Worcester's Weblog - Too early to call?

Having entered election year pundits, journalists and political anoraks alike are beginning to speculate about the result of the election, and some are even prone to predicting the exact result of what will happen when over 22 million people vote in 650 constituencies in an election three months away. In post-election recalls in the past, we have found that around one person in ten has decided whether or not to vote, or changed their mind on for which party to vote, in the final 24 hours of the election.

The first thing to make clear is that opinion polls –especially this far out from a general election – do not seek to predict the outcome of an election but represent a snap shot of public opinion at the time it was taken. The only poll that any self-respecting polling organisation should consider a ‘prediction’ is their final poll before the day of the election. But as Mike Smithson (www.politicalbetting.com) reminded his loyal band of bloggers, I often say “polls don’t predict, but sometimes pollsters do”. But they shouldn’t, without putting ‘health warnings’ on their predictions, as I did Thursday night to the Harvard Business School Alumni dinner in London. I’m sure they did, but they weren’t published.

Health Warnings

1. Lord Ashcroft’s millions. Most seat calculations are made based on uniform swing, which means taking a nationwide opinion poll and using that to predict the results of all the seats across the country. Clearly this is not an exact science as each seat has its own mitigating circumstances. One such circumstance which is not easily picked up on in national opinion polls is the heightened campaigning that Conservatives have been conducting in targeted marginal seats – funded by their wealthy donor Lord Ashcroft. If you can believe what you read, he’s funded private polling in the Conservatives’ target seats, and this intelligence will be useful to win over marginal seats.
2. ‘Events dear boy, events’. The well known saying first uttered by Harold MacMillan coupled with that other well used political cliché, my late client Harold Wilson’s ‘a week is a long time in politics’, both underline that with three or so months still to go before election day anything can happen, and anything may influence the way the public votes. What effect will Tony Blair’s appearance at the Chilcot Committee today have on the election? Could the announcement of a ‘double dip recession’ affect the outcome? Will the public warm to Cameron? Will any more expenses stories damage senior politicians? Internal squabbling on either side could destabilise party unity. And of course, party manifestos are still to be published!
3. Leaders debates. This is the real wildcard of this election. Never before have the leaders of the major parties stood side by side and debated live on television before a general election. Will Cameron manage to live up to the high expectations? Will Brown be able to hold his own or even impress the public, or will he come across as awkward? Will Clegg benefit from the increased air time? It is likely of course that the debates will have little effect, but anything could happen. Remember George H.W. Bush looking at his watch as then Governor Clinton was answering a question, or Richard Nixon looking uncomfortable and his stubble showing through the make up?
4. Turnout. Turnout will be key. 59% in 2001, 61% in 2005, what will it be 2010? Will the expenses scandal galvanise the public to turnout and vote out an incumbent? Or, as is possible, the public feel so fed up with politicians that they won’t bother voting at all? The lower the turnout the greater benefit for the Tories. If turnout is in the low 50s, it could be a Tory majority of 50 or more; in the mid to high 50s and there would be a Tory majority of 20 to 50; in the low 60s and it falls to under 20; in the mid 60s and we get into hung Parliament territory but with the help of the Ulster MPs the Tories would be sitting on the front benches. If turnout hits the high 60s or low 70s a Labour/LibDem coalition could be on the cards, and if it is in the high 70s then a Labour overall majority is the most likely result.

Watch the Share, not the Lead

The media are fixated on reporting the lead, the gap between Labour and the Tories. Often it is the most misleading figure of all, but it makes for easy headlines.

Watch the shares for each party, not the lead. And first thing have a look at the lead for the opposition party. In past elections it was Labour. Now it is the Tories. If the Tories don’t achieve 40% of the vote share (and there were nine polls during October and November showed them beneath that mark) then they won’t have an overall majority, based on a uniform swing.

What’s the bottom line? The election’s on the 6th of May (at least that’s my best guess, but Gordon could call it any time). Cameron will be in No. 10 the next day. I’ll go that far, based on what I know now and my judgment of what may happen over the next three months.

Will Cameron have an overall majority? I don’t know, nobody else does either.

That’s my best guess.

Sir Robert Worcester is the Founder of MORI.