Introducing the new swing voters: Purple Labourites and Rainbow Liberals

For a long time it looked as if two factors were going to deliver the keys to Downing Street to Ed Miliband. The first was that in the first six months of the parliament, around two-fifths of the Lib Dems’ 2010 vote switched to Labour and appeared firmly embedded there. The second was that a little later, starting in 2012, an increasing share of the Conservatives’ 2010 vote was peeling away to UKIP. By the middle of the parliament, these two effects were producing a net swing of close to 7% from Con to Lab despite there being next to no direct movement between those two parties at all.

In fact, it was never quite as good as that for Labour. While the LD-Lab swing has been frequently remarked upon, less noted has been that share of the ex-LD vote that’s gone to the Tories. While a good deal smaller than Labour’s gain, it’s still been responsible for consistently adding some 3% to the Tory share and so reducing the net benefit to Labour of the Lib Dems’ splintered vote.

Since 2012, things have moved on again. The figures in the graphs above are taken from the five YouGov polls at the end of this month, of the same time two years ago and of the end of January 2011 (YouGov didn’t publish a subset by the 2010 vote in October of that year). They represent the shares of the electorate (not of that party’s vote) to have moved between the parties in question.

In those two years, the benefit of the LD-Lab voters has dropped from 9.5% of Labour’s share to 6.7%: still substantial but no longer election winning by itself given that the Tories’ former Lib Dems are still putting 3% on the Blues’ total. In other words, the net effect of the Lib Dems’ collapsed vote share is a Con to Lab swing of less than 2%.

By contrast, the leakage of the 2010 Con vote to UKIP has increased; something that is unlikely to be stemmed if Rochester becomes the second UKIP by-election win this year (a feat which would, incidentally, make UKIP only the fourth GB party to win two by-elections in the same parliament since WWII, counting the Lib Dems’ family tree as one block). More than 7% of UKIP’s vote share is made up of those who’ve switched from Con since the election; a larger number than the LD-Lab defectors.

To these well-established swing groups, however, there are now three more to consider, who will play a significant part in determining how next year’s election goes: Purple Labourites, Soft Labour Nats and Rainbow Liberals. During the first half of the parliament, Labour’s losses to UKIP were minimal. No longer. Labour’s relatively comfortable win in the South Yorks PCC election should not mask the fact that UKIP won nearly a third of the vote there. It certainly shouldn’t overwrite the near-loss of Heywood & Middleton, nor UKIP’s strong polling in Labour areas during the local elections of May 2013 and 2014. While two and a half per cent isn’t yet a huge loss for Labour, it is reaching the stage where it’s becoming meaningful.

UKIP isn’t the only party to be eating into Labour’s vote either: the Greens are nibbling but as this week’s polls have revealed, the Scottish Nationalists are tearing chunks out of their share north of the border; something which could deprive Labour of two or three dozen MPs.

But if the smaller parties are an annoyance for Labour then they’re a far more potent threat for the Lib Dems. Partly that’s through relative size – UKIP overtook them nationally two years ago and the Greens are now within touching distance – but it’s also about direct loss. Perhaps surprisingly, more people have moved from the Lib Dems to UKIP than from Labour, mostly in the last two years (a fact which may explain Labour’s loss of ex-Lib Dem voters since 2012: they’ve moved allegiance twice since the election). Similarly, Lib Dem losses to the Greens have trebled from one to three per cent in the same time period. For the first half of the parliament, the 2010 LD vote sat almost entirely in the Red, Yellow or Blue columns; it’s now all over the place.

The common thread of course is that all the major parties have lost support to the minor ones. With so little direct switching between Con and Lab, the election looks to be won not by the party that can appeal to traditional floating voters but to that which can best keep its fractious internal coalition together.

David Herdson