He points out a few political constraints, in particular:
If a John Lewis style primary school were a floperoo, would all the teacher-shareholders be sacked, or only the head? A resolution procedure for failing co-ops that didn't harm pupils - or patients of community nursing teams - would plainly be essential.In fact, this is probably a constraint on all private provision of public services. The co-op structure is not really an issue either way, except that it sounds nicer and less cut-throat than straight privatisation.
I'm a fan of market solutions and have generally been well-disposed towards the idea of private companies bidding to do public work - why shouldn't a company be able to offer medical treatment or training if it can do a good job for less than the public sector?
But what if the inefficiencies of the public sector do not actually originate in the incentives of staff, or the lack of innovation in civil service bureaucracy? What if the high costs of public provision are all down to the need to cover every citizen and every eventuality?
One of the main ways in which private firms make money is by picking the most profitable customers or the most attractive markets. This is why we have universal provision rules in the telecoms, postal and water industries, all of which make much less money as a result. It's relatively easy to cover the first 95%, but covering that last 5% of any market gets really difficult and expensive. Look up 'increasing marginal cost' in your economics textbook - there's a reason the supply curve slopes upwards.
Inevitably any public service will suffer political pressure to cover everyone - including the hard-to-reach, hard-to-service individuals - and to provide enough spare capacity to step in if a particular institution (school, hospital) fails. Perhaps there is no solution - public or private - to provide the level of universality that we demand in our public services without the high costs and low average productivity that we see in the public sector.
How depressing that would be.
Update: Chris Dillow has found some completely different reasons to be cynical about this idea.