Yes or no

[altariel]

OK, f'listers, let rip. In 100 words or less, and without reference to the other case, tell me why I should vote either 'yes' or 'no' to the following question:

"Do you want the United Kingdom to adopt the 'alternative vote' system instead of the current 'first past the post' system for electing Members of Parliament to the House of Commons?"

Non-UK perspectives welcome.

Comments

[iainjcoleman.livejournal.com]

FPTP breaks when there are more than two candidates. AV is the minimal change that fixes this.

Under FPTP, many voters vote tactically: "My preferred candidate won't win, so I will vote for my second preference who can win", and so on. The problem is that the voter's knowledge of which candidates can/cannot win is limited and subject to error, based on previous election results and perhaps some knowledge of political changes since then. AV is essentially automatic tactical voting: if your preferred candidate genuinely cannot win this election, then your vote is automatically shifted to your next preference, and so on. It allows a voter to accurately express their preferences without risking wasting their vote.

(Yes, I've ignored "without reference to the other case" because the point of AV is to fix problems with FPTP, and I can't be arsed with the circumlocutions involved in discussing the advantages of the former without reference to the latter.)

[altariel.livejournal.com]

Oh yes, I quite understand the logic of it. However, this doesn't address my concerns how the system will work in practice. What I'm really concerned about is whether or not it does deliver the horse-trading and behind-the-scenes dealing that opponents are anxious to claim that it does. Because that seems to give more power to our political classes, and not necessarily of a healthy sort.

[iainjcoleman.livejournal.com]

I think it's important to have a system that more accurately reflects the voting preferences of the electorate. Whether the result of any given election is one that I would like is up to the voters, and I wouldn't choose an electoral system on that basis.

However, to answer that point, and some of the concerns you've mentioned in other responses:

It's quite hard to say how previous elections would have gone under AV, as FPTP doesn't give sufficient data about voters' preferences. What we can say in general is that AV tends to reward parties/candidates that manage to broaden their support beyond their base and who avoid being widely disliked. So if AV had been used for recent general elections, it would have tended to benefit the Lib Dems (who would do well from transfers) while the Conservatives would do a bit less well as they tend not to get so many transfers. (This can be seen in the 2007 local election results in Scotland.) But of course, the ability of the parties to pick up preferences may well be different in 2015, particularly with the Lib Dems having lost a lot of their soft supporters. What is certain is that AV massively disadvantages the BNP, which is presumably why they're dead against it.

As far as "kicking the bums out" goes, I gather that AV tends to magnify the effects of landslides, so that (for example) Labour would probably have gained even more MPs in 1997.

AV does make safe seats less common. This is an important part of reducing the power of political parties (as they are less able to parachute favoured candidates into safe seats) and makes individual MPs more readily held to account. My experience of safe seats is mostly second-hand, as my party doesn't have any, but my experience of living in Labour-dominated central Scotland makes me think this is a definite advantage of AV.

As for No2AV's wittering about "behind-the-scenes deals", It's bollocks. AV is not a proportional system, and will not necessarily lead to more coalition governments than FPTP.

[Digression: In reality, most politics involves behind-the-scenes deals anyway. If a future election does result in another hung parliament, then all that means is the deals are much more open. Compare the Con-Lib coalition agreement to the Blair-Brown pact. The political model in which a governing party spends its parliamentary term implementing its manifesto points is woefully inadequate and naive. It's good that parties should develop detailed plans prior to the election (rather than waiting till afterwards), but what really matters is their main priorities. Whenever Clegg was asked about hung parliaments prior to the last election, he responded with his four main priorities, which have found their way into the coalition document and now into practice. It would be very useful for voters if all leaders were asked to be explicit about their priorities in this way during an election campaign, as this would provide a much better guide to their actions in office than a set of manifestos that no-one reads.

But as I say, that's a bit of a sidetrack. It would be more relevant if we were discussing STV or list PR, but AV is unlikely to lead to a future of endless coalitions. It's a bit of a red herring.]

Short version: whether you like or loathe hung parliaments, AV doesn't make them more likely. It does mean candidates have to work harder to appeal to a broader range of voters, and sitting MPs have to pay more attention to more of their constituents. Extremist parties do worse under AV, parties with broad appeal do better, and there are fewer safe seats.

[altariel.livejournal.com]

Yes, I've now read John Curtice's number-crunching from BES data on second party preferences that suggests AV would have marginally inflated the 1987 and 2007 majorities, and that the main gains would have been for the Alliance/Lib Dems. (Particularly in 2007, when they would have been within spitting distance of the Tories and therefore have had a much stronger case to present themselves as a creditable opposition.)

So I do understand the LibDem case. But I'm genuinely struggling to see what this system offers over FPTP.

[iainjcoleman.livejournal.com]

I'm genuinely struggling to see what this system offers over FPTP.

I'm not normally so egotistical as to go around quoting myself, but I really can't do much better than:

1. AV is essentially automatic tactical voting: if your preferred candidate genuinely cannot win this election, then your vote is automatically shifted to your next preference, and so on. It allows a voter to accurately express their preferences without risking wasting their vote.

2. AV does make safe seats less common. This is an important part of reducing the power of political parties (as they are less able to parachute favoured candidates into safe seats) and makes individual MPs more readily held to account.

That's what AV offers over FPTP. Wasted votes and an overabundance of safe seats are two problems that arise under FPTP when there are more than two candidates per seat. AV fixes those two problems. With the trend in recent decades away from effectively two-party politics, fixing these problems is timely.

[altariel.livejournal.com]

I'm not normally so egotistical as to go around quoting myself

Now you’re just being modest.


AV does make safe seats less common

I’m trying to find evidence to support this. In the last election, the candidate received more than 50% of the vote in 217 seats, and therefore would have been elected with first preference votes. Thinking it might be reasonable to assert that these very safe seats might have poorer turnout, I eyeballed the data on turnout, which had a reasonable spread from high 50s to low 70s. Expanding the range, in 386 seats, the candidate received more than 45% of the vote.

This leaves a substantial number of safe seats into which favoured candidates could be parachuted.

I then looked at comparative data from Australian legislative elections. This did suggest that seats change hands more frequently under AV, and needed a smaller swing in order to change.

However, what it couldn’t tell me is: Which of these had been very safe seats; whether seats that changed were primarily marginal seats; whether Australia simply has more marginal seats.

At this point, I could reasonably conjecture that decades of AV have brought about a greater culture of marginal seats in Australia, and that there might be longitudinal effects from introducing AV in the UK. Obviously conjecture, but a reasonable working hypothesis. Someone should do a study.

Therefore, AV seems to have little effect on very safe seats but, based on the Australian data, it’s likely that it will make some seats more marginal.

All of this seems congruent with the findings from the Curtice BES study of second preferences, which suggests that main party majorities would have been largely untouched under AV in elections held between 1983-1997, and that the landslide victories would in fact have been slightly inflated. The Australian data is also congruent with his finding that Lib Dem representation would have been increased: this would presumably arise from second preference votes in marginal seats, and if more seats became marginals.

So I can’t conclude that AV would have an effect on very safe seats, or, therefore, on two-party politics. I would agree that it makes marginal seats more up-for-grabs, and make some less marginal seats more marginal – but not seats that were previously ‘safe’. If you have other evidence, it would be good to see it.


That's what AV offers over FPTP. Wasted votes and an overabundance of safe seats are two problems that arise under FPTP when there are more than two candidates per seat. AV fixes those two problems. With, the trend in recent decades away from effectively two-party politics, fixing these problems is timely.

So the evidence, as far as I can see, is that AV would not affect extremely safe seats or indeed two-party politics; it is likely to provide a better barometer of opinion in marginal seats; it is likely to expand the number of marginal seats. I can imagine a long-term effect in which more seats become more marginal.

Based on the Curtice study, this might favour the Lib Dems; it’s also possible that they will be the main targets for anti-Coalition sentiment in the next general election.

I would also question a bald assertion that we have seen a ‘trend’ away from two-party politics in Westminster elections. Certainly the Alliance/Lib Dem share of the vote increased markedly after the Labour implosion of the 1980s. However, it’s remained roughly steady since (some decline during the 1990s, not yet back to the 1983 high). As such, it seems to be part of the longer-term cycle of reconfiguration of voting patterns for and against the Labour Party. Obviously devolution has brought other parties to prominence in Scotland and Wales.

Nobody has yet made what seems a straightforward pro case to me that introducing AV might have a Hawthorne effect and raise turnout for a couple of elections.

[katlinel.livejournal.com]

Nobody has yet made what seems a straightforward pro case to me that introducing AV might have a Hawthorne effect and raise turnout for a couple of elections.

This would be one thing I really would like to see from electoral reform - an increase in voter turnout, but I would like it to be a sustained increase.

[altariel.livejournal.com]

I agree. And I think the problems are more structural, and fiddling around with the ballot isn't going to make a blind bit of difference.