I remember that last week during the presentation by Axel Leijonhuvud concerning monetary policy and the financial crisis, there was some animated discussion regarding the usefulness of DSGE models as a robust tool for guiding the formulation of monetary policy.
So I was interested to find by coincidence this recent working paper from the BIS concerning that very topic, which I thought that other people, who like me are on the 'dark side' of economics (i.e. macro), might be interested in. To spoil the ending of the paper, it seems to offer the perspective that whilst DSGE models are certainly useful tools which provide information relevant to the policy making process, they are subject to a number of caveats and limitations which need to be borne in mind in interpreting their results.
PL and Dany - sorry for knocking the ad for the BBL tomorrow from the top of the page!