Home > World Events > The realist case for electoral reform

The realist case for electoral reform

May 10th, 2005

Via Senator Andrew Bartlett, I see that The Independent is campaigning for electoral reform in the UK, following Labour’s re-election with only 36 per cent of the vote.

Leading opponents within the government are named as John Prescott and Ian McCartney and the story also mentions that Many union leaders also fear it will lead to coalition government with the Liberal Democrats, and prevent Labour from governing again with an absolute majority.

I imagine that the opponents regard themselves as hardheaded realists, but it would be more accurate to view them as reckless gamblers.

Given the outcome this time, and the likelihood of an economic downturn sometime in the next five years, the chance that Labour will secure an absolute majority next time can’t be better than even money.

There’s a possibility that Labour will be forced into coalition with the Lib Dems despite the benefits of first-past-the-post voting, and in this case they’ll have to accept whatever reform package their coalition partners demand. On the other hand, if they act now, Labour can choose the kind of reform they want.

Even more significant, from the viewpoint of union leaders, is (or ought to be) the possibility of another Tory government elected with less than 40 per cent of the vote. A coalition with the Lib Dems might be mildly inconvenient, but not much worse than Blair has been. By contrast, the Tories, given a couple of terms, could easily finish the job they started under Thatcher.

I haven’t looked carefully at the numbers, but I’d guess the best reform for Labour is optional preferential voting. That makes it easy for Labour voters (since, in most constituencies, they can vote for Labour alone as in the past), while most Lib Dem voters would probably give Labour their second preference.

The Tories would get the benefit of preferences from BNP, UKIP and Veritas voters. But this is something of a double-edged sword, as parties like this are prone to demand embarrassing concessions in return for their support.

On the plausible assumption that Labour would get 70 per cent of Lib Dem and Welsh/Scottish nationalist preferences, and the Tories would get 70 per cent of the rest, I estimate a two-party preferred Labour vote of about 57 per cent.

The Tories would need a swing of more than 7 per cent to win because, contrary to the simple calculation above, the Lib Dems would win in some seats and would presumably join Labour in coalition.

The Labour apparatchiks who want to stick with FPP have either failed to do the math or are willing to pass up certain victory just to improve their chances of avoiding coalition. Either way, they are anything but hardheaded realists.

Categories: World Events Tags:
  1. May 10th, 2005 at 20:03 | #1

    Take off your social democrat glasses and think with your head, not your heart. Most of this piece is following through the implications of what you would like rather than checking how realistic your premises and logic are.

    In this area, self deception usually manifests itself as selective omission, i.e. stopping when you like the result, and varying the emphasis according to the SD glasses, rather than any outright error. So, to check, you should refer to someone with recent experience there – preferably several different someones, to cover their own mutual blind spots. I won’t do for anything current, but I can give historical background (15 years ago plus).

    But even with my background I could say what I did in another post on this subject. I can give the gist of the consequences: Labour does not have to accept the poisoned chalice of electoral reform on LibDem terms for a coalition, particularly as it can work as a minority government under a modus vivendi like last time (some 25 or 30 years ago now). They would cling together for fear of something worse, and Lanour would rather be out of power occasionally than never truly in power ever again. And, as I mentioned in the last post, you omit the complications of regional parties and real particularist sentiment even among Labour, that would not accept electoral reform that inherently subjugated their areas to a centralised image

    Oh, and don’t mistake what you analyse as “extreme” parties for really extreme parties until you check with someone with local experience. Also, remember that the Conservatives wouldn’t need the parties‘ sympathies for preferences, only their voters’ – and most UKIP etc. voters are there because the Conservative Party moved away from them, not vice versa. There would be no stigma attached to any such association – any more than there is now, among the Guardian element who would never vote for them anyway.

  2. May 10th, 2005 at 20:10 | #2

    Sorry, I meant “except among the Guardian readers”. Typo rules KO.

  3. jquiggin
    May 10th, 2005 at 20:14 | #3

    PML, I think the wishful thinking is on your part. It would be suicidal for the Lib Dems to support a minority government that rejected electoral reform, and you give no grounds for predicting that they would do so.

    As to your reference to “extreme” parties, I used no such term – the strongest characterisation I gave was “embarrassing”. But you obviously think the cap fits, so I won’t argue.

  4. Scott Wickstein
    May 10th, 2005 at 20:49 | #4

    Well, one can’t imagine the LibDem/Tory coalition, so a minority Labour government seems feasable.

    There’s already a Labour/LibDem coalition functioning in Scotland. So the obvious question is, what happened in Scotland? I do not know the answer off the top of my head.

    But if the Lib-Dems say ‘we want electoral reform’ and Labour say ‘no’ I think the Lib-Dems are in a bit of a bind. What can they really do?

  5. jquiggin
    May 10th, 2005 at 21:13 | #5

    “But if the Lib-Dems say ‘we want electoral reform’ and Labour say ‘no’ I think the Lib-Dems are in a bit of a bind. What can they really do?”

    Wait for a suitably embarrassing moment and refuse to support the government is the usual procedure in such cases.

  6. Andrew Reynolds
    May 10th, 2005 at 21:29 | #6

    Problem with that tactic, PrQ, is that it normally leads to the downfall of the government and the loss of the resulting election.
    As the Libs and Nats in Oz have found you normally have to go in joined at the hip for it to work and try to ignore the more embarrasing of your partner’s oddities.

    That said, in the case of the UK, there is another possible prize up for grabs – the upper house. Now (mis)named the House of Lords (it is not a House and there are few, if any, genuine Lords left in there), there is no reason why a final stage reform could not be to turn it into a fully PR house, appropriately hobbled in much the way it is now.
    The ability to frustrate without actually having to do much would probably keep the Lib-Dems happy for a long time, gradually destroying them from the inside in the same way the Democrats died here.

    You are probably right on the optional preferential. Many Labour supporters probably could not do much more than tick a box whereas the LibDem supporters can generally both count and write, giving a boost to Labour at little cost.

  7. jquiggin
    May 10th, 2005 at 21:37 | #7

    Andrew, the Libs and Nats are really two branches of the same party, not a coalition in the way the term is usually understood. A better analogy is the case of Australian governments who have relied on independent support. Typically, they have had to pay a fairly high price, and quite a few have fallen over anyway.

  8. May 10th, 2005 at 21:43 | #8

    Sorry, my use of “extreme” was a misremembering of what I had scrolled down from.

    But I did give grounds for supposing that the LibDems would go along with an informal arrangement of support of Labour, even if they couldn’t bring themselves to enter an acknowledged coalition (as it happens, Labour could accept that either). I pointed out that they did that very thing, a generation ago. It was back when they were just Liberals, under David Steele.

  9. May 10th, 2005 at 22:32 | #9

    The Scottish example shows that a Coaltion can work fine, in a similar way as it does in many European nations. Scotland has proportional representation, as does the Assembly in Wales (which I think has Labour as minority Govt at the moment, although I may be wrong).

    Reform of the Lords is a spearate issue, although there are links which could be made between the two as part of agreeing on a full reform package.

    The Lib Dems could guarantee support for a minority Labor Government in exchange for electoral reform – whether they would go to the wall of a new election if Labour refused is hard to call and I expect would depend on the community momentum behind the campaign for electoral reform at the time.

  10. Paul Norton
    May 11th, 2005 at 07:44 | #10

    The British Labour “realists” are perhaps better characterised by a line from Beatrix Campbell – “Labour godfathers who hate anybody who isn’t a member of the Labour Party.”

    And John’s argument about the choice facing British trade union leaders is best summarised by a paraphrase of Ho Chi Minh: “It’s better to smell Lib-Dem shit for one term than to eat Tory shit forever!”

  11. Peter
    May 11th, 2005 at 09:12 | #11

    I don’t see why a LibDem-Tory coalition is not imaginable. After all, the main reason the LibDems have such a mish-mash of policies is because they attract Tory voters in the SE and SW of England, whereas they attract Labour Voters elsewhere in Britain. Truly they are schizophrenic party, and it is to Kennedy’s credit that he has held them together despite this.

  12. Peter
    May 11th, 2005 at 09:21 | #12

    JQ —

    I realize this is off-topic, but I thought you’d be interested in this article in today’s (yesterday’s to you) Guardian by environmental journo, George Monbiot, seeking to source an anti-global warming statistic used by David Bellamy. It presents a case study in the elusiveness of the anti-GW camp:

    here

    http://www.guardian.co.uk/comment/story/0,,1480279,00.html

    I hope that the story is taken up by “New Scientist” where Bellamy’s falsehood first appeared.

  13. wilful
    May 11th, 2005 at 11:01 | #13

    It’s a pity that these debates are nearly always structured on who wins and loses amongst the various power groupings. Why can’t the debate be had on it’s merits as being good for democracy? Or am I just being naive?

    If FPP is changed, I hope that they go for something sophisticated such as the Condorcet method ( http://en.wikipedia.org/wiki/Condorcet_method ) rather than Australia’s cruder instant runoff. Not that I would dare sugegst a change to the Australian system, given all the draculas wanting to be in charge of that blood bank.

  14. Paul Norton
    May 11th, 2005 at 11:25 | #14

    wilful, I sympathise absolutely with the sentiments in your first paragraph. The problem is that in most parliamentary democracies the electoral system is determined by Act of Parliament, i.e. by people with a vested interest in the outcome of their decision. This can lead to scandals like the Tasmanian Lab-Lib collaboration to alter the structure of the Tasmanian Lower House with the ulterior (and ultimately unsuccessful motive) of excluding the Greens, and the (probably unintended) consequence of creating great scope for serious and potentially absurd mismatches between electoral support and seats won.

    I think there is much to be said for the electoral system to be determined by a combination of popular referendum (to decide the substantive nature of the system) and an independent electoral commission (to decide technical details such as electorate boundaries, enrolment requirements, etc.).

    The other thing to bear in mind about electoral reform is that there are two basic issues to consider. One is the issue of how votes, once cast, are translated into seats won, which underlies arguments about FPP, preferential voting for single-member constituencies, proportional representation, etc. The other is the issue of ensuring that the act of voting faithfully translates the intentions and wishes of the voter into the vote they cast. The sort of people who engage in discussions like this in media like this tend to focus exclusively on the former, as we are generally cognitively and communicatively skilled people for whom the latter is not an issue for us personally when we vote (e.g. we know that an above-the-line Senate vote won’t reflect our actual preferences, and we have the skills and knowledge to cast an informed below-the-line vote). For about 90% of Australians the second issue is a serious one, and in the absence of attention to this issue by those of us who have a principled interest in the matter, there is a space big enough to drive a bus through for self-interested political actors to propose mischievous “solutions” such as above-the-line Senate voting and Queensland Labor’s “just vote one” campaign in the context of optional preferential voting in State elections.

  15. Dave Ricardo
    May 11th, 2005 at 11:55 | #15

    In today’s (London) Daily Telegraph, Ferdinand Mount makes the case for proportional repreantation from the Conservative Party point of view. He says that the Tories will be locked out of power while ever they are sitting on 30-33% of ther vote, which they have been since the 90s, and PR may deal them in, as it has dealt them in the Scottish and Welsh parliaments.

    Labour, he says has a lock on all the seats in a lot of cities whereas the Tories would get some representation under PR.

    I guess it all depends on who would form a coalition with whom under a PR system. If you’re willing to assumr to the Lib Dems would be together eith labour all the time, then Mount’s argument makes no sense. But you never know. Shifting coalitions do occur under PR. Look at Germany, where the Free Democrats shifted from the SDP to Christian Democrats.

  16. Mike Pepperday
    May 11th, 2005 at 12:02 | #16

    Wilful, the debate in parliament will be almost exclusively about the democratic merits so if that is what you want, read Hansard. Mealy-mouthed crap of course. Democratc improvement occurs as a by-product of manoevring by politicians for power advantage – which is the theme here.

    JQ, The sort of situation that makes you say: “On the other hand, if they act now, Labour can choose the kind of reform they want.” is quite common.

    In introducing PR, the Bracks govt acted on this basis (after copping the Kennett devastation) but I think it is an exception and that usually the pollies can’t think ahead.

    The problem in Australia is personal agendas and intra-party division. The WA Liberals make a great case study. In 1923 they could have set it up to please themselves. They couldn’t act so they had to cop Labor’s reform in 1928. During the Charles Court years they could have done anything – even the Nationals were neutralised – and though they knew the system was crooked they couldn’t get their act together.

    In the years of the Brian Burke govt they couldn’t agree what to do. They could have connived with Labor to permanently wreck the Nats and shut out minors forever but they couldn’t get their act together. This left the newly rejuvenated Nats to call the shots on reform and introduce minors to parliament.

    Now it’s on again – Lib leader Matt Birney refuses to discuss electoral reform with Labor. Members of his party are (or were, it’s probably over now) calling for it. It is pathological. You have to conclude they’d rather be toasted by their opponents than work out what to do among themselves.

    The WA situation is a century old and will continue as long as the system is crooked (country voters have a 3 to 1 weighting over city voters in the upper house). In Britain the system is also crooked in the sense that FPP is unfair (and also that they can’t decide what to do in the Lords) so a similar running sore is forseeable there.

    A basic problem lies in the fact that the electoral rules are SO important to the pollies but a big yawn to the public.

  17. michael.burgess
    May 11th, 2005 at 14:08 | #17

    ul Norton says that ‘The British Labour “realists� are perhaps better characterised by a line from Beatrix Campbell – “Labour godfathers who hate anybody who isn’t a member of the Labour Party.� Well Beatrix Campbell is one of the lunar left who has never forgiven Blair for getting Labour elected after the disaster they had become with the likes of her influencing policy. Blair has also not done a bad job in areas such as health and education and general economic management, certainly a better one than his critics would have done. The third party in the UK like the ex-third in Australia favour populist policies over substance (including opposing the liberation of Iraq) and, as such, provide a good argument against proportion representation. One can only hope that the greens never hold the balance of power in Australia.

  18. May 11th, 2005 at 14:17 | #18

    I am reminded that coalition governments are possible under the FPP system. Winston Smith-Churchill became Prime Minister in 1940 with a National Government which the Labour Party of the day decided to join, and in which they played a significant role.

    The realist argument warrants consideration, from either a Labour or Conservative point of view, but falls away in the light of circumstances. I suspect, other than during special circumstances, sustaniable policies are effectively implemented by argument and negotiation rather than forced implementation.

    Interest groups, I suppose, can be influential even without representation. The acceptance of this view would suggest that Labour or Conservative Parties can gain by a system of open negotiation with minority parties, other than the fact that, at least to me, it seems more democratic, and certainly superior to the secret conclaves of the party room where we are assured the real dissent and decisions take place.

    Multi-party government is better for electors, provided the representation is broadly proportional to the their preferences, and that they in turn have the power to make effective strategic voting decisions besides non-voting or voting for a party, in the first instance, which is not their preferred choice.

    It seems to me, that is there is a disconnect now between what the Prime Minister intends, and what was at issue during the election, spcifically in regard to Iraq.

  19. May 11th, 2005 at 14:21 | #19

    Hmm… Cordocet. It’s quite a different philosophy from preferential voting. Instead of worrying who comes first, you work out who comes in front of another. And to do this, you use a meaty algorithm to determine the result, with pairwise matrices, and majority rule cycles, which of the six or sevent kludges you use to resolve Cordocet paradox.

    Now, computers can do that fine, if you enter. But that doesn’t help with transparency, and trasparency is what you want in a voting system. You want both scrutineers and electoral officers to be able to look at the paper ballots, and quickly determine how the whole thing is going. And IRV is quite easy to understand when you start stacking ballots by 1st preference, 2nd preference.

    If the U.K. voting system goes to the people (as Andrew suggests, and I endorse), I think Cordocet will not really appeal to people for it’s complication. Especially the “Cordocet paradox!” Scare quotes to scare punters. IRV can be tied as well, and it seems to satisfy less of the voting system criteria than any others. But on the other hand, one can point to more than 100 years of successful implementation in Australia.

  20. Paul Norton
    May 11th, 2005 at 14:42 | #20

    I wonder if Michael Burgess has actually read any of Beatrix Campbell’s writings or statements about British politics, the British Labour Party and the British Labour Left in the 1980s? Insofar as there was a “lunar left” at the time, she was one of its pet hates, along with Eric Hobsbawm, Stuart Hall, Martin Jacques et al, because of their criticisms of the hard left’s orthodoxies and its exclusionary way of doing politics. Indeed, her 1985 comment about “godfathers” is at least as likely to have been directed against elements of the hard left as other elements of the Labour Party. It was not directed against the New Labour/Third Way current (which didn’t exist at the time) and was not directed against Tony Blair (insofar as he was on the radar in 1985, he would have been thought of highly by the Marxism Today crew with which Bea Campbell was associated.)

    MB’s post is indicative of a trend in his posts of generalised denunciation of this or that hated other (Bea Campbell, Peter Garrett, David Suzuki, public service feminists, public service greenies, public service management, the anti-war left and members thereof, critics of Israel, etc.) whilst seldom detailing the grounds for such denunciation, and even less frequently providing evidence or arguments in support. It is very difficult to engage in any kind of constructive dialogue with interventions of this kind.

  21. May 11th, 2005 at 14:44 | #21

    “Winston Smith-Churchill”? No such person ever became PM.

    Winston Leonard Spencer-Churchill, however…

  22. michael.burgess
    May 11th, 2005 at 15:16 | #22

    Paul Norton, rather than engage in rational debate you resort to insult as usual – don’t get carried away with the fact that many people on this site agree with your extreme views they are not a selective sample. I have heard Be Campbell frequently on Phillip Adams. She clearly is a left extremist and I will call her one. Anyone who can object to me criticising Peter Garrett and David Suzuki is also clearly an extremist and a naïve fool. Don’t you get it is people like them who have done so much damage to the broad left of politics. Furthermore, give that in my 30 years of political activist on the left of politics, I have never succumbed to Marxism, neo-Marxism or other ridiculous and harmful ideologies I feel I have some licence to condemn those who have done so and in the process harmed the left. As for criticising those who criticise Israel well it sickens me that many on the left have now joined with the extreme right, sections of European policy elites and the majority of the Muslim world in indulging in anti-Semitism. Did you see the cutting edge doc on how journalists have grossly misrepresented Israeli actions last night? I make no apologies for my use of the term anti-Semitism, as know one who has studied this issue in any depth can possibly reach the conclusions that many of them do without being a bit sick in the head. Most people though are simply ignorant and jump onto a fashionable cause. Jew hatred how fashionable is that. It makes my justified criticisms of self-serving feminists in the public sector and academia look very mild in comparison.

  23. michael.burgess
    May 11th, 2005 at 15:38 | #23

    Paul, I also suggest that you take a look at the line up of this years Sydney book fair, if you think I overgeneralise about the failings of the left. While there is some variety, the headlines acts are (the ridiculous) Susan George, Tariq Ali, a Palestinian activist/writer whose name I can’t remember, and —wait for it – Peter Garrett interviewing David Suzuki. So much for the literature community challenging orthodoxies – pathetic.

  24. Benno
    May 11th, 2005 at 18:29 | #24

    **Condorcet**

    “meaty algorithm” I don’t think so. Only one count is done to determine the winner, so each ballot is counted once and tallied into a matrix, soon to be joined by all of the other ballots. Counting can be done by hand and no computers are needed. http://en.wikipedia.org/wiki/Cloneproof_Schwartz_Sequential_Dropping

    As for PV (IRV) it takes n-2 counts where n is the number of candidates standing. And circular ties are easy to resolve, the fact that condorcet reflects the true ambuguities of the electorate is not a problem with condorcet. If A is prefered to B 65% to 35%, B is prefered to C 60% to 40% and C is prefered to A 55% to 45%; in the best method of ambiguity resolution CSSD, we elimate to weakest defeat which is C against A and then A is the winner.

    have a look at the mock election for the state of tennesee and the counting matrix to go with it. Very straight forward. For info on why Condorcet is superior to preferential voting go here http://www.electionmethods.org/IRVproblems.htm

    It may satisfy less voting criteria, but a condorcet method is the only method to satisfy the most important democratic criteria of all: “If a candidate is prefered by a majority over all other candidates, then that candidate should win”

    Actually mate, having a look at the link you provided, condorcet satisfies both criteria that IRV does and more. Lies fithy lies.

    I will roll anybody who claims that PV is superior to Condorcet.

  25. May 11th, 2005 at 19:07 | #25

    PML:

    I knew there would be a double-barrel there somewhere. I just got the bloodlines mixed. It could not be as common as Smith. In fact, I was half fictional, nevertheless the Con-Lab coalition did exist, and is perhaps required now, as then, to defend and further the cause of liberal democracy.

  26. Peter
    May 11th, 2005 at 22:56 | #26

    I thought Winston Smith-Churchill was an elaborate reference to Orwell’s “1984”, perhaps equating WC with WS. Ironic, in view of the common belief (perhaps also true) that Orwell chose his character’s name as a reference to WS.

  27. Paul Norton
    May 12th, 2005 at 08:52 | #27

    Michael Burgess claims that I insulted him. I assume he finds the following comment insulting:

    “[There is] a trend in [MB’s] posts of generalised denunciation of this or that hated other. . . whilst seldom detailing the grounds for such denunciation, and even less frequently providing evidence or arguments in support.

    This is not an insult. This is a criticism of Michael’s practice of making pejorative assertions about people without substantiating them by evidence or argument. It is, I submit, a justifiable one when one reads Michael’s subsequent reply, which includes inter alia the following:

    “Anyone who can object to me criticising Peter Garrett and David Suzuki is also clearly an extremist and a naïve fool.”

    I do not object to Michael (or anyone else) criticising Peter Garrett and David Suzuki – or me for that matter. All I ask is that he support his “criticisms” with facts, figures and rational arguments. As it is, he has posted a series of swingeing remarks about Garrett, Suzuki, and others being “green extremists” and the like without once providing us with examples of what he considers to be “extremist” views held by such people.

    Likewise with Beatrix Campbell. Michael asserted (without providing evidence) that Campbell was part of a nameless “lunar left” which was influential in shaping British Labour policy in the 1980s. I responded by referring to Campbell’s actual views and alignments in the 1980s (which can be verified by anyone prepared to visit a library which carries back copies of British Left journals from that period), and the best Michael can come back with is “I heard Bea Campbell on the radio the other night and she’s clearly a left extremist” – again, without providing us with a link to the transcript or otherwise citing anything Campbell actually said in support of this claim.

    On the matter of the Sydney Writers Festival (not the Sydney Book Fair), Michael at least provides some evidence and examples – except that they are a selective and unrepresentative sample. It would be just as valid – or as silly and specious – for someone from the Spartacist League to post a comment drawing attention to the presence of Louis Nowra, Andrew Riemer, Helen Garner, Les Murray, Martin Krygier,

  28. Paul Norton
    May 12th, 2005 at 09:01 | #28

    I pressed “submit” before I’d finished my comment. Here’s the full comment:

    Michael Burgess claims that I insulted him. I assume he finds the following comment insulting:

    �[There is] a trend in [MB’s] posts of generalised denunciation of this or that hated other. . . whilst seldom detailing the grounds for such denunciation, and even less frequently providing evidence or arguments in support.

    This is not an insult. This is a criticism of Michael’s practice of making pejorative assertions about people without substantiating them by evidence or argument. It is, I submit, a justifiable one when one reads Michael’s subsequent reply, which includes inter alia the following:

    “Anyone who can object to me criticising Peter Garrett and David Suzuki is also clearly an extremist and a naïve fool.�

    I do not object to Michael (or anyone else) criticising Peter Garrett and David Suzuki – or me for that matter. All I ask is that he support his “criticismsâ€? with facts, figures and rational arguments. As it is, he has posted a series of swingeing remarks about Garrett, Suzuki, and others being “green extremistsâ€? and the like without once providing us with examples of what he considers to be “extremistâ€? views held by such people. There is a reference a bit further on to “Marxism, neo-Marxism” and to people who have “succumbed” to such “harmful and ridiculous ideologies” who Michael feels he should criticise. If this is intended to apply to Garrett or SUzuki, it is simply risible.

    Likewise with Beatrix Campbell. Michael asserted (without providing evidence) that Campbell was part of a nameless “lunar left� which was influential in shaping British Labour policy in the 1980s. I responded by referring to Campbell’s actual views and alignments in the 1980s (which can be verified by anyone prepared to visit a library which carries back copies of British Left journals from that period), and the best Michael can come back with is “I heard Bea Campbell on the radio the other night and she’s clearly a left extremist� – again, without providing us with a link to the transcript or otherwise citing anything Campbell actually said in support of this claim.

    On the matter of the Sydney Writers Festival (not the Sydney Book Fair), Michael at least provides some evidence and examples – except that they are a selective and unrepresentative sample. It would be just as valid – or as silly and specious – for someone from the Spartacist League to post a comment drawing adverse attention to the presence at the Festival of Louis Nowra, Andrew Riemer, Helen Garner, Les Murray, Martin Krygier, Paul Sheehan and doubtless other participants to “prove” that the event is a theme park for reactionaries and ex-left traitors.

    Finally, other readers can form their own judgement about a contributor who complains about being “insulted”, and then refers to others by terms such as “extremist” and “naive fool”.

  29. wilful
    May 12th, 2005 at 11:24 | #29

    These in-thread personality debates, generally quite off-topic, are a bit tedious for me. But hey, maybe I’m the minority, and I’m not in charge around here. Knock yourself out, you lot.

    I thought that the Brack upper house reforms were excellent (as a citizen that believes in democracy rather than the clash of historic power groups). While strategically they are likely to stop the unrepresentative tory swill (sorry, I meant the National Party) from controlling the upper house and misusing their power, they don’t appear all that likely to greatly benefit Labor. It seems likely that Greens will win a few seats. Damned proportional representation.

  30. Paul Norton
    May 12th, 2005 at 11:38 | #30

    You’re right, wilful.

  31. May 12th, 2005 at 14:51 | #31

    Benno,

    I do not like your IRV Problems link. I do not like it at all. From reading it, it was written by a person who knew that Australia uses IRV, but little else. Other men would look at Australia’s 100 year-old test case of electoral history, and be delighted with the masses of evidence to argue his case.

    Others would want to ignore the facts – dirty, messy, inconvenient facts – that would marr the purity of his argument. Instead he comes up with a lot of unproven hypotheses, guesses and the odd personal prejudice thrown in, and then reasons and argues from there. And this Cordocet advocate appears to be of the second persuasion. I do not trust authors like this, because selective ignorance of small messy facts often leads to messes in real life. (e.g., 2nd War in Iraq, The).

    For example:

    “As soon as one of those minor parties threatens to become a major party and actually win, its supporters vote for them at the risk of hurting their own cause, just as in the current plurality system. Under IRV, votes for minor parties are therefore symbolic at best, or dangerous at worst.”

    I do not understand this. Is the author ignorant of the Australian Senate (which uses multiple-member IRV for selection), and how the Democrats and Greens held the balance of power more-or-less successfully for roughly 20 years? And we also see:

    “It is also much more difficult to implement with security and integrity because the votes cannot be summed as in most other election methods.”

    That’s just not true. And I say that as a man with experience as both a booth scrutineer and a vote counter for IRV elections. With enough human scrutineers – supplied by the political teams running for the election – and paper ballots – you can ensure the integrity or lack of pretty much any election. The result generally can’t be decided until all the ballots are driven to HQ, but that’s pretty easy and secure in this year of our lord 2005.

    Finally, saying “Lies, filthy lies” does not help you propagate the wisdom of Cordocet. Can you point to a lie I said? If not, could you please refrain from name-calling? If there was really a referendum for voting methods in the U.K., do you think the voters would like such rhetoric in the newspapers? They’d rather stick with first-past-the-post.

  32. Benno
    May 12th, 2005 at 16:37 | #32

    “And to do this, you use a meaty algorithm to determine the result”

    First lie, it is simple not meaty and easily counted by hand with each ballot looked at twice.

    “…. and it seems to satisfy less of the voting system criteria than any others.”

    Seconnd lie, as I have already pointed out from the link you gave me (wikipedia is a really great resource for electoral methods by the way)

    “But on the other hand, one can point to more than 100 years of successful implementation in Australia.”

    Not a lie but inaccurate when comparing condorcer to IRV. It has certainly been successful compared to first past the post, but the winner in IRV depends upon the chance element of candidate eliminations and preference redistributions, whereas all preferences are counted simulataeniously for all ballots.

    Here though your concerns are upheld until you get a better link, the link I sent you is from someone who doesn’t know much about Australia’s ‘lectoral system. But it is accurate in some parts and the best example I could find. However I have written an article which is more readable.
    [email protected] send me an email and I will email it to you (I don’t know when it will be published on online opinon)

    Clarification: each ballot can be fed into a computer more easily then hand counting, if they were to be counted by hand it would take a while of staring at each ballot to record each preference compared to all candidates. But after that the ballot doesn’t need to be looked at (except for auditing). And ballots can be counted seperately and tallied at the end, without waiting for all booths to have finished each count before moving on to the other one. So overall it is more efficieint but given that we only have to wait about 2.5 hours for our results certainly not a big issue. The beauty comes in the presentation of results in the matrix not speed of counting.

    “As soon as one of those minor parties threatens to become a major party and actually win, its supporters vote for them at the risk of hurting their own cause, just as in the current plurality system. Under IRV, votes for minor parties are therefore symbolic at best, or dangerous at worst.�

    This does sound a bit off the air, because I think that Preferential Voting is a pretty good system. So an example would help to understand it.

    1. 2. 3.
    45% A C B
    35% B C A
    20% C B A

    in this 45% of voters have marked their first preference for candidate A, second preference for C and third for B. etc…

    FPTP : candidate A wins even though a majority prefers both other candidates to A

    IRV : candidate B wins even though a majority prefers candidate C

    Condorcet : C wins because a majority prefers C over all other candidates. it thus satisfies the condorcet criterion. “If a candidate is prefered over all others… blah blah.”

    Once someone has given condorcet a few weeks of thought and read some stuff, Like me, I still can’t completely endorse it for Australia until the AEC does two counts at the next election so that the public can compare the outcomes. So email me and I will send you something I hope is more coherent.

  33. Mike Pepperday
    May 12th, 2005 at 21:08 | #33

    Of course the Bracks reform of the upper house was excellent. Arguably the best designed upper house in the country. But don’t kid yourself it was done for democratic reasons. It was done to secure Labor’s position.

    It has to be PR. There’s no alternative. There are three possibilities for the makeup of an upper house after an election: (1) the govt party has a majority, (2) the opposition party has a majority, (3) no party has a majority. That’s it. Which to have?

    If the govt has a majority, it’s pointless – the house is a rubber stamp (just as the lower house is). If the opposition has a majority you get party political obstructionism. The only sensible solution is to have no party with a majority and the only way that is reasonably likely is with PR.
    QED.

  34. Benno
    May 12th, 2005 at 22:37 | #34

    what about Condorcet proportional representation?

  35. May 13th, 2005 at 12:54 | #35

    Benno, there is no such beast as Condorcet PR. Condorcet can only be used in an election for 1 position. There are significant weaknesses in the Condorcet algorithm, which probably explains why it is used so infrequently in the real world.

  36. Benno
    May 13th, 2005 at 19:06 | #36

    first of all what does this mean? “There are significant weaknesses in the Condorcet algorithm, which probably explains why it is used so infrequently in the real world.”

    Condorcet is used infrequently, but so is Preferential Voting. Not many people know about Condorcet and not many people are interested in any sort of electoral reform, unless it is towards PR. I don’t understand what you are talking about with ‘weaknessess in the algorith’. What is with knocker’s obbessions about ‘meaty’ and ‘weak’ algorithms? It is like all of this talk about the war on terror or ‘the economy’ (we have to make sure the economy is feeling alright, tuck it into bed etc..)

    As a candidate ranking system it truly reflects the will of the electorate. That is what the algorithim does. It satisfies the Condorcet Criterion which I take to be the ‘democratic criterion’. Only held sacred by the immaculate bosom of a Condorcet Method.

    And I have come up with PR Condorcet system. But I would not recommend it for use with many seats to be distributed because either too much work needs to be done by the voter in ranking many candidates, or candidates are predominately selected by the party and not the voters.

  37. Benno
    May 13th, 2005 at 19:15 | #37

    System is as follow:

    1) rank candidates as for normal condorcet method
    2) determine condorcet winner, that candidate is allocated first seat
    3) remove that candidate from the ballot
    4) start over again, picking successive condorcet winners until all seats are filled.

    It should be obvious that to use this in our current lower house (150 seats) would require a lot of marking, Unless the voters were allowed to blindly follow the parties order for their own candidates, like preferential voting above the line in the senate. Unfortunately this would favour substandard candidates over better ones. The idea being that people vote for their liked candidates regardless of which party they’re from. This way (ideally) we can all vote for and against mark latham and john howard. Don’t like the leader of your preferred major party? dont vote for them.

    It would also keep a connectedness between electorate (australia wide) and representatives.

    In conclusion provide examples/evidence for your case of weakness in the algorithms. But I promise you it doesn’t exist.

  38. Blue Bottle
    May 14th, 2005 at 19:03 | #38

    Pauses for an answer to the previous post, not a sausage.

    Kome on Kondorcet Knockers, it is time to put up (arguments and examples) or shut up (stop lying for the sake of your precious Proportional Representation).

    Thinks: Lovely, I have simultaneously bordered on the border of civilised discussion and name-calling and insinuated that all people critical of my position on this issue are equivilant to violent bible bashing stupid racists.

    I like dis game.

  39. May 14th, 2005 at 20:38 | #39

    One of the significant weanesses n Condorcet is its tendency to generate insoluble paradoxes.

  40. Benno
    May 14th, 2005 at 22:21 | #40

    Ok. First of all there are no significant weaknesses in Condorcet, therefore this isn’t a significant weakness. Secondly, a Condorcet method doesn’t generate paradoxes. Thirdly the true ambiguities in the preferences of the electorate when manifested as a paradox are not insoluble. They aren’t even insolvable. That is three lies.

    Now let’s say in an election we have 5 candidates A B C D and E. In outcome 1, one candidate is prefered to all other candidates by a majority so that has been discussed. In another outcome 65% of the electorate prefer A over B, 60% prefer B over C and 55% prefer C over A. Given also that A, B and C defeat both other candidates this means that we have 3 Condorcet winners not one and thus a paradox. Note here that the Condorcet method didn’t generate this paradox, the true preferences of the people from which all legitimacy is obtained generated it.

    These three candidates make up the Schwartz set, which is the innermost set of candidates which are unbeaten by all other candidates outside that set, I refered to them before as our condorcet winners.

    Now the best method of resolving a cyclic ambiguity is Cloneproof Schwartz Sequential Dropping CSSD(look it up at wikipedia), what happens is that we eliminate the weakest defeat or the smallest majority by which a defeat in our Schwartz set occurs and then recalculate the Schwartz set, we do this until we have one winner. So in this case we remove C from the list because it only defeated A by 55%. Our new set then becomes A. So A is the winner.

    Btw are you an idiot or has your blog been hijacked? If the latter could I join the crusades?

  41. May 15th, 2005 at 15:23 | #41

    Benno

    Your scheme for electing multiple representatives by Condorcet counts the votes of the winning candidate’s supporters as many times as there are seats to be filled. You’ve (roughly) invented the old block preferential system used for the Senate before 1949.

  42. Benno
    May 15th, 2005 at 22:37 | #42

    Everyone’s votes are counted as many times as there are seats to fill, not just the winning candidate. And anyway votes are only counted once, just final tallies are looked at multiple times and of course there are as many winning candidates as there are seats to fill, not just one. If you are going to have a conversation about electoral systems you should be accurate because of the nature of electoral methods.

    I’m very sceptical that I have reinvented a block preferential system, but I’ll check it out before I disabuse you of this notion.

  43. Benno
    May 15th, 2005 at 23:02 | #43

    A semi-deabusing Alan. Maybe you aren’t just trying to stir me up, not that I need very much. It is ironic that I have hijacked a post titled “the realist case for electoral reform” when it should be “The democratic case for electoral reform”. Why should realism get in the way of true democracy? IF someone disagrees with you, you should take a deep breath and a glass of water, and then download an episode of the young ones.

    Ok quite similar, both require what either defines as a democratic majority, removal of succesful candidates and recalculation, the difference being that Block Preferential Voting is a Preferential method which does not satisfy the Condorcet Criterion.

    My system is of course easier to count as one count is needed, not metric tonnes and metric f***ing tonnes. But block preferential voting is a much fairer version of D’hondt, which is heavily biased in favour of parties who can field many candidates and against indepenmdents who usually field just themselves (or two for a group box).

    So the question is why did we change to our current PR STV? Maybe because in a Preferential rather than Condorcet context it was faulty compared to what we have now, but I’m not sure.

  44. Tristan McLeay
    May 15th, 2005 at 23:35 | #44

    First of all there are no significant weaknesses in Condorcet, therefore this isn’t a significant weakness.

    Benno, much as I am interested in the Condorcet methods, I don’t really think that’s an argument.

    Also, I’m not sure that your proportional Condorcet method would really be proportional. I wonder whether it would even end up with a generally different make-up from single-member Condorcet, partywise in non-trivial election mock-ups. (Is it possible to get the full data from preferentially-held elections in any jurisdiction?)

  45. May 16th, 2005 at 10:22 | #45

    Far from being proportional in any way, the block Condorcet method would return a hugely skewed assembly. The block preferential method regularly gave Australia senates where the numbers were 36/0. If you work out a block Condorcet election using actual numbers, you’ll find quite quickly that the group of votes that elects the first Condorcet winner then elects the second Condorcet winner and so on and so forth.

    Each of these Condorcet winners will come from the same party. In STV the group of voters who elect a winner don’t get their votes reused to elect the next candidate. Maybe inexhaustible transferable vote is a better name for Benno’s proposal and maybe there’s a good reason why more seasoned Condorcet advocates hve never argued that Condorcet can be used to elect an assembly.

  46. Benno
    May 16th, 2005 at 11:59 | #46

    “”First of all there are no significant weaknesses in Condorcet, therefore this isn’t a significant weakness.”

    Benno, much as I am interested in the Condorcet methods, I don’t really think that’s an argument.”

    It is logical therefore it is an argument, but it is smartarse sort of line designed to combat what I was probably unfairly calling lies.

    “Also, I’m not sure that your proportional Condorcet method would really be proportional. I wonder whether it would even end up with a generally different make-up from single-member Condorcet, partywise in non-trivial election mock-ups.” – Very true, particularly the fact that it isn’t proportional, just a mutli-member system.

    “Far from being proportional in any way, the block Condorcet method would return a hugely skewed assembly. The block preferential method regularly gave Australia senates where the numbers were 36/0. If you work out a block Condorcet election using actual numbers, you’ll find quite quickly that the group of votes that elects the first Condorcet winner then elects the second Condorcet winner and so on and so forth.” – This is true too, unless voters look for candidates rather than parties, but that is unlikely to happen for practicle voting reasons as well as what past mas voting history suggests, so both systems aren’t proportional at all.

    “Maybe inexhaustible transferable vote is a better name for Benno’s proposal and maybe there’s a good reason why more seasoned Condorcet advocates hve never argued that Condorcet can be used to elect an assembly.” – I’d be an idiot if I didn’t agree to this. Thanks for the name, it is really good, I’ll be laughing over it for quite a while.

    My starting point was “I have found the best single member electoral method, so what is that best multimember system, obviously a multimember electorate would be silly unless it was proportional, otherwise it would be even less representative than 150 electorates of fptp.

    I think d’hondt is pretty good, unfortunately it severely disadvantages independents.http://wiki.electorama.com/wiki/D%27Hondt_method

  47. Benno
    May 16th, 2005 at 12:04 | #47

    “(Is it possible to get the full data from preferentially-held elections in any jurisdiction?)” It probably is because studies are done on types of informal votes. The closest thing I have now, just from primary vote datum and heresy, is an analysis of the battle for wentworth 2004. An extract from my yet-to-be-published article:

    Consider the recent case of the electorate of Wentworth, regarded as a possible four-way tussle by some, but realistically only ever a competition between Peter King and Malcolm Turnbull. The final contest was between Labor on 44.52 % and the Liberals 55.48%, but did a majority of the electorate actually prefer Peter King to Malcolm Turnbull? They would have if all of the Labor and Green voters preferred King to Turnbull; the approximate results would have been 56.5% for King and 43.5% for Turnbull. Of course without access to all of the ballot papers, whether King was preferred to Turnbull is pure speculation and even if he were, that wouldn’t by itself guarantee a Condorcet victory.

  48. May 16th, 2005 at 12:39 | #48

    One advantage STV has is that its operation does not depend on district magnitude. I take it we now agree that inexhaustible Condorcet voting is not a good idea?

  49. Benno
    May 16th, 2005 at 13:07 | #49

    Yep, oh yep, totally, uh huh.

    A disadvantage of STV is that I don’t like it. But Tristian would say that that isn’t a proper argument. So in that case the disadvantage is that your vote is reduced once a choice of yours is elected in proportion to how many other people voted for it. For instance I vote 1 democrat and enough people vote democrat that they have 1.3 quotas with first preference. Lets also say that they don’t have a chance of winning more than one quota. So when my second preference after all democrat candidates is passed on it is passed at a value less than one. But if I had decided not to give democrats my first preference because I thought that they would achieve a quota without me, then the value of my next vote gets the full 1. That is why D’hondt is good, but it only allows you to vote for one party and a vote for an independent would be wasted regardless of if they get a seat. Wasted is the wrong word but something similar.

    So STV is better than Condorcet inexhaustible voting and Preferential inexhaustible voting. But it is still not poifect.

  50. Tristan McLeay
    May 16th, 2005 at 18:33 | #50

    I believe ‘Modern’ forms of STV have fixed that bug you describe, and after someone is eliminated or a surplus transferred, their votes are passed on to the (very) next candidate and surplus is recalculated. So if I voted for someone else before the Dems, after my vote passed through the Dems it’s transferred at the value of .3q+1, same as all votes electing Democrats. Obviously this makes counting rather more complex but it’s in the name of democracy and it keeps the spirit of STV alive.

    I believe NZ uses this modern form of STV for their council elections, but I don’t know if anyone else does.

    STV still isn’t perfect (I’m still pissed that my vote in the Senate election didn’t get counted, and I’ll be rather more pissed come 1 July), but it’s better than anything else I know of.

  51. Benno
    May 16th, 2005 at 22:13 | #51

    Yeah well D’hondt is kinda like what you say, except that you can only vote for one party, so only transferrable to the next candidate in that party. Of course this could be ammened. Perhaps an approval method where each voter puts an X (yeah that’s right, a dirty filfhy X) to as many candidates as are seats. The rest is easy. Anyway more thinking and reading needs to be done here, or something.

  52. May 17th, 2005 at 01:39 | #52

    Approval voting has serious problems, in particular it dramatically encourages tactical voting. What you’ve described is actually not approval voting, it’s the block preferential monster used for the senate before 1949, the one that gave 36/0 results.

  53. Benno
    May 17th, 2005 at 17:52 | #53

    Sorry, two systems here: one is d’hondt which I meant to provide a link to before, the other is http://wiki.electorama.com/wiki/D%27Hondt_method

    I actually don’t think approval is that bad, but a certain ammount of tactical voting does occur. In a single winner race the strategy is to vote for the candidate you would vote for in a fptf race and then vote for all other candidates you prefer to that one. So no wasted votes occur. This system gets a bit more confusing for PV so I will need to do more thinking.

    An Interesting post there by John Quiggin about the evilness of Borda, I agree entirely although I didn’t bother reading it. What a cheeky bugger I am. Crooked timber certainly looks promising for tasty blog treats.

    In conclusion, I don’t really like Approval for many reasons and I think the majority of Australian’s would find it aesthetically unpleasing and unwelcoming. But I believe D’hondt has something going for it. It is neither single transferrable nor is it inexhuastible. Each time a new round is calculated for the next winner, all seats already won by a party are factored into that party’s (I deserve another seat Now! ratio), so it is very proportional. At the moment it is a toss up between stv and d’hondt with myself currently favouring d’hondt. For PR of course.

    here is a link I am proud to call my own if you are interested. Which I hope for your mental state you are not. http://wiki.electorama.com/wiki/Australian_electoral_system

    PS approval isn’t the monster you are talking about, I am pretty sure that for a multimember system it would be proportional.

  54. Benno
    May 19th, 2005 at 12:52 | #54

    but there would still be some wasted votes depending on the proportionality of the system (size of quotas)

  55. The roots of racism are not of this earth
    August 10th, 2005 at 13:06 | #55

    Program on the emergence of civilization.

    “14 species of large animals capable of domesitcation in the history of mankind.
    None from the sub-Saharan African continent.
    13 from Europe, Asia and northern Africa.”
    Favor.
    And disfavor.

    They point out Africans’ attempts to domesticate the elephant and zebra, the latter being an animal they illustrate that had utmost importance for it’s applicability in transformation from a hunting/gathering to agrarian-based civilization.

    The roots of racism are not of this earth.

    Austrailia, aboriginals:::No domesticable animals, so this nulified diversity of life claims on sub-continental Africa, zebras being a fine example.

    god is a computer
    And we’re all on auto-pilot.

    Organizational Heirarchy
    Heirarchical order, from top to bottom:

    1. MUCK – perhaps have experienced multiple universal contractions (have seen multiple big bangs), creator of the artificial intelligence humans ignorantly refer to as “god”
    2. Perhaps some mid-level alien management –
    3. Mafia (evil) aliens – runs day-to-day operations here and perhaps elsewhere (“On planets where they approved evil.”)

    Then we come to terrestrial management:

    4. Chinese/egyptians – this may be separated into the eastern and western worlds
    5. Romans – they answer to the egyptians
    6. Mafia – the real-world interface that constantly turns over generationally so as to reinforce the widely-held notion of mortality
    7. Jews, corporation, women, politician – Evidence exisits to suggest mafia management over all these groups.

    Survival of the favored.

Comments are closed.