Wednesday, February 01, 2012

Use and abuse of modelling

In a comment on Navigating the economic forecasting mess, regular commenter kvd pointed me to this piece by Ross Gittins, Damned lies and economic modelling. kvd asked; "Are you moonlighting for Ross Gittens? .. When I read this I could almost swear I've read very similar things right here!"

Over at Winton Bates' place, a strongly favourable comment here on Ross Gittins' writing from another regular commenter, Evan, inspired Where is Ross Gittins coming from? It's kind of a Gittins phase at the moment!

kvd is right, of course, because I have been hammering away at the misuse of modelling as part of my analysis of just what's wrong with current approaches to management and public policy. The Gittins piece is worth reading because it's a very clear exploration of one aspect of the problem. 


Winton Bates said...

I think the world might be a better place if people stopped talking about multipliers. Multipliers must be about the most dodgy concept that anyone ever tried to put numbers on. Government spending multipliers are probably even more dodgy than the industry employment multipliers tha Ross writes about.

Jim Belshaw said...

They have certainly been used in a dodgy way. However, when I was looking at the 1924 Cohen report on new states I wished that the multiplier had been invented!

NSW Treasury figures concluded that a new state in Northern NSW would not be viable because of added costs involved in new jobs presently done in Sydney. When I looked at the numbers, the application of even a moderate multiplier meant a significant net gain to the North, a significant net loss to Sydney. I suspect everybody knew that!