The prison population has fallen by five a day since the 2015 General Election, Richard Garside writes. What's needed is a long-term plan to deliver a sustained fall.
The state of the prison system in England and Wales is so grim, so awful, it is easy to feel at a loss over where to start and what to do about it.
Earlier this week the Prisons and Probation Ombudsman, Nigel Newcomen, reported that prison violence was now at a 'wholly unacceptable level'.
Six prisoners were killed in the 2015-16 year, he observed. This is up from four last year and is the highest number since records began in 2000.
'Buildings are crumbling, infested with rats and cockroaches', The Economist notes this week. 'In the year to June 2015,' it points out, '105 prisoners killed themselves, compared with 59 in 2010'.
The solution offered by the former Justice Secretary, Michael Gove, was to establish 'Reform Prisons'. These semi-autonomous institutions, supposedly freed from the dead hand of Whitehall micro-management, would apparently be able to deliver safe and decent institutions for staff and prisoners alike.
In truth, reform prisons were and are a tragic distraction from the real challenge facing our prison system, as I argued earlier this year. We need more of a focus on prison reform. Less of an obsession with reform prisons.
Recent hints that Liz Truss, Mr Gove's successor, might be rethinking the reform prison plans is therefore to be welcomed.
The problem of prison numbers
At the heart of the current malaise in our prisons is the sheer number of people unnecessarily imprisoned.
Last year, the prison population stood at around 85,000 people. In 1985, at the height of Thatcherism, there were around 46,000 people in prison.
The rate of crime and lawbreaking was not notably different last year than it was 30 years ago. What has changed is the way we deal with suspected and convicted lawbreakers.
We are better at criminalising people, at giving longer sentences to those who are convicted and at making it more difficult for people to get out of prison and away from ongoing supervision and surveillance.
If we imprisoned today at the same rate as Mrs Thatcher's government did, we'd have at least 30,000 fewer people in prison.
Mrs Thatcher was hardly a bleeding heart liberal on law and order matters. On prisons she was far more liberal than the most liberal of mainstream politicians today.
How might the government go about achieving the very necessary goal of a downsized prison estate?
The prison population grew slowly, and inexorably, over a number of years. It is likely that any attempt to downsize prisons will require a similiarly long-term approach.
In 1995, the prison population in England and Wales was around 51,000. It grew to 76,000 by 2005. Last year it stood at 85,500.
Over that 20 year period the population grew by around 34,500, or by five prisoners a day.
What if we reversed this trend and pursued policies that reduced the prison population by five prisoners a day?
What if, shortly after the last General Election, Michael Gove had committed the government to a slow and sustained reduction in the prison population?
By now, that policy would have delivered a reduction in the prison population of around 2,200. The prison population today would stand at some 83,300.
The actual prison population today is 85,321.
A reduction of five prisoners a day would deliver a prison population of around 82,000 by mid 2017, and 80,000 by mid 2018.
By the time of the 2020 General Election - if the current parliament lasts that long - the prison population would stand at around 76,500; about the same level as prevailed in 2005.
By time of the 2025 General Election, the prison population would stand at some 67,300. By 2030 it would stand at 58,200.
It would fall below the 50,000 mark by the time of the 2035 General Election.
The power of small changes
People can get used to just about anything, if it happens slowly enough. We have got used to a bloated prison population precisely because it grew slowly, almost imperceptibly, over years and decades.
A target to reduce the prison population by a mere five prisoners a day will strike some as overly cautious.
Getting the prison population down to around 50,000 should also not be the end of our ambitions.
Yet we are where we are because of the slow, but relentless, rise in the prison population over many years.
It is plausible, realistic indeed, to work on the basis that unwinding the current mess in our prison system will require patient, sustained work over a number of years.
After all, the power of small, incremental reforms lies in their capacity to deliver important, long-term gains.