Boris Johnson’s popularity boost won’t last long

John Rentoul, The Independent

Markets in political commentary overshoot. The trade among hyper-political obsessives has swung from “Keir Starmer had an amazing start as leader of the opposition” to “Labour is doomed” in just a few weeks. One opinion poll putting the Conservatives 13 points ahead and Labour’s sky has fallen in.

The truth lies between the extremes. For most of last year, Boris Johnson had a hard time and was thought to be handling the COVID-19 pandemic badly. This year, the vaccination programme has gone well and he gets some of the credit. Starmer’s fortunes have largely reflected those two phases. Most normal people, who pay little attention to politics, have noticed that he isn’t Jeremy Corbyn, that he seems competent — if a bit dull — and that he tends to complain a lot.

That means he is doing better than most opposition leaders. Labour under Starmer is ahead of the average opposition party since 1983, according to an analysis of Ipsos MORI opinion polls by Dylan Spielman.

The average path of the opposition in those nine election cycles is that its level of support rises from the previous election by 15 percentage points until it reaches a point two-thirds of the way through the time between elections. If Starmer follows that path, and assuming the next election is in 2024, Labour would be at 48 per cent in the polls at the end of 2022. The bad news for Labour from Spielman’s figures is that, on average, opposition parties then fall back 10 points in the remaining period before the election, and end up only 5 points ahead of where they started. This is probably because governments control policy and try to maximise support at election time, and because voters switch from protesting against the existing government to making a choice between alternative governments. Still, a 5-point gain might be enough to wipe out Johnson’s majority, and it would put Labour within range of forming a minority government in a hung parliament.

Those averages summarise a wide range of variations, but provide a better context for analysing today’s politics than a single YouGov poll.

In the longer view, Starmer is likely to continue to make progress, not because he will “articulate a compelling vision” or whatever it is that commentators are urging him to do, but because the Conservatives will continue to govern. To govern is to choose, and to choose is to create losers, which is why governments tend to become more unpopular until the turnaround point two-thirds of the way through their term.

I boldly predict that Johnson’s popularity boost from the success of the vaccines will not last. Partly, this is because life is likely to return to pre-coronavirus normal more quickly than seems possible while we are still in the coronavirus abnormal, but partly just because voters tend to be ungrateful. They are pleased about the vaccines now, and give Johnson some grudging credit for them, but the jabs are unlikely to be a big feature of the next election campaign.

The main way in which the pandemic is likely to be an issue in that election is its effect on the public finances, and the struggle waged by Johnson and Rishi Sunak to try to prevent the word “austerity” from sticking to them. We have a premonition of that struggle now, with the backlash against the proposed 1 per cent pay rise for nurses.

I suspect Johnson will change the proposal, and end up with a 2.1 per cent pay rise, as originally planned. I doubt if this is a deliberate ploy, but if the government had simply proposed 2.1 per cent in the first place, the howls of protest would have been almost as great. This way, Johnson can say nurses are getting more than twice as much as the number he first thought of. This is just the start of a long and difficult period in which pressure for more public spending will be hard to resist, and every attempt to resist it will be portrayed as “austerity”. Many of those attempts will fail, meaning the government will both be tarred with the A-word and end up looking fiscally incontinent.