Overcentralisation and the commons

The Archdruid has it right: 

The more distance between the managers and the commons they manage, the more likely failure becomes, because two factors essential to successful management simply aren’t there. The first of them is immediate access to information about how management policies are working, or not working, so that those policies can be adjusted immediately if they go wrong; the second is a personal stake in the outcome, so that the managers have the motivation to recognize when a mistake has been made, rather than allowing the psychology of previous investment to seduce them into pursuing a failed policy right into the ground. Those two factors don’t function in an overcentralized system. The center cannot hold, 'The Archdruid Report', 6 February

Overcentralisation is a difficult problem to solve under the current ...overcentralised regime. It's not at all obvious when and how to intervene to stop it happening, nor who should do so (as some of the commenters on the Archdruid's post say). Perhaps Social Policy Bonds can solve the problem: instead of trying to work out the optimal level of centralisation for any social or environmental intervention, it would seem simpler to target a specific outcome, reward the most efficient ways of achieving it, and let the market for the bonds determine who makes decisions and at what level of aggregation. A bond regime would encourage diverse, adaptive solutions of that an overcentralised bureaucracy cannot - and would be likely to see as a threat. In short, Social Policy Bonds would deal with the problem of overcentralisation quite neatly.