How is Labour doing?

I confess to continued puzzlement by a sense among the commentariat that it is all going wrong for Keir Starmer, the Labour leader. We have had a few weeks of this gloom for Labour. There is annoyance, of course, from the left of the party about his tilt towards the centre. But centrists are worried, too. For example, here is Bagehot, the Economist magazine’s pseudonymous UK columnist:

"He has appointed a shadow cabinet that has done the seemingly impossible and underperformed the worst cabinet since the second world war. After months of level-pegging in the opinion polls, the Conservatives have established a solid lead.”

The article frets that Starmer should do better. It is all of a piece with the grumbling that Stephen Bush identifies in a recent column: the commentariat and fast-news consuming classes think Starmer has been doing poorly. I am not wholly clear why.

First, to take the point raised by Bagehot,  I am not sure I think the Conservatives have a “solid lead”, so much as a "lead". When you look at the volatility of these time series and add in the prospect of systematic bias in polling, I do not think the margins are "solid".

Second, just look at the change in Labour's position under Starmer. This is unusual. Labour seems to have done a lot of repairing the damage from a once-in-a-generation clobbering through to a pretty narrow margin. That strikes me as an achievement worth noting. (Change is a more reliable thing to monitor in polls than comparing parties' levels of support.)

Yes, he appears to have hit a ceiling for now on Labour's support - and so the period of rapid improvement has come to a close. I would not yet be too worried about this. He has largely held onto his gains.  

The reason why some people are unsettled is they have a certain idea of the political cycle. The thought is, roughly: Starmer is not keeping pace with the leads run up by previous winners at this point in the cycle. I would caution against reasoning like this - not least since this way of thinking usually takes the experience of Blair or Cameron and then extrapolates wildly.

It is not clear why the path to victory requires them to be ahead only one year into the term. It is akin to football fans worrying about shots-on-target statistics ten minutes into a game when no goals have been scored. Let this Crystal Palace fan tell you: that is not how this works. But maybe especially not now.

The idea of the "political cycle" relies on two basic ideas. The first is that people's appetite for switching parties is limited. That means you need to use the whole parliament to bank them and rack up your score. I would not put a lot weight on this assumption.

As a way of conceptualising this, here is a frequency graph showing the distribution of how much Labour's vote share tends to move. It shows, for every week since 1980, how far Labour's vote share moved in the previous 6 weeks. The dotted lines mark the 25th and 75th percentiles. Half of all six-week periods see no more than a 1.3 percentage point change.  

The "old" political cycle is one where there are relatively few voters up for grabs in campaigns. That was true. I have marked on, with dotted blue lines, how far things moved in the six weeks prior to the elections from 1997 to 2015.

Here, though, are 2019 and 2017.

The 2017 campaign is the sort of thing that McKinsey types would describe as a once-in-two-millennia event. I use that daft maths to make the point that the conventional cyclical model of political volatility is pretty uninformative at the moment.  We have some soft evidence that huge numbers of people are more willing to switch rapidly as elections loom.

The second idea underpinning the idea of the "cycle" is that performance in the first x days of the parliament is a good indicator of how you will do later. But I am very unclear that you can learn much about the trajectory of the next few years from Labour's performance in 2020.  It is very difficult to oppose during a national crisis.  And, as Rob Hutton argues in the Times, Starmer has spent a non-negligible portion of his leadership locked in his own house and literally banned from meeting members of the public.

To answer another criticism of Starmer, I am also unclear how wise it would be for Starmer to go horizon-gazing about his future plans when the immediate crisis is so remarkable and extreme. In any case, the pandemic plus Brexit make this an unusually volatile time to make policy.

Tom Hamilton, a former member of the Labour nomenklatura, argues:

"In three years’ time we will know a lot more than we do now about how well the UK has recovered from the pandemic and its accompanying recession, about the effectiveness or otherwise of efforts to “level up”, about Boris Johnson’s ability to satisfy a fractious and ideologically confused parliamentary party, and about whether or not Britain has made a success of Brexit. Keir Starmer’s job is to ensure that Labour is well placed to compete when voters come to deliver their verdict on the world as it is then."

This sort of argument is pretty much always true. Cameron refused to be pinned down in 2005-7 on lots of areas of policy - at that was a time when the governor of the Bank of England was banging on about how placid conditions were. But during a period when we are fighting a pandemic, adjusting to our new terms of trade and in a political age when voters seem more prone to making their minds up in the run-up to elections, the case for patience seems stronger than usual.

This is not intended as a defence of Starmer so much as a slightly bemused shrug. Labour would obviously like to be doing better. Their current polling position does not imply they would win an election now - but it would take a fairly heroic effort to have got there so rapidly from the recent drubbing. In short: we are just over a year into this very weird parliament. Labour types - relax.