We are a normal-sized polling error away from a Labour government

There is distinct movement to Labour at the Liberal Democrats' expense. 

NS

Sign Up

Get the New Statesman's Morning Call email.

Is it happening? After weeks of what looked like noise within the margin of error – thanks in part to the fact that very few pollsters are now doing polls more than once in week, which makes it difficult to draw out trends – we can now say with certainty that Labour are gaining ground in the polls. They are not doing so at the rate they did at the last election yet, and the Conservative remain ahead in the polls, but there is a clear trend: of Labour gaining ground thanks to defections from the Liberal Democrats and Greens.

As I’ve written repeatedly, you should pay very little attention to the headline figures, because they might be wildly wrong in either direction. As it stands, we are a normal-sized polling error – that is to say, if either or both of the parties are being systematically under- or overestimated by three percentage points – from either a Labour minority government or a Conservative landslide.

The good news is that a broken scale can tell you if the weight on it changes even if it can’t accurately predict the level. Analysis based on trends (whether in 2015 that the focus on the SNP was helping the Conservatives at the Liberal Democrats’ expense, or in 2017 that Labour was gaining support compared to its performance in the 2017 local elections) tends to age better than analysis based on levels.

And the polls are showing that the trend is of Labour gaining ground at the expense of the Liberal Democrats, and the Conservatives consolidating the Leave vote at the Brexit party’s expense. Are they right?

Well, that’s certainly the impression on the ground from both big parties – though the pattern in constituencies that the Liberal Democrats are heavily targeting is of course very different. CCHQ has instructed some local parties to switch from targeting households that are likely to be in the Conservative Remainer, Conservative Leaver and Brexit party supporter groups and have told them to pay more attention to Labour and Liberal Democrat voters – suggesting that the pattern in the polls, that they are at or near the peak of what can be achieved solely via squeezing the Leave vote, is accurate, at least as far as what the Tory party is hearing on the ground. And purely anecdotally, the return of voters who backed Labour in 2017, but the Liberal Democrats in May looks real to me based on my own conversations with voters in this group.

I said last week that the story in the polls so far had been “Conservatives extend lead, impact under first past the post unclear”. That increasing lead was based on the Tory party’s success in making inroads into the Brexit party vote, and the Labour party’s failure to make concomitant inroads into the Liberal Democrat vote. Just at the point when that Conservative lead looked to be reaching the point when the implications were clear, the Liberal Democrat vote has visibly begun to fall to Labour’s benefit: taking the parties to the point when a hung parliament again becomes a possibility.

Stephen Bush is political editor of the New Statesman. His daily briefing, Morning Call, provides a quick and essential guide to domestic and global politics.