Ambitious? The housing targets don't even begin to reduce the backlog

The preferred option shows how it is possible to build 366,000 homes in the study area by 2031, creating 293,000 jobs while protecting valuable agricultural land and conservation sites. As well as a major expansion of the current growth areas of Milton Keynes and Northampton, the plan will assist regeneration of the less prosperous towns of Luton, Bedford and Corby. It includes proposals for providing adequate affordable housing, improving public transport so as to shift away from car-borne journeys and investing in advance on the full range of community infrastructure.

These proposals should be warmly welcomed, especially by the housing lobby. Over the past 20 years the number of new homes being built has slumped. We now face a huge shortage of housing across southern and eastern England. It is not a choice between the alternatives of more housing on greenfield sites or more urban regeneration. We need the major expansion proposals and to build on all feasible brownfield sites and the higher densities proposed by the government and to stem the outward drift from northern cities.

The targets now being set may look ambitious, but even these are likely to be too low. Not only did the government cut back the target for new homes in the South-east proposed after the public inquiry by its own inspector, but neither those regional planning targets nor the London mayor's strategic plan for London include provision for reducing the existing backlog.

The failure to provide for this has for years been one of the most serious failures of planning policy. The "backlog" is actually the housing crisis we see around us – the thousands of homeless families spending years in temporary accommodation and the single people with nowhere to stay but friends' floors or in emergency hostels. Leaving it out of the estimates of need is rather like a hospital ignoring its waiting list and only planning to carry out operations for people who become ill in the future.

The Housing Commission, which I chaired for Ken Livingstone, calculated that London needs an extra 11,000 homes every year, in addition to the 21,900 required to meet the demand from its growing population. This amounts to almost 10,000 a year more than the 23,000 provided for in existing government plans.

Yet at present only 16,000 homes are built in London a year. Even with higher densities and with the new proposals for building over shops and petrol stations and bringing derelict brownfield land into use – especially in the Thames Gateway – it is unrealistic to expect that more than 23,000 homes a year will be built. Some of the sites proposed for new homes have serious drawbacks and too often those reserved for affordable housing are on the least attractive land – bordering railways, busy roads or unsightly pylons.

Housing associations play second fiddle to private developers, where local authorities have negotiated an element of affordable housing. The roles should be reversed

It is fashionable to decry the "dispersal" policies that led to the post-war New Towns and the "out-county" estates. My own experience tells a different story. Thirty years ago, as a young community worker in north Islington, I made a formal objection to the inadequacy of the housing policies in the Greater London Development Plan. Invited to give oral evidence, I found myself cross-examined by a leading planning QC, before the panel chaired by Sir Frank Layfield and the government's ex-chief planning officer, Professor Jimmy James. Anxious to strengthen my case for the GLC's ability to provide more homes, I decided to go to the public meeting being held in Tring in Hertfordshire, where the chairman of the GLC housing committee was defending proposals for a new housing estate in the town, in the face of vehement criticisms from residents.

Twenty years later I came back to Tring to live here. The "GLC estate" had been built. For years its residents had been blamed for everything that went wrong in the town. But today the Silk Mill is an integrated part of the local community. It has helped to save Tring as a vibrant market town and avoid it turning into a lifeless commuter suburb. It means the local secondary school is a genuine comprehensive. As a practical illustration, it is home to our dog walker, who spends every day taking more than 20 dogs for their exercise, in groups of four or five, and has become one of the most common sights around the town – and probably as valuable for community safety as any beat police officer would be.

Hertfordshire is one of the successes of post-war planning. Its three new towns have settled successfully into stable communities, including far more affordable rented homes than the traditional Home Counties towns. Urban sprawl has been contained within the envelopes of the towns and villages, so that between every settlement are several miles of unspoiled countryside. The Grand Union Canal offers delightful recreation for boats, anglers and walkers.

We need to learn from these experiences in planning for the future. In both urban and rural areas we need more socially balanced, mixed-tenure communities. We need policies that meet housing need, provide jobs and protect the environment. Sadly, housebuilders and private developers have lacked the vision to seize those opportunities. Most have focused only on minimising risk and maximising short-term profit.

At present housing associations play second fiddle to the private developers, where local authorities have negotiated an element of affordable housing. There is a good case for reversing those roles, with housing associations playing the leading part in partnership with the local authorities. The Rowntree Housing Trust's plans for a mixed-tenure community at York provide one example of how this can be done. The Peabody Trust is taking a similar lead role for a development in the Paddington Basin.