Clutching her baby, Kate walked into our project on the Easterhouse estate, Glasgow. She wept as she told us that, after paying her debts, she had £30 a week to keep a family of four. She didn't think she could go on. While Kate told us her story, volunteers nursed her child. We tried to sort out the debts.
Like so many others, Kate had not turned to government agencies, but to her neighbours. Community projects such as ours are locally controlled, staffed by volunteers and exist for the people of a particular area. The Community Development Foundation estimates that these neighbourhood schemes involve more than two million citizens - yet the makers of social policy consistently undervalue them.
Projects such as ours take on a variety of services: food co-ops, credit unions, youth clubs, second-hand shops. They are characterised by their ability to respond practically and quickly. Recently, a 17-year-old was fatally stabbed. His mother faced a funeral bill of £2,000 from her income of £64 a week. We negotiated a grant for her.
Two 20-year-olds from Easterhouse were asked on television why they had not got into drugs and crime. They replied that, at the age of 12, they had chosen the "club stream" rather than the "gang stream".
What explains these achievements? For a start, neighbourhood schemes are rooted in the community. The committee of our project is elected. Members know what residents want - be it ping-pong or skilled drugs counselling. They show that low-incomed people can form policies, handle budgets and manage staff. Staff are long-term. They do not move on after three years; they stay and are trusted.
Community Links has 1,600 current local projects on its database, yet finds it hard to attract official support. Central government will not fund it, and local authorities increasingly direct grants (in the form of contracts) to already wealthy national voluntary bodies. Some progressive charitable trusts do favour local schemes, but their resources are limited.
One might expect the Labour government to give such projects a more central role, given its stated intention to transform Britain's 3,000 most deprived areas. The Social Exclusion Unit (SEU) launched its £800m New Deal for Communities soon after Labour came to power. The unit invited "partners" - for example, local authorities and health authorities - to bid to become pilot pathfinder areas. Now action and analysis have been drawn together in an SEU report, "National Strategy for Neighbourhood Renewal: a framework for consultation". In it, Tony Blair reiterates Labour's commitment to "renewing and revitalising poor neighbourhoods". It is clear that the SEU is going to plough more resources into such powerful partnerships, which might also include housing associations, private companies and national voluntary societies. Large-scale interventions from outside will no doubt lead to sports complexes, better housing management, environmental improvements and so on.
Yet the approach is not unlike that of the Thatcherite City Challenge and urban development corporations, whose elitist, top-down methods did much for private enterprise, but less for the benefit of residents of poor neighbourhoods.
The missing element in the consultative document is any hint that neighbourhood schemes will be players in the regeneration game. The usual platitudes are voiced about the importance of empowering residents, and the partnerships will no doubt put some money into local groups. But, insultingly, the report mentions grants of £50-£500 at the same time that the SEU is lavishing 21 potential partnerships with £200,000 each for management costs to beef up their bids for New Deal millions.
There is no consideration of a mechanism to allow residents - and residents alone - to allocate and administer the millions for their neighbourhoods. There is no recognition of the central roles played by low-paid members of neighbourhood groups. A few crumbs are thrown at the socially excluded, while the feasting is left to outsiders. Why?
The SEU approach is a classic top-down one: the powerful decide what is best for the poor. Ironically, a unit set up to tackle social exclusion starts by excluding the excluded. It copies the widespread pattern in which the controlling members of quangos, health boards, citizenship schemes, national voluntary bodies and so on are appointed from among the privileged, who are distanced socially, financially and geographically from those at the hard end.
My proposal is that the SEU finance a National Neighbourhood Fund, which would allocate money to neighbourhood trusts in deprived areas which, in turn, would make grants to locally run projects. The neighbourhoods would elect the members of the trusts, who would, in turn, elect the national fund members.
Low-income residents would thereby control national and local fund distribution. There are many advantages: neighbourhoods would gain the services that they choose; more preventative services would mean fewer children removed from their families; local jobs would be created.
I am not suggesting that the SEU's policy of creating large outside partnerships be abolished, but that the number of areas to be helped by partnerships be matched by a similar number where the same amount of money goes to neighbourhood projects as to the main vehicle of change. At the end of a set period, their effects on neighbourhoods and their residents can be compared.
For government to trust poor people with millions of pounds goes against the modern grain, because it involves denting the power of the privileged. But if Labour really wants to bring about a transformation in the most deprived areas, it has to look at the ways in which single parents such as Kate can be helped to deal with their debts.
The writer's Kids at the Door Revisited is published by Russell House (£9.95)