No-one cares what you think about AV

There’s an interesting discussion in this week’s Bagehot about the rival campaigns in the coming referendum on changing the voting system used in UK general elections. Bagehot’s key argument is that, rather than engaging with the substantive issues thrown up by the proposed switch from First Past the Post (FPTP) to Alternative Vote (AV), the ‘yes’ and ‘no’ groups have focused instead on often-nonsensical mudslinging.

The Yes to AV campaign poster

First we have the ‘yes’ campaign. Its official campaign literature argues that “A YES vote means: We’ll make MPs work harder…we’ll get a stronger voice…[and] we’ll tackle ‘jobs for life’ in Westminster”. On each count, it is flawed. There is no evidence that MPs in general are not already working hard, and indeed as the BBC reported this week, some otherwise inclined to support the ‘yes’ campaign have been alienated by its repetition of tired tabloid stereotyping in this regard. Dismissing all MPs as worthless because of the crimes of a handful and the insensitivity of others (most of whom are, rightly, no longer MPs) is a great way of deterring otherwise excellent candidates from stepping forward. It similarly isn’t clear that AV brings a stronger voice to the electorate. If anything, it makes its views even more difficult to decipher than they already are: remember the complete inability of any commentator to predict accurately the outcome of the 2010 election, and then imagine the need to factor in multiple, hierarchical, and often intransitive preferences. Finally, the ‘jobs for life’ argument, probably the strongest of the three, which holds that the number of ‘safe seats’, in which an MP of a particular party is highly unlikely to be defeated because of the proportion of the electorate that supports their party under any circumstances, would decrease under AV. Bagehot references a New Economics Foundation report which is described as estimating “that AV would merely trim the number of safe seats, so that 16% rather than 13% of seats would change hands at a typical election”. The NEF predicts an increase from 81 to 125 in the number of “very marginal” seats. Hardly an eradication of safe seats, as a spokesman points out: “whatever the outcome of the referendum, politicians will still largely ignore voters in safe seats”.

But surely this latter argument misses the point. The problem with safe seats is not their monopolisation by single parties, but by single candidates. What is needed to make MPs more accountable is an increase in the power of internal party mechanisms, and in the willingness of local activists to hold representatives to account for their actions. No shift in the voting system that retains a link to a geographical area would change the fact that some areas of the country are predominantly urban and poor, and thus full of ‘natural’ Labour voters, while others are rural and better off, and thus more naturally inclined towards the Conservative party. Indeed, even a ‘full’ PR system on a national basis, eradicating the link between individual MPs and a particular geographical constituency, would presumably result in some safe seats being held by those members ranked at the top of a party list system.

As for the ‘no’ campaign, surely the most comical of its arguments is the one advanced by David Cameron, that AV is dangerous because it gives smaller parties more power and may lead to hung parliaments and more coalition governments. Apparently, democracy is damaged when a party with 36.1% of the national vote, and one with 23%, must each compromise some of their plans in order to come up with a programme supported in some way by 59.1% of voters. It would be more democratic if just one party got 40% of the vote and then could do what it liked. What David Cameron isn’t telling you is that the Conservatives’ main worry is that Labour and Liberal Democrat voters will be more inclined to give second preference votes to each other than to the Conservatives, effectively limiting Tory prospects in any constituencies where they currently win with less than 50% of the vote.

How big an impact might this be? Using the 2010 General Election Results, it is possible to find out how many MPs each major party would have gained just on first preference votes alone under AV. The Conservatives would have performed best, with 126 candidates gaining over 50% of the vote in their constituency. Next were Labour, with 75, then the Liberal Democrats on 12. Another 21 seats would have been won by ‘other’ parties, 18 in Northern Ireland, 1 in Scotland, 1 in Wales, and 1 the Speaker of the House of Commons’ seat in Buckinghamshire, not normally contested by the other main parties.  So that’s a total of 234 of the 650 seats decided on the first round, exactly 36%. The remaining seats would have been decided according to the second, third, fourth etc. preferences of those whose first preference was eliminated early on. And here’s where making any sort of prediction gets complicated. Although we can look at the 2010 data to see whose first preferences would have been eliminated first, we cannot tell for whom those voters would have expressed a second preference, if any. We also can’t see how many people voted tactically; such as Labour supporters in Liberal Democrat – Conservative marginals who voted for the former to keep out the latter.

So in truth, nobody knows for definite what the impact of changing the voting system would be on politics. Probably it would still be possible for one party to gain an overall Westminster majority. Possibly we would see more coalition governments, including the terrible injustice of a party which won 36% of the national vote being forced to compromise with a party holding a mere 23%. Democracy could yet be undone by MPs gaining at least the grudging support of at least half the electors in their constituencies, or by governments doing the same with the national electorate. But we won’t be sure unless we try it. And this is another way in which the two campaigns don’t care what you think. Neither is willing to offer you a considered choice, or to listen to what you think. Both seek to lecture, to hector, and to persuade through image and simplification rather than fact. This is to be expected. Neither campaign cares what you think. They care what you can be cajoled into voting for.

My polling card, sans personal details

Anyway, I’m voting Yes. David Cameron says voting ‘yes’ is a bad idea. This implies it would be against his interests. This, in turn, probably means that voting ‘yes’ would benefit me.

Note: Predicting the outcome of a referendum is an acceptable goal of opinion polling. The UK Polling Report website maintains a rolling aggregate of various polls conducted using the actual referendum question. The ‘yes’ group is holding a slender lead in most polls, albeit one generally within the margin of error. Interestingly very few people are admitting that they ‘would not vote’. If just 1% of eligible voters stay home on May 5^th, as the latest Yougov poll predicts, it would be a resounding shock. People lie to pollsters.