This panel session from the Modernising Criminal Justice conference 2026 brought together architects, a former prison governor and someone with lived experience of the CJS to make the case that the physical design of prisons and probation buildings directly affects behaviour, safety and reoffending.
Gerry Guerrero, Global Civic Director at HDR Architecture, introduced the session alongside colleagues Heather Lorenzo (Justice Planner + Research) and Tommy Sinclair (East Region Civic Director), plus Gethin Jones, Founder of Unlocking Potential and former prison governor Paul Baker. Gerry noted HDR's justice work increasingly draws on the firm's healthcare design expertise, given that a large proportion of people in custody have some form of mental health need.
Heather set out the framework HDR applies: trauma-informed design, trauma-responsive design, evidence-based design and biophilic design. Trauma-informed design starts from leadership down to frontline staff, recognising that different people - and different conditions, from ADHD to autism to chronic illness - respond to different triggers. Trauma-responsive design translates this into physical choices: textured walls that autistic residents can use to self-regulate, a choice of outlook from a window (trees rather than a wall and fence), furniture that isn't bolted down, and control over lighting. Tommy added that evidence-based design means applying proven research to space planning from day one of a project - even something as simple as colour choice has a measurable effect on emotion, and needs to work for colour-blind residents too. Circadian lighting that mimics sunrise and sunset is now used in rooms without windows.
Paul Baker, who ran HMP Norwich, HMP Leeds and later oversaw around 40 prisons as a group director, was blunt about the culture he inherited: risk-averse to the point of self-defeat. His example - a prison that chopped down every tree prisoners climbed, rather than asking why they were climbing in the first place - drew a direct line to Halden Prison in Norway, where reoffending runs at 20-22% over two years compared with around 40% in the UK. Paul argued that well-designed environments improve safety, reduce self-harm and violence, and cut staff burnout - benefits he said are increasingly backed by evidence, not just aesthetics. He also acknowledged the challenge of retrofitting these ideas into existing UK prison stock built for a different era.
One of the session's most powerful moments came from Gethin Jones, who described a childhood of trauma that led to his first conviction at 12 and repeated incarceration by 15. He said the system responded to his trauma with punishment, which only deepened his distrust of it. Everything changed when he moved to Sheville, a redesigned unit at a prison in the early 2000s - light-filled, with garden access and staff who wore polo shirts and used first names. Gethin was careful to say the unit didn't undo his trauma, but it started a process that eventually led him from 56 offences and heroin dependency to founding an organisation that has now worked in over 50 UK prisons and internationally.
Audience questions repeatedly returned to money. Gethin noted that of the roughly £50,000-£60,000 spent per year keeping someone in prison, he estimates 90-95% goes on infrastructure and as little as 5-10% on education and rehabilitation. Gerry and Paul argued that "hard" design choices - stainless steel toilets, heavy-gauge doors - cost roughly four times as much as standard alternatives, and that better design doesn't necessarily mean higher cost once maintenance and reoffending savings are factored in. Paul pointed out that matching Norway's reoffending rate would save billions.
Asked whether recently opened prisons reflect these principles, Gethin - who works within one of the newest prison contracts - said progress is limited: new builds are largely replicating existing templates rather than testing genuinely different models. His call was for government to commission a live comparison between a conventional new build and a smaller, differently designed alternative, with strong political leadership behind it.
And even better would be "a double investment," addressing both prison environment and custodial settings in order to drive prevention and stopping the "pipelines" into prison in the fisrt place.
The panel's evidence suggests design choices that feel like "nice to haves" - natural light, colour, choice over small daily decisions - are linked to measurable operational benefits: fewer assaults, lower maintenance costs and reduced staff burnout. The open question for delivery leaders is whether new-build programmes will be given permission to test genuinely different models, rather than repeating existing templates at scale.