On Twitter yesterday, Daniel Davies asked,
If AV is so god damned simple, why can nobody explain convincingly to me whether it screws the LibDems or not?
This seems like a fun question to work through at longer than Twitter length, even if it is purely hypothetical, since the No side is going to win.
One obvious answer is that as long as the Liberal Democrats are polling 10% the voting system won’t make a lot of difference. Another obvious answer is that if the Lib Dems recover at all, then AV would seem to help them. There will be plenty of seats, such as Oxford East which they lost under FPTP, but would have a very good chance of winning under AV.
But if AV in England plays out in a similar way to how AV played out in Australia, there is a big risk to the Lib Dems. They could lose a huge portion of their vote to the Greens.
Let’s start with a little story about a large English-speaking constitutional monarchy. At one time the two party system, one roughly centre-right, the other a trade unionist party of roughly centre-left persuasion, was disrupted by the formation of a new party, the core of which came from centrist defectors from one of the major parties. The second word in the name of the party was Democrats.
Initially the party appealed to moderates disaffected with the two major parties. But over time their policies drifted leftward, so that on many issues they opposed the trade unionist party from the left. (This was helped by the trade unionist party drifting rightward over time.) They got a large boost in this when they opposed military intervention in Iraq, something that both major parties supported. Although polling on this was never completely clear, it seemed their supporter base consisted largely of people who supported these left-wing views, or who were disaffected with the major parties.
Then one day the party decided to support the radical fiscal policies of the centre-right party, at the time when that centre-right party had much more support than its trade unionist opposition, but needed their support to get their policies through parliament. This opened up divisions in the party, and led over time to their support cratering. Within a couple of elections they were out of the parliament, and most of their support had gone to the Greens.
End of little story.
The story was about Australia, and the party was the Australian Democrats. But with a little work, you could read it as being about the UK and the Liberal Democrats. Now I’ve had to strain a little to get the analogy between the Liberal Democrats and the Australian Democrats to work. Most notably, the Liberal Democrats have a much bigger, and much more diverse, supporter base than the Australian Democrats ever had. So I don’t think we will see a scenario where the Lib Dems cease to win parliamentary seats within two terms of doing their big deal with the large conservative party. But I do think that they have something to fear from the Greens, and AV would make these fears a little more realistic.
It’s true that with the current voting system, i.e., plurality voting, the Greens do not do well in House of Commons elections. They only got 1% of the vote last time. Though amazingly they did win a seat. But under a different voting system for European elections, they got 8.6% of the vote. And I would guess that if AV were in place for House of Commons elections, and voting Green didn’t take away one’s ability to express a preference between the larger parties, the Green vote would go up by several percent.
If that’s right, a big question becomes where the vote would come from. I would bet that much of it would come from the Lib Dems. That is, I would bet most voters who would identify as Greens (in the German or Australian sense) would currently vote Lib Dem rather than Labour. In part that’s because the Lib Dems have many sensible environmental policies (such as no new runway at Heathrow). And in part that’s because many (not all!) Green voters tend to be antagonistic to big legacy parties like Labour.
One thing we’ve seen in Australia, and potentially one big effect of AV in England, is that once AV lets a party round up the votes of its core supporters without them having to worry about wasted votes, there can be something of a bandwagon effect. With AV, the Greens could easily get 5-7% of the vote in a Commons election. And once they get 5-7% of the vote in a Commons election, they might attract more voters, especially voters who are less sympathetic to fringe parties.
In short, just like the Australian Democrats ceased being the natural home for the We hate both those bastards voters, and they lost that mantle to the Greens, the Liberal Democrats could similarly lose voters who drift to them out of disaffection with the two big parties to the Greens. It won’t happen under plurality voting; the Greens won’t reach the needed “critical mass” of voters to be a natural home for disaffected voters. But it could happen, and could happen quickly, under AV.
Of course, since AV isn’t actually going to win, these are mostly unfalsifiable predictions. Which is too bad, especially if I’m right. I’d love watching the Liberal Democrats be undone by their own attempt to tilt the system in their favour.
fn1. I have no idea how AV would affect voting in Scotland and Wales. There’s no obvious precedent I know of for what happens under AV with at least four parties having a large portion of the votes in many districts. If I’m right that AV would help the Greens, there might be many Scottish and Welsh seats where we have five major candidates. Once that happens it gets very hard to predict who will get eliminated at each stage of voting, and hence who will win. If I were in Scotland or Wales, I’d be tempted to vote No just because it would be good for someone else to be the guinea pig for using AV in a many party system. It is much easier to speculate about what will happen in England, where Australian examples provide better analogies. So this post is just for speculation about the consequences of AV for England.
fn2. Northern Ireland, on the other hand, seems easy to predict. We have lots of examples of how runoff elections work when there is a large ethnic or religious or racial divide between two large groups, plus some political disagreement within the two groups. (See, for example, lots of runoff elections in the US South.) The short answer is that the kind of upset we almost saw in Fermanagh and South Tyrone will be even less likely than under the current system, since Unionist candidates will generally win all and only seats where a majority of the voters are Unionist.