‘Prison governors have been ordered to cut the cost of holding inmates in England’s bulging jails by £149m a year, as part of a radical programme designed to slash the costs of incarceration by £2,200 a year per prison place’, reported The Guardian recently. The ‘savings’ were to be found through economies of scale (closing smaller prisons and building bigger ones), replacing prison staff with CTV cameras, and an enforced ‘benchmarking exercise’ designed to drive prisoner costs in higher-cost jails down to the level of the lowest. Meanwhile, the justice secretary has responded to two cases of high-profile prisoners absconding from open jails by tightening the licensing and temporary release rules.

Taken together, these are among the most depressing pieces of news in the last six months. They are a veritable compendium of stock management responses and assumptions (size, cost, technology, control), all of which are misleading, counterproductive or wrong. They’re back to front, stating the answers before the question. It’s management on auto-pilot, without thinking or learning, a continuation of the policies that have seen the prison population double to 85,000 since the 1990s, thus causing the cost explosion that the current initiative is supposed to mitigate, only on steroids.

These are the same outdated industrial strategies, born in the mass-production factories of the mid-20th century, that continue to wreak havoc across the NHS and the rest of the public sector. The benefits of scale are always overestimated, partly because the wrong things are being measured, and the diseconomies ignored, for the same reason; managing cost always raises cost because it starts from the wrong end – lower costs are the earned consequence of redesigning the system to work better; technology is invariably used to do things that should be done by humans and vice versa; tighter controls focus on doing the wrong thing better at the expense of doing the right thing. It’s a locked-and-bolted certainty is that as with A&E units or local authority benefits offices, the remedy will make things worse and more expensive.

Starting with cost per prisoner year is a classic case of management being led astray by an irrelevant measure – one that calculates activity cost but gives no indication of how well the system is performing overall against its purpose. Only measures related to purpose are any use to anyone, whether politicians or managers, trying to figure out how to do things better.

Let’s assume for argument’s sake that the purpose of prison is to protect communities and enable the maximum number of people to live stable lives. In that case, even a cost per prisoner year of zero would be a false economy if achieving it means putting offenders in larger prisons which have higher reoffending, suicide and self-harm rates than smaller local establishments where the offender can stay in touch with their family, friends and local services and thus stands a better chance of reintegration into the community when they come out. Larger prisons, such as Oakwood, the G4S-run jail that sets the cost benchmark, are often drug-ridden, impersonal, violent and hard to control, making them perfect academies for criminals and breeding-grounds for addictions and dependencies that pass on massive costs to other public agencies.

Consider the prison population (‘demand’) as a three-tiered triangle of which the apex comprises a small minority of unreconstructed serial offenders who absorb the bulk of the prison resource. At the bottom (the biggest section) are those who are in prison through bad luck, temporary misfortune or because their lives have fallen into chaos. In the middle are prison’s ‘floating voters’, offenders who could go either way: either joining the ranks of the hard cases or with a bit of help getting their lives back on track so that they can rejoin society without endangering themselves or their communities.

Since between 50,000 and 70,000 offenders rejoin society every year (it’s self-evidently too expensive to keep them in for ever), the effects of being locked up really matter. In all three categories, the only conclusion is that traditional prison does a terrible job – not just wasting resources, actually exacerbating the problems it was meant to solve. Recidivism statistics suggest that prison creates more violence than it prevents. Hardened criminals get harder and more prolific (hello, Skullcracker). Prison is poor at deterring or rehabilitating – nearly half of those who leave are back behind bars within 12 months, more in the case of those on shorter sentences, and, shockingly, three-quarters in the case of young offenders. As for the lowest-level demand, much of it occurs because, as with A&E departments, prisons are a plughole of last resort, the recipient of all the miscellaneous problem demand which has nowhere else to go. Most offenders in this category suffer from problems that prisons aren’t designed or equipped to solve – very few don’t suffer from one of mental, learning or other disability, drug or alcohol dependency, homelessness or unemployment or a combination of the above. They shouldn’t be there at all.

From all this it’s fairly obvious that the current prison cost-cutting exercise is an irrelevant response to the wrong problem. Crime prevention and treatment cries aloud for the kind of holistic, locality-based approaches to social policy issues that are being tried out with promising results by councils such as Stoke, Bromsgrove and Camden, along with many other agencies in the public sector. In the US, in the rare places where such therapeutic initiatives have been applied, they predictably end up saving many times their cost, by reducing overall demand. The best rehabilitation programme? Prison degree courses, which have been ‘shown to be 100 per cent effective for years or decades at a time in preventing recidivism’. Gardening also shows promising results. As for prisons, it’s amazing that we tolerate the continued existence of institutions with such a grotesque and expensive failure rate – literally throwing good money after bad. As a US professor of psychiatry puts it, in time ‘prisons will come to be seen as a well-meaning experiment that failed, rather like the use of leeches in medicine.’ But no politician would dare to say that.