Imagine that the people who benefit from public support were asked to pay back in some way, not with money but by helping others. That wasn't how the welfare state was intended to work - but it is an idea being explored at the liberal end of US politics, by philanthropists and professionals. Their answers may turn out to be relevant to one of the central questions facing the "Blair project": why, despite big injections of cash, are the problems that the giant welfare bureaucracies were set up to tackle as bad as ever? Why have nearly six decades of the welfare system made so little difference to poverty, youth crime, ill-health and school failure?
This is a revolution in its early stages. It barely has a label yet - though its practitioners are starting to call the new thinking "co-production" - but it emerges out of the failure of communitarian writers such as Amitai Etzioni and Robert Putnam to come up with much of a solution to the social breakdown they identified.
Put simply, "co-production" means that welfare programmes, policing and health need to be equal partnerships between professionals and clients. Politicians have been paying lip-service to the idea for a generation. What makes this latest thinking more subtle and revolutionary - and why policy-makers in the US and Japan are getting excited about it - is that "co-production" has come up with practical and often surprising ways in which professionals can transform their relationships with their clients.
Such as the poorest neighbourhoods in downtown Boston that work alongside Harvard professors to eliminate drug-resistant diseases. Or the 16-year-olds from some of the most notorious housing in Washington who run their own courts under licence from the District of Columbia. Or the pupils in failing Chicago schools who earn themselves recycled computers by tutoring younger children - with the result that the schools now record consistent gains in maths and reading among pupils who used to fall further and further behind each year.
Co-production first emerged at the University of Indiana in the 1980s as a way of explain-ing how policing fails without community support. Its ideas received dramatic support in 1997 from the Harvard School of Public Health. In a study of more than 300 Chicago neighbourhoods, the academics concluded that the main determinant of the crime rate wasn't income or employment but trust: whether or not people were prepared to intervene if they saw local children hanging around.
The concept was extended by the former civil rights lawyer and speechwriter for Robert F Kennedy, Edgar Cahn. He accuses professionals of creating dependency: a dependency of a peculiarly corrosive kind which convinces people that they have nothing worthwhile to offer, and which undermines what systems of local support still exist. "Professionals must learn how to say: 'I need what only you can do as badly as you need what I can do'," Cahn says.
In America, where philanthropy still plays at least as important a role in welfare as the state, most of Cahn's work has been done with charitable foundations. He argues that most philanthropic money doesn't get to the poor. It goes on the infrastructure and the professionals, and the hidden message is that people can gain access to help only if they have problems and crises. "Charity has to become a two-way street," he says. "Money, philanthropy, programmes alone can't cure social problems if we can't enlist those being helped as partners and co-workers."
US professionals who flirt with such ideas are often treated with suspicion by colleagues. For example, the child psychologist Jilly Kinney founded the charity Homebuilders in the hope that by sending professionals to work with at-risk families, she could prevent children being taken into care. Once the professionals had gone, neighbours (or people who had faced similar problems) might provide permanent support, she proposed. At this suggestion, Kinney was banished by her own charity and forced to set up another.
Sometimes, however, the issues are so intractable that there really is no alternative to getting ordinary people to do what once seemed the sacred preserve of professionals. Take multi-drug-resistant tuberculosis (MDR-TB). Medical opinion says that the disease requires such expensive drugs - and such complicated safeguards to make sure that courses of the few powerful antibiotics are completed - that only the very wealthiest communities can afford to start to tackle it. Health agencies advise developing countries not even to try. But in Lima, Peru, where MDR-TB is widespread, the US charity Partners in Health trained local people to supervise the drugs in patients' homes. Local people also designed individual treatments for each patient, with great success. Partners now achieves cure rates of 80 per cent - as good as anything in the US but at a fraction of the cost - and it has taken the lessons home to Boston. It has also been asked by George Soros's Open Society Institute to take control of Russia's TB treatment programme.
Cahn himself first used co-production to explain his approach to training lawyers. Students at the University of the District of Columbia School of Law are trained on the job by providing legal support for people and communities who need it but cannot afford it. But they don't do it for free. They charge out their time in a currency of Cahn's invention that he calls "time dollars" or "time credits". The recipients of legal advice pay off their bill either by passing on what they have learnt to somebody else or by helping out in the community in some other way.
In St Louis, a regeneration programme has put a similar principle to work, so that locals earn time credits for each hour of mutual support. These are then used to buy help from other neighbours, to pay doctors' fees, to take courses in the local college or to buy basic food. In the same way, a UK "time bank" at the Rushey Green GP surgery in Catford, south London - which involves patients in befriending, visiting or doing small repairs through a DIY scheme - has resulted in health improvements for both givers and receivers. It's a whole new conception of the NHS - one which sees patients as vital and wasted assets that can make all the difference between illness and recovery for neighbours.
Co-production turns welfare on its head. The main focus is no longer the failures of the person before you at the desk, it is their capabilities and how they can put them to use. Feeling useful is a basic human need that can transform lives. Public services designed like this would have a different focus from what we are used to, with different targets and working styles. There would be time banks or similar in every branch of a public institution.
Will it happen? The word co-production has been cropping up in Downing Street seminars where policy wonks desperately search for a solution to the sheer intractability of public service failure. Watch this space.
David Boyle is the author of Authenticity: brands, fakes, spin and the lust for real life (Flamingo, £12.99) and an associate at the New Economics Foundation