Statistically speaking, Keir Starmer will probably fail

Or: why Labour leader and Conservative leader are not the same job.

There’s a line Peter Mandelson has taken to using about the Labour party’s record. Since 1979, he says, it runs “Lose, lose, lose, lose; Blair, Blair, Blair; lose, lose, lose, lose”.

There are several things that are annoying about this. One is that, by highlighting that only Tony Blair has led Labour to victory over the last 47 years, it risks suggesting that he might be the hero the party needs now – as if the Blair of 2021 is still the one of 1997, and not, well, this guy.

Another is that it implies, without ever quite stating it, that the solutions which succeeded in the mid-1990s are still the right ones today, when rather a lot has changed in the intervening quarter of a century, and neither the country nor the electorate still look like they once did. Yet another is that I can’t work out what song that slogan should be sung to. (Suggestions on a postcard.)

But he does have a point. Since the previous Labour leader to win a general election, Harold Wilson, stepped down in 1976, Labour has cycled through no fewer than nine leaders, of whom only three have been Prime Minister. Those do not sound like great odds, but they sound a damn sight worse when you consider that the two who weren’t Blair got the job by inheritance rather than electoral success. It’s no doubt wrong to imagine that either Blairism or the 2021 vintage Blair could save the Labour party today; but it’s undeniable that the 1997 version did manage something which nobody else has for nearly half a century.

Things look rather different on the other side of the fence. I have dim memories of hearing William Hague speak, when he was Tory leader, c1998, about how exciting it was to have the opportunity to rebuild the most successful party in the world or some such. Little wonder he was excited: every single one of his predecessors – all the way back to the party’s foundation, in 1834 – had become Prime Minister. He must have thought he was a shoo in for Downing Street.

He was not: Hague went on to become the first Conservative leader in history who did not become PM. His two immediate successors, Iain Duncan Smith and Michael Howard, didn’t manage it, either; then the historic pattern reasserted itself. How much this temporary disruption to senior Tories’ career plans can be blamed on/credited to Tony bloody Blair again is left to the reader to decide. I’m not touching that with a barge pole. 

More to the point: I’m not really here to talk about Tony Blair. I’m here to talk about the fact that being Labour leader is fundamentally different to – and a hell of a lot harder than – being Tory leader.

There are, I think, four easily quantifiable ways to measure the electoral success of a top-two party leader: whether they become Prime Minister, whether they win a majority, whether they win an election even if they don’t get a majority, and whether they manage to increase their party’s number of seats. A truly successful leader will manage all four of these things. An abject failure will manage none of them. 

From these we can get an extremely rough score out of four for each leader. I don’t think these can really be used to rank leaders in any meaningful way, but nonetheless if one party is getting a lot of 0s and 1s, and the other a lot of 3s and 4s, then it feels like something worth noting.

Here’s how Labour’s leaders, since it first joined the top rank in 1922, have scored. (For ease of reference: green means yes, red means no, grey means “didn’t get the chance”.)

*Arthur Henderson did lead the party to an election at which it increased seats. It was in January 1910, so before the period we’re considering, but after some debate I decided we still had to count it. 

The average score there is just over 1.6. That is... not great.

Compare the Tory leaders, over the same period:

*Bonar Law’s majority victory and increase in seats were at the 1918 election, so again before the Labour/Conservative duopoly kicks in. But, again: we probably do need to count it.

Also, if you’re wondering about Hague’s single point: at the 2001 election the Tories managed to gain a single seat. Banter.

The average score this time: 3. Oh – and not a single leader this time failed on every measure. (The closest is Iain Duncan Smith, who MPs forced out before he could lead the party to an election; but given his record of complete and utter failure in everything else he has ever done, it would hardly be a surprise if he’d managed the same here.)

In fact, on every measure that can be quantified via an election, Tory leaders outperform Labour ones. They’re more than twice as likely to become Prime Minister, of course; but they’re also more than three times as likely to win a majority, and more likely to lead the party to an increase in seats, too. 

Most of all, remember, if a politician becomes leader of the Conservative party, they’re extremely likely to become Prime Minister. If they become leader of the Labour party – well, they have a greater shot at it than you or I, but the odds are still not great.

Two conclusions present themselves from all this. Firstly, it is much, much harder to be a successful Labour leader than a successful Tory one – and for all the talk of the Red Wall and so forth, that is not a new problem. There may be all sorts of reasons for this: the press, the electoral system, the small-c conservatism of the electorate. But whatever is at the root of the problem, the two parties are playing different games. Describing it as a “two-party system” flatters the Labour party’s position.

Secondly: the odds are stacked against Starmer, as they’ve been stacked against everyone else who has done his job. That doesn’t excuse his unforced errors, of course, and it doesn’t change the fact he seems, fundamentally, to just suck at his job. But it does at least suggest that any argument of the form “If only the leadership did X, it would win” is, statistically speaking, extremely likely to be wrong.

I still don’t think Mandelson’s comments are especially useful: they imply that either Tony Blair or how he responded to the political conditions of the 1990s have something to teach us today, and I’m not sure they do. But they do highlight the fact that, whoever is leading the Labour party, odds are they’re going to fail.

Still, cheer up, the sun’s out.

If you would like me to write more of this sort of thing, why not become a paying supporter? For just £4 a month, or £40 a year, you will get a newsletter every week containing some notes on the week’s news, links to things I have found interesting, a feature on a nerdy topic, and some notes on a map; plus the warm, self-satisfied feeling that you are sticking it to The Man by funding independent journalism. You can sign up here.

And if you don’t want to do that but simply want to comment…

…or share…

…well, you can do those things too.