A low point in William Hague’s otherwise excellent conference speech on Sunday was his cheap shot at the Alternative Vote system and its role in the outcome of the Labour leadership election. The result, he implied, was some sort of mathematical anomaly; somehow unfair. But there’s nothing unfair about the election of Ed Miliband. If the election were repeated tomorrow using the first-past-the-post system, with only the two Milibands standing, the result would be the same. What the AV system showed was that a majority of the Labour party electorate preferred Ed over David. And there’s nothing unfair about that. (Whether the electorate should include unions is a separate matter.)
There’s no such thing as a perfect voting system. That’s given some mathematical basis by Arrow’s impossibility theorem, which takes a set of reasonable sounding conditions for a perfect system and shows that there is no way to satisfy them all.
The first-past-the-post system fails a test about being a “non-dictatorship”. If everyone else’s votes are equally balanced, then one voter can take his pick of the candidates, and is effectively the “dictator” of the election. That’s a rather unlikely scenario. More relevant, though not explicit in the theorem, is the way that it forces voters to consider tactical voting. One of Labour’s more effective electioneering tactics in this year’s general election, particularly in Scotland, was to claim that “a vote for anyone else is a vote for the Tories”. In an AV system, LibDem voters in a Labour-Tory marginal would no longer have a dilemma about whether they should actually vote for their second choice of party, and Labour would no longer be able to use fear of the Tories to drag voters away from other parties.
It’s also worth noting that the popular Lib Dem line – albeit a less negative one – that “it’s a two horse race” would no longer be useful.
That wouldn’t be a bad thing. We should all be in favour of a voting system which allows people to express their true opinions without first having to guess how everyone else will vote.
The alternative vote system, on the other hand, fails a technical-sounding test about “pareto efficiency”. In Labour’s election, the outcome only told us about the electorate’s preference between the Milibands. But if we were to look at the ballot papers again, it could actually be the case that a majority of voters would prefer Ed Balls to Ed Miliband. And, in a paradoxical circle, it might also be the case that a majority of voters would prefer David Miliband to Ed Balls. With the counting system that was used it’s impossible to know. But it is possible.
However, the possible unfairness of such a case would be rarer than the very obvious unfairness of the first-past-the-post system, and we should welcome the fact that the Labour party now has a leader who recognises this.
Follow @libdemvoice.org on Bluesky
Like us on Facebook
Subscribe to our feed
Sign-up for our daily email digest
Six council by-elections, spread over two days
Carers remain forgotten in government's welfare overhaul
Steffan Aquarone's Commons debate on coastal communities
Getting things done - Saggar Street - an update #dundeewestend
Loving our neighbours
Getting the basics right
Roland
Joseph Bourke
Stephen Nash
Margot Wilson
4 Comments
@Malcolm Wood
“If the election were repeated tomorrow using the first-past-the-post system, with only the two Milibands standing, the result would be the same.”
Sorry to point out the obvious, but by that tortured reasoning you could say that if the general election were held again tomorrow with just the Conservative and Labour, the Lib Dems wouldn’t be in coalition.
Now, what would the result be if the election was held again tomorrow using FTP and all the runners were standing (which should be the true criteria for comparing AV/FTP)?
A minor point, but that’s not actually what the dictatorship criterion means, and FPTP (like almost every voting system) does not fail it. FPTP fails IIA in a big way (what the media calls the “spoiler effect”).
I am baffled by the statement that “it could actually be the case that a majority of voters would prefer Ed Balls to Ed Miliband”. I assume that this is based on the theory that David Miliband’s second preferences went overwhelmingly to Balls (impossible on any other basis). However:
1. Inherently, this is very unlikely. Indeed, because Labour MPs had their votes published, we can know that it wasn’t the case for them.
2. It is not “impossible to know” with the AV voting system whether this is the case. As long as the votes are retained, you can find it out by running multiple two person run-offs. (With FPTP, you can’t find it out because you don’t know people’s preferences between different candidates).
3. It is impossible to avoid this as a theoretical possibility in a single constituency (whether FPTP or AV). However, the chances are much reduced as a result of AV whereas the article gives the misleading impression that it is only an issue with AV.
All in all, I don’t think this article was very helpful for the AV cause and the headline positively misleading.
@MarkG, fair point about the headline. My headline was “Looking for a perfect voting system”, but our editor unhelpfully changed it. I meant that this was a theoretical but very unlikely failure mode of AV. It’s “impossible to know” because I don’t have the ballot papers, not because the data doesn’t exist.
@Andrew, thanks. Yes, I did misread the non-dictatorship test. And the absence of IIA is what forces tactical voting, which I hadn’t realised on first reading (while quickly trying to add more about Arrow’s theorem at the request of the editor).