Reprinted from The Observer, Sunday, 11 April 2010
If there is anything remarkable about the national polls this year, it is how stable they have been. In more than 70 polls – involving 100,000 electors – the Tories’ share has not fluctuated beyond 3 per cent above and below 38 per cent. Not once. Nor has it risen above 40 per cent.
Below that ‘magic 40 per cent’ the Tories are in some trouble. Not because they can’t win, but they are unlikely to win by enough to secure a sufficient majority to guarantee a full term. The stability of public opinion also makes it difficult to see how Labour could stay in power. But there is still almost a month to go.
There are three “known unknowns” yet to come. Most crucial are the three televised debates. Six out of 10 prospective voters in battleground marginals say the performance of the leaders will be important to them when deciding how to vote. That is according to last week’s Ipsos MORI survey of 1,008 electors in 54 constituencies for Thomson Reuters.
The debates are not like party election broadcasts or prime minister’s questions: those are turn-offs. These debates, at least the first one on Thursday, will be a turn-on.
It remains to be seen whether anyone other than the political junkies will sit through the entire four-and-a-half hours. But the first programme is likely to have a massive audience, with the attention of most of the 60 per cent or so of the public who will be turning out to vote. The rest just aren’t interested in politics.
Expectations favour Cameron – it’s his to lose. Nearly half – 45 per cent – of people in marginal constituencies, defined as those seats that would be gained by the Tories with a swing of between 5 per cent and 9 per cent, think Cameron will gain most public support as a result of the debates. Nick Clegg gets 22 per cent and Gordon Brown lags at 17 per cent .
Another of the “wild cards” in this election is the effect of party and media attention on these key seats. Already we’ve seen the prime minister start his campaign in the Medway towns, a move featured, above, on this page.
The eight national polls carried out after Brown called the election suggest a swing to the Conservatives from Labour of 5.5 per cent since the last election. On a uniform national swing this would leave the Conservatives just over 20 seats short of an overall majority. The swing was also 5.5 per cent in the latest marginals’ poll, which shows that in the first week of the election there has been no boost to the Tories in these battlegrounds, where their expenditure has been heaviest. The third factor is turnout on the day. And of course there are the “unknown unknowns”, whatever they may turn out to be.
Some academic psephologists and pundits have, predictably, been saying that the polls are inconsistent and inadequate. They hark back to 1992 – reminding us that we got it wrong. So we did, and every polling organisation has taken the steps to correct their methodology since.
At the last election, all seven pollsters were within 1 per cent of the Tory share of the outcome and all seven were within 2 per cent of both the Labour and the Liberal Democrat share. On the night, the exit poll forecast broadcast on ITV and the BBC was that the Labour majority would be 66. It was. We hope to be as close this time.