Why Gordon Brown should go for an early election

Over on UK Pollling Report, Anthony Wells has a superb summary of the arguments for and against the idea of an early election.

As it did last year, the decision boils down to this - if Brown runs he may lose, but that is true at any time. So isn't this (by which I mean early next year) the best time to go?

Let me add one more reason to those advanced by Anthony as to why I think it probably is the best moment he will get. The reason is risk aversion.

In January the conventional wisdom was that a looming economic dowturn would further add to Gordon Brown's troubles. I questioned this wisdom, believing that the downturn might help him. One part of my argument concerned risk aversion:

A global downturn also brings into a play one of the strongest forces in human affairs - risk aversion.

A fashionable area of economic research is the exploration of the way that people make decisions under uncertainty. Repeated experiments demonstrate our reluctance to accept a bargain with an uncertain payoff rather than another bargain with a more certain, but possibly lower, expected payoff.

When faced with a choice between the devil they know - Gordon Brown - and the devil they don't - David Cameron - voters may prefer to stick with Mr Brown.

Part of this research, employing experiments with gambling and such like, shows the much greater impact of losses on human beings, and the prospect of losses compared with gains.

In an ordinary election two parties might fight over which has the best offer, which provides the bigger gain. The risk involved in making the wrong choice - sacrificing a gain you haven't yet made - doesn't loom all that large. Losses, according to many of these new school economists, are a different matter. If you make a mistake and lose more than you bargained that you would, it is much more painful.

Voters in elections where economic losses are in prospect are likely to be more risk averse.

The critical words in the last sentence are "in prospect". People are risk averse when losses are "in prospect". When the losses have already happened they became more willing to experiment.

So now is the moment when "it's no time for a novice" is at its strongest against "it's time for a change". I don't think that, even now, it would win. But this is the best time to go.