In December 2002 the American Economic Review, one of the most prestigious economic journals, carried a paper by Sven Bouman and Ben Jacobsen called "The Halloween Indicator" in which they showed that selling equities in May and buying on Halloween had paid off very well. Anybody who had followed this advice would have sold shares in May 2008 and thus saved themselves a massive loss*.

This is not the only way in which economics predicted the crash. In 2007, I warned that a good lead indicator of equity returns - foreign buying of US shares - was predicting big losses.

In these two senses, economics did foresee the crash. Contrary to Andy Haldane, there was no "Michael Fish moment".

So, does this mean I agree with David Miles that economics has done just fine?

No.

For one thing, these two examples suggest that he is wrong to say that crises are "virtually impossible to predict". They did indeed predict a crisis. They suggest that aggregate stock markets are not wholly rational. Such a view isn't terribly heterodox: it was Paul Samuelson, the granddaddy of orthodoxy, who said stock markets were ""micro efficient but macro inefficient**."

We must distinguish here, however, between explanation and prediction. As Jon Elster stressed, these are two different things. These indicators predicted a crisis, but didn't explain it.

There is, though, a second reason why I disagree with David. His correct claim that standard economics has models of crises and bank runs overlooks the fact that, as Dani Rodrik says, the art of economics lies in applying the right model to the right situation. And many economists failed to do this in the md-00s. Rather than point to increasing bank leverage as a potential cause of trouble, they were talking of the "great moderation." And they perhaps focused too much on DSGE models that explained stability - or at least regarded shocks as exogenous - and too little upon Minsky-type models in which crises were endogenous.

However, although I'm unhappy with some defences of economics, there's something about attacks upon the subject that troubles me.

Quite simply, economists were not, for the most part, responsible for the 2008 crash - except perhaps insofar as their talk of the great moderation emboldened banks to take more risk. It wasn't us, for example, that said it was a good idea to take over ABN Amro (quite the opposite!). Instead, the crisis was caused by bankers. Criticisms of economics deflect attention from this truth - and in doing so, they serve a reactionary function.

* Updated and ungated pdf here. The rule would have meant missing out on decent profits in the summers of 2003 and 2005, however - although the UK market performed as well as cash in 2004, 2006 and 2007.

** Actually, the defensive and momentum anomalies suggest they aren't wholly micro efficient either, but let that pass.