The interesting thing about Nick Clegg's victory in the Lib Dem election is that it was much closer than early betting suggested, implying that he fought a lacklustre campaign. But being a poor campaigner is surely is a huge drawback in a leader. Which raises the question: is there an alternative way of electing a leader, which gives greater weight to campaigning skill?
In theory, yes. The party could have two votes, one at the start of the campaign and one at the end - in the same way that many debates do, such as those run by Intelligence Squared. Then, the winner would be not the candidate with the most votes, but rather the one who gained the most votes between the two votes. This would identify the best campaigner.
Granted, there are some flaws with this:
1. Campaigning skill isn't the only desirable feature in a leader - personality, other abilities and maybe even ideas also matter.
2. Aspiring leaders would have less incentive to work hard for their party, as they know that initial support will mean little.
You might think these are fatal objections. They're not. For one thing, they could be accommodated by a more complex voting arrangement, which gives weight to both the final result and the change in the vote.
And for another thing, these objections are offset by another potential flaw in the system - the incentives it gives voters to misrepresent their initial preferences.
Had the Lib Dems used this system, Clegg's supporters would have had an incentive to vote for Huhne in the first ballot, intending to switch later, thus increasing the change in Clegg's vote.
But this restores the importance of non-campaigning skills and gives aspiring leaders an incentive to build early support in the party, thus solving our two problems.
I mention this not so much to advocate the system, but to show that there are more ways of choosing leaders than parties might think. Politicians under-rate the importance of the question: how should we choose?