Sentencing is changing. What happens on release matters more than ever.
· Ezra Sutherland, Founder, The Quiet Orbit · 7-minute read
The Sentencing Act 2026 changes who goes to prison and when they come out. But a sentence doesn't end at the gate, and that's where too many people are still set up to fail.
I've seen the justice system from more than one side. I grew up around crime, and it took me 15 years to get out of it. Later I worked with the Probation Service, and with G4S in secure training centres. Today, The Quiet Orbit works with people who want to actively change their lives for the better, wherever they're starting from. For some, that starting point is a prison cell.
So when sentencing changes, I don't just read the headlines. I think about the people it will touch, and the environments they'll walk back into.
Where we are
Prisons in England and Wales are close to full. The population was 87,342 at the end of March 2026. Even with this year's reforms factored in, the Ministry of Justice expects it to keep rising, to around 95,900 by late 2032.
In response, the Sentencing Act 2026, which became law in January, changes two big things:
- Fewer short prison sentences. Since March 2026, courts have been expected to suspend prison sentences of 12 months or less, unless there's a good reason not to, such as a significant risk of harm.
- Earned progression. Instead of most people being released automatically at the halfway point, eligible prisoners will be able to earn release earlier, from around a third of their sentence, based on their behaviour and engagement. Those who don't engage can be held longer. Some serious violent and sexual offences are excluded.
You can argue about the politics. I want to talk about what these changes mean in practice.
Short sentences: too long to keep a life, too short to change one
The case against short sentences isn't new, and the numbers are stark.
Of adults released from prison after a sentence of less than 12 months, 67.1% were proven to have reoffended within a year. For sentences of 12 months or more, the figure was 23.2%.
(These figures are for people released between April and June 2024, published by the Ministry of Justice in April 2026. The MoJ cautions that the groups aren't directly comparable, because different people get different sentences. The gap is still hard to ignore.)
I've watched what a few months inside can do. It's long enough to lose your tenancy, your job and your place in a family's routine. It's nowhere near long enough to build anything new. You come out with less than you went in with, into the same environment that led you there in the first place.
If fewer people go through that cycle, that's a good thing. But a suspended sentence only works if there's something real in the community to hold onto. Without that, we've just moved the problem from one place to another.
Earned progression: engage in what?
I support the idea behind earned progression. Rewarding effort and responsibility makes sense. It's how most of life works.
But it raises a question I'd want every prison to answer honestly: what exactly are people meant to engage with?
You can't earn your way forward by taking part in purposeful activity that isn't there. When regimes are stretched, staff are short and education places are limited, "engagement" can come down to luck: which wing you're on, which week it is, whether the session runs.
Our work is built on a simple idea:
What someone achieves depends first on who they are, their natural traits, temperament and skills, and then on whether the environment around them provides the opportunity for those skills to be used.
Earned progression puts a lot of weight on the first half of that sentence: the individual's behaviour. It needs the same weight on the second half. If the environment doesn't offer real opportunities, we'll end up measuring access, not effort, and calling it character.
The gate is where sentences succeed or fail
Whatever the sentence, nearly everyone comes home. And the first days after release decide a great deal.
Think about what a person might be facing on day one:
- No ID, and no bank account to pay wages into
- Nowhere settled to live
- Probation appointments to keep, and licence conditions to follow
- Employers who won't look past a conviction
- The same people and pressures that were there before
If any one of those goes wrong, everything else gets harder. When it falls apart, it's usually recorded as the person's failure. Often it's the plan that failed, or there wasn't one.
That's why release can't be an afterthought. It has to be designed, just as carefully as anything that happens inside.
What we're doing about it
Groundwork, our enterprise programme, was built around this problem. It works with people in the last months of a sentence who want to start their own business. Self-employment can be a route around employers who won't look past a record, and a way to work around licence conditions, family and probation.
Over eight weeks, participants build a real business plan, pitch it to a panel, and leave with a release plan for day one, the first 30 days and the first 90 days. We then stay in touch at one, three and six months after release, to see whether the business is registered, trading and lasting.
It's not the answer for everyone. But it's built on what I believe the whole system needs:
- Start planning for release early, not on the last day.
- Make every activity produce something real, such as a plan, a skill, a reference or a contact, that a person can use outside.
- Follow people through the gate. Support that stops at release stops right when it's needed most.
- Measure what lasts, not just what's easy to count. Attendance and completion are a start. What matters is what happens six and twelve months later, and why. That's what our evidence system, S.C.L.E.C.I. (Sutherland's Core Longitudinal Evidence and Change Intelligence), is designed to track.
- Look at the environment, not just the person. If someone builds real skill and confidence but has nowhere to use it, that's not their failure. It's a gap we can close.
The real test
Sentencing reform will be judged on prison numbers, at least at first. That's understandable when prisons are this full.
But the real test is simpler, and harder. A year from now, are the people who passed through the system safer, more capable and more connected than they were? Or did they leave with less than they came in with?
I know which answer most have lived. I know which one we're working towards.
Ezra Sutherland, Founder, The Quiet Orbit
Work with us on release and resettlement
Bring Groundwork to your establishment, or talk to us about a programme built around your outcome.