There’s an old joke in Whitehall about government’s commitment to “policy-based evidence making”, and in the criminal justice world, no subject has been as evidence-free as sentencing policy. So the statement made last week by the minister for sentencing policy, Rory Stewart, deserves more attention than it got.

Referring to a published analysis of the impact of short custodial sentences, the minister said:

“This further evidences that community sentences can do more good than short prison sentences, particularly in reducing re-offending”

The analysis is not an easy read. Much of it describes the arcane methodology required to separate out the real impact of a particular sentence from what was probably going to happen anyway. Over half of the bulletin is given over to detailed statistical tables. Its conclusions are cautiously stated, and most of the impacts it describes are worthwhile but modest, rather than startling.

But not all.

For the most prolific offenders, a short period in custody makes the odds of re-offending 36% higher than a community penalty. If that person also has significant psychiatric problems, those odds are increased to 67%.

In other words, it is precisely those cases where the defendant seems forever to be before the court – in which sentencers often say they feel they have “no option” but to impose custody – that the futility of resorting to custody is most marked.

This new analysis also confirms what evidence has shown for many years, that community penalties overall have a better chance of reducing reoffending than short spells in prison. That’s not a surprise, given the harm prison does to the prospect of having a job, a home and a family to return to, but it is significant that the numbers also consistently prove it to be true.

Of course, it’s no co-incidence that Mr Stewart is also the minister for prisons and probation. His prisons are in crisis, and his probation service is withering on the vine as courts continue to use prison where a community order would work better. Introducing a presumption against the use of custodial sentences of under 12 months, as the Scottish government has committed to do, would deliver some really significant operational relief for both prison and probation services. It would shrink the workload of hard pressed gaols while concentrating probation’s effort on the work it does best. The obstacle is neither practicality or money – quite the reverse. Rather, it’s the anticipated parliamentary and press reaction to anything that might appear to be “going soft” on crime.

An unevidenced narrative has been allowed to develop that courts have lost confidence in community penalties because the probation service was privatised. Misguidedly, some call for harsher enforcement policies to restore that lost confidence. But the truth is that the decline in the use of community penalties pre-dates privatisation, as does the evidence on the futility of short prison sentences. The existing range of community-based options is poorly understood and used – not least the mental health treatment requirement order which this latest analysis shows to be especially effective. And probation is still full of skilled people who can be trusted to exercise sensible discretion on whether someone really needs to face gaol because they are struggling to meet all the many demands a community order is placing upon them.

So hats off to the minister for putting some facts before us. They should serve as a reminder that we already have many of the tools we need to create both a more humane and effective system of punishment. The task now is to turn that knowledge into action.