As part of efforts to place a greater focus on the education and rehabilitation of young offenders, the government is pushing ahead with plans to establish pilot secure schools. Draft guidance published in June by the Ministry of Justice (MoJ) sets out the expectations and requirements for prospective secure school providers.
While elements of the proposals have been welcomed by the sector - the Association of Youth Offending Team Managers (AYM) praising the "fresh and child-focused approach to custody" - a number of concerns persist.
Controversially, the draft guidance reveals that the "expected student cohort" for secure schools will include girls.
The prospect of mixed provision has proved unpopular in the past when it was included as part of previous government proposals for "secure colleges". Campaigners warned that girls held alongside boys would be placed in danger of sexual assault and exploitation.
However, the proposals for secure colleges, which would have held as many as 320 young people, went on to be dropped in July 2015.
Unlike the proposals for secure colleges, the proposed secure schools do not require a change in the law. They will be established under using existing 16 to 19 academy and secure children's homes legislation, meaning there is no need for a parliamentary vote on the matter.
Size is also an issue. Although the plans are now for each establishment to have between 60 and 70 places - making them far smaller than the previously envisaged secure colleges - there are still fears that this will be too large to deliver the kind of provision intended.
Carolyne Willow, director of Article 39, points out that although the guidance document for prospective providers indicates they will be based on secure children's homes, the size does not compare.
"Secure children's homes in England and Wales look after 17 children on average," she says.
"The staff to child ratio is usually between 1:2 and 6:8. Yet secure schools are envisioned to hold up to 70 children. Admittedly this is smaller than regular penal institutions, but we must abandon prisons as a point of contrast."
It is a concern shared by AYM. "We would like to see more thinking on the size of the establishments, evidence for best practice points to smaller locally managed placements ensuring children maintain links to their own communities promoting better opportunities for community integration," Andy Peaden, AYM chair says.
Meanwhile, it is not clear where the secure schools will be located. Willow believes that Medway Secure Training Centre (STC) in Kent, where there were allegations of abusive behaviour in 2016, is under consideration.
The STC currently offers secure accommodation for up to 67 males and females aged between 12 and 18 years. The concern is that unless secure schools are purpose built and represent a clean break for the existing estate, it will simply represent a rebranding exercise, rather than delivering a change in culture.
"The use of former prison buildings would contaminate the culture from the start," Willow says.
There are also concerns that the initiative will be led by the MoJ, rather than having Department for Education or cross-governmental involvement. "The Ministry of Justice should not be in charge of child welfare institutions, if this is really the plan," adds Willow.
A new model for custody is essential, particularly one that endorses the phasing out of YOIs and STCs
By John Drew, senior associate at the Prison Reform Trust and former Youth Justice Board chief executive
The consultation document makes a clear commitment that these new schools will be child-focused, keeping visible security features to a minimum, run by experts in educating and caring for children, and will be much smaller than the current prison-like young offender institutions. So far so good.
But there are problems with the new model. Most important, there are only two fleeting references to black, Asian and ethnic minority children who will very likely be a majority in these schools.
Reducing and responding to racial disparity are the most important issues to address in respect of children in custody. I know just how difficult these issues are. But they will need much greater attention.
I remain certain that the schools are too large at 60 to 70 places each. Everything we know about institutional care of children tells us that residential centres much larger than 35 are at major risk of being too large to provide the individualised response that is central to their success.
It is telling that the recent work to turn around Medway Secure Training Centre has been with this number of children. There's a big lesson here.
Then there are critical issues like restraint, training and retaining staff, making genuinely transformative relationships between adults and children (that can continue post-release), and the use to be made of "temporary release", where the model is as yet ill-defined.
Underpinning all of this is the difficult issue of the cash available to each school. Money is not everything, but "cheap as chips" rapidly leads to corners being cut and children contained rather than cared for.
I remain convinced that a new model for custody is essential, particularly one that endorses the phasing out of both YOIs and secure training centres. So I remain cautiously optimistic. But until we are clearer on these difficult issues, it will remain hard to conclude that these secure schools are the answer.