Digital Cabinet

Polevaulter Donkeyman's rants, raves musings and flame wars

FPTP is just as bad!

with 14 comments

  1. PolvolterDnkymn
    @sabhlok All the ills described in the linked article happen in India which uses First Past The Post

Sanjeev Sabhlok excerpts from an article in The Australian by Janet Albrechtsen[1] on why proportional representation (PR) is bad (for Australia). He uses this article as to why India should not adopt PR for elections. I counter that all the evils of PR as propounded by the article happen in India too which uses the first past the post (FPTP) voting system

Albrechtsen’s salient points are:

  1. PR ensures extremist (I read as non-major)[2] parties get representation
  2. No centrist party gets a majority and thus has to enter coalition government with the smaller extremist parties
  3. Thus these smaller extremist parties hold the balance of power
  4. Policies not desirable to the greater centre have to be adopted so as to placate the extremist party.
  5. Under PR, voters cannot know, when they vote, what the future governing coalition will look like
  6. It takes months of horse-trading and backroom deals to form a new government

Refutation of Sanjeev Sabhlok’s contention

Each of these above scenarios occur in India too which uses the FPTP system

Table 1. Tally of Seats won by INC and BJP 1989-2009
Lok Sabha[3] INC BJP Total/% of seats
Ninth (1989)[4] 195 89 Not calculated[5]
Tenth (1991) 252 121 373/68
Eleventh (1996) 140 163 303/56
Twelfth (1998) 142 183 325/60
Thirteenth (1999) 118 189 307/56
Fourteenth (2004) 159 147 306/56
Fifteenth (2009) 210 117 327/60

  1. Non-major centrist parties get elected to the Indian Parliament.
    • The major parties in India closest to the “centre” are the Indian National Congress (INC) and the Bharatiya Janata Party (BJP). The last time the INC ever had a majority in the lower house of the Indian Parliament was in 1984.[6] The BJP has never achieved a majority on its own (it was formed in 1980)
    • Thus even with FPTP 40-45% of the seats go to the non-major parties
  2. Thus the smaller parties hold the balance of power
    • In 1999 the governing coalition was brought down by a minor partner[7]
    • But it is not the smaller parties only which are guilty. In 1991 and 1997, the INC which was supporting the coalition from outside withdrew support resulting in elections
  3. Smaller parties in India have also forced the government to adopt policies at variance with the major party.
    • The Indian government’s policy to allow foreign direct investment in the retail sector was derailed by the TrinaMool Congress (TMC) which has 20 (4%) of seats in the Lok Sabha and is a member of the current governing coalition.
    • The Indian government barely survived a confidence motion in the Lok Sabha after the Left Front[8] having 60 (11%) seats withdrew support over the Indo-US civilian nuclear agreement[9]
  4. It is laughable to say that under PR voters cannot know, at the time of voting, what the future governing coalition will look like. Such phenomenon occur in FPTP too.
    • In the UK general elections in 2010 no party had a majority resulting in a coalition between the Tories and the Lib Dems. At the time a tory voter or a lib-dem voter voted, s/he had no idea that there would be a future governing coalition (let alone what it would look like).[10]
      • Voters who voted for the Lib Dems, after the party pledged not to raise tuition fees, wouldn’t have known that the Lib Dems would vote for a tuition fee increase when in government as a coalition with the Tories.
      • Voters who voted for the Tories, believing its manifesto promise to be tougher on Europe, wouldn’t have known that the Tory PM Cameron would make a U-turn on his campaign promise to support a campaign by the European Parliament to reduce its monthly “travelling circus” to Strasbourg.
    • In India, due to the emergence of non-major parties on to the national scene, parties usually contest elections as part of larger coalitions, which one may think gives the voter an indication of what the future governing coalition would look like. However such coalitions are extraordinarily fluid and regularly lose and gain members.
      • The United Progressive Alliance (of which INC is the major member and which is now in power) has regularly lost members.
      • The National Democratic Alliance (of which BJP is the major member and which is now in opposition) has also regularly lost members.
      • In fact most of the smaller parties keep switching between the two major parties. Thus even in an FPTP system voters cannot know, when they vote, what the future governing coalition will look like.
  5. The last objection to PR is that it leads to months of horse-trading and backroom deals to form a new government. Such horse trading and backroom deals are nothing new in India.
    • Aaya Ram Gaya Ram politics in India have been going on for decades.
    • Horse trading in Uttar Pradesh has a long and (un)distinguished history.
    • In 1993 certain MPs of a small party were given “donations” of money to vote for the government and against a no-confidence motion (which the government survived).[11]

Thus given the above evidence I believe, unlike Sanjeev, that India does not need to be wary of proportional representation because any “ills” it has are already manifested in India with its FPTP system. Sanjeev’s position therefore stands refuted.

A Theoretical and Philosophical argument againt Albrechtsen

Turning to a more theoretical and philosophical discussion regarding PR and FPTP, I wonder why Albrechtsen is hostile to small parties. Is it because as a supporter of one of the major Australian parties (the Liberals) she is does not like the feeling of having to negotiate and compromise with other duly elected representatives? It seems to me that the article is arguing that the centre should be allowed to ignore the non-central opinions, that the 60-70% of the electorate has the power to ignore the remaining 30-40% which does not agree with them.

Turning to the issue of smaller parties preventing the adoption of good policies by the centrist parties, what is preventing the centrist parties to come together in support of the good policy and freezing out the smaller parties. e.g. in a 11 seat legislature let’s assume A has 5 seats, B has 4, and C and D have 1 each, with C and D being the non-centrist parties and A being in a coalition with C. If A is pushing a policy opposed by C with C threatening to leave the coalition why doesn’t A solicit support from B? If it is a reasonably centrist (and thus desirable according to Albrechtsen[12]) policy then I don’t see why B and A cannot negotiate some sort of acceptable compromise legislation. It is a failure of the major parties to come together to pass centrist policies and this failure is being disguised, by the likes of Albrechtsen, as the unreasonableness of the smaller parties.[13]

As for smaller parties forcing through undesirable policies, who is letting them? The major parties should be blamed for kowtowing to the smaller parties in their lust for power. If the policy is undesirable to the major party what is stopping it from telling the smaller party to take hike? The fear that it will lose a no-confidence motion? So is staying in power more important to the major party than opposition to bad policy? And instead of blaming the major parties Albrechtsen is blaming the small parties?

Australian Context

One point that should be made is that Albrechtsen wrote this piece in the Australian context. In Australia the upper house is elected by a PR system incorporating a single transferable vote with an “above the line” system. In this system a voter instead of individually ranking each candidate, ranks slates of candidates (each slate comprising of all the party candidates). Since the parties are in possession of these preferences they can then trade them with each other. While such trading agreements are published in advance, they are complicated enough such that it is difficult for the average voter to easily determine the fate of his or her preferences. In such a context parties get enormous power on how to direct the individual voter’s vote. Thus it makes the parties powerful and also it abrogates the link between the elected official and the voters and weakens accountability. I agree with Albrechtsen and Sanjeev that such a system which gives so much power to parties is bad for democracy.[14]

Personally I prefer the Instant Runoff Voting System. This allows voters to show their support for smaller parties without the risk of a major party losing because of a divided vote. Given that the Freedom Team of India (FTI) is a fringe party I am surprised Sanjeev[15] would not be in favour of a system which will allow people to vote for FTI without fears of a wasted vote.[16]

A Note on Language

The last important point I want to note is the language used by Albrechtsen. She uses the word “extremist” and “fringe” to denote the non-major parties. This a point worth noting. By labeling the non-major parties as extremist and fringe Albrechtsen is attempting to confine them to beyond the pale. However what is left unexplained is on what basis should the major parties be respected? Because they are supported by a majority of the public? The smaller parties are to be ignored because they are not supported by the majority? That logic is no different from one justifying the tyranny of the majority; so why constrain[17] the power of a government duly elected by a majority? But is being extremist wrong? In a polity dominated by major parties which do not believe in free trade[18] a position supporting free trade is by definition extremist. In a world where mainstream policy favours protectionism support for free trade is extremist. In a polity where the major parties do not believe in personal liberty[19] a position believing in personal liberty and autonomy is by definition extremist.

Extremism in the defence of liberty is no vice. Moderation in the pursuit of justice is no virtue.[20]

A corollary to the above is the combination of Australia’s compulsory voting system with Albrechtsen’s view that fringe parties are not worthy of representation. It is akin to forcing people to go to the voting booth and then making them choose between alternatives which are both repulsive (one may be slightly less so than the other): “You must buy a car and it could be any colour you want as long as it’s vomit green or feces brown.”

By Albrechtsen’s (and by extension Sanjeev’s) logic parties such as Lok Satta and FTI that are fringe (and extremist given the pro-statist ideologies of both the INC and the BJP) do not deserve legislative representation.[21]

And the very last point. Majority government does not magically provide good policy and governance. A majority government is just as likely as a coalition to promulgate bad policy. It even finds it easier to ride rough-shod over individual freedom and liberty because there is no party in the legislature to challenge it. The only good government is a small government (whether minority or majority), constrained by a constitution with enough space for economic freedom and personal liberty to unleash the power of the free markets and free minds — the surest engine of human growth and progress.


[1] The Australian is the newspaper of the conservative establishment in Australia. Albrechtsen is a columnist for it. It is no surprise that the article is a paean to the establishment.

[2] This is a major point which I will come to later. As of now I will restrict myself to pointing out that the label of “extremism” is used to delegitimise the smaller parties.

[3] As of now its strength is 545 members

[4] The second largest party in that election was the Janata Dal which thereafter split repeatedly

[5] Because the INC and BJP were not the two largest parties

[6] 426 seats; The BJP had 2 seats.

[7] The minor partner was the AIADMK which had 18 MPs in a house of 545 (3%). It withdrew its support to the governing coalition because certain demands were not met e.g. dismissal of the then Tamil Nadu government run by AIADMK’s arch rival DMK

[8] A conglomeration of communist parties

[9] The said agreement is a most complicated agreement and on which I am not an expert. However it goes to original anti-PR point that small parties have a disproportionate influence on policy. My point has been that such disproportionate influence exists in FPTP too.

[10] One may object to this argument on the basis that it is not an apt analogy because in PR voters know that there would be a coalition, they only don’t know what it would look like and that this was not the case in the UK, since the voters expected their party to win (not the lib-dems surely?). However final polls before the start of voting show that no major party was close to getting a majority of the seats, thus raising the spectre of coalition.

[11] See also It’s official; political bribery is tax-free. Ask the taxman,

[12] Coalitions comprising of fringe parties produce “lower-quality policy and politics”. Supra note 1

[13] The issue of FDI on retail in India is the example of such failure by the major parties. As has been detailed above the TMC held the governing coalition hostage. The BJP could have supported the governing coalition as it had supported the policy when in power. Of course needless to say it dropped such a policy when in opposition.

[14] One rationale for why above the line voting was adopted is that since Australia enforces compulsory voting it behoves the administration to make voting as easy as possible. However above the line voting is an attempt to mitigate the impact of a bad policy viz. compulsory voting (based on the same rationale as conscription) by another bad policy such that the net result is even worse.

[15] He is a member and supporter of the FTI. His writings on FTI.

[16] Some may object that IRV denies the “one person one vote principle” but I disagree. While it may seem that the voters whose first preference candidate loses get to vote a second time for another candidate, nobody is preventing any voter from ranking any number of candidates. If a voter declines to choose any candidate apart from his/her first preference, it is no different from an eligible voter abstaining from an election in an FPTP system.

[17] Courtesy Sanjeev Sabhlok.

[18] Only 25 members of the US House of Representatives in the 108th Congress voted consistently in favour of free trade Free Trade, Free Markets: Rating the 108th Congress

[19] The House of Commons passed, by large majorities, the Health Act 2006 which banned smoking in pubs. A private member’s bill to exempt pubs from the 2006 ban was defeated in 2010

[20] This is a paraphrase. The actual quote is “I would remind you that extremism in the defense of liberty is no vice! And let me remind you also that moderation in the pursuit of justice is no virtue!

[21] I have purposefully referred to Sanjeev’s writings on the two major parties of India given that his endorsement of the Albrechtsen’s viewpoint would lead to their entrenchment in the Indian polity.

14 Responses

Subscribe to comments with RSS.

  1. PD – like many of your views, i totally agree with this one as well…

    in my view, FPTP is the simplest but probably the worst of the lot. PR is unnecessary maligned and IRV is the way to go for India (useful and non-controversial). See my short blog post on IRV.

    Let’s move from FPTP to IRV…useful and non-controversial electoral reform
    May 11, 2012

    India’s current system is first past the post (FPTP) where every voter selects the most preferred candidate and the candidate with the highest votes wins. Very simple.

    We need to move to instant run-off voting (IRV) where every voter ranks the candidates.

    Say, there are 10 candidates, the voter knows 5 of them and likes 3 of them. In this case, he should ideally press 3 buttons (first for his 1st preference candidate, then for his 2nd preference candidate and then for his 3rd preference candidate). If a voter doesn’t understand ranking/preference or is unaware, he can simply press 1 button (same as now).

    Counting of votes is complicated (Start with the 1st preference votes, eliminate the candidate with the least no. of first preference votes, redistribute his votes using the second preference of those voters and so on) but easy to automate /leave to experts.

    The cost of moving to IRV is one-time: upgrading the EVMs to allow ranking rather than a single choice.

    The benefits of IRV vis-a-vis FPTP:

    * encourages more people to vote by allowing them to articulate their choice better (its easy to curse people for not voting but currently the vote is too weak for many rational people to take the effort of using it. We need to incentivize the voters by strengthening the vote – introducing right to reject is another way to strengthen the vote)
    * eliminates tactical voting: Currently, instead of voting for their most preferred candidate A, many people vote for a more famous candidate B so as to decrease the winning chances of their least preferred candidate C. B wins despite being less preferred than A. Under IRV, people can choose A as their first preference and B as their second preference
    * fairer outcomes eg: a candidate with a 30% vote but hated by the other 70% might win under FPTP but will lose under IRV
    * encourages more people to contest elections not just by ensuring fairer outcomes but also by eliminating the risk of being a ‘spoiler’. eg: Currently, if there are 2 candidates – a liberal and a leftist. I am also a liberal. Even though I do not like the liberal candidate for some reasons, I do not contest elections for fear of splitting the liberal vote.

    Proportional representation is an even better method but generally opposed since it is believed to substantially increase the likelihood of minority/unstable governments – I don’t think it does but that is a slightly complex argument that I will make some other time.


    July 10, 2012 at 00:41

    • D4All,

      I am intrigued why you believe PR is better than FPTP and even IRV. My major problems with PR:

      1. It requires (usually) multi seat constituency where each constituency is represented by more than one elected representative. This weakens the link between the constituency and the representative and thus the accountability. If the single representative fails to give effect to his/her constituents’ desires then s/he can be easily replaced in the next election. In a multi-member constituency each representative can only focus on the desires and wishes of his/her supporters and ignore the rest. In a single member constituency the representative has to, by necessity, appeal to a large plurality or majority of the electorate.

      2. You may make a valid point that in an FPTP system, in a safe seat the winner can always ignore the minority of the electorate. However s/he knows that she cannot alienate a sufficient number of people because then s/he will lose. In a PR the representative need not have any such worries. S/he only needs to keep a minority happy.

      3. A list-PR system cedes even more power to the party from the voter. Voters vote for the party and then the party chooses who the representatives in the legislature are going to be. This means that for a person to become a legislator his/her primary constituency are not the voters but the party high command. This is similar to the situation in India where to win a ticket to contest an election the candidate has to please the party HQ but is even worse because in India the candidate has to fight the election on his/her name too. In list-PR systems the link between the voter and the legislator is completely severed. It makes the party HQ all powerful (arguably even more so than FPTP where a representative well liked in his/her constituency can challenge the party HQ). Do you think that is a good outcome in India?

      4. Even certain open list systems such as in Australia use the above-the-line-voting system which nudge the voter into voting for the party slate rather than for individual candidates, again empowering the party at the expense of the voter (as I touched upon in the “Australian Context” section in the main post).

      These are some of the reasons I think PR is worse than FPTP and definitely IRV.

      As for the majority government vs minority govt, one should note that both are capable of corruption and abuse of power e.g. 1975 Emergency was imposed by a majority government. Stability in government is not a desirable quality when that government abuses its power in derogation of the rights of the people. The more effective way to fix the government, compared to tinkering with the voting system, is to bind the government down by dividing the powers and by enumerating the powers so that the government cannot make a grab for more powers. As long as the government’s powers is not limited, it will be abused whether by a majority government or by a minority government.

      Polevaulter Donkeyman

      July 10, 2012 at 16:18

      • Hi:

        I think IRV is the way to go for India. PR is better but controversial and the decision to move from FPTP to PR is too complex and time-consuming for our slow political system. Also, I admit my understanding of PR is not enough to be able to answer everything clearly.

        Why PR is better?

        The fundamental big picture case for PR (vs. plurality) is simple and very strong…its based on equity… get around x% say in decision making if you get x% of votes. Under plurality: 1) the winner takes all and 2) the winner is ‘sticky’ and, over time, people just end up trying to choose the less bad amongst the 2 BIG parties. Once you agree with this, there are numerous rules (even if arbitrary) you can use to design a system that gives you a high degree of proportionality (not full proportionality) while ensuring other desirable features.

        response to your points:

        1. PR has numerous variations and is possible to have a single seat constituency (just my thinking…don’t have time to prove…so, ok to ignore assertion). Further, even multiple reps in a constituency is not necessarily bad…a citizen has more than one option to get work done rather than just hope that the next election will give him a better/more preferable rep. Multiple reps can lead to election-like competition throughout the term.

        2. Not sure I understand this point correctly. Under FPTP, slightly higher no. of votes gives you disproportionate power. So, as a corollary, when you lose those extra votes, the disproportionate power will move from you to the new ‘winner’. Why is this a good thing?

        3. I am not necessarily advocating a list-PR system but more power to parties in selecting candidates is not necessarily bad. Under all systems, party HQ have more power in deciding national policy. Why shouldn’t they have power in deciding their candidates? The problem with FPTP is that it leads to a 2 party system such that the party HQs care neither about people nor their own members (since FPTP’s high barriers to entry help maintain their parties’ lead positions). PR systems lower the barriers to entry and true competition among parties (for votes and members) make party HQs accountable.

        4. I think point 3 answers this

        many thanks for your time.




        July 11, 2012 at 07:47

        • D4All–

          1. The issue with multiple winner constituencies is that a candidate that appeals to a narrow slice of voters will be elected. Given that, what incentive does this candidate have to appeal to a broader cross section of the electorate? Especially if this winner has the same legislative power as a winner with a much a larger cross section of support. IRV and FPTP force candidates to appeal to larger cross section of the electorate. Don’t you think a certain democratic legitimacy is granted when a broader cross section of the electorate supports a candidate? (This shouldn’t be construed as support for the “majority is always right”). As for the voter having a choice of multiple members to ask for help, this also dilutes accountability. When a member is not solely responsible and accountable to the voter the incentive for him/her to ignore the voter is also higher than otherwise.

          2. Not sure I understand this point correctly. Under FPTP, slightly higher no. of votes gives you disproportionate power. So, as a corollary, when you lose those extra votes, the disproportionate power will move from you to the new ‘winner’. Why is this a good thing?

          Accountability. This is a good thing because it keeps the incumbent on their toes. They know they have to appeal to a broader cross section to win. This has a moderating influence (partisans may disagreee on whether moderation is good but don’t you think incremental change is more stable and has more staying power over time rather than radical change?) Do you think a legislator should still have that power when s/he declines in popularity?

          3. more power to parties in selecting candidates is not necessarily bad.

          ?? So you think a candidate should primarily be accountable to the party HQ rather than his/her electorate? A party HQ comprising of a small elite (using elite not in the sense of ability e.g. Sonia, Rahul), situated far away (not spatially but mentally also) should have more power on who represents the electorate of a constituency than the voters of that constituency?

          Under all systems, party HQ have more power in deciding national policy. Why shouldn’t they have power in deciding their candidates?

          Lets use a model. There exists a legislature L with 100 members which are allocated proportionally on a list-PR system. A, B, C, D are the parties. In election 1 A = 40% B = 30% C = 20% D = 10% so 40 seats in L are allocated to party A, 30 to B and so on. The members who fill up those seats are based on the party lists produced before election. If A had a list of 100 members it can only choose 40 out of those 100? If party HQ has all the power who among those 100 will be the chosen 40?

          In election 2, A = 35% B = 25%, C = 25%, D = 15%. Now A has lost 5 seats. Which 5 of its earlier 40 legislators would be dumped now?

          This model shows how for a legislator to continue in the legislature his/her primary constituency is not the electorate, it is the party HQ. Therefore it is in the legislator’s interest to kowtow to the party HQ and not rock the boat. So who are the legislators representing in the legislature? The people or the party HQ? Do you think that is good from a democratic viewpoint?

          Atleast in the FPTP system in India the party HQ can only allocate tickets for contesting the election. The candidate still has to work to win the election. In a list-PR the party actually allocates winners. The incentive for corruption is even higher in such a system. (This is no different from “safe seats” in FPTP where the candidate does not have to work to win)

          Note: The model above is very similar to how seats in the Knesset are allocated.

          I am not saying that party HQ should have absolutely no say in who should represent the party in the general election. If you think of parties as brands then certain candidates could exist that are so far out of what the party wants to represent, so that it makes sense for the party to dissociate itself from the candidate. But ultimately it should be up to the electorate to decide who should represent them in the legislature and not the party HQ (Which is why I am also in favour of party primaries to determine the candidates for the party in the general election).

          Polevaulter Donkeyman

          July 11, 2012 at 15:19

          • This is a very interesting discussion and I think one can have productive discussions with you. I think a live call is much more efficient. Do you mind getting on a call with me? My no. is +852 60791436 (am based in HK…2.5 hrs ahead of India…anytime is fine with me). Alternatively, give me your no. and a preferred time, I’ll call.


            July 12, 2012 at 03:33

          • by the way, you can read about ‘Mixed member proportional representation’ methods on wikipedia to see how Germany, New Zealand use proportional representation while maintaining MPs direct link/accountability to voters. This is to substantiate my earlier assertion: “PR has numerous variations and is possible to have a single seat constituency (just my thinking…don’t have time to prove…so, ok to ignore assertion)”


            July 12, 2012 at 09:18

          • “Lets use a model. There exists a legislature L with 100 members which are allocated proportionally on a list-PR system. A, B, C, D are the parties. In election 1 A = 40% B = 30% C = 20% D = 10% so 40 seats in L are allocated to party A, 30 to B and so on. The members who fill up those seats are based on the party lists produced before election. If A had a list of 100 members it can only choose 40 out of those 100? If party HQ has all the power who among those 100 will be the chosen 40?

            In election 2, A = 35% B = 25%, C = 25%, D = 15%. Now A has lost 5 seats. Which 5 of its earlier 40 legislators would be dumped now?”

            In open list system, voters get to influence the MP to be selected. But even in a closed party list system, the party list pre-election is already ranked . So, if you are a good leader but your party ranks you too low on that list (or mistreats you in any other aspect), you simply join another good party or start your own party pre-election and ask your voters to vote for your new party. FPTP doesn’t give you a meaningful choice since 2 parties dominate the scene. In PR, you have meaningful choice since any party’s % of seats equals its % of votes….so, low barriers to entry and exit and high competition


            July 13, 2012 at 01:27

            • 1. It is not easy to leave a party and start one’s own party. It takes time and money to obtain some mindshare of the voters, to familiarise them with oneself, one’s new party, one’s new party symbol, one’s new manifesto and how one is different from the parent party. Therefore the best course for a candidate is to continue pleasing the party HQ.

              2. As for defecting to another party, there too one has to appeal to that party’s HQ to give one preference over other endogenous candidates.

              Ultimately in a closed party list system the party HQ has tremendous power.

              One way to reduce the power of the party HQ (even in a closed party list) is to have primaries for who the candidates would be in the party list.

              But this still isolates the candidates from the voters and does not reduce the power of the party. For democratic accountability it is essential that the constituency representative be intimately linked to the electorate. PR destroys this linkage and thus destroys the accountability.

              Polevaulter Donkeyman

              July 16, 2012 at 18:52

              • thanks for your response PD….i disagree but my apologies, I can carry this discussion further only in a live chat. The current mode, despite its advantages, is too inefficient for this complex issue to be swiftly discussed and resolved by someone as technologically challenged as I am.


                July 17, 2012 at 03:25

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: