Last week’s electoral contests across the United Kingdom have split Labour opinion. Devotees of Jeremy Corbyn’s leadership say that the results are passable, only one year after a devastating and unexpected General Election defeat; their opponents reply that, taken in context, these elections show that Labour has absolutely no chance of forming a government in 2020. The party is of course loudly and deeply divided, and such a difference of opinion is to be expected when no landslide or rout lights up the landscape in utterly unmistakable ways: but which of these groups’ claims is better rooted in the evidence, and backed up by the data?
The first thing to note is that one reason why opinions genuinely differ is the avalanche of statistics that last Thursday produced. Elections are complicated, and hard to interpret – especially if they use multiple voting systems, take place across a patchwork quilt of jurisdictions, and involve a number of different party systems. The United Kingdom is steadily becoming a much more fissiparous place, politically speaking, and what succeeds electorally in (for instance) Bristol and Birmingham and England’s smaller towns and cities look nothing like each other – to say nothing of the glaring differences that are so apparent when we look at politics in Cardiff, Edinburgh and London. Outside Scotland, where Labour lost 9 per cent of the constituency vote, 13 members of the Scottish Parliament and its status as the official Opposition, there were so many confusing results that it was easy to miss the big picture.
Away from Holyrood, the match between votes and the headline results was not so clear: the next reason why arguments will persist is that the fall in Labour’s actual seat totals often did not reflect the retreat in their share of the popular vote. Their numbers of Welsh Assembly members and English councillors has only fallen a touch, by one and 18 respectively. That masks a very large fall in Labour’s share of the vote since the last time these seats were fought: in Wales, by eight points in the constituency voting, and five points on the regional list, and in England by a rather similar six or seven per cent (depending on who’s doing the counting). In a number of cases, Labour held on by small amounts that did not reflect the magnitude of the swing against them: their vote spread in Wales, for instance, was highly efficient and gained them a couple more seats than the polling (which was broadly accurate) predicted.
If we ignore numbers of elected representatives, and look only at votes, then Thursday night was clearly a bad one for Labour. In terms of what this means for the next general election, there are political science models ready to hand that report likely outcome in 2020 if past historic relations hold. Chris Prosser of the British Election Study has argued that the result then might be something like 30 per cent to Labour, and 37 per cent to the Conservatives. Matt Singh, of Number Cruncher Politics, reckons that these results might mean a slightly larger eight- or nine-point deficit between Labour and the Conservatives. These are key findings: Singh used just this relationship to highlight the danger of a polling failure before last year’s fateful exit poll, and has argued repeatedly that local elections are the best indicator we have in terms of performance during the next battle for Westminster seats.
On this basis, we can now for the first time make a tentative forecast of the vote share and seats result next time: if nothing big changes, Labour is likely to be between seven and nine points behind the Conservatives, and to face a Conservative Prime Minister enjoying a majority of between 20 and the mid-30s (on existing boundaries) or 50 and 70 (on the likely new boundaries to be proposed by the Boundary Commissions). Labour would probably lose seats in any of these projections, and up to 40 in a worst case scenario.
This Labour result was extremely poor by historic standards. Labour turned a six-point deficit last May into a one-point lead this year: a seven-point move that was very weak compared to most Opposition’s first years, and bears comparison since the 1980s only with William Hague’s poor performance in 1997-98, his first year as Opposition leader. Labour’s share of the vote, at 33 per cent according to Colin Rallings and Michael Thrasher of Plymouth University, was also extremely low for an opposition. Michael Foot managed 42 per cent in 1980; Neil Kinnock 37 per cent and 38 per cent in 1984 and 1988; and Ed Miliband 37 per cent in 2011. Labour is now doing worse than all three of those leaders who never won a general election, a situation which the rise of multi-party politics cannot quite explain. Ukip gained 12 per cent last week, against the Liberals’ 13 per cent in 1980, the SDP and Liberals’ 21 per cent in 1984, the Social and Liberal Democrats’ 18 per cent in 1988 and the Liberal Democrats’ 16 per cent in 2011 (they gained 15 per cent this time).
It is the gap between Labour and their main opponents, not the splintering of votes into various camps, that is the most glaring element here. Labour has just finished a paltry one point ahead of the Conservatives: in 1980, the figure was two points, while in 1984 and 1988 Labour lost by one point, all very similar results to this time around. They lost all of the subsequent General Elections. In 1994 under John Smith, on the other hand, Labour led by 12 per cent: they went on to win the next Westminster contest for Tony Blair by a slightly bigger margin. Oppositions performing at the relative level observed last week are extremely unlikely to go on to victory. It is as simple as that.
This is not to say that Mr Corbyn’s first real electoral test was all bad. Far from it. In relatively young, student-friendly, urban, liberal and cosmopolitan areas, Labour did move forward a little. Results in Bristol, Cambridge, Norwich and Oxford were all quite good, with the Greens falling back slightly and Labour picking up some of their votes: the unexpectedly huge margin of Marvin Rees’ election as Bristol’s Mayor, and the party’s capture of overall control there, was probably the most pleasing advance on this front. London was, of course, the jewel in the crown of Labour’s night: Sadiq Khan’s victory in the capital’s Mayoral contest was the clearest and best example of Labour’s ability to win, even while in the doldrums.
There is also a suspicion that Mr Corbyn’s brand of socialism might have fired up some activists in council campaigns where Labour hung on in the South – in Harlow, Southampton, Crawley and the like. In a low turnout election, and given the appalling state of the Conservatives’ activist base, just a small uplift in these places can make all the difference. Note, too, that Labour broadly held what it had (outside Scotland), at a time when social democratic parties are enduring terrible polling and results across most of Europe. Labour’s poll rating is historically very low, last month running at between 30 per cent and 35 per cent; but the German SPD’s poll rating is presently plumbing historic depths in the low-20s.
Most of this is, however, exactly the pattern that local by-elections since last September should have led everyone to suspect all along. In Britain’s cities, and wherever there are students, young people and professional workers, Labour is able to put up a fight: it is possible, though as yet informed speculation, that such people’s moves out of a now hyper-expensive London might explain some of Labour’s hold-the-line performance in surrounding towns such as Reading. Labour is staging a managed retreat, on exactly the same lines as it did in the 1930s and the 1980s: pulling back to its urban fastnesses, and sheltering among its own sympathetic ‘Labour people’. That is perfectly natural and understandable, will preserve the party’s self-image and self-confidence, and no doubt one day will lead to its re-emergence as a national party of government. What it will not do is lead to national electoral victory any time soon – and probably for a decade to come.
Here’s what I would say about last week: Labour are resilient, and the party has deep roots, but it is in deep electoral trouble made worse by the denialism that has spread through parts of the party. It is important to pay attention to and be careful about language here. The best characteristation of Labour’s performance in these elections is something like ‘very poor’, ‘weak’, ‘troubling’ and ‘ominous’. Had they made the 150 or so losses predicted by some political scientists, that would have merited use of the word ‘disastrous’. More than 200-250 English council losses, or falling to an AM count in the mid-20s, would have attracted the description ‘catastrophic’ – a word that probably can now be used in the context of Labour’s Scottish debacle. But not doing quite as badly as you feared does not equal doing well, or even standing still.
Labour have gone backwards everywhere on their performance in the 2010-15 Parliament, an experience that hardly seemed to cover them in glory even at the time. They have lost ground in England, in Wales and – most alarmingly – in Scotland, where they now face a threat to their very existence as a viable political force. They have just put in one of the palest, most anaemic performances by an Opposition in living memory. Their vote share has only just flickered above the levels it reached even in the last General Election, and in Scotland and Wales they managed to go into reverse even on last year. No amount of explaining, ‘context’ or special pleading can change the fact that Labour are in deep, dark electoral waters, and in a crisis of meaning, identity and trust far wider and worse than the endless debates about its leader’s ‘electability’. The sooner they accept that, the sooner they can start to do something about it.