“Complexity and the art of public policy” is an important book. If you are a politician or a senior civil servant making policy you absolutely must read it. Its authors - David Colander and Roland Kupers - say that our world is becoming an increasingly complex, highly-interconnected social system - one in which everything affects everything else. But the economic and other models on which government bases policy-making do not reflect this; and we are suffering as a consequence.

The book explains how complexity science - the study of complex adaptive systems - is being applied increasingly to good effect in the harder sciences but not much in the social ones. It particularly singles out economics as laggardly, with out-of-date simplistic models and entrenched industrial-age mindset.

Its discusses the ineffectiveness of direct policy action which is increasingly leading to unintended consequences and often expensive failure. They argue that governments needs to shift their focus towards creating healthy ecosystems that will enable society to solve its own problems and away from fixing those problems directly. Government must think of itself more as an influencer and less as a controller.

I notice discussions of "wicked problems" (another name for complex problems) cropping up in government circles more often nowadays and last year I even attended a seminar run by one of the departments. I was delighted to bump into a chap at the event who turned out to be the department's permanent secretary. I quizzed him about how complexity thinking was affecting his policy-making. Sadly, I got a very blank look back.

David and Roland are quite right that complexity has little visibility at the policy level. Frustratingly, I'd say that the complex problems are the only ones worth fixing. I've written about this separately: the problems of the Universal Credit programme are much to do with a failure to understand the implications of complexity - see Universal Credit and the need to think more ‘Grow’ (and less ‘Build’) and Government doesn't get complexity.

Complexity is such an important thing for government to get. Read the book.

Other reviews by...

• Herbert Gintis
• David Wilson