There’s a great piece by Richard Denniss over at The Monthly, on the uses of economic modelling in policy debates. I don’t agree with everything Denniss says, but this is a great read.
I think a lot of it has to do with power (which is pervasive in politics), but also with quantification, and how we think about things that are quantified (there’s a discussion here at The Guardian, although I don’t think it quite articulates the issue right).
This is an important one, but I don’t think I’ve seen the underlying problem clearly articulated yet.
EDIT: Also worth a glance at this piece over at The Age; sad that to be taken seriously, things have to be valued in economic terms. It’s worth recognising, I think, that while economics is a useful discipline, it’s limited in many ways. Putting a value on things that aren’t traded in a market is an area of weakness. And there are many valuable things which aren’t traded in a market; quite often for good reasons.