I stayed up for the result last Thursday night and toasted Gareth Snell with a year-old bottle of Orval. I still had some beer when the Copeland result came in, but if I knocked it back it was only so that I could get it over with and get to bed. It wasn’t surprising – both results were what the bookies had effectively been predicting – but the Copeland result was very disappointing.
But then, the Stoke-on-Trent result wasn’t that great. On the plus side, we sent Paul Nuttall homeward to think again (not that he’ll be welcome there); if the result has revealed the irrelevance of UKIP to a wider public, that will be something to celebrate. But Labour’s share of the vote went down – again. And, although the Lib Dems came back, and although the Kippers profited from the Lib Dem collapse in 2015, the Lib Dem revival seems to have been largely at the expense of Labour: the UKIP vote share actually increased. The fact is that we held on thanks to a divided opposition; if the Tories had done a Copeland and appropriated most of the UKIP vote, they could even have won.
So what’s going on here? Let’s look at some pictures.
What you see there, clearly, is a thumping, election-winning 43.2% Labour vote share dribbling away over the next three elections – and dribbling away in all directions: to the Greens (in light green), the Lib Dems, the Tories and even UKIP (and other right-wingers; in purple). Whoever was responsible for letting that much support slip away ought to be thoroughly ashamed of themselves. There’s a bit of an uptick in 2015, but not nearly enough. In that election it’s the turn of the Lib Dems to shed voters in all directions – UKIP very much included. Not many of those votes seem to go to Labour, but the big rise in the nationalist vote (dark green) was largely at Labour’s expense, so I think we can assume that Lib Dem->Labour switchers partially compensated for those losses and stopped the picture looking even worse. In short, 2015 was marked by two one-offs – the implosion of Labour in Scotland and of the Lib Dems throughout Britain – whose effects on the Labour vote largely cancelled each other out. The trend from 1997 to 2010 is very clear, and it’s not likely that the 2015 figures would have marked a dramatic break from it in the absence of those factors.
Or you can look at it this way:
Yes, that’s a log scale; there is a reason for that, as you’ll see later on. What we’ve got here is the number of votes for each party as a proportion of the number of votes for that party in the previous election. The colours represent the same parties or groupings as the previous chart, apart from the grey bar which represents the total number of votes cast. As you can see, Tory vote growth has been ahead of total growth since 2005, and the Lib Dems could say the same until the 2015 collapse. The purple slump in 2001 represents what happens when you compare UKIP + BNP + odds and sods (2001) with the Referendum Party campaign in 1997; what’s interesting here, clearly, is the way the Kipper vote romps away after 2001. Now, look at Labour. 79% of the people who voted Labour in 1997, voted for the party in 2001; 89% of the people who voted Labour in 2001, voted for the party in 2005; and so on. Even the 2015 uptick is – well, it’s barely there.
If you’re wondering what proportion of the people who voted Labour in 1997 voted for the party in 2005 (and so on) – and especially if you’re thinking you’d rather not find out – I’m ahead of you.
Same system, except that all the figures are calculated against the 1997 figure for the party. As you can see, turnout fell pretty substantially in 2001 and still hadn’t got back to its 1997 level in 2015 (despite population growth in the meantime). What had got back to its 1997 level – and rather beyond – was the Tory vote. The UKIP vote, for its part, reached four times the level of the Referendum Party vote in 1997. (Don’t get too excited about the Green/left vote in 2015; yes, it was eight times the level of the 1997 vote for Green and left parties, but that vote was one eighth the size of the Right vote the same year.)
And, again, look at Labour. 79% of the 1997 vote; 71%; 64%. There was a really dreadful loss of support over those years; a massive drift away from the party, and one in which – I feel fairly safe in saying – Jeremy Corbyn played no part.
How did all this play out in our by-election constituencies? Here’s Stoke-on-Trent Central:
The same steady downward slide: from one election to another, the Labour vote seems to have trickled through the party’s fingers, opting for the Lib Dems, UKIP, local independents (in grey), even the Tories. Then, the same boost for the Greens and the Kippers in 2015 as the Lib Dems collapse, with the same very small uptick for Labour. And then the by-election. Lower turnout makes this difficult to analyse (although see below), as does the disappearance of most of the ‘independent’ field. But it does look as though, with the Lib Dems a player once more, Labour have again shed support in all directions – to the Lib Dems, to UKIP, even to the Tories.
Vote growth figures aren’t so different from the national figure: Labour are behind the trend in every election but one, and even then they only beat it by 1% (97% of 2010 Labour voters voted for the party in 2015, as against only 96% of total 2010 voters turning out in 2015). The cumulative picture is woeful:
What’s interesting here is how depressed the votes for all the major parties were in 2001, and how both the Lib Dems and the Tories recovered. Labour, not so much. To have got less than half of the 1997 vote in both 2010 and 2015 is shocking. The by-election figure of 29%, meanwhile, might be excused on the grounds of overall low turnout, were it not by far the lowest of the three main parties. The Tories managed to score three-quarters of their 1997 vote on the basis of half the overall turnout; I’d say respect was due, if it wasn’t the Tories we were talking about.
Finally, how fares Copeland?
Pretty darn ill fares Copeland as far as Labour’s concerned: once again we see the vote trickling away, election by election. Unusually, in this constituency the Lib Dem crash doesn’t seem to benefit Labour at all, unless there are offsetting movements from the Lib Dems to Labour and from Labour to UKIP. And then the by-election, where – as in Stoke-on-Trent – the Lib Dem revival seems to cost Labour rather than UKIP or the Tories; UKIP do get squeezed, but here again none of the benefit goes to Labour.
In Copeland the Labour vote growth, election to election, wasn’t ahead of the overall growth in turnout in any of these elections – 0/5, compared to 2/5 for the Lib Dems and UKIP and 3/5 for the Tories. At least here there was some vote growth – between 2005 and 2010 – but Labour growth was outstripped by overall growth, and by every other party except the Lib Dems.
One more and we’ll be done.
Interestingly, the post-1997 vote depression effect doesn’t seem to apply here – with the sole exception of the Labour Party. At four out of five elections the Tories exceed their 1997 vote; the Lib Dems do it twice and come close a third time. And look at the rightmost cluster: that’s the by-election. Even with only 75% of the overall 1997 turnout, the Tories managed to exceed their 1997 vote. Labour got less than half of theirs.
What do I conclude from all of this? A number of things. First, Labour have got a big problem – we seem to be losing support much faster than we’re gaining it. Second, whatever the problem is, it’s been there for a very long time – there’s very limited evidence to associate it with the current leadership, and plentiful evidence to associate it with Blair and Brown. We need to take a proper look at New Labour and how they achieved those three election victories; a steadily declining Labour vote plus a divided opposition is a trick that can only be worked so many times. But the most important point, as well as the most alarming one, is the third: nobody seems to know how to halt that downward trend – and in that I do include the current leadership. Unless we can put that twenty-year trend into reverse, neither Labour’s policies nor the party’s leadership will matter very much, as the party’s never going to be in power again.