I've been hearing a lot of pretty stupid arguments recently on electoral reform. They have largely been based on very marginal, technical cases and negative arguments rather than positive reasons to support one of the two systems on offer. In my view there are strong positive reasons to vote for AV.
My argument rests on three examples; they are made-up but (unlike most electoral reform arguments) they represent situations that frequently occur in real life.
EXAMPLE 1
Situation: Labour are the party with the most supporters in Constituency X. The Conservatives and Lib Dems are runners-up by a decent margin, but have a majority of voters between them. The Conservatives and Lib Dems in Constituency X hate Labour and would prefer anyone to win but Labour. They are very supportive of each others' policies.
Result under FPTP: Labour win a convincing majority. The most popular party won. However most voters now have an MP they actively oppose.
Result under AV: Whichever of the Conservatives or Lib Dems got the most first choices wins, possibly by a relatively slender margin. The most popular party did not win, but most voters now have an MP whose policies they can support (though probably not their first choice).
Comment: This is AV doing what it does best. In a situation where one group's votes are split between two parties both of which they support, under FPTP a third, less popular party can easily win. AV ensures that no party loses because there is another, similar party competing with them.
EXAMPLE 2
Situation: The Conservatives are the party with the most supporters in Constituency Y, but by a very slender margin over the Lib Dems. There is a significant Labour presence, but without much hope of ever taking the lead. Labour voters strongly prefer the Lib Dems over the Conservatives in this constituency.
Result under FPTP: Barring tactical voting by Labour voters, the Conservatives win with a very slender majority. The (marginally) more popular party won. However, the majority of voters get an MP they actively oppose.
Result under AV: The Lib Dems win a convincing majority. The party that won is quite popular, but (marginally) not quite the most popular. The majority of voters get an MP they support.
Comment: Again, this is AV at work. Under FPTP a tiny lead translates into victory despite the fact that most voters would prefer the runner-up to have won. AV ensures that the unpopular party can't cling on in this way. Under FPTP the only way to get this result is tactical voting, which ensures that Labour can never be a contender in this constituency even if their underlying support increases. "Vote for us because Labour can't win here" becomes the main argument used by the Lib Dems, instead of promoting their own policies.
EXAMPLE 3
Situation: Labour are the largest party by a decent margin. Conservatives are the main opposition, but in second place by a decent margin. There is a small but significant Green presence. Labour supporters are pissed off at some of Labour's policies and many would ideally like an alternative, but they strongly prefer that neither the Conservatives nor the Lib Dems win.
Result under FPTP: Safe Labour seat. Despite an underlying discontent with Labour they will never get voted out because it's better to not risk victory by parties seen as "the enemy". Alternatively, if people get very annoyed with Labour, the Conservatives might win - but only because people hated Labour, not because they supported the Conservatives; indeed, it might be that this would lead to a Conservative victory despite popular opposition to them.
Result under AV: Probably still a safe Labour seat. However, as discontent increases, Labour voters may begin to put Labour second, opting for the Greens as their first choice. This could ultimately lead to a Green win, over time; and would certainly put pressure on Labour to make some concessions to the Green position.
Comment: FPTP prevents a minority party ever gaining dominance. It actively prevents political change, except in unusual circumstances, e.g. Brighton in 2010 or Tatton in 1997. AV means people can vote for a minority party without risking a hated opponent getting in. Arguably this is good for minority parties, but remember the minority party (in this case the Greens) cannot win unless it has stopped being a minority party and gained majority support! NB this could equally be the case for a Tory constituency with a UKIP minority, or similar.
CONCLUSION
You've all heard the reasons not to vote for FPTP or AV, but very few positive arguments are put forward for either.
These are not artificial examples of the kind you get chucked about in the pointy-head debates about electoral reform. Most of us have found ourselves in one of the above situations. We might hate a party but know that they'll never lose because the vote against them is split. We might dislike a party but be forced to vote for them tactically to avoid an even worse alternative getting in. We might like the look of a minority party but be unwilling to risk voting for them in case our least favourite party wins.
In my view the above examples show that AV can significantly improve on FPTP, by enabling a party to which most voters are opposed to be ejected (first two examples), and by opening up the chance, over time, for political change from a dominant party who are losing support to a rival alternative - instead of simply letting in their political opposite by default.
I haven't addressed the standard (negative) arguments about electoral reform here, but happy to chew them over if anyone wants to. In my opinion, they are a distraction, most of them based on marginal considerations or even illusory. It's the positive arguments that should win out. AV lets you express who you want to win - and who you don't want to win. That's why I'm voting yes in May.