The debate around the Alternative Vote (AV) referendum is clouded by vague terms like “reform” and “fairness” and by accusations of telling lies that fly around like confetti thrown by both sides. The aim of this article is to try and raise the quality of this debate by presenting one simple argument against AV using a real data to show how it works.
The proposition I prove here is how AV gives some people more votes than others.
Richmond Park constituency
I am using Richmond Park constituency as an example partially because I live there but also because it is a marginal (it changed from Lib Dem to Conservative last time) and so is one of the constituencies where AV could make a real difference.
The result in 2010 was:
Conservative. 29,461
Liberal Democrat. 25,370
Labour. 2,979
UK Independence Party. 669
Green. 572
Christian People’s Alliance. 133
Independent. 84
I appreciate that the numbers would have been different if we had AV but this is a meaningful profile to work with. The example works equally well with others,
Let’s look at what happens when we add some specific voters to the mix to see what impact their votes have under AV.
A Tory voter
Clearly a large number of constituents are true-blue Tories and they are not going to vote for anybody else and as they are one of the two largest parties there is no point in specifying anything other than one preference for the Tory.
Impact, Tory vote goes up by one.
A leftie (like me)
I have voted both Green and Labour previously and AV lets me vote for both. My Green first preference raises their support by one but they still get eliminated so my vote gets transferred to Labour and their vote goes up by one also. Both parties will welcome my vote and claim me as a supporter.
Impact, Green and Labour votes both go up by one.
At this point the Tory voter is, rightly, feeling a little aggrieved that I have had more impact than them, as I get counted twice and they only get counted once, but other voters have it even worse.
A Green voter and a Labour voter
As with the Tory we mentioned earlier, there will still be many people who will vote just Green and others who will vote just Labour.
Now if we add both a Green only and a Labour only voter to the equation we see that the Green voter adds one to the support for Greens and likewise the Labour supporter adds one to the support for Labour. But that is exactly the same impact that my one vote had.
Impact, two voters have the same say as one voter.
Conclusion
These are examples of real voters so this will happen many times over under AV. Some voters will get more votes than others. And that’s unfair.
The proposition I prove here is how AV gives some people more votes than others.
Richmond Park constituency
I am using Richmond Park constituency as an example partially because I live there but also because it is a marginal (it changed from Lib Dem to Conservative last time) and so is one of the constituencies where AV could make a real difference.
The result in 2010 was:
Conservative. 29,461
Liberal Democrat. 25,370
Labour. 2,979
UK Independence Party. 669
Green. 572
Christian People’s Alliance. 133
Independent. 84
I appreciate that the numbers would have been different if we had AV but this is a meaningful profile to work with. The example works equally well with others,
Let’s look at what happens when we add some specific voters to the mix to see what impact their votes have under AV.
A Tory voter
Clearly a large number of constituents are true-blue Tories and they are not going to vote for anybody else and as they are one of the two largest parties there is no point in specifying anything other than one preference for the Tory.
Impact, Tory vote goes up by one.
A leftie (like me)
I have voted both Green and Labour previously and AV lets me vote for both. My Green first preference raises their support by one but they still get eliminated so my vote gets transferred to Labour and their vote goes up by one also. Both parties will welcome my vote and claim me as a supporter.
Impact, Green and Labour votes both go up by one.
At this point the Tory voter is, rightly, feeling a little aggrieved that I have had more impact than them, as I get counted twice and they only get counted once, but other voters have it even worse.
A Green voter and a Labour voter
As with the Tory we mentioned earlier, there will still be many people who will vote just Green and others who will vote just Labour.
Now if we add both a Green only and a Labour only voter to the equation we see that the Green voter adds one to the support for Greens and likewise the Labour supporter adds one to the support for Labour. But that is exactly the same impact that my one vote had.
Impact, two voters have the same say as one voter.
Conclusion
These are examples of real voters so this will happen many times over under AV. Some voters will get more votes than others. And that’s unfair.
Every vote is counted every round. If you weren't counting the votes of those that hadn't moved they would disappear.
ReplyDeleteOr maybe you look at it as additional votes? Fine, so those who get their second preference get an additional vote, in exchange for having their previous vote nullified. How many votes do they have on the table at any one time? One, the same number as anyone else involved at the same time.
To repeat: Either everyone has a vote in each round, with those who no longer have their previous candidate able to move it to someone else...so everyone's vote is counted every round....OR those who change preference get an extra vote, but their last vote is nullified and void....so everyone only ever has one vote on the table at any one time.
It's really simple maths, it's astounding that it's being abused this way to misconstrue the realities of how votes are counted.
I'm grateful to Lee for showing how stupid the #yes2av arguments are. Clearly the statement "their last vote is nullified and void" is a lie. The vote is counted and reported afterwards. It does not disappear.
ReplyDeleteIt's recorded, but for the purposes of counting it is void. They are two separate things
ReplyDeleteIf it's recorded then it's counted and the parties will know that the number of votes they had at the end of each stage. It's just that mine vote gets counted by two parties and the other voters' votes are only counted by one each. Therefore my vote counts more.
ReplyDeleteThis is interesting, I would have thought "my vote counts more" meant that you vote had more influence on the election, like in the Labour Party's leadership election where an MP's vote is worth many times more than an ordinary members vote. Under AV do you think that the vote of a person who votes for a smaller party has more influence over the election than one of the larger parties?
ReplyDeleteMatthew, if you want to learn from the horse's mouth how UK elections would be conducted and counted under the proposed AV system then you can read all about it in the text of the Parliamentary Voting System and Constituencies Act 2011: http://bit.ly/eWpfD7.
ReplyDeleteIf you're interested in a legal opinion as to whether AV gives certain voters more than one vote, then take a look at: http://www.ekklesia.co.uk/node/13996.
Ultimately though, I think your concern is that minority parties whose candidates are eliminated in early rounds of counting will claim that they have support, when in fact by the time of the final round and the determination of the eventual winner, that support may have migrated to other candidates.
Of course nothing's going to stop political parties claiming anything they want. As Antony Green says, "Politics is sometimes the art of saying black is white and keeping a straight face." (http://bit.ly/fJ7Emk)
Nevertheless I believe you can resolve this apparent unfairness by distinguishing between support and approval, and separating issues of support from the business of electing representatives. We probably agree that support is indicated by first preferences. AV attempts to select the election winner on the basis of majority approval, rather than (as FPTP does) on absolute primary support. The justification for this is that it prevents the election of a candidate who might be actively disliked by a majority of voters. Under FPTP such a candidate can win, and arguably, there are examples of the phenomenon currently sitting in parliament.
I think I see your point. When a candidate is eliminated, the votes that they had get reallocated according to the voters' stated preferences (or else put on a pile of "exhausted" votes). The returning officer will eventually detail the number of votes that eliminated candidates had, along with the counting round they were eliminated at, and how many of their votes were reallocated.
ReplyDeleteAlthough those votes went elsewhere, the eliminated candidate will still claim that level of support.
So yes, you have a point, and there certainly seems to be scope for confusion and mischief associated with the number of votes that eliminated candidates received.
However, it's definitely not true that there are any extra votes floating around, or that certain voters get extra votes when their lower level preferences are used to help decide the election.
I believe you are confusing votes with support, or at any rate with what parties (& particularly parties that get eliminated) might well claim as indicating support for them. Howcver, the only thing that gets an MP elected is the votes that count for him/her at the election, and at any time during the count of an AV election, each voter has a maximum of one vote in play. That's objectively true. Here are some links that I think give good explanations of the system:
http://en.wikipedia.org/wiki/Alternative_vote
http://www.ekklesia.co.uk/node/13996 - references a relevant US legal precedent
http://www.electoralcommission.org.uk/news-and-media/public-awareness-campaigns/public-information-on-5-May-2011-elections-and-referendum - Official information leaflets from Electoral Commission
http://ilovefairervotes.blogspot.com/2011/01/distinction-that-makes-no-difference.html - Simon Dowden's blog
Or you can read the gory details in Schedule 10 of the Parliamentary Voting System and Constituencies Act 2011 - see http://www.legislation.gov.uk/ukpga/2011/1/contents/enacted
Under your existing system, the Conservative candidate has only gained 49.268% of the primary vote. That means that more than half of the constituency didn't vote for them. That is simply undemocratic and unrepresentative of the people's will.
ReplyDeleteIf you were to count second preferences as we do in Australia, all of the number 2s from the Independent would be added to the tallies and so forth until 50%+1 votes are gained.
Antony Green from the ABC wrote a brilliant piece on why the AV works better than the FPTP system:
http://blogs.abc.net.au/antonygreen/2011/03/how-av-builds-a-majority-for-a-candidate.html
The single-party loyalists you use in the above analogy may well represent a significant, if minority, proportion of the population. However, the fact that they choose to cast only a first preference vote is their own choice!
ReplyDeleteAlso, your analogy doesn't prove that plural voters receive more votes than single-party loyalists. Each voters receives a number of votes commensurate with their voting practice. The single-party loyalists, in using only his/her first preference is making a choice to abstain from voting in rounds following the elimination of their chosen candidate. The pluralist voter, in choosing to use their second preference is making a choice to transfer their vote (rather than abstain) to another party still in the running, once their first preference has been eliminated.
What you have identified, is an issue which affects the first part the posts electoral system as much as it does the alternative vote. That is, the lack of a 'none of the above' option on the ballot paper. A single-party loyalist is left with little choice but to either transfer their vote in later rounds (after their first preference candidate is eliminated) or to abstain. This is a dilemma faced by many non-partisan electors, who have lost faith with the current political class and have no means of casting a protest vote without choosing a 'best of the worst' or 'least likely to win' candidate as a protest vote.
Where a 'none of the above' option added to the ballot paper (either under FPTP or AV) it could be used to a) gauge public dissatisfaction, and/or b) force a re-election with different (better?) candidates. The latter of those two comes with significant downsides; assuming uptake of the 'none of the above' option is widely used, it could leave a constituency with no representation or even cripple government. But then such would be a consequence of a truly representative democracy.
Votes don't have equal value now. You should have a look at this http://www.voterpower.org.uk/
ReplyDeleteIt really isn't hard. I don't believe you FPTP people are as thick as you make out.
ReplyDeletewhat a rubbish argument
ReplyDeleteyou completely miss the point that it is a series of rounds and people only have one vote per round. The chances are that in the scenario you set out the Conservative and Lib Dem voters will be counted in every round. By the (mistaken) logic of your Green / Labour argument, if I am a Tory voter, the result of my vote is that in two rounds of counting the Tory vote goes up by two.
ReplyDeleteI am no fan of AV and have no axe to grind over this. There are plenty of good arguments against AV, not least its potentially disproportional effects, but this argument is mistaken.
Many of the comments here are on other aspects of AV v FPTP and while I am happy to have those discussions this is not the place for that.
ReplyDeleteSo far nobody has done anything to contradict the main conclusion of my argument that my one vote has exactly the same impact as two other people's votes combined, and that's unfair.
Danny helps me here by pointing out that the Tory vote is not actually counted twice as some AV advocates would claim.
Sorry - i just re-read my own comment and realise that what I was saying may not have been very clear (bad editing on my part). I'm afraid I'm not saying that Matthew.
ReplyDeleteAlthough it's true that the leading candidate's votes are not literally counted more than once as far as I know, the effect is the same. To explain:
The idea of AV is to re-run the ballot until someone can get 50%. The difference is that instead of having to go back and vote again (like they do in French Presidential elections), people are required to express their preferences all at once (the Supplementary Vote system used for mayoral elections in London and other parts of the country is a variation on this).
That means at each ballot (or in each round), ALL VALID VOTES are counted, INCLUDING the first preferences of the leading candidates. Therefore, if you vote Tory or Lib Dem in the scenario above , your vote WILL be (effectively) counted more than once and thus no one's vote is worth any more than any other.
Again, I stress that I do not support AV. I am a politics academic with a side interest in electoral systems.
Isn't the question of AV really more than just do we value second preferences? It is rather, who are we trying to prevent gaining power when we cast our vote (or votes). My problem is that, under AV, if I vote for the most popular candidate, my second preferences will never be considered - or my "who I want to avoid getting in" preferences aren't. The problem with AV is counting 2nd and 3rd preferences as equivalent in value to a first preference. I understand people voting for AV from their own, selfish political motivation, but to have the cheek to claim it is a fairer system is appauling!
ReplyDeleteYou just make a really weird, irrelevant strawman (parties will claim me as a supporter!!) and then blather on about whether or not people choosing not to have any other preferences will or will not have 'impact' (making up voter caricatures from nowhere in particullar).
ReplyDeleteIt's not the most compelling argument.
Too late now, but just to clarify, I thought I'd post this little passage from the (impeccably neutral) Political Studies Association briefing paper on AV about the 'multiple votes' fallacy (http://www.psa.ac.uk/PSAPubs/TheAlternativeVoteBriefingPaper.pdf):
ReplyDelete"The claim that AV gives some voters extra votes is a fallacy
Many supporters of FPTP have argued that AV gives some voters extra votes. This is wrong. Under AV, each voter’s vote has exactly the same value.
In the first round of counting, everyone’s first preference is counted as one vote.
In the second round, if your favourite candidate is still in the race, your first preference still counts for one vote. If your favourite candidate was eliminated, your first preference now counts for zero but your second preference counts for one vote.
From each ballot paper, only one vote is being counted. This remains true at each stage of the counting process."
Still haven't decided which way I'm voting. This is a new experience for me. If you are voting against please don't do it because of the nonsense that some people have more votes than others. It's not true.