Millions of people turned away from Labour during its 13 years in power. There’s only one way it can win them backby Peter Kellner / October 17, 2012 / Leave a comment
Published in November 2012 issue of Prospect Magazine
A builder reads the Sun on 30th September 2009, when the newspaper dropped its support for the Labour party after 12 years
A single, stark statistic ricocheted round Labour’s annual conference this autumn: that during the party’s 13 years in power it lost five million votes. In the Blair landslide of 1997, 13.5m people voted Labour. By 2010 the figure was down to 8.6m.
The challenge now is to win the defectors back. How can this be done? Labour-supporting blogs offer different ideas. A new pressure group, “Five Million Votes,” was set up in July. A growing number of activists are joining the debate. All of them face the same problem. They have no firm evidence on which to base their plans. Has Labour lost votes by diluting its progressive ideals? Or has it not done enough to secure the centre ground from David Cameron’s assaults? Has the party suffered from too much New Labour thinking—or too little? Has the time come to bury the politics of triangulation or to revive it? The argument rages, but the data has been absent.
Until now. At YouGov, we have set out to fill the gaping empirical gap. In recent weeks we have been asking tens of thousands of our panel members how they voted in 1997. We have matched their answers to their vote in 2010 and current party support, alongside their attitudes to a range of political issues. This has allowed us to provide the fullest analysis yet of Labour’s lost voters—who they are and what they really think.
First, though, a word of warning. Memory and mortality impose limits on even the most rigorous inquiry. Not everyone remembers accurately what they did 15 years ago, and not even YouGov is able to poll the great suburbs in the sky. However, I’m confident that had we managed to keep tabs on every Labour voter from 1997—who had lived and who had died; who had stayed loyal and who had switched sides—the data would be similar, and the conclusions identical.