I heard an awful lot about vote share at both the Conservative and Labour conferences. Both parties used this to argue that the 2017 election was a type of victory, and that things were going well.
It’s true that Labour got close to their 1997 performance, while the Conservatives bettered even their 1983 high watermark as far as vote share goes. However, the British electoral system isn’t about vote share. It’s much more important to remember that in 1997 Labour got 418 seats, while in 2017 they got 262. In 1983 the Conservatives got 397 seats and a landslide majority, in 2017 they got 317 and no majority.
It’s more instructive when assessing if an election was a “good” result or not to look at the situation that the relevant parties went into the election in. Labour went into the 2017 election on the back of a defeat that was so bad, as I wrote during its 2015 leadership election, it had realistically lost two elections in one night.
The Conservatives had made phenomenal in-roads into the ethnic minority vote, among social liberals and young professionals. Many former marginals had fortress-sized majorities. To get a majority of one, Labour needed a 1997-level national swing. There were only really 30 arguable marginals and only 15 seats to be gained under a “normal” swing.
After the 2017 election, the Conservatives essentially reversed their Cameron-era gains with ethnic minorities and social liberals. They lost a swathe of marginals, including ones where Labour were well-beaten in 2015. To win the last election, Labour had to do something they have done once in their history. To win the next one, they need to pull off a swing that is below their historical average. They can gain 30 seats with a swing of just 1 per cent.
Talking about vote share is just a bit daft, frankly. It’s also worth remembering why vote share was up: because Ukip collapsed, the Liberal Democrats failed to recover from the damage their participation in the Conservative-led coalition did to them, because people who voted for the first time in the referendum kept voting in the general election, and because people who had never voted before voted for Jeremy Corbyn.
Why did it happen? Largely, because of decisions taken by, in no particular order: David Cameron, Nick Clegg, Tim Farron, Boris Johnson, Paul Nuttall, Nigel Farage and Jeremy Corbyn. Note how the name “Theresa May” is not on that list? It was their victory in the referendum that wounded Ukip, and the ineptitude of Nuttall as their leader that helped to ensure that the wound was mortal, or at least severely damaging.
The only active politician who can take any credit for their vote share is Corbyn, and even then, we’re really only talking about perhaps an extra percentage point on the Labour share. (This is still a big deal – that percentage point was decisive in many Labour gains throughout the country.)
There is a counterfactual argument that only May could have won over quite so many Ukip votes, but the important thing to remember is that thanks to first past the post, a lot of these votes were useless. The price they paid for May’s rhetoric was scaring off a bunch of actually useful votes that David Cameron won, most importantly social liberals – which helped lose them Stroud, Bristol North West, Kensington and Canterbury among other places – and affluent ethnic minorities, which helped lose them Cardiff North, Peterborough, Croydon Central, and Battersea.
So talking about vote share correctly tells you that the 2017 election went better for Labour, but is a pretty odd way of working that out – rather like deciding that getting full marks in an exam was a good result because you used the same ballpoint pen throughout.
On the Tory side, talking about vote share tells you that actually the Conservatives did very well in 2017, that the forward march of Corbynism is actually some kind of victory for centre-right ideas, and that everything is rosy in the Conservative garden. That is to say, it’s nonsense.