As the election approaches, one talking point has been the relatively low share of the two largest parties. Their combined share has been difficult to measure, due to the substantial differences between pollsters, despite their reasonably strong consensus regarding the difference between their vote shares. Confusing the matter further is that a surprising number of commentators erroneously refer to the 2010 share for the UK – opinion polls are for Great Britain only, so on a like-for-like basis the number to compare with is 66.6% (CON 36.9%, LAB 29.7%).
After dropping substantially during 2014, the combined ‘big2’ share seems to have ticked up this year, at least with some pollsters. The uptick for Labour looks less notable than it might, due to the wider leads that many pollsters showed in December. I argued at the time that this looked like it might have been due to festive distortions and was likely to reverse itself once the holiday period was out of the way. But both Labour and the Tories may now be starting to gain slightly.
Are we seeing the start of the smaller parties getting squeezed? It’s too early to be sure, but one pattern has been for UKIP’s vote share to drop among phone pollsters. Compared with December, UKIP has gained a point with ComRes, but dropped 2 points with Ipsos-MORI, 3 points with ICM and 4 points on average with Lord Ashcroft. There was only one Ashcroft National Poll in December, and its 19% may have been something of a blip, but January’s average was still a point lower than November. The picture was more mixed for the Lib Dems and Greens.
YouGov has seen quite a bit of day-to-day volatility for UKIP, but the overall trend now seems to be incrementally lower too. Let’s not exaggerate, there is no sign of UKIP (or any party) collapsing, but some element of ‘squeezing’ ahead of a first-past-the-post election shouldn’t be a surprise. The question is whether it continues over the next three months.