1. What is the probability of Labour winning under your preferred candidate, relative to the probability under your second preference?
2. How much superior would be a government under your preferred leader to that under your second choice?
3. What are the confidence intervals surrounding answers 1 and 2?
I suspect that honest answers to these questions would be: small, little and wide. But in this case, the outcome of the election just isn’t that important.
Instead, the result of the next election, and the shape of the next Labour government, surely depends more upon circumstances outside of Labour’s control than it does upon the character of the leader.
Take, for example, Paul’s endorsement of Ed Balls, on the grounds that he prioritizes economic growth over deficit reduction.
His support would be entirely reasonable, if we were looking for a new government today. But we’re not. The next Labour leader will - at best - only determine policy after 2015. And in this context, Balls’ words are less important. Let’s say he’s right, and that Osborne’s deficit fetishism does clobber the economy and - in doing so - leave a big deficit. It will then be clear to everyone that a change in policy is needed. Whoever the leader is will therefore adopt a Balls-style policy - because this will be the only option. Balls’ support now for such a policy will make him look perspicacious - though no more so than any other Keynesian - but it does not greatly affect the course of the next Labour government.
In this sense, fact, I fear that the leadership contest is reinforcing the widespread fundamental attribution error that gives us the over-personalization of political issues. This is an especial danger, given that the Labour party has a bad record in judging the character of its future leaders: few of those who think Blair a lying warmonger today thought this was part of his make-up in the 1990s, and most of the party were over-optimistic about Brown's ability to be PM.