The claim
“The last Conservative Government had run out of steam. It was time for change and our voting system helped to deliver that change.”
David Cameron, 18 April 2011
The background
The Prime Minister made a surprising admission as he cranked up the “No to AV” referendum campaign on Monday.
Casting his mind back to Tony Blair’s landslide victory over John Major in 1997, the Tory leader admitted that it had been a “painful” occasion.
But he went on to concede that on reflection, “it was right” that the Tories had been handed a drubbing and Britain’s First Past the Post (FPTP) electoral system helped secure a decisive shift in the balance of power.
Mr Cameron’s wider point was that FPTP is “more decisive” than the Alternative Vote (AV) system and gives the electorate more of a chance to kick out governments they don’t like.
He repeated the familiar refrain among opponents of AV – that it will result in more coalition governments – and said that if last year’s election had been decided by AV, Gordon Brown could still be Prime Minister.
The analysis
Since 1983, researchers at the British Election Study (BES) unit at the University of Essex have been simulating how AV would have changed the outcome of British General Elections.
They carry out surveys of a representative sample of voters shortly after the poll and ask them how they would have voted if they had ranked the candidates in order of preference on their ballot paper.
The results are handily summarised by Dr Alan Renwick for his briefing paper on AV for the Political Studies Association.
There are serious caveats to be borne in mind, as researchers at the Institute for Public Policy Research (IPPR) think tank point out, saying: “Speculation as to the likely effects of electoral system change should be treated with caution, as it is difficult to predict accurately how a change to the rules governing an electoral contest will alter the way in which parties and voters approach an election held under those new rules.”
Bearing that in mind, BES research does suggest that, far from being less decisive, AV would have made the Labour landslide that brought Tony Blair to power in 1997 even more pronounced.
Under AV, Labour would have won 445 seats that year, 27 more than they actually did. The Lib Dems would have been up too, by 69 seats, while the Conservatives would have seen their power base reduced to just 70 seats, 96 fewer than the actual number.
So the New Labour victory would have been even more crushing, and that’s a pattern that is repeated for most of the elections analysed by the BES. In 1987, 1997 and 2001 the landslides would have been heavier. And the 2005 election would have looked more like a landslide for Labour under AV, with the system giving them a majority of 108 – 44 more seats than they really won.
Only in 1983 would Margaret Thatcher’s landslide victory have been slightly less resounding, with the Conservative majority cut from 144 to 132.
Mr Cameron also mentioned Mrs Thatcher’s famous victory in 1979 as a similar example of a pleasingly decisive election, but unfortunately the BES research doesn’t go back that far.
So far so bad for Mr Cameron’s claim that FPTP is more likely to produce clear-cut victories. What about the notion that Gordon Brown could still be in Downing Street under AV?
In 2010 the BES found that AV would have cut the Tories’ share of seats from 306 to 284, while Labour would have been down 10 to 248 and the Lib Dems up 32 seats to 89.
We also have a second AV simulation to put alongside the BES study. The Electoral Reform Society – who, it should be acknowledged, are active supporters of the “Yes to AV” campaign – carried out their own survey and the numbers were slightly different: Tories 281, Labour 262, Lib Dems 79.
But the effect of AV would have been the same under both studies: with less of a gap between Labour and the Tories, a Labour/Liberal Democrat coalition would have been far more plausible.
As it was, the Lib Dems could only create a coalition with a clear majority by teaming up with the Tories. If Gordon Brown and Nick Clegg had formed a partnership, they would have had to recruit smaller parties to form a majority. That wouldn’t have been the case under AV.
So according to both projections, Mr Cameron’s warning about Labour still being in power despite failing to win the support of most voters does carry some weight (although Mr Clegg would still have had to get over the political problem of “propping up” a Labour Party which had lost its majority in Parliament and finished second in the election).
As for the supposedly increased risk of more hung Parliaments, the BES results show that under AV – as with FPTP – only in 2010 would we definitely have seen a coalition in power (though it’s possible that the Tories’ reduced majority in 1992 might have produced a hung Parliament).
The IPPR pointed out that Australia has had just two hung Parliaments since it brought in AV in 1919, compared to five in Britain since the beginning of the 20th Century. FactCheck picked some holes in that comparison in a previous post, but Australia’s experience does show us that a strong two-party system helps produce fewer coalitions whatever the voting system.
In Canada, FPTP has led to 12 hung Parliaments since 1900. As Britain moves in a similar direction away from two-party politics, academics like the (pro-AV) LSE professor Dr Patrick Dunleavy think that could be the shape of things to come here.
The IPPR concludes that hung Parliaments are indeed more likely in the future. But that will be the case under both AV and FPTP, and it reflects the long-term growing power of the Liberal Democrats more than any inherent feature of either voting system.
The study concludes: “The most important source of indecisive outcomes in the UK will not be a switch to AV or otherwise, but rather the trend towards greater third-party representation, which starves the two main parties of the seats they need to comfortably form a majority Government.”
The verdict
We can’t say whether David Cameron’s argument is fact of fiction because we’re all in the business of making predictions and projections when it comes to AV.
But the best academic research we have suggest a change of voting system won’t make big landslides a thing of the past or make hung Parliaments more likely.
There is some credibility, however, to the Prime Minister’s claim that, under AV, he might not have made it to Downing Street at all.
It might take some mental gymnastics to imagine Gordon Brown actually surviving electoral defeat, but a Lib/Lab coalition would have been at least an outside possibility.
By Patrick Worrall
