Friday 31 December 2010

AV and the magic of geographic dispersion

"AV will cause more coalitions in this country!", "FPTP is the only way to ensure strong government!"

Nonsense.

It's time to put this myth to bed once and for all. No voting system that purely asks people to vote for an MP to represent them in their constituency is able to cause or limit the amount of coalition governments that form over the years. It may slightly enhance or hide the realities of support in each constituency, but in itself neither FPTP nor AV can cause coalitions.

Take, for example, our last general election result. The Tories had 36% of the vote, Labour 29% and Lib Dems 23% (in popular terms). Let's think about the realities of what forms coalitions. They are caused when the number of MPs elected for each party is in significant enough number that there is no majority to be had.

To form coalition governments in single member constituency voting systems (or to at least see Hung Parliaments) you need three parties with at least a modest popularity each, depending on how many smaller parties have strong localised support. The reality is that the cause of our latest coalition, under FPTP no less, is the increased support for the Lib Dems over several elections mixed with a redress of the geographical dispersion of the vote for each of the main parties.

Let me just be clear about that. It is how spread out or concentrated differing political opinions are across the UK that determines the likelihood of coalitions being formed.

Take a highly unlikely example. If the population was perfectly dispersed based on political opinion, to the degree that every constituency reflected the national picture of C 36%, L 29%, LD 23% then there would be only one result possible under FPTP....a parliament made up 100% of the Tories. Under AV it would be different, again assuming opinion is the same we would have either a 100% Labour parliament or 100% Tory parliament depending on where Lib Dem's would tend to put their second preferences.

But what if we switch it around? Let's say that the 36% of Tory support are localised entirely in 234 constituencies but in 0 others, that Labour similarly have 100% support in 189 constituencies, and Lib Dems the rest (not counting the constituencies that would be filled with the independent, national and Irish MPs). In this situation, still with the same voting system we'd have a proportional representation in parliament.

The same number of people voting for the same party they did in 2010, but depending on where they live you can have anything from an elected dictatorship to a fully proportional parliament forming a coalition.

Our current situation with a hung parliament is based mainly on the divide of class in this country, where more rural and higher earning areas tend to vote Tory, middle class and student (though we'll see how long that lasts) areas vote Lib Dem, and Working class to middle class areas vote Labour.

Without mixing these classes in to the same constituencies we have concentrated blobs of party support throughout the UK, and it is this segregation, combined with a greater than two dimensional political opinion, that led us to the coalition government we have today.

How will AV change this? I will come back to this issue in a future piece; However, it may well be that AV increases the hyper-concentration of support, making it less dispersed, and causing a greater hung parliament. Yet it is also the case that the fine balance of the geographic dispersal of political opinion could result in a bloating of the vote for (most likely) Labour, giving them a disproportionate (based on 1st preferences, though not in terms of absolute support) majority that helps to avoid a coalition while more accurately reflecting the political leaning of the population.

If you're against coalition politics then I'm afraid your only recourse to achieve this will be to try and return this county to a system of two party politics and to try and stop people from believing in political ideals they believe in.

Voting against AV won't stop coalitions from forming while three party politics takes hold, it's time for the "No" camp to accept this reality.

20 comments:

  1. This is a fine analysis of coalitions in general, which I believe could work fine (just as well really as I'd be a bit out of place as pro-PR if I didn't). However the No campaign aren't talking about coalitions in general, they're talking about one in particular, that is the current government. This coalition is a bizarre mismatch that is hardly representative of the sorts of coalitions that would be commonplace under PR. This coalition came about because of extraordinary circumstances, and I've already blogged on why it could only have come about under First Past The Post. The no campaign seem to be trying to use this coalition to scare people off electoral reform, but the reality is we need reform so that we don't end up with this again. Here's the blog post I wrote (it's the second of two parts). http://takebackparliament.ning.com/profiles/blogs/so-what-does-av-mean-in-terms-1

    ReplyDelete
  2. The Jenkins Commission also observed: “There is not the slightest reason to think that AV would reduce the stability of government” Chap 5 Section 81 of their report :)

    ReplyDelete
  3. Cheers for the link Ben!

    And yes, whether or not coalitions are workable or good things are a separate issue. A lot more needs to change than just our voting system if the attitude of three party politics is going to remain, governments will have to be much more able to gauge what policies are getting MPs in to power.

    ReplyDelete
  4. I've been discussing this on Twitter as @TheoEsc, but there seems to be more space here. I disagree with your assertion there that AV does not make hung parliaments more likely. In your words "Coalitions are decided by dispersion of political opinion, not voting systems."

    In a vague way, you say here it COULD be the case that a mysterious detail of the "fine balance of the geographic dispersal of political opinion" creates a disproportionate majority for one party.

    But you don't present any reasons for why this actually WILL happen.

    YouGov have conducted an opinion poll with an AV ballot paper, which bears out the common sense view that AV boosts the Lib Dems at the expense of the Conservatives and Labour, which makes hung parliaments more likely.
    http://www.today.yougov.co.uk/commentaries/guest/how-would-election-tomorrow-look-under-alternative-vote

    Now this is a national poll, which does not consider any effects of uneven distribution of support. Uneven distribution could make hung parliaments either more or less likely. But REGARDLESS of that issue, the way people's preferences operate on a national level makes a hung parliament more likely.

    There are two different effects here.

    1. Nationally, the way both Tory and Labour voters are more likely to give Lib Dems their second preference, makes a hung parliament (with Lib Dems the kingmakers) more likely.

    2. Regional effects of the distribution of votes may make a hung parliament either more or less likely.

    You seem to be attempting to argue that effect 2 will operate in a particular direction, enough to compensate for effect 1.

    But you have not given any evidence to support this claim.

    ReplyDelete
  5. Cheers for the longer comment, you're right it's better for here.

    "In a vague way, you say here it COULD be the case that a mysterious detail of the "fine balance of the geographic dispersal of political opinion" creates a disproportionate majority for one party."

    Yep, I'm theorising and showing how the same number of voters, dispersed in different ways around the UK, could come up with polar opposite results.

    "But you don't present any reasons for why this actually WILL happen."

    I never said anything will happen (or won't happen), it's a tough job to try and hypothesise how voter intentions will change over 4 years in each constituency, with their own local factors, in such a way as to affect the likelihood of coalition.

    "YouGov have conducted an opinion poll with an AV ballot paper, which bears out the common sense view that AV boosts the Lib Dems at the expense of the Conservatives and Labour, which makes hung parliaments more likely."

    It's possible, I'm not saying it isn't. What makes it more likely is the factors I describe in the blog post...the reality that as a nation we're dispersed unevenly.

    "1. Nationally, the way both Tory and Labour voters are more likely to give Lib Dems their second preference, makes a hung parliament (with Lib Dems the kingmakers) more likely."

    They are indeed, the two sides don't seem capable of voting for each other now do they? If there is ever a hung parliament it is always going to have the Lib Dems involved it seems...this isn't a problem or choice of the Lib Dems, but a choice of both Labour and Tories to avoid coalition with each other. It's not like any other option is on the table now, is it?

    "2. Regional effects of the distribution of votes may make a hung parliament either more or less likely."

    Precisely...

    "You seem to be attempting to argue that effect 2 will operate in a particular direction, enough to compensate for effect 1."

    I'm not saying it will, I'm saying it can, and that due to a lack of information on voting intentions we do not know

    I'm quite happy to say I understand that the Tories are likely to lose some seats, and Labour and Lib Dems are likely to win some seats. But then boundaries will have changed by then and with it the constituency make up of political views away from anything we have previously understood.

    Until the boundary changes are finalised, until we see what has happened in the next 4 years, and until the changes in intention on a constituency level are seen for the first time we simply cannot say that AV will increase or decrease the chances of a coalition.

    My point with this blog post is only the one thing, and that's to say that AV itself could move us towards a deeper hung parliament, or away...it has the potential to do both under the array of changes coming up, and the only cause of that is spread of opinion. Simple as that.

    ReplyDelete
  6. Well, effect 2 has the potential to move either way, depending on which second preferences dominate in which marginals. But effect 1 can only move us towards more hung parliaments. So over the long haul, the average effect is going to be more hung parliaments, even if occasional elections flout the rule.

    Specifically in the UK, Labour supporters are more likely to give their second preferences to the Lib Dems than the evil Tories, and Conservative supporters more likely to give their second preferences to the Lib Dems than evil Labour. The Lib Dems are fewer in number, and their second preferences are less important, as well as more evenly split. So, the effect will generally be to strengthen the Lib Dems at Labour and Tory expense on average, whatever geographical weirdness happens.

    But more generally, the argument that's usually put forward for AV or PR is that it allows people to vote for who they want, instead of holding their noses and voting for the Big Party they disagree with least. Now if that is to be true, it is always going to weaken the Big Parties and strengthen the smaller parties.

    So, any electoral reform that has the effect you want is going to make hung parliaments more likely, since it weakens the big parties. That's not a side detail: it's a consequence of the very thing you want.

    AV doesn't guarantee hung parliaments. But it makes them more likely.

    You yourself said on Twitter that you prefer hung parliaments, so that's not a problem for you.

    It's a bit weak to say "we do not know". First, we have a pretty good idea from the polling. Second, you're supposed to be advocating for this change: "vote for it because we don't know what it will do" isn't a terribly persuasive argument.

    ReplyDelete
  7. "But effect 1 can only move us towards more hung parliaments"

    Your effect 1 is intrinsically linked to effect 2. This is what this blog post states. If political views are such that constituencies become more uniform in their opinion (perhaps more equal in their Labour vs Lib Dem share for example) then AV actually reduces the amount of chance of hung parliament.

    Take a rather basic example of 5 constituencies, they have the following (very) rough make up....

    1: Tory 60%, Labour 30%, Lib Dem 10%
    2: Tory 40%, Labour 30%, Lib Dem 20%
    3: Tory 30%, Labour 45%, Lib Dem 25%
    4: Tory 25%, Labour 30%, Lib Dem 45%
    5: Tory 20%, Labour 45%, Lib Dem 35%

    Under FPTP this returns Tory - 2, Labour - 2 and Lib Dem 1. Hung.

    Under AV, we assume the second preference always goes to Lib Dem from Tory or Labour, but from Lib Dem it always goes to Labour

    The new result would be Tory - 1, Labour - 2, Lib Dem - 2. Hung as well, though potentially with a much strong "massive" coalition of Labour and Lib Dems.

    Now see what happens if the constituencies change to the following. Note that I'm not changing the proportional share of voted "nationally" (or across all constituencies), only how they're balanced in each...

    1: Tory 44%, Labour 41%, Lib Dem 15%
    2: Tory 51%, Labour 39%, Lib Dem 10%
    3: Tory 30%, Labour 45%, Lib Dem 25%
    4: Tory 15%, Labour 30%, Lib Dem 55%
    5: Tory 35%, Labour 25%, Lib Dem 40%

    FPTP would now return Tory - 2, Labour 1, Lib Dem - 2

    But under AV we'd now have Tory - 1, Labour - 1, Lib Dem - 3.

    Notice how in the first example the hung parliament is the same, just with different people in power. But in the second example AV actually moves us AWAY from a hung parliament.

    Don't get caught up on the names I've used, you can interchange them, they're not meant to be representative of any real constituencies or views...the fact is that the number of people in each example voting for each of the three parties hasn't changed...the absolute opinion of the UK hasn't changed, only their position geographically.

    And it is this geographic change that made my second example return an absolute majority under AV that would have been hung under FPTP. You could do a third example that would show a FPTP majority making result turned into a hung parliament by AV, again with the same "national" vote share.

    So there...I've explained it properly (again) for you...do you understand now I'm talking the truth when I say AV isn't the cause of coalitions, nor does it necessarily make them more likely?

    ReplyDelete
  8. "It's a bit weak to say "we do not know". First, we have a pretty good idea from the polling. Second, you're supposed to be advocating for this change: "vote for it because we don't know what it will do" isn't a terribly persuasive argument."

    We don't have a good idea from polling, go back 4 years and see what the polls were saying compared to in May 2010, see how different many different political effects can affect our decisions in an election. We also have ZERO data right now about how the new constituencies are going to be formed.

    It's absolutely honest to say we don't know if AV will cause a coalition to be more likely or not, an honesty you won't find in the No2AV camp, but it's also irrelevant...as we're not campaigning for hung parliaments or against hung parliaments (my comment above proves why it'd be idiotic to do so in this referendum), we're campaigning for more of a voter's opinion to be counted in each vote to ensure a more representative candidate with a known mandate is returned to parliament. That's the point of AV...accountability and mandate for an individual constituency's MP.

    ReplyDelete
  9. I'm looking at your second set of constituencies, but don't get it.

    Constituency 1. No overall majority, Lib Dems weakest, they are assigned to Labour giving Tory 44%, Labour 56%. Result: LABOUR

    Constituency 2. Tories have an overall majority so it never goes to a second round. Result: TORY.

    Constituency 3. No overall majority, Lib Dems weakest, they are assigned to Labour giving Tory 30%, Labour 70%. Result: LABOUR

    Constituency 4. Lib Dems have overall majority so it never goes to a second round. Result: LIB DEM

    Constituency 5. No overall majority, Labour weakest so their vote goes to Lib Dems: Tory 35%, Lib Dem 65%, Result: LIB DEM.

    So under FPTP it's Tory 2, Labour 1, Lib Dem 2.
    Under AV it's Tory 1, Labour 2, Lib Dem 2.
    No change to the hung parliament.

    Where are you getting the 3 Lib Dem wins from? Are you forgetting that if one party has an overall majority it doesn't go to a second round?

    ReplyDelete
  10. Let me sort out that second example again, I misplaced a couple of the figures!

    1: Tory 44%, Labour 41%, Lib Dem 10%
    2: Tory 56%, Labour 39%, Lib Dem 10%
    3: Tory 25%, Labour 45%, Lib Dem 30%
    4: Tory 15%, Labour 30%, Lib Dem 55%
    5: Tory 35%, Labour 25%, Lib Dem 40%

    There you go, Tory, Tory, Labour, Lib Dem, Lib Dem under FPTP, but Labour, Tory, Lib Dem, Lib Dem, Lib Dem under AV.

    ReplyDelete
  11. OK, that time you get the results you want.

    But let's look at the nation you've created.
    Average Tory percentage is 35%
    Average Labour percentage is 36%
    Average Lib Dem percentage is 36%

    In this nation, the Tories are the third party, Labour and the Lib Dems the biggest.

    The assumption you're making is that Lib Dem second preferences always go to Labour. I'm not sure this is true. But even if it is, in the real world it's not as significant as the Tory and Labour second preferences, because there aren't as many Lib Dems voters.

    So sure, you can create an artificial circumstance where AV creates a bigger majority in a particular election. I've never disputed that in certain cases it's theoretically possible. But it that doesn't tell us whether that's LIKELY to happen.

    I notice that to do this you've had to create several unlikely circumstances.

    First, you've had to create an extremely close election result overall: 35%,36%,36% is a two-way tie, close to a three-way tie. Sure, in an almost tied election regional pecularities are going to be significant, but in the real world that's not going to happen very often.

    Second, with only 5 constituencies, freak results like this are easy to get. With 600, they're less likely. To get the supermajority, you're relying on the odd AV phenomenon where the 2nd most popular party still wins. In this example, that happens for constituencies 1 and 3 where it benefits Labour and the Lib Dems. They're boosted by this effect, the Tories are completely shut out from this effect, so the Tories are stuffed. But with the 600ish constituencies in the real world the Tories would be likely to get some benefit from this effect too.

    Third, the Tories in this setup are a pariah third party. They're the least popular party, and they don't get anyone's second preferences at all. Labour's second preferences always go to the Lib Dems, the Lib Dems always to Labour. Not quite sure why everyone hates them so much when they never have a shot at power.

    So sure, you can create a freak circumstance when AV can give a majority when FPTP would be a hung parliament. But they're not very likely, and you haven't given any reason to think they will occur. I repeat that opinion polls with AV ballot papers show that in the real world, hung parliaments are more likely under AV.

    On Twitter you've said things like:

    "AV cannot increase or decrease likelihood of coalitions, only specific dispersal of political opinion, that's a fact like it or not"

    That isn't a fact. AV can and does increase the likelihood of coalitions. It is likely to reduce the number of Labour and Tory MPs, and boost the number of Lib Dem MPs, and so create more coalitions: reducing the two-party duopoly is part of its design. Sure, you can create some freak circumstances where the opposite occurs, but that tells us nothing about what is more likely.

    ReplyDelete
  12. "The assumption you're making is that Lib Dem second preferences always go to Labour. I'm not sure this is true. But even if it is, in the real world it's not as significant as the Tory and Labour second preferences, because there aren't as many Lib Dems voters."

    The point of this isn't to state where we are right now, or where we will be. The future political landscape can change. It may well happen that AV for the next election increases a chance of a coalition...however in 5 years after that it could have the opposite effect. It is not set in stone because the system doesn't define if it's more or less likely to happen.

    I've explained this over and over, in quite a great amount of detail, so if you're still failing to understand the realities then that is a shame.

    ReplyDelete
  13. I understand perfectly well, it's just that you're wrong.

    The circumstance that you've given relies on a three-way near-tie: 35%,36%,36%.

    In real life, two-way near-ties are much more likely. In a three party system, three times as likely in fact, since there are three possible combinations (Lib Dem-Labour, Lib Dem-Tory, Labour-Tory).

    In the situation of a two-way near-tie, AV increases the likelihood of a hung parliament, since it weakens the biggest parties, and strengthens the smaller.

    In the situation of one big party being ahead of two smaller ones, AV does the same thing.

    So, you've chosen the unlikely combination of a three-way near-tie to eliminate that effect. (Along with some unlikely assumptions and small-sample fluctuations).

    But there are many more combinations where either one party is ahead, or two parties are ahead. In those cases, AV increases the likelihood of a hung parliament.

    Endlessly harping on about one unlikely combination doesn't make it likely.

    ReplyDelete
  14. "The circumstance that you've given relies on a three-way near-tie: 35%,36%,36%."

    It's actually 36%, 35%, 27% if you count, which is roughly what the polls were saying at one point close to the election in May, so not so unbelievable.

    "In real life, two-way near-ties are much more likely."

    In constituencies? If a "near tie" is less than 5% (a sizeable swing) then only 14% of all constituencies are a two-way near-tie. I'd hardly call this more likely.

    If you mean nationally then what you're saying is irrelevant as, as I've explained, it's not vote share nationally that matters, it's where that share resides and how evenly spread it is.

    "In the situation of a two-way near-tie, AV increases the likelihood of a hung parliament, since it weakens the biggest parties, and strengthens the smaller."

    And AV's effect has nothing to do with two-way near-ties, AV's effects are felt wherever there is currently not a majority verdict, and this can go a multitude of ways.

    The reality is that you could have a even national spread of 40%, 30%, 20% and 10% others and still have both a hung parliament or a coalition depending on where people live. You can pretend this isn't true, of course.

    "So, you've chosen the unlikely combination of a three-way near-tie to eliminate that effect. (Along with some unlikely assumptions and small-sample fluctuations)."

    Well, I didn't, so that kind of blows your argument out of the water considering I used a two-way near-tie ;)

    "Endlessly harping on about one unlikely combination doesn't make it likely."

    It's one combination of an "infinitely" large amount of combinations, I think you've kind of missed the point by latching so strongly on to my example, how many constituencies I used (and no, using 5 or 650 doesn't matter, scale the result type's up by multiplying my example by 170 and you still get the same results), and which names I used for each party (feel free to swap all the names around, they're only there for familiarity, not likelihood of actually occurring, as I stated before).

    It's been good discussing this with you, but so far you've very much utterly failed to counter-act the basic point here...the same national vote share, distributed in different ways, can mean AV is both good and bad for coalitions depending on those scenarios.

    ReplyDelete
  15. Heh, made a slight mistake with my figures. As did you, so we're even on that score.

    But you're still not addressing my point.

    On Twitter, you made these very strong statements:

    "The system, between AV and FPTP, doesn't decide coalitions. Our geographic placement does."

    "AV cannot increase or decrease likelihood of coalitions, only specific dispersal of political opinion, that's a fact like it or not"

    "Some years FPTP may have greater chance of causing coalition, others AV, it is nothing to do with the system."

    You repeatedly asserted there that switching to AV does not change the LIKELIHOOD of hung parliaments.

    Now, you're changed to a different assertion:

    "the basic point here...the same national vote share, distributed in different ways, can mean AV is both good and bad for coalitions depending on those scenarios. "

    In you're latest comments here, you've stopped making the strong claim that AV does not increase the general likelihood of hung parliaments. Instead you've retreated to a much weaker claim, that theoretical circumstances, of unknown likelihood, exist in which AV could produce a majority when FPTP would not.

    This means your earlier statements on Twitter about AV not increasing the likelihood of coalitions now have no basis.

    The circumstances you've used are very contrived: a pariah party that gets no second preference votes, a tiny number of constituencies, carefully hand-crafted constituencies rather than a proper model.

    So let's look at how AV works in the real world. From Wikipedia:
    http://en.wikipedia.org/wiki/Instant-runoff_voting
    'In actual use, jurisdictions such as Australia, Canada, India, and the U.K. all currently (2010) have "hung parliaments" and all have four or more parties with seats in their parliaments. Of these, only Canada is not governed by a formal coalition.'

    Both the YouGov UK polling, and worldwide results of AV in practice, show that AV makes coalitions more likely.

    You've denied that on Twitter, but presented no evidence to support your claim. An artificially contrived circumstance shows nothing about what is more LIKELY. Polling and real-world results do.

    ReplyDelete
  16. "You repeatedly asserted there that switching to AV does not change the LIKELIHOOD of hung parliaments."

    AV may deliver a more or less hung parliament, the CAUSE of that shift is due to political opinion spread nationally, not AV. If the same system can one year increase chances of a hung parliament, then in another decrease chances, then basic logic tells you it is not the system that is the determining factor in likelihood of a hung parliament.

    This has been my assertion at every point, but it's nice to see you're trying every which angle to disprove an absolute reality ;)

    Think of it this way... FPTP can one year deliver a more hung parliament than a previous election under FPTP, even if the national vote shares for all parties are exactly the same, depending on the distribution of political opinion. Perhaps 2015 FPTP will cause more coalitions than 2010 FPTP?!

    "Both the YouGov UK polling, and worldwide results of AV in practice, show that AV makes coalitions more likely."

    UK, India and Canada don't use AV for parliamentary elections.

    "The circumstances you've used are very contrived"

    If you say so, it's perfectly easy to map the May 2010 scenario, take the national vote shares, use the data from the survey you linked to above on second preferences, and prove I'm right with that, very much real, data.

    Knock yourself out and let me know when you've realised your mistake. I shouldn't need to hold your hand all the way through this process.

    ReplyDelete
  17. First, these AV systems are in use, just not in institutions called "Parliaments". I note that this seems to be your only argument in rejecting this evidence from the real world.



    Second, I note again that you're changing your assertions after I disprove them, but refusing to admit you've done so.

    I'm quoting you directly from here:
    http://twitter.com/Niaccurshi/statuses/22273093841981440

    "AV cannot increase or decrease likelihood of coalitions, only specific dispersal of political opinion, that's a fact like it or not"

    Now you've switched to this:

    "If the same system can one year increase chances of a hung parliament, then in another decrease chances, then basic logic tells you it is not the system that is the determining factor in likelihood of a hung parliament."

    Events can have multiple causes. For example, dousing your house in petrol doesn't guarantee that it will catch fire, but it makes it more likely. My contention has always been that introducing AV makes it more likely that there will be coalitions, not that it guarantees it. AV is therefore one cause, but not the only cause.

    Your initial statement "AV cannot increase or decrease likelihood of coalition" rules out AV as being even one of many factors contributing to a hung parliament.

    You've now switched to a very different statement about "the determining factor".

    Your current "logic" is this. You are now trying to imply through language about "the determining factor" that the event of hung parliament can have only one cause. You then argue that there is another cause which affects the outcome, and therefore AV cannot be the one cause or "determining factor".

    This is logically false because events can have multiple factors that cause them, not just a single "determining factor".

    So to continue the analogy, on Twitter you initially said "Dousing your house in petrol cannot increase the likelihood of it catching fire". Having realised that's a bit silly, in your later comments you retreated to the weaker position "Dousing your house in petrol doesn't guarantee it will catch fire".

    But you're unwilling to admit that your original position was wrong. So now you're making the argument "If striking a match causes your house to catch fire, then basic logic tells you it is not dousing your house in petrol that causes your house to catch fire, so therefore it's safe to douse your house with petrol".

    This is logically false.

    ReplyDelete
  18. "First, these AV systems are in use, just not in institutions called "Parliaments"."

    Your quote was as follows...

    "'In actual use, jurisdictions such as Australia, Canada, India, and the U.K. all currently (2010) have "hung parliaments" and all have four or more parties with seats in their parliaments. Of these, only Canada is not governed by a formal coalition.'"

    I stated that three of these countries don't use AV for parliaments, so your quote...which specifically talks about parliaments...is irrelevant.

    "Second, I note again that you're changing your assertions after I disprove them, but refusing to admit you've done so."

    Given you've started using false analogies to "prove" and "disprove" I think it's time to agree to disagree. I've always been consistent in my stance (though on twitter the wording is always more clunky), and the evidence is above as to just how much of a "cause of coalitions" AV is, that is to say...not at all.

    BTW, if you want a real analogy that works with what is being said here, put a ball on a wall. The wall can run south to north, or east to west. Depending on which way the wind blows the ball may only fall off of one wall, or it may fall off both, or if the wind isn't strong enough it may not fall off either. The cause isn't which way the wall is facing, it's the way the wind is blowing today. You can try and build the wall every time the wind changes, or stick with your wall design because you believe it works best for you given the trend of wind directions...but it is ALWAYS the wind direction that is the cause of the ball to fall.

    ReplyDelete
  19. Oh dear, oh dear.



    First. The term "hung parliament" in that article is used in quotation marks to denote the situation where no party has an overall majority. True, those bodies are not called "parliaments". But you're playing with words to avoid the issue here. In 2010, every real AV-elected body in the world had the hung parliament / no majority situation that you claim AV won't make likely.

    The real world disproves the assertion of your blog post.



    Second, if my analogy is false, use your logic to show why it is false. Show me, don't just assert it.

    Your logic was "geographic distribution is a cause of hung parliaments, therefore AV cannot be a cause". I disproved this by pointing out that events can have multiple causes, not just one "determining factor". To make it simpler, I illustrated this with an example of another event with multiple causes.

    Now you've responded with an analogy of an event depending on a single cause. If hung parliaments are events that can have only a single cause, your analogy is appropriate. If hung parliaments are events that can have multiple causes, my analogy is appropriate. So prove to me that hung parliaments can only have one cause ("determining factor") and I'll accept your analogy.

    ReplyDelete
  20. Theophile: I'm just going to leave your last comment as is, I'm honestly wondering now whether or not you're intentionally acting this stupid, it's a beautiful example of utter ignorance. When you eventually manage to understand the proof has been in front of you all this time it'll be a wonderful day!

    ReplyDelete

Got something to say about my post? I'd love to hear it!

Try to keep it civil, I don't delete comments unless obliged to or feel the thread is getting too out of hand, so don't make me do it.