Results 1 to 2 of 2

Thread: BERES: Alternatives to First-Past-The-Post system

  1. #1
    Join Date
    Oct 2008

    BERES: Alternatives to First-Past-The-Post system

    Is FPTP system inherently fraud-prone?

    Primary tabs

    First Published:10:14pm, Jun 27, 2013
    Last Updated:8:10am, Jun 28, 2013


    'Witnessing history' by Wong Chin Huat

    UNTIL the 2013 general election, fraud in the electoral process – for example, phantom voters, vote buying, fraudulent counting or tallying – is often discussed independently from the flaws in the electoral system.

    However, the truth is that electoral system may determine the degree of fraud and manipulation. Granted, all electoral systems may be compromised by fraud or manipulations but their vulnerability varies significantly.

    The benefit-cost analysis of fraud

    People do not cheat simply because they are bad, but because cheating is profitable. If for the same quantity of benefit, fraud incurs a greater cost than honesty, then even the devil will opt for honesty.

    The benefit in cheating can be understood as the gain in seat share caused by such a degree of cheating. Similarly, the cost of cheating can be understood as the increase in vote share needed for such a degree of cheating.

    What then determine the possible sizes of benefit and cost? The former is determined by number of seats elected by a constituency – the terminology is "magnitude" – while the latter is determined by the size of electorate.

    Determinant 1: Number of seats

    For the First-Past-The-Post (FPTP) system, the magnitude is one. Because winner gets 100% of the seat and loser gets nothing, hence the benefit of successful cheating is 100% in seat gain.

    Not all electoral systems opt for single-member constituencies.

    Many European, Latin American and African countries, as well as some Asian polities like Indonesia and Hong Kong, use the party-list proportional representation (Party List) system whereby voters choose between parties and not individuals.

    Under this system, every constituency will have a certain number of seats, from the smallest 2 (as in Chile) to the entire national parliament (as in the Netherlands), so that the parties can divide the seats according to the percentage of their votes.

    The single transferable vote (STV) system used in Ireland, Malta and certain legislatures in Australia allows the voters to list down their preferences amongst the candidates – first, second, third and so on – and wants a variety of representatives to represent such preferences. So, this system too uses multi-member constituencies.

    Both these systems use some kind of mathematical formulas to decide a quota of vote share that a party needs to claim a seat.

    Surely, instead of all or nothing in single-member constituencies, seat change in multi-member constituencies will be incremental. For example, for a 10-member constituency, seat change happens in the unit of 10%.

    In the language of criminal profitability, fraud yields lesser benefit in multi-member elections than in those single-member ones.

    Determinant 2: Electorate size

    The cost of fraud is determined by the number of votes the fraud-plotter needs to rig or buy. That can be measured in absolute numbers or percentage.

    For the FPTP system, the minimum it takes to effect a 100% change in seat share, is one vote, if the electorate size is an odd number, or two votes, if it is an even number.

    In reality, paper-thin margins like 1% – for example, 50.5% versus 49.5% in a straight fight – is all possible.

    But how big is 1% of votes?

    That really depends on the constituency size. Here again, the FPTP elections become attractive to electoral criminals because it normally uses single-member constituencies, which naturally tend to have smaller electorates than Party List or STV constituencies that may have 10 or 20 members.

    When we combine bigger electorate size (high cost) in Party List or STV constituencies with the size in incremental seat change (low benefit) in those elections, the profitability of fraud further depletes.

    The peril of small-sized single-member constituencies is tremendous.

    Take a state constituency with 20,000 voters. A 5% real margin will mean 1,000 votes. To turn around the election outcome, the trailing party only need to do the following:
    (a) buying over 501 voters;
    (b) planting 1,001 phantoms;
    (c) spoiling 1,000 votes;
    (d) combination of the methods above.

    Gerrymandering and voter transfer

    If a state has 50 seats and all 50 seats are so competitive with a 1,000-vote margin, cheating will still pose a financial and logistical challenge, because you will have to buy 25,025 votes, plant 50,050 phantoms or spoil 50,050 votes.

    Here's where gerrymandering can do wonders for whoever can control the constituency redelineation process. It can cut down the number of marginal constituencies and within these constituencies, cut down the size of margin.

    Hence, the total cost of fraud can be sharply reduced, making electoral fraud profitable.

    Conceptually, transfer of voters is actually the surrogate of constituency redelineation in between the actual exercises.

    While redelineation moves the boundaries with voters staying put, transfer moves the voters with the boundaries staying put. Either way, the composition of constituency electorate changes.

    Mal-apportionment and electoral targeting

    To control a state, you only need to win half of the legislature plus one constituency, by winning half of the electorate plus one vote in all these constituencies.

    Even if the constituencies are equally apportioned, that means you can win a state by winning minimally just above 25% of the voters.

    This means if the incumbent party can identify specific groups of voters, it can target them with either election goodies or media control. By doing so, it may actually win with just creating non-level playing fields, without resorting to phantom voters or other tactics of vote rigging.

    Now, if you can mal-apportion the seats, then that minimum size of necessary votes can drop beyond 25%.

    In the 2013 parliamentary election, BN won its smallest 112 seats with just 20% of the total votes. It is therefore perfectly possible that Pakatan wins 60% of votes yet still lose the 14th general election.

    Fraud under FPTP – low cost, high benefit

    Proportional representation (PR) systems like Party List and STV certainly are not free from their own problems. However, these systems do not induce gerrymandering and mal-apportionment because they strive to produce vote-seat proportionality.

    While one may still consider cheating under such systems, the payoff will just not be attractive enough. A small swing in vote share (low cost of fraud) will only cause a small swing in seat share (low benefit of fraud), and a large swing in seat share (high benefit of fraud) will require a large swing in vote share (high cost of fraud).

    Fundamentally, the FPTP system is fraud-prone simply because a small swing in vote share (low cost of fraud) may cause a big swing in seat share (high benefit of fraud).

    Attached to Penang Institute, Wong Chin Huat is a political scientist by training and a political activist by choice. He believes that like it or not, we are witnesses to history. We can choose to shape it or be shaped by it. This is the personal opinion of the writer and does not represent the view of

    This article is part of a series on electoral reforms written by Wong. Watch out for the next instalment next week. Wong thanks Prof Sarah Birch, his PhD supervisor at University of Essex, UK and a leading researcher in electoral fraud, for her guidance and inspiration both in his doctorate and the writing of this piece.

    Read more:

    Alternatives to First-Past-The-Post system

    Primary tabs

    First Published:10:37pm, Aug 28, 2013
    Last Updated:9:00am, Aug 29, 2013


    'Witnessing history' by Wong Chin Huat

    • FPTP is the simplest one of all but as we have seen, hardly the best one.

    IF THE First-Past-the-Post (FPTP) system is problematic, what options do we have?

    Based on the works of political scientists Douglas Rae (1967) and Andreas Blais (199, an electoral system has six elements:

    (a) number of votes
    (b) types of votes, whether it is nominal, ordinal or numerical
    (c) object of votes, ie, individuals or teams
    (d) constituency nature, ie, one or many constituencies
    (e) constituency magnitude, ie, number of seats; and
    (f) formula, whether it is plurality, majoritarian, or proportional.

    Under this framework, FPTP, or officially known as Single-Member-Plurality, is an electoral system where a voter is given one ballot, a nominal ballot to choose between individual candidates, in a single-member constituency, and a candidate needs only a plurality to win.

    For legislative elections, FPTP will have many constituencies.

    FPTP is the simplest one of all but as we have seen, hardly the best one. By varying these elements differently, we can then have different electoral systems.

    I will introduce two of them which may be featured in future debates.

    The first is Australia's Preferential Voting system advocated passionately by Prof Clive Kessler, a renowned Malaysianist from Australia in his chapter in the book "Elections and Democracy in Malaysia" (edited by Dr Mavis Puthucheary and Prof Norani Othman) as well as some Australia returnees.

    The second is Mixed Member Proportional (MMP) system in Germany and New Zealand, which the Election Commission has expressed interest to study and emulate.

    The Australian option – Preferential Voting

    Also used in presidential elections in India and Ireland and parliamentary elections in Papua New Guinea and Fiji, the Australian system is similar to our FPTP in three ways: first, it has many legislative districts; second, each district elects only a single member; and third, the voters are to choose amongst individual candidates.

    The system differs from FPTP in the ballot structure and the electoral formula.

    Firstly, the Australian ballot is ordinal, where candidates are ranked. When it comes to the number of candidates to be ranked, there are two variants.

    Under the Alternative Vote (AV) variant, applicable for the state elections in New South Wales and Queensland, the voters can choose only as many candidates as they like.

    Under the Compulsory Preferential Voting variant, applicable for Australia's federal and other state elections, the voters have to rank all candidates.

    Secondly, to win one must obtain a majority, not just plurality – in other words, the winner must have more supporters than opponents.

    If a candidate wins more than half of the first-preference votes, he or she will be declared the winner, much like our FPTP. However, if no candidate does so, the weakest candidate will be removed and his/her votes will be redistributed to the remaining candidates based on the second preference of these votes.

    This process will continue until a candidate which commands a majority support is produced. You bet, the counting will take some time.

    The German option: Mixed Member Proportional

    Applied in New Zealand since 1996, the Mixed Member Proportional (MMP) System originated from Germany is simply a 50-50 hybrid of FPTP and Party List Proportional Representation (Party List) system.

    Party List system is commonly used in many European and Latin American countries, but also in Indonesia and Hong Kong.

    To allow different parties being represented in the Parliament, Party List have multi-member constituencies, with candidates contesting not as individuals but as members of part lists, and seats are then allocated to party lists based on some proportionality formulas.

    For example, in Hong Kong, where a method called Hare Quota is applied, to win a seat in a five-seat constituency will require only 1/5 of votes.

    Under the MMP, the voters will have two votes, one for the FPTP parliamentarian and another for the party list.

    Half of the Parliament will come from the FPTP constituencies and the other half from the party list. The percentage of party votes will determine the total seats a party will eventually get.

    Assume we have a 200-seat Parliament, with 100 FPTP seats and 100 party-list seats. If Party A gets 35% of the party votes, it should get 70 seats in total. Should it already win 50 out of the 100 FPTP seats, it will be given 20 party-list seats.

    However, if it wins only 20 FPTP seats, it will be given 50 party-list seats.

    The MMP has a watered-down variation called Mixed Member Majoritarian (MMM) which is adopted by many countries including Japan, Korea, Taiwan, Thailand and the Philippines.

    It differs from MMP in two ways. First, the FPTP and Party List components of the parliamentary seats need not be equal, with often the FPTP seats making up a vast majority.

    Second and most characteristically, the party votes only decide the allocation of party-list MPs, not that of the entire Parliament.

    In Thailand, for example, out of 480 seats, 400 are elected by FPTP constituencies, and the remaining 80 (16.67%) are by party list.

    In other words, MMM tries to check the excessive vote-seat disproportionality of FPTP but shies away from full proportionality, hence producing semi-proportional outcomes.

    Endless possibilities

    To borrow Datuk Seri Najib Razak's latest slogan, there are "endless possibilities" in the choice of electoral system. Just because we have been eating Mackerel yesterday and the day before yesterday, it doesn't mean we have to eat Mackerel today and tomorrow.

    Attached to Penang Institute, Wong Chin Huat is a political scientist by training and a political activist by choice. He believes that like it or not, we are witnesses to history. We can choose to shape it or be shaped by it. This article is part of a series on electoral reforms written by Wong. Watch out for the next instalment tomorrow. This is the personal opinion of the writer and does not represent the view of

    Read more:

  2. #2
    Join Date
    Oct 2008
    Choosing the electoral system for tomorrow's Malaysia

    Primary tabs

    First Published:7:01am, Aug 30, 2013
    Last Updated:8:43am, Aug 30, 2013


    'Witnessing history' by Wong Chin Huat

    THERE is not really such a thing as the best electoral system. Rather, we just have electoral systems offering us different strengths and different weaknesses. In that sense, in considering the best alternative to the First-Past-The-Post (FPTP) system, we should set out with some criteria or goals in mind.

    Reflecting on our experience and context, I would propose the following criteria:

    1. It should be representative and fair, ensuring the democratic legitimacy of government.

    2. It should deter fraud and manipulation.

    3. It should encourage moderation.

    4. It should encourage elected representatives' accountability to voters.

    5. It should facilitate diversity in legislature.

    So, how do FPTP, Preferential Voting and MMP fare under these criteria?

    Criterion 1: Representativeness, fairness and democratic legitimacy

    FPTP generally fails in ensuring representativeness as amplifying the strength of the winner – ensuring a strong government – is in fact its purpose. In fact, FPTP may produce minority winners at both the constituency level – as in many cases of multi-cornered fights – and the national level – as we have been experiencing since May.

    In this sense, Preferential Voting is an improved alternative to FPTP, as there can never be a minority winner at the constituency level. At the national level, the Australian system remains vulnerable to the problems of excessive vote-seat disproportionality and minority government in term of votes.

    In comparison, MMP can pursue maximum proportionality at the national level but at the constituency level, its FPTP component will of course inherit the problem of minority winner.

    Criterion 2: Deterrence of fraud and manipulation

    Both FPTP and Preferential Voting are vulnerable to fraud and manipulation from malapportionment, gerrymandering, tampering of electoral rolls to vote buying for the same reason: the national outcome will be determined at small local constituencies, hence, it is possible to pursue marginal gains in selected constituencies.

    Of course, when it comes to polling and counting, the complicated Preferential Voting poses good obstacles as compared to the FPTP.

    In comparison, because the final seat share will be determined by party vote share, MMP eliminates completely the incentives to malapportion and gerrymander constituencies.

    The parties may still cheat but it will have to cheat by increasing its party vote share in commensurate portion of the additional vote share it desires. And the crime can hardly be profitable.

    Criterion 3: Promotion of moderation

    This is one of the most important criteria in deciding an electoral system, especially for a diverse society like Malaysia.

    The conventional view would place FPTP at an advantage as its winner-takes-all characteristics will force political groups to consolidate into two blocs which will then compete in the middle.

    Malaya/Malaysia's decades of authoritarianism are of course an antithesis to this myth.

    In comparison, Preferential Voting may claim this as its strongest point. To ensure victory, candidates are wise to not take extreme positions and alienate any groups so that they may still hope for their second, third, …nth preference votes even if the first preference is out of question.

    This may further encourages coalition building by encouraging candidates to endorse each other for the lower preference votes of their own supporters.

    MMP, like other forms of Proportional Representation (PR) systems, is often accused of causing party proliferation, political fragmentation and eventually, political instability.

    What Germany and New Zealand do is to impose a 5% threshold of party votes or winning a certain number of FPTP seats, as a threshold before a party can claim party list seats.

    However, as we have argued previously, the real key to moderation is actually the opportunity to rule or share power. Hence, if we introduce decentralisation and local elections to expand the opportunities of various parties come into power and form coalitions, their behaviours will be better moderated than under the current FPTP system.

    Criterion 4: Elected representatives' accountability to voters

    On one hand, FPTP, Preferential Voting and MMP all provide geographical representation so that the constituents' local needs can be attended to.

    On the other hand, the three electoral systems differ significantly in the control of political parties – vis-ΰ-vis the voters – have over the candidates.

    FPTP, in the rigid context of coalition politics in Malaysia, hands the power to the coalitions' and parties top leadership to decide which component party and which leader to run in a constituency.

    Not having to worry about consequences of vote splitting, voters in the Preferential Voting system can afford to support candidates dropped by party leadership.

    On the surface, MMP would give party leadership the power to decide candidates on both FPTP and Party List sections.

    However, put in Malaysia's context, party votes would allow competition between allies within a coalition. Also, dropped candidates can have a better chance to appeal to voters for either party votes or FPTP votes.

    Hence, in contrast to FPTP, both systems in the Malaysian context would somehow weaken party leadership's control on candidacy and strengthen democracy within parties or between parties and supporters.

    Criterion 5: Diversity

    There is no coincidence that most countries that have high percentage of women in Parliament are countries with Party List systems. The nature of Party List system simply facilitates reservation of seats for certain demographic groups.

    For example, if the state wants to have one-third of women amongst parliamentarians, it just needs to legislate that every one of three candidates by a party must be women. In this way, even the most patriarchal party would have to at least nominate one woman after two men in its list and the more seats it gets, the more women will be elected in. (See Table 1 below)

    The same method can be used to ensure representation of other groups such as aboriginal peoples or disabled.

    By virtue of having half of the parliamentarians from the party list section, an MMP Parliament can therefore have a more diverse composition to represent the population.

    As a matter of fact, parties can have division of labour between FPTP parliamentarians who have to serve the constituent needs and Party List parliamentarians who are there to represent their demographic groups and/or to contribute their expertise.

    Hence, parties can fill their party list lawmakers with professionals, scholars and experts to ensure better policy making.

    A debate is needed

    The electoral system can have vast influence on our political life. It is good that the Election Commission is now studying the electoral systems of other countries. It would be tragic however if it or the government alone decides what electoral system we should have tomorrow.

    That should be a conscious choice of all Malaysians after a comprehensive debate on the pros and cons of the proposed options.

    Attached to Penang Institute, Wong Chin Huat is a political scientist by training and a political activist by choice. He believes that like it or not, we are witnesses to history. We can choose to shape it or be shaped by it. This article is part of a series on electoral reforms written by Wong. This is the personal opinion of the writer and does not represent the view

    Read more:

Visitors found this page by searching for:

Nobody landed on this page from a search engine, yet!
SEO Blog

Tags for this Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts