fact decided who would be France’s 81 members of the European Parliament.
Each of the election committees had between 12 and 25 members. So it was these members, fewer than 100 in all, who in fact decided who would be France’s MEPs for five years, and not the voters, who just went through a charade of voting.
It need hardly be said that in this situation it is very easy for a wealthy supporter of a party to bring pressure on an election committee to put one or more candidates in winning positions on the party list, or to put a potential MP low down on the list if the supporter does not want him or her to be elected.
Countries where PR is used often realise how easily the system can be manipulated. They therefore frequently modify the system so it is less proportional. For example, in several countries there is an electoral threshold. This means that a party or candidate must receive at least a given percentage of the total vote in his constituency or region to be able to win any seats. Some systems deliberately cut the country up into regions or large constituencies, which cannot be as proportional to total votes as a large electoral region like a country. Then there are many methods of counting the votes cast for parties or candidates. Some give seats to smaller parties that would not have won seats if the count had been truly proportional.
Counting votes
Best known of these systems is the D’Hondt method that divides the total number of votes for a party by the number of seats it has already won, plus one. Seats are allocated one after the other, with the first seat being allocated to the party with the most votes. The total votes of this party is then divided by the number of seats won plus one, namely 1 + 1 or 2. So this party’s total is halved. The next seat is given to the list that now has the highest number of votes.
In the example below, the total of the second list to win a seat, 23, is then divided by the number of seats that list has won, plus one, namely 1 + 1 or 2, leaving 11.5. The totals of all the lists are again compared, and the next seat to be won is by the party that now has the highest total of votes, namely party A that won the first seat. The total vote of party A, 39, is then divided by 2 + 1 or 3, leaving 13. And so on, until all the seats have been elected.
Seven MPs are elected from this fictitious region, where there are seven fictitious parties, A to G, and seven rounds of counting, each count to result in one candidate – the one with most votes at that stage – to be elected.
The first MP to be elected is from party A, which has most votes, namely 39. This number is then divided by (1 + 1) 2, leaving the party with 19.5 votes. The party with the largest total of votes is now F, which has 23 votes, so the second seat goes to party F. This party’s total vote (23) is divided by (1 + 1) 1, giving the party 11.5 votes.
The third seat is allocated to the party now with the highest number of votes. This is again party A, which has 19.5 votes. The party’s original vote, 39, is now divided by (2 + 1) 3, leaving a new total of 13.
The totals of all the lists are again compared, showing that party G now has most votes, with 17. So it wins the next seat, and its total is then divided by (1 + 1) 2, to leave it with 8.5.
The lists are again compared, to show that party B now has the largest number of votes, namely 16. It wins the next seat, and its total is divided by (1 + 1) 2, leaving it with 8.
The lists are compared yet again, and it is now party A that has the largest number of votes, with 13. It wins the next seat, and its original total (39) is divided by (3 + 1) or 4, leaving a new total of 9.75.
The lists are compared for a seventh time, and it is party F that now has the largest total, 11.5, so it wins the seventh seat.
So, using PR and the D’Hondt method of counting, party A wins 3 seats, party F wins 2, and parties G and B win one each. If seats were allocated on a simply proportionate basis, party A would win 3 seats, and parties B, E, F and G would each win one seat – quite a difference.
D’Hondt is just one method of counting. There are other systems that have other divisors, so that instead of dividing the total of a list that has just won a seat by the number of seats plus one, one divides the total by 3 + 1 or by 7 + 1, which both produce quite a different result. The larger the divisor the greater the number of candidates from small parties or independent candidates who get elected.
German electoral system
Another very political electoral system is the German Mixed Member Proportionate (MMP) system, which elects half the total number of MPs by DR from individual constituencies, and the other half of MPs from party lists by PR, in proportion to the votes their parties received. At first glance one is tempted to think the system is fair, as it ensures that MPs represent all parts of the country and that the total number of MPs is in proportion to how the electorate voted.
However, at first glance one does not spot the system’s worst characteristic: that it punishes political parties that manage to win a good number of constituencies, and it rewards those that won only a few or none.
If a party has managed to win, say, 25 out of 50 constituencies but it only has 15% of the vote for the PR lists, it will not win any PR seats. This is because 15% of a 100 seat parliamentary chamber is only 15, which is what the party would normally be entitled to win. It is allowed to keep the 25 seats its candidates won, but it has no extra seats allocated from the PR lists. Another party that won, say, 8 constituencies but received 15% of the vote for the PR lists wins seven seats from the PR lists. This is because it is entitled to 15 MPs. As it only won 8 constituencies, it is given 7 PR seats, that bring its total MPs up to 15.
Also, this system has a 5% electoral threshold, so no party with less than 5% of the vote can win any of the PR seats. This was written in to the constitution specially to make it very difficult for a dictator like Adolf Hitler to become the ruler or chancellor of Germany. The limit is effective in discouraging all small parties from being represented in the Bundestag.
On two occasions, in 1969 and 1998, Germany’s third party, the liberal Free Democrats, only just scraped into the Bundestag. In 1990, the next party, the Greens, did not get 5% of the vote in West Germany. However, the rules were temporarily changed to put together their vote in the former East Germany with their vote in the West to enable the Greens to have some seats in the Bundestag.
This electoral system has much to recommend it, although to be fair the PR seats should be awarded only according to the votes parties get with electors’ second votes for the PR lists, irrespective of how many constituencies were won by that party. This would avoid the present situation of punishing parties that win many constituencies by not giving them any list seats, or fewer seats than they would be entitled to according to the PR list vote, and rewarding them if they win few or no constituencies.
Conclusion
India’s present electoral system, Direct Representation or First Past the Post, has had several criticisms made against it. However, most of these criticisms have been on account of the fact that India does not run the elections correctly. For example, it allows a significant part of the population to be almost unrepresented in parliament.
Another criticism of DR in India is that several of the candidates are criminals or otherwise unsuited to be MPs. This is not a criticism of DR, but of the way it operates in India. The authorities should tighten the rules as to who can stand for election.
A further criticism is that there is corruption, with people or organisations financing a party or one or more of its candidates in exchange for the party or its MPs doing something in parliament after the election. This is also a criticism of the way DR operates in India. The authorities could simply apply rules that such misbehaviour cannot happen.
The best that India could do now is to examine why and where DR is not working properly and to propose corrections to the system. In particular, it should ensure that every citizen has a vote and that every citizen is able to go and vote at elections.
Meanwhile the proponents of change should be asked to specify exactly what electoral system they would prefer, and why.
If it was found that it was too difficult, or not possible, to correct the present system, that would be the time to have a country-wide discussion and consultation about the best alternative electoral system. This could be followed by a referendum, in which the electorate was asked which election system it would prefer.
This would give India the time it needs if it decides to change its electoral system.
