The last couple of weeks have seen a potential Labour rout turn into a mere defeat, as the party climbs in the polls to stand some way clear of its 2015 share of 31.2%: at the moment, an average of the last six polls taken before the tragic events in Manchester puts Labour on 33.8%. This is a long way behind the Conservatives’ score 45.3%, to be sure, but that represents a swing to the Government of only 2.5% or so since 2.5%.
It is important to be clear about where these voters are coming from, as we seem to live in an era when ‘classic’ swing voters who alternate between the Conservatives and Labour seem few and far between. Although the Conservatives’ vote share is being swollen by defectors from the United Kingdom Independent Party, Labour are benefiting from smaller numbers returning to them from the same source, and from some left-leaning Liberal Democrats and Greens attracted by Jeremy Corbyn’s new-look Labour Party.
So although the Conservatives have surged to a share of the vote they have not really dreamed of since the 1980s, Labour has been chasing them upwards.
The danger to Labour comes from some of the implications involved in those complex voter flows. For they seem to be doing fairly well in London, and across the South and South West of England, but disproportionately badly across the Midlands and the North. This is exactly what we would expect given the surge of UKIP voters over to the Conservatives.
Wherever there are large numbers of Greens and Liberal Democrats to draw on – so in urban seats in large cities, and in university towns – Labour might actually do very well. They seem much less likely than they did to lose (say) Clive Lewis’ seat of Norwich South, or Ben Bradshaw’s in Exeter. But in the great expanse of small towns that characterise so many seats across the Midlands and North, Labour might get into deep trouble as they fail to get all of ‘their’ voters back from UKIP.
Take a look at some of the latest regional voting figures we have, from YouGov. These are a little old now, and pre-date Labour’s latest polling improvements. Nevertheless, the figures come from 24 April to 5 May – a period when on average the Conservatives were on 46 per cent, and Labour 29 per cent, in four sets of the same company’s Great Britain figures (for comparison, the latest four datasets give them 46 per cent and 32 per cent).
That five per cent swing was exceeded in England’s North East (where the swing was 10 per cent), North West and Midlands (both seven per cent), and in Wales (eight per cent) and Scotland (9.5 per cent). It was much smaller in London and the South of England (both two per cent), and the South West (where there was almost no swing at all).
That’s good for Labour in London, where the gap between that two per cent figure and the average five per cent swing means that they would save five seats (they would still lose five): but not so good in the South and South West, where they are only defending a paltry eight seats.
Of these it looks uncertain whether Hove and Bristol East will fall to the Conservatives – the latter seems the most endangered, given the relative strengths of UKIP and the Greens in 2015. So Labour may lose one or two seats here – gaining nothing at all from actually seeing a slight increase in their vote in Southern England.
As in the early 1990s, they are now so weak across great swathes of the country that any improvement in their fortunes goes without reward. In comparison, that 7 per cent swing against Labour in the North West alone would lose them sixteen seats.
At the time this data was collected, there was about a five per cent swing away from Labour overall (that would now be about 3.5 per cent). But the relative splits are probably still there: indeed, they may have become more pronounced as Labour piles up votes in London and across the South of England.
Only in Wales, where the latest YouGov figures show an extraordinary Labour recovery, do we yet have any evidence that this pattern is being reversed. That means that – whatever the national picture – the size of the Parliamentary Labour Party is probably always going to be a little bit worse than the national figures imply.
These intuitions are backed up by the latest figures from ICM, which seem to show about a six per cent swing against Labour in their marginal seats. The last time YouGov looked at this, they also found a higher swing against Labour in similar seats (of 7.5 per cent against 6.5 per cent, though that first figure shrank if you asked voters about their own MP and local issues).
That swing from the ICM result could lose Labour sixty seats, even if their vote share rises slightly on 2015 – a perverse outcome, perhaps, but one that is quite possible under Britain’s First Past the Post electoral system.
Overall, Labour looks like it might be able to match or exceed its 2015 vote share. It will not have fallen away overall in the manner many of us feared. Returning Labour voters with nowhere else to go, a canny mix of popular policies and a leader more comfortable with stump speeches than leading the official Opposition might mean that they hold the line. But the signs as to how many MPs they can return still look very worrying.
If Labour’s civil war resumes on 9 June, both sides may now back up their arguments with statistical evidence that lies close to hand: and that is no recipe at all for its deep-seated divisions to heal any time soon.
Glen O’Hara is Professor of Modern and Contemporary History at Oxford Brookes University. He is the author of a number of books on modern Britain, including The Politics of Water in Post-War Britain, published this month by Palgrave Macmillan. A frequent contributor to the national press, he blogs at Public Policy and the Past.