The coming referendum on electoral reform is about more than a simple “Vote yes if you want a foot in the door for electoral reform” and “Vote no if the status quo suits the party you support”.
Below are some questions about the Alternative Vote (AV) system, and an outline of the debate around those questions, with the First Past The Post (FPTP) and Single Transferable Vote (STV) systems discussed by way of comparison.
Don’t assume my sources below are neutral unless you read to the contrary.
How do they work?
FPTP is used for constituencies to elect a single representative from a number of possible candidates. It’s very simple – candidates are listed on a ballot form, voters make a single mark on the ballot paper for the candidate they most want to vote for, and the candidate with the most votes wins, regardless of whether or not they have the majority’s support. Norm provides a scenario to demonstrate how this can elect candidates without majority support:
‘In Lower Zogby by the Fen 35 people vote Tory first, 33 vote Labour first, and 32 vote LibDem first. But the Labour voters would prefer the LibDem to the Tory, and the LibDem voters would prefer the Labour candidate to the Tory. As is, with first-past-the-post, the preferences of 67 out of 100 people to have a candidate elected other than the Tory are nullified, where with AV Labour would win.’
Like the current FPTP system, AV is used for constituencies to elect a single representative. Like the STV system, AV is a preferential voting system, but a much simpler one which works as follows:
- you (the voter) rank the candidates on the ballot form in order of your preference;
- if a candidate gets more than 50% of the first preferences, they are elected outright;
- otherwise the candidate with the fewest first preference on the ballot forms is ruled out. If you voted for that ruled-out candidate, your ballot paper’s second preference is then redistributed, with the same value as your first preference;
- then there’s a second round of counting to see if any of the candidates are now polling over 50% – if so, that candidate is elected;
- if not – and let’s say you voted for the candidate with the least votes in that round – then your second-preference candidate is ruled out and your ballot paper’s third preference is then redistributed;
- and so on, until a candidate polls over 50% and is elected.
See the pro-AV Electoral Reform Society’s AV Questions Answered pamphlet [PDF]. As Arieh Kovler puts it, unlike our current FPTP system:
“…votes under AV aren’t necessarily rival – a vote for one candidate doesn’t always mean a vote against another candidate. That vote might only be with your first choice for one round, but could stick with your second choice all the rest of the rounds until the end.”
STV won’t be an option for the coming referendum. STV systems are used to vote for candidates in multi-member constiuencies. Rather than polling a majority, successful candidates have to reach a quota – a minimum number of votes – to be elected. This tends to be the number of valid votes cast, divided by the number of available seats plus one. If a candidate exceeds their quota, their surplus votes are redistributed. The best way to avoid chance here involves weighting the second-choice votes of any candidate who exceeds their quota at a fraction of the value of the first choice votes to reflect that they have already been used. This is explained on page 1 of this ERS information sheet [PDF].
What is a ‘wasted vote’ under these systems?
For FPTP it’s often said that a wasted vote is any vote cast for any candidate who doesn’t have a chance of winning the election. The contest is between the parties who are doing well in the opinion polls and stand a chance of winning – any other vote effectively disenfranchises that voter.
Under the STV system, Arieh Kovler points out:
“…it’s also a wasted vote to vote for a popular candidate who’s going to be elected anyway. If Candidate A gets elected with 1000 votes but less-popular Candidate B gets 500 and both are elected to the block, then votes for Candidate A were only worth half of those for B.”
Under the AV system, the more support your first choice candidate has from other voters, the less likely your second, third, fourth etc votes are to be counted. Norm encourages us to view an AV election as a series of one-on-one contests between each of the candidates.
Would the extremist vote – e.g. BNP – be empowered under AV?
This concern is shared by people who are liberal first, democratic second.
The BNP do not want AV. Part of this is that while their supporters are empowered, the smaller parties themselves do not stand to gain. However, this fact hasn’t stopped the Green Party campaigning for AV on grounds that it demonstrates an “appetite for change”. Unlike the Greens, the BNP thrives on disempowerment and the anger it breeds. The danger AV poses for the BNP is that if extremist voters were empowered, the empowerment may well have a mitigating effect on the extremism by removing the need for a protest vote (anecdotally, this is the view of some local Labour canvassers I met in the pub today). So while the BNP detests the current system and wants a bloc proportional representation system, unlike the Greens it is not prepared to lose ground under AV to get it.
What about the BNP’s supporters, the nearly 564,000 people who voted BNP in the last election? The worry is that their second preferences could have significant effects in the reallocation process. Marco Attila Hoare argues that because voters who preferred extremist parties would have their second, third, fourth, etc preferences counted more than those who preferred the leading parties, in effect these minority voters would be privileged with more votes. What are the implications here?
IPPR research suggests that BNP voters’ second preferences would not ultimately affect the outcome:
“Given the marginality and distance from 50% for both the first and second placed candidates it is true that BNP supporters’ second or third preferences will be counted in the 35 seats listed by the ‘No to AV’ campaign.
However, the BNP vote is still very small in each of these seats, averaging a vote share of just 4.5% – yet the average distance from 50% for the winning candidate is 11.3% and 14.2% for the runner-up. Even if we assume all BNP preferences go to a single candidate (which they wouldn’t) they would still require more than twice the number of BNP supporters to win under AV. BNP voters cannot therefore single-handedly change a result.”
In addition, Norm quotes Brian Barder, an anti-AV campaigner who nevertheless points out:
“All the valid votes are counted again at every recount. Those giving their first preferences to the two candidates who come first and second, and who are therefore never eliminated from the next recount, don’t get their second and lower preferences redistributed and counted, but that’s not a disadvantage: their first preferences continue to count right to the last round.”
Lastly, since I risk forgetting that BNP support isn’t merely a function of an electoral system, here’s a list of other things associated with BNP support.
What does tactical voting look like under these different systems?
New systems mean new games.
The FPTP system tends to render votes for smaller parties wasted votes. Accordingly the Liberal Democrats has tended to encourage their supporters to ‘go long’ and think of a vote for the Lib Dems as an act of incubating the party, an investment for the future. However, the leading parties work hard to deter voters from ‘splitting the vote’. Consider a scenario in which Conservative is the likely winner, followed by Labour, followed by UKIP. Say a Conservative Party supporter with strong anti-immigration views is considering giving their vote to UKIP; the Conservative Party would argue against this on grounds that it splits their vote and improves Labour’s chances of winning – Labour is even less anti-immigration than the Conservatives.
Under STV, tactical voting is very complicated, needs to be precise, and tends to be worked out by election geeks for various factions and parties running models on surpluses and transfers, making deals, and then telling their supporters how to vote.
For tactical voting under AV, here’s Arieh again:
“Imagine that at the next election there was a ‘no to cuts’ party which opposed Government spending cuts. All the party would do is talk about how bad the cuts were and how much better it would have been if they didn’t happen or were slower. They’d also call for a second-preference vote for the Labour party. It would get Election Broadcasts. Its candidates would appear at hustings and be interviewed on the TV where they’d put their messages across and call for people to vote Labour (second).
A new party like this probably wouldn’t win any seats, but that’s not what it’s trying to do. If people voted through all their preferences then a new party like this could help the Labour party get its message across and pick up more votes. Equally, I could have given examples that would benefit the two governing parties instead, e.g. a Taxpayers Alliance Party.”
And that second preference vote for the big party would stick for far longer than the first preference one. There’d be a lot more horsetrading between politicians in advance of an election according to AV, and minor or single-interest parties would gain attention as they announced their second preferences.
So if I were the BNP (a pariah whose second preference votes nobody wanted) I might consider starting some single issue parties – on environmental issues, social housing, import and export – and having them recommend a second vote for the BNP. But I think that’s pretty unlikely – the barriers to contesting an election entail significant expenditure of time and money.
Does a No to AV amount to a No for electoral system reform?
Conscious that rather than being enamoured of AV, the Greens are anxious to indicate an appetite for reform itself, I wonder what the effects of a No vote will be. Norm quotes from a Political Studies Association briefing paper [pdf]:
“A “no” vote in a referendum is always followed by what Professor Lawrence LeDuc calls a “battle for interpretation”. Those who support the status quo argue that the people have spoken and that the issue should be left alone. Supporters of change, by contrast, argue that the referendum has not decided the issue: they might say, for example, that voters were offered the wrong reform option or that a better information campaign should have been launched.
This will happen in the event of a “no” vote in the UK too. Supporters of FPTP will say that the people have decided in favour of the status quo. Supporters of change will argue that AV was the wrong reform and that a more substantial change should be offered.
The question is, who will win this battle? Given that the issue of electoral reform has not caught the public imagination and that few voters understand the intricacies of electoral systems, it is likely to be difficult for reform supporters to convince many that another reform should now be considered. Such was the experience of reform supporters after recent referendums in three Canadian provinces: the battle of interpretation was decisively won by the supporters of the status quo.
It is clear that changing the electoral system is easier where change has already recently happened: the idea of reform is no longer so radical…”
So if you want any kind of reform, vote Yes to AV. Hear that, BNP?
Will AV improve voter turnout?
The Yes campaign says “We can’t promise the earth with AV”.
Professor of politics at the University of Strathclyde John Curtice told BBC Radio 4’s Today Programme (Thurs April 14th) that, based on the literature, a modest improvement in turnout is associated with proportional representation but since AV is a disproportional system, there’s no reason to suppose it will help in that respect.
(There’s a better way to solve that problem, anyway: treat not voting as the personal travesty it is and outlaw it.)
Are coalition governments more likely under AV, and what are the implications?
“There are also other unanswerable objections to AV. By increasing the number of seats won by third party candidates, it would make hung parliaments much more frequent, and thus produce more coalitions or minority governments, which in turn undermines the convention of the party manifesto mandate and the public accountability which that entails.”
Have to save that one for another day – but it’s an important one and I can’t decide how I’ll vote without it.
A selection of more baroque / diversionary reasoning:
From one of Waterloo Sunset’s comments on Bob’s blog, a means justifying ends argument: ‘No’ vote for the AV referendum will break the coalition and precipitate an election which will return Labour to power.
The vote counting machines will cost more [see update below]. Small price to pay if it’s a more democratic system.
That’s it for now – run out of time…
“In his speech [Cameron] also suggested that if AV was used, “we may have to buy and install electronic voting machines to make sense of all the different outcomes and possibilities”.
But when the topic was raised in a question session after Mr Clegg’s speech, the deputy PM dismissed it. He said in Australia, which uses AV, votes were counted by hand. Reports it would cost millions to administer AV were “wildly inaccurate”, Mr Clegg said, adding he hoped the No campaign would not “create a whole barrage of scare stories and myths about this”.