prisonsPrisons are expensive and there is little evidence they work. It costs £37,000 per year to keep a person in prison – ten times that spent on a secondary school pupil – yet six out of ten prisoners are illiterate and six out of ten re-offend within two years. The annual cost of failure is estimated at £11bn.

Eight years ago I argued that it is time to end the cycle of failure and rethink the basic structure of prison life. Working under the auspices of my organisation the Do Tank, with an inter-disciplinary team that included senior members of the prison service, prison governors, criminologists, educationalists and the architects Buschow Henley, we developed an alternative model: a prison that would cost no more to build and would save money in the short to medium term as a learning programme based on proven international best practice would bring down re-offending rates. Our model – Learning Works – is based on two years work with prisoners and prison officers in England and on international best practice.

At the heart of our proposals are new architectural designs which maintain the highest levels of security while freeing up staff time and prison budgets. Currently at least 80 percent of resources are absorbed by the prison infrastructure leaving less then 20 percent for programmes proven to reduce re-offending. The Learning Works proposals reverse this ratio whilst maintaining the highest levels of security.

Maintaining Victorian buildings, which occupy valuable urban land, is increasingly costly. Escorting prisoners along sprawling wings to exercise, see their lawyers, or attend a class is labour intensive. The proposed ‘learning prison’ would be at least a third smaller than existing jails. It has been designed to reduce the movement of prisoners within the jail, enabling scarce resources to be switched from security to rehabilitation.

The learning prison is made up of 11 houses, in each of which an accountable group of 36 prisoners will live in a small community. Cells are grouped on three floors around a central atrium, ensuring a secure inside and outside space where the prisoner is continually visible and can move unescorted. In every detail capital investment in infrastructure has been planned to support social outcomes.

The houses are networked for working and learning. There are screens for individual intranet learning – an important resource for many who are ashamed to tackle their learning issues in groups. Common areas are also available for working, learning, counselling and exercise. Prisoners would be expected to work an eight-hour day and take part in community activities such as cleaning, cooking and budgeting. These too are learning activities, central to the prisoners’ ability to structure their lives after release. Inmates will have free run of a secure space, but they will be closely supervised by staff teams located in the house.

The role of the prison officer would be transformed. Officers would become facilitators, with opportunities for formal and on-the-job training, for learning and career progression. This innovation, like others we suggest, builds on existing best practice. Reduced rates of reoffending offer the potential for increased salaries.

Is this still a prison? Yes – prisoners are deprived of their liberty and free contact with family and friends. The surroundings are no longer mentally and emotionally repressive, but they are spartan. This is a place that you would not choose to visit more than once. That single visit should equip you with skills, internal discipline and the potential for personal transformation.

In 2002 the proposals were welcomed by Hilary Benn, then prisons minister and Martin Narey, then Director General of the prison service and championed by providers in the private sector who argued that they would be only too happy to provide such a prison, if only the government would not repeatedly tender on a low cost, old model basis. Why has it not happened? The answer is in part malaise. Reforming the prison system, however important, will never win many votes. But it is also about some of the deeper structures and mindsets that are limiting our attempts at public service transformation more generally.