Welcome to the Public Accounts Committee on Monday 27 January 2025. The prison system has been operating at close to full capacity since autumn 2022, with many prisons severely crowded. In 2021, the Government committed to build 20,000 new prison places in response to rapidly increasing prison demand forecasts. The NAO reported that this is not expected to be delivered until 2031, five years later than planned, and at an increased cost of £4.2 billion. In response to extreme pressure on available prison spaces, the Government have enacted emergency measures, taking a reactive and costly approach to try to urgently increase capacity. Today, we will be questioning witnesses about how they are managing the increased demand for prison places, the impact of the crisis on prisoners, staff and the public, and their long-term strategy for delivering a prison estate that is fit for the future. To help us with all of that, we extend a very warm welcome to our witnesses, firstly to Dame Antonia Romeo DCP, Permanent Secretary and Clerk of the Crown in Chancery. I am not quite sure what that is, but perhaps you will tell us in a minute, Dame Antonia. You were appointed accounting officer of the Ministry of Justice in 2021. Ross Gribbin is director general of prisoners, offenders and international justice. Ross was permanently appointed as director general in the Ministry of Justice policy group in April 2023. It is his first appearance before the PAC, so a special warm welcome to you. From HM Prison and Probation Service, we have Amy Rees CB, who was appointed director general chief executive of HMPPS in August 2022. Last but by no means least, we have Jim Barton, executive director HMPPS change. He is responsible for the prison expansion portfolio, the electronic monitoring service, and overall oversight of the HMPPS change portfolio. Dame Antonia, since I raised the question, perhaps you would tell us exactly what the Clerk of the Crown in Chancery is.

Traditionally, the Permanent Secretary at the Ministry of Justice is also the Clerk of the Crown in Chancery, which is an office that oversees the Crown Office. In this role, I am the King’s sign-manual for letters patent. I also take the formal record of any coronations. Thank you for asking.

We have learned something this afternoon—or at least I have, anyway.

Good afternoon. Thank you for joining us. The reality is that you have known about the lack of capacity in the prison estate for many years. Dare I say you have been good at forecasting the fact that there would be a future shortfall, but is the reality not that you have presided over, effectively, a ticking timebomb, where demand would eventually outstrip supply at some point?

You are right to say that it was known for some time that the lines would be crossing, as we call it, which is where demand exceeds supply. As you will have seen from our annual capacity statement that we published last December, we also see the lines, absent significant intervention, crossing again. The prison service has been operating at near to full capacity for many years. This is a good opportunity to thank the excellent leadership and also the staff for the fact that they have kept the system going in that period. Successive Governments have used, essentially, two methods to address this: the supply side, which is the prison build programme that I am sure we will be going on to, so that could be newbuilds, house blocks, rapid deployment cells, etc; and the demand programme, which is about policy measures, of which the previous Government and this Government have done a number, and I am sure that we will go into that too. A huge amount has been done in order to keep the system sustainable and keep the whole of the criminal justice system going.

We will get into the details of the supply and demand question. Just to end my little section, if I can draw your attention to paragraph 3.13 in the NAO Report, it states that, effectively, back in 2021, the projection that demand would outstrip supply was a known quantum. It was a known reality that the spending review in 2021 would leave a capacity gap of nearly 1,000 places. How did that come to fruition? Surely that was some shortcoming in the settlement process.

We have been bidding for money to build the prison places required to deliver that. That has happened over a period. As you will know, until about 2020, the forecast was roughly flat for prison projections, but then it started to go up significantly, partly to do with new police officers that the previous Government were planning to recruit and partly to do with the results of Covid. There were other things that exacerbated this, including more and longer sentences, and people spending longer on sentence in the population. As a result, we bid for more money to have a big prisons programme. That is the 20,000 programme that we are here to talk about. It was known that we were going to be under a lot of pressure. We have been negotiating funding settlements with the Treasury in successive spending rounds to try to have the money both to build new places and importantly—I am sure that we will get on to this—to maintain those places that we have in order to reduce dilapidation and the number of cells that need to be taken offline. It has been known about and we have done a lot to address it.

Just to close, was it, therefore, that you were not so good as a Department at bidding for money, or was it a question of Ministers at the time not being forthcoming with the funding settlement that was needed to stop this shortfall from arising?

You will not find many Perm Secs who will say that there is a correlation between being good at bidding and the money that you then end up being given.

It was the latter, so Ministers did not give you enough money to tackle the crisis.

We were given money for the 10,000 programme in SR19 and then for the 20,000. A number of things have meant that the costs of these programmes have gone up hugely. It is a combination of a range of things. It is not about whether we could have been given more money at the outset. I am sure that we will get on to this, but delays in planning have had a massive role in this and have led to significant increases in the costs of the supply programme. There are a number of unexpected things as well that have happened, so it is not as straightforward as saying, “If we had been given more money, we would not have found ourselves in the position that we have been in”, although money has sometimes been a limiting factor.

Dame Antonia, just to clarify, the Lord Chancellor, in her 10-year forward strategy announced in December, codified it for the first time—or at least it is the first time I have ever seen it—by saying, “We still project the prison population to increase by an average of 3,000 annually over the coming years”. Are all the projections that you are going to tell us about today taking that into account?

Yes. This was the first of its time. As the Lord Chancellor says in her foreword, this is a crucial moment, because it is one of the ways to ensure that there is full transparency about both the prison build projections and what we expect in terms of demand. The demand projections have been pretty accurate over the time. Generally speaking, the central projection has been where we have ended up, but, of course, there is always, inherently, some uncertainty around forecasting. It is fair to say that this annual publication is going to be essential, because it is going to allow us to see, on an annual basis, where we are on demand and supply, and to forecast when the lines might cross again.

Does that figure take into account demand-side measures, for example what might come out of the sentencing review or the reducing of remand from 50% to 40% of sentence served?

In line with all Government-published figures, we have not included changes due to policies that have not yet been made or legislated for, so what may or may not come out of the sentencing review has not been taken into account. In fact, we will need the Government’s response to the sentencing review to help us manage the gap that the document that you are referring to indicates. It includes measures already taken, for example what we call SDS40, which is the reduction of the SDS point from 50% to 40%.

Thank you for that helpful remark. Before I call Clive Betts, I should just extend a warm welcome to Andy Slaughter, who is Chair of the Justice Committee. You are very welcome to guest on this Committee today, Andy.

Thank you very much.

In 2021, you committed to build 20,000 new prison places by the mid-2020s. Do you regret having made that commitment?

No. The Government would not regret making an ambitious commitment to delivering something that was required. The key point is about all the things that happened since then, due mostly to planning permission, but also to other things, which meant that there was a degree of slippage in the delivery of the 20,000 and that they were not delivered according to the original timetable set. The Government would not regret making a commitment.

A degree of slippage might be a polite way of addressing a major failure, would it not?

Garth has been 40 months in planning. There is no way that that could have been predicted. Is it a major failure that a prison has spent that long in planning? I would say that it is just part of the process of building major infrastructure programmes.

How many new prisons had planning permission when you made the commitment?

The team will correct me if I am wrong, but I think three had it and three did not. Of the three that did not, two now do and one is subject to a further challenge.

Would it not have been reasonable to assume at the beginning that planning might be an issue that delayed some of those projects?

Yes.

So that was the mistake.

What was the mistake?

Not making assumptions about delays in planning, which were probably fairly foreseeable.

Jim will be able to give more detail on this, but, essentially, the story is this. The original timetable was set out during a period of something called Project Speed. It was a Treasury and Cabinet Office-led initiative to, essentially, streamline and massively speed up planning approvals. All infrastructure projects at the time were working on the basis that Project Speed would deliver. This is referred to in the document. That was not, in fact, implemented, so as a result there were delays there. Aside from that, there is the degree of delay that we have seen in the planning decisions. Even the fact of having PPA—I cannot remember what that stands for, but the early planning approvals that we were given in two of those three prisons—did not then lead to planning approval. We found ourselves in a far more delayed situation than can reasonably have been anticipated, albeit I agree that, in terms of the original planning under Project Speed on the basis of a speeding up of planning processes that did not materialise, it would, in retrospect, have been better not to have planned for Project Speed to have delivered what it said it would deliver.

So it was the Treasury’s fault, then.

Absolutely not. I am pointing out that it was the Government’s policy position that the project should plan according to the principles of Project Speed.

Who was in charge of Project Speed and making it work?

It was a Government-wide initiative.

Somebody has to be responsible in a Government-wide decision.

It was Treasury and Cabinet Office-led.

So it was Treasury and Cabinet Office who were responsible.

It is not helpful to blame. I am stating the facts.

Trying to find out who is responsible is perfectly reasonable, surely.

I am just stating the facts. We set the project plans according to Project Speed principles. Those Project Speed principles were not, in the end, delivered by the Government at the time. Therefore, we have re-baselined the programme, which, among other things, has been a reason for the delay.

Before Mr Betts goes on to his next question, it is not just planning permission, is it, that delays prison building? It is the contract itself. What assessments have you made of that? You now have planning permission on three out of four of the new prisons. From the moment of having got planning permission, how long will it take to get a brand-new prison into operation?

I am going to hand over to Jim, but I would say that it is about three years.

Of the three further prisons that we are building at Garth, Gartree and Grendon, one will open, essentially, in each of the years 2029, 2030 and very early 2031. We are in the ground and preparing the sites at one of those three sites already, in anticipation of getting a positive planning decision. That was subject to very careful consideration as to whether that was money well spent in anticipation of receiving that outcome. I might just add, though, that, within the 20,000 portfolio, the area where we have had greatest success, when we have got through the planning process, has been in new prison builds. Both Fosse Way and Five Wells, which are now open and full, were delivered to time. It was critical that they were delivered to time, given the acute pressures that we faced in prison capacity over the course of the last couple of years. I am not in any way suggesting that this is the antidote to some of the criticisms set out in the NAO Report, but Fosse Way has recently won programme of the year across the civil service. That is a testament to the fact that we know how to build prisons when we get through factors that are outwith our control.

Well done. I have just one supplementary to that question. Do you have what the health service is striving for, which is what I call a model prison design, so that you can replicate from one to the other?

Yes.

Is that one of the reasons why you have been able to build it on time and on budget?

Yes. You could go across those six prisons, once built, and you might struggle to spot the differences between them. There are differences; we are improving it every time, but the essential design is the same.

In fact, we were all on a visit to Millsike, which is a prison that is about to open in spring of this year. They were saying that, at this point in the previous two projects, they had many more snaggings than they have now. That is because we are replicating a design and learning every time. There are some site-specific issues, but it has made us much slicker.

So you have ironed out problems such as line of sight, ancillary buildings and all that sort of thing in a standard design.

Yes, exactly.

Well done.

In terms of the problems with planning, you are saying that there were no problems with these newbuild prisons that you have just mentioned.

Of the three prisons that we have built, two are open and one is about to open. Five Wells and Fosse Way are on sites where we previously had prisons and knocked them down to replace them with larger prisons. Millsike is adjacent to an existing prison. We got through that process much more easily than we experienced with the other three, which partly goes to the point that the Permanent Secretary has made that it was reasonable to assume, as we did, that getting through planning in 26 weeks, which was the planning assumption that we baked in, was possible. The view across Government at that time was that the best-case version of that was 13 weeks. That felt very speedy, no pun intended, so we included 26 weeks. For two of those three prisons, the local authority planning official recommended that planning be approved at the first event. If that had happened, we would have secured planning well within that 26-week window.

They are on sites that already had a prison.

Sorry, I was moving on to talk about the further three. Apologies for not making that clear. Of the further three, we assumed that 26 weeks in our plan would be sufficient to secure full planning.

Were these sites that were earmarked in the local plan for such development?

Not at the outset, no, but we had extensive engagement with the local planning officials. As I mentioned, prior to the planning committee meeting, the recommendation on two out of the three was that the application was approved.

Without knowing all of the details, it sounds to me extremely optimistic to think that you can suddenly announce that a prison is going to be built on a site, with no reference to it in the local plan, and that it is going to go through just like that.

They are all adjacent to existing prisons.

Nevertheless, it is a newly identified site. It just seems to me very optimistic, which it proved to be at the end of the day, did it not?

You can take different reference points. If the challenge back is that we should credibly have included in our plans an assumption that it would take three and a half years to go through a planning process to secure permission to build a prison, which the local planning official supported, adjacent to an existing prison, that does not feel like a credible basis upon which any project, Department or Government could proceed.

It is also worth adding that we go for planning only where we think that it is reasonable and proper to build a prison, and that it can meet with all the local planning requirements, etc. Those three were whittled down from a long list of places that we could have built and were the ones that we thought it best to go for.

With your assurances about the future and your new commitment to build the number of places required by 2031, all the potential problems have been built into those forecasts, have they? Are they absolutely certain now that you can deliver on that?

For the one where we are still subject to an appeal, one of many things that we have learned is that we would be insane to suggest that that is definitely not going to have any problems now. It is still subject to an appeal.

So the commitment to 2031 is still not certain, then.

There are a huge number of factors that go towards building a project that costs many hundreds of millions of pounds over six to seven years. I would not, of course, feel confident sitting here today and saying, “Absolutely everything is within our control and that can be achieved”, but we make allowance for it within our planning. To give you an indication of that, we use a tool called reference class forecasting to try to learn from other projects across Government in different jurisdictions and to understand how likely it is that things will go as we expect them to do. We then weight our plans in terms of both time and money to take account of that, so we build more contingency into the timeline, where it is more risky. As an example, we are planning to extend a category D prison at HMP Ford. We assess that as being P43 on this index of 0 to 100. That means that we have added 50% contingency into the time schedule, so we take account of that risk in the way that we plan.

On your particular point on planning, it is worth adding that 95% of those remaining 14,000 places have outline planning, but that would include Garth, for example, which now has a judicial review. Sir Geoffrey, you alluded to things beyond planning. We have had two firms that have gone bust and into administration. These are the sorts of things that it is almost impossible to forecast at that time.

We will come on to ISG and come back to planning towards the end of the hearing, but thank you for that for now.

It is fair to say that costs have increased a lot. According to the Report, costs have increased between 80% and 93%. When we look at figure 4, we can see that the biggest cost increase is in the rapid deployment cells. Mr Barton, the specific cost increase here is between 247% and 259%. I think I am right in saying that the rapid deployment cells have not been deployed very rapidly. Could you tell us why it is that the costs have increased so much?

Yes, absolutely. As of today, we have opened about 770 rapid deployment cells across the estate that are currently being utilised, so just over half the total that we are expecting to deliver in full. There are, essentially, three factors that have contributed towards that cost increase. The first is that we have developed the design from the point of that original quote. We have worked with our contractors to ensure that we have facilities that best meet the needs of prisoners and of the local establishment. That has added costs through the process. These are not portacabins that we buy from a generic supplier. They are tailormade to our needs.

So you think that the original spec turned out not to be appropriate for what you needed.

As we do across the 20,000, we have continued to revise and review that spec as we move closer to production. These are just the very early cost estimates. As you get closer to having finalised plans, you have an understanding of the full cost of that design. Of the three factors that I referenced, that is the smallest in terms of its component impact. The second has been site-specific costs. For a lot of these projects, we are inserting RDCs into the corners of existing prisons, many of which were built over 60 or even over 100 years ago. They are incredibly complex sites. Just as with some of the other 20,000 projects, as you get into the ground, we have uncovered challenges that have added to cost, such as asbestos on sites and issues around ground conditions not being as stable as we had anticipated and, therefore, requiring more underpinnings. That has added significant cost. At the more expensive end of the RDC projects, the accommodation itself makes up something like 25% to 30% of the cost. The remainder is going on everything else required to get them installed and make them operational. The third factor takes us back to the conversation earlier around delays. One of the other issues that we have experienced in projects across the 20,000, but which has impacted several RDC sites specifically, is delay linked to nutrient neutrality. These are regulations that require us to be able to offset the impact of additional grey water running into local waterways. The challenge that we have in some of our projects is that there are no schemes established for us to be able to offset the impact of those projects by buying nutrient neutrality credits. That has led, in the worst case, to delays of close to a year at some of those RDC projects, because we simply cannot lawfully proceed to full construction until we have been able to remedy those nutrient neutrality issues. We have been working with colleagues across Government on that and have lots of work under way as to what the right policy solutions to that might be. Right here, right now, we have to work within that regulatory framework.

Which of those three issues that you identified has been most relevant to the costs going up on the RDCs?

Site conditions.

So the challenges around asbestos and ground conditions.

Yes. If I may, it is a factor across the 20,000, in that we started, as any sensible person would, with the more straightforward sites. The sites that were quicker into approval and construction were where we had more space, fewer operational constraints, and more confidence in the conditions in which we were building. As we have moved through the programme and added further projects to get us towards that very necessary target of 20,000, we have taken on projects that were further down our priority list because we were aware that there may be challenges with them. We are working through those challenges, but they contribute towards that increase in cost.

Dame Antonia, the Report highlights that there is now a very wide spread of cost per prison place. In some parts of the programme, the cost per place is only £200,000, but, in a small number of individual projects, it might cost up to £1 million. What is the maximum amount per place that the MOJ should be prepared to pay for?

As Jim says, the way that the process works is that you always go for the highest value for money first. Also, as you get to a point where you still have to have more places being brought on stream, you have to go, inevitably, for more expensive places. In some cases, there is also a sunk cost element.

I can understand, on that logic, there being a range, but there must surely be an upper edge at which you say, “This is now too expensive and we are not prepared to pursue this option”.

It is really quite complex, because, in a lot of those cases, that will not have been the number. All business cases will be signed off by Amy, as AO, and then by me as principal accounting officer. There will have been a value for money test on all of those. In a situation such as RDCs, which were very ambitious and completely new at the time that we were first designing them, it was something new that could be brought on quickly. It was a totally different type of accommodation that we thought could be put in the ground more quickly. Therefore, by the time that we were on tranche 4 or 5, it did not seem that some of those costs would be worth incurring. There was, in fact, a tranche of RDCs that we, as AO and as principal accounting officer, did not agree to because of the cost. It is not necessarily about cost per place, because they can end up being quite expensive when you have already sunk quite a lot of that cost. There is no cutoff where one says, “More than this is too expensive”. It all depends on proximity to full capacity requirement, how quickly they can be brought on stream, the cost that is going to be paid and the reason for that cost.

If you look at the Lord Chancellor’s table on page 8 of her 10-year strategy, it does look as though these rapid deployment cells are a relatively small amount of the total capacity that you are going to bring on stream. Are they really the last resort rather than the first resort? You use them only when you need to keep up your delivery, if you like, of new places. In normal terms, because they are expensive and have a life of only 10 years, you would not normally use them.

I will say some things, and Jim and Amy might want to comment as well. As originally envisaged, we were going to bring on 1,000 RDCs when they were first identified. In an ideal world, if you are going to build your big prison builds, you are going to build really good, new, green, smart prisons, but there is a limit to how many of those you can do and you have to operate the estate that you already have. Small, secure house blocks are a good example. There are lots of different types of accommodation. RDCs were envisaged to be a new type of accommodation that would be quicker to bring in. They are not suitable for all types of prisoners. They are only for cat C or D prisoners and they do not last as long. They last for something like 15 years, so they are not as good in that regard because, by definition, if they are going to last for only 15 years, they are not going to be as good value for money, but they can be brought in more quickly and can be more flexible. The other thing to say is that, at the time that they were initially being designed and thought of, the fact of bringing in a new type of accommodation was worth something in itself, because it gave us much more flexibility. As you can imagine, we always try to think about how we can be more innovative in how we provide accommodation in ways that could, over time, deliver much higher value for money, and they represented as that. As I say, by the time we got on to what we thought was going to be the fifth tranche, we had, in fact, decided that, because of the site conditions, they were not going to be better value for money. In principle, they had a lot to offer and, in fact, as Jim said, we have brought on 770 and are going to bring on another 1,000, so they are going to offer quite a lot in terms of capacity.

Straightforwardly, just to answer your question operationally, no, they are not a last resort, because they have brought something really additional and helpful to the prison estate. They are really high-quality accommodation. It is some of our best accommodation out there in the estate. As the Permanent Secretary says, they are only for risk-assessed cat C and D prisoners. They offer a really good step, if you like, from coming out of normal prisons, with very big landings and all the things that happen, into living more independently as a step towards being released or perhaps going to the open estate. When we are trying to transition people into resettlement, there are only certain places in the country that you can do those things. People need to be able to access jobs, etc, which really means that, operationally, we need to be able to use quite small parcels of land in existing sites, which is where this has offered something genuinely new and helpful.

Can I just check one thing? We have not got into a situation now, because they are so specialist, that we have only one supplier. There is a bit of a competition in the market to keep the price down, is there?

We contract with two suppliers.

Have you thought about asking any of the Scandinavian countries, which produce a lot of these things, to compete in this market?

As with all of our procurements, we have gone out to open market and taken in any bid and tender with a company that has an interest. The two suppliers that we went with were the best providers on price and quality at that point in time. Going back to the earlier question, the cost of the unit itself is not really the factor that is driving some of the points of observation and concern within the NAO Report.

Sir Geoffrey, you are right to identify that the tightness of the supply market is one of the issues, to Ms Olney’s point, not just in RDCs, but in costs going up generally. That is one of the big issues that we face.

Mr Gribbin, one of the issues identified by the report was the fact that, because places were needed so urgently, that, in itself, contributed to costs increasing, for example through needing to pay a larger margin to contractors to accept higher levels of risk. Do you have a sense of what additional costs you have incurred because the need for places became so urgent?

We give transparent estimates of costs. There is quite a range of reasons for why the places were needed. As your colleague said, we predicted this at the start of the last Parliament. There were then some years where the plans were in place. Could you give us a bit more as to what exactly you are looking for here?

Just the extent to which the need to provide places urgently was a standalone reason for driving costs upwards. If projects had not been delayed and, therefore, the need for the places was not so urgent, to what extent has that had an impact on costs?

I now understand the question. We have not disaggregated just the costs for it being required urgently, because, as you have heard from my colleagues, there are a number of reasons why these places were required. As has just been observed, the different categories of accommodation all have different costs, so I am not sure whether it would be possible to give you that disaggregation that you are looking for.

Do you accept that the increasing urgency of the need for places has had an impact on the increase in cost?

It is certainly the case that, if it is necessary to do something quickly, it will likely cost more than if you had a long time to sort it. On the other hand, some of the inflation cost increases that have taken place over time have added to cost as well, so it is not quite as straightforward as just saying that things done quickly are more costly than things done over a long period.

I totally agree with that. Delay leads to cost, not least because the inflation that we faced in the construction market has been so significant. Over time, costs go up significantly, but it is also the case that the more swiftly you need to bring something online, the more often it can lead to increased cost in itself. As we have demonstrated with the RDCs, you start with the ones that are the best value for money, or the easiest, which often end up being the best value for money, because of ground conditions. Over time, you might expect, as you have to bring more and more new places online, costs to go up, but it is complicated.

Where we can accelerate delivery, we will. To give an example, a new house block at Rye Hill is due to open in the next few weeks. We have opted to pay slightly more for the delivery of that project, months earlier than it would otherwise have come online, through weekend working and longer shifts. That was subject to a very detailed assessment around, “Is the value of having those places for the weeks’ or months’ acceleration that we get worth it, rather than just money being thrown at contractors?” We look actively at where we can accelerate delivery through those sorts of measures.

Ms Rees, what will you do when the temporary accommodation that is currently being used to house prisoners reaches the end of its intended life?

What we will do is what we do with the rest of the estate. We will assess the condition and whether it can reasonably and safely go beyond 15 years. When it cannot, it will come out of dilapidations. Our previous version of RDCs was temporary modular units. We had to take some of those out in the last five years because they reached the end of their serviceable life. We will always assess it for condition of assets and then decide whether it is reasonable to keep it in play. In the end, that is a straightforward value for money judgment.

Ms Rees, I hope that we do not get into the situation that the Department for Education has got itself into with Elliott buildings, which were designed for 25 years. I still have some in my constituency after 40 years.

It is certainly true to say that the modular units that I just referred to went for a lot longer than their original lifespan. When they came to the end of being somewhere I could safely keep prisoners in and our staff could work, we took them out of use. That led to a reasonable spike in our dilapidations over the last five years, because we had to take these modular units out of use.

Finally, Mr Barton, ISG, which was one of your main contractors, went into administration last September. To what extent is that going to delay delivery of some of your projects?

It will, but those delays are reflected in the forecast that we published in December, to the best of our ability to forecast it. We had a significant exposure to ISG. It was contracted across 13 projects within the 20,000 portfolio, adding up to about 3,500 places in which it had some involvement. The projects have been delayed. To give you an example from Birmingham, where I live, the Victorian house blocks at HMP Birmingham are being refurbed. They are stripped back to bare brick at the moment. We had a tower crane in the middle of the prison. We had to close that bit of the prison overnight when ISG went into administration to make it safe. That is now done. We are now in the process, with commercial colleagues, of looking to retender those contracts across those 13 projects. We are doing that in as speedy a way as we can, while keeping a clear eye on value for money. We expect that it will have a significant impact in terms of both cost—our estimate is circa £300 million, which is in the figures in the NAO Report—and delays. The delays are quite site-specific, because it depends upon the stage of completion that they are at, but some projects could be delayed by up to a year. Again, I would just emphasise that our assumptions around those delays are built into the forecasts that we published before this year, but it is fair to say that ISG going into administration has been a significant issue and challenge for 20,000. It may be something that we should come back to later in the session, but it has also had a significant impact on fire safety maintenance.

Our exposure was more on the maintenance side than it was on the 20,000.

Can I just clarify that, Mr Barton? Paragraph 3.8 of the NAO Report says that there are places at risk of enforcement action by the Crown Premises Fire Safety Inspectorate. Was ISG involved in that contract? Can we be assured that bringing cells up to fire safety standards will not be put at risk because ISG has gone out of business?

I will give you some straightforward answers. Yes, it was involved. Yes, that will put at risk the completion by 2027, but it will not change the outcome, because, if those places are not remediated by the end of 2027, we will take them out of action.

Without ISG, is there capacity in the market that you are contracting from to deliver the remaining project, or is that going to be a constraint?

We believe that the capacity is there around 20,000. Amy may want to add something around fire safety and maintenance. It is fair to say that that market has its challenges, which are not at all specific to building within the context of prisons, but are quite well reported on within the trade press.

There are two issues. Maintenance capacity will be an issue in terms of capacity in the market and how we do that. The other issue is that I have to carefully sequence how I take places out of use. Now, of course, that has been messed up, for want of a better word, and so I will have to figure out how we re-sequence where we take places out of use in order to maintain our headroom, make full use of the estate and manage the different geographic parts of the estate. That means that we may not have completed the remediation by the end of 2027, but our commitment is that we will take them out of use until such time as they can be remediated.

The collapse of ISG is going to impact the provision of prison places quite significantly, but it is not primarily due to not being able to build the new ones; it is about maintaining the older ones and the impact that that is going to have on the existing estate.

It is a mixture of both. We are exposed on both, but there are more cells on the maintenance side than there are on the newbuild side.

To pick up on what we were just discussing, I am going to refer to paragraph 1.25 in the NAO Report, which says, “Since the Government’s 2010 prisons strategy, HMPPS has indefinitely closed 4,151 cells due to dilapidation. This is more than the combined capacity of the two newest prisons in the estate (Five Wells and Fosse Way)”. My question is to Dame Antonia and Ross Gribbin, if I may. How do you consider the trade-offs of investing in prison maintenance versus prison expansion?

I will say something, and then Ross will want to add. The first thing to say is that there is no perfect algorithm for the marginal benefit of keeping a cell online versus building a new one, but, generally speaking, it is better to do maintenance. Amy and Jim will correct me operationally if this is not right, but, once you have the place, if you can maintain it and stop it going out of use, it is much more expensive, as per Ms Olney’s questions, to have to build a new prison, a new house block or a new RDC. You cannot plan for always being able to maintain your cells. As Amy has said, she has to take them out of use in order to maintain them, so we have to do both. So we have a big programme of building and, alongside that, we have a big programme of maintenance, so we do both things.

Antonia has already said that we do both. Of course, some of those decisions were taken before the projections that you are now looking at were in place, and some of the reasons that the population is now growing are new decisions that have been taken since, so we should be careful about judging those things in hindsight. In addition, it is not always the case that all places can be maintained safely, and Amy and Jim might want to say a bit more on that. Where places can be maintained, the Department does it. Where they cannot, we build more. In both cases, we are looking ahead to see how many places we need.

What assurance can you give us that you have that balance right between maintenance and expansion?

As you will know, the investment into maintenance is £220 million this year, and £300 million next year. Meanwhile, we have £2.3 billion over those two years for the newbuild programme. We are running at full pelt to do both, because the lines crossing means that we have to do the maximum amount that we can. It is not so much about a balance. We are never defunding one thing to do the other. We are making judgments. Amy has a lot of very particular operational red lines, such as statutory responsibilities on fire safety, that she has to worry about when she is thinking about cells being kept online and maintaining them, etc. Meanwhile, we have a big prison build programme, which has to deliver over 20,000 new places, with 14,000 still to come. We are doing both. We have a detailed programme plan for both. We are not really seeking to balance in that sense, unless I have misunderstood you.

No, you have not misunderstood. Thank you.

Just operationally, straightforwardly, if you are asking, “Would I like a lot more money on maintenance and would I like to do more maintenance?” the answer is a resounding and absolute yes, but money is not the only constraint on me managing maintenance in an existing estate. There are two others. One is the capacity of the market to do much more maintenance, because, on the whole, coming into a working prison to do maintenance is quite a longwinded job. It is quite a lot of exposure. If you are an organisation, you have to have security clearances for everyone working there, etc. The other is that I have to have the headroom in the estate to take places out of use to maintain them. There are quite a lot of constraints to how much maintenance we can do. The Permanent Secretary alluded to my red lines, one of which was that we would keep a minimum number of 1,500 places offline to keep my essential maintenance going. To go beyond essential maintenance and into nice-to-do things such as flooring, I would need to take more places offline, which has proved very difficult over the last few years.

Nevertheless, Ms Rees, the Permanent Secretary is right that it is much cheaper to maintain than build new places. Given that the NAO says, in paragraph 1.24, that a backlog between £900,000 million and £1.8 billion has built up, it would seem sensible to surcharge this maintenance somewhat. I would ask you, therefore, Permanent Secretary, whether £200 million a year on maintenance is enough, given that scale of backlog.

It is not enough. As HMPPS has revealed, we need £2.8 billion to totally remediate. In the current SR, we were given, specifically, £220 million and £300 million. We can all do the maths. We are going to need a lot more money in the next spending round to begin to make inroads into the maintenance.

Just to be clear, not to repeat myself, I would need more headroom and more market capacity to go much beyond £300 million.

That is absolutely understood. Can we take it, Permanent Secretary, without wishing to divulge any confidences, that you will be making strong cases in the spending round for more money for maintenance?

I look forward to the PAC report suggesting that more money should be found for maintenance in prisons.

It is a two-way thing. We can help each other.

Just following on from that and going back to the newbuild situation, are you confident that you are going to have the resources and the finance to build the number of places that you are committed to build by 2031?

First of all, you cannot prejudge upcoming spending rounds. At some point, it all depends on the money that I am given in the spending rounds. We work very closely with the Treasury on the plans. It is a crucial part, as is the IPA, of our programme planning teams and of all levels of assurance, and it signs off all the business cases. As Amy said, over 90% now have planning and the work has started, so I am as confident as I would want to be.

We have explored this in questions before to an extent. You think that the money is going to be confirmed at the spending review. With the money available, are you going to deliver the places, for certain, by 2031?

At the risk of repeating the comments that I made before, we are as certain as we can be. We make allowance for the fact that there are inherent uncertainties in building complex things such as prisons, particularly where we are inserting capacity into existing prison sites. That is where it is more challenging, not the new prisons. We allow for that in our planning assumptions in the way that I described before.

To Sir Geoffrey’s point earlier, the thing that we have all definitely learned on this side of the table is that, the more you do these things, the more you replicate the design and the more we have a very clear route to it, the slicker and better the chances that we will deliver it on time. Stop/start is the worst thing that we can do for the market, for the design and for how sure we are that we are going to deliver it.

Have you made any changes yourselves in terms of processes, procedures or resources that are needed to deliver? Have you learned lessons from the previous period when delivery was not achieved?

There are quite a few things that we have done, which I can speak about and others will have thoughts on. The first is that we have done a number of internal reviews in order to learn the lessons of what has happened on the supply side. Some of the reviews are referred to in the Report. In particular, in 2022, we had a peer review by Network Rail. We also had a review by Stanhope. I asked the IPA to do a review particularly into RDCs and our delivery of them. We already had, I would say, reasonably good governance in some of these areas, following a governance review that I had done in 2021 of our delivery of major programmes, but we significantly strengthened both the governance for the programmes and the assessment of delivery following those reviews. I can say a bit more about that, if that is helpful. One of the things is that, every month, the investment committee looks at prison supply and the investments that we are making to ensure value for money. We have a portfolio committee and a delivery board, which are overseeing this. That is on top of the regular programme boards, Amy’s audit and risk committee in HMPPS, and my audit and risk committee in MoJ. The governance to flag issues is significantly improved, which gives me some confidence. The 10-year prison capacity strategy and the annual statement of capacity are things that now set out strategically what we intend to do and how we are going to deliver those places. Just the transparency of having published now what the plan is, I think, is another example. Finally, to talk about capability, we have a large multidisciplinary team. There are something like 100 FTE equivalent of commercial and property staff from MoJ supporting the programme team. As Jim said, the Fosse Way programme team recently won the civil service award, to give a quick shout-out to them. We also have about 340 professional services resources supporting on things like technical assurance, risk, contracts, and health and safety. We have a lot of capability in this. We hope to get the money secured. We have been working with the Treasury on that and, in the next SR, we think that we will. We have set out our published plan. We have given ourselves as strong a chance as we could, albeit, in all of these big infrastructure projects, there is always the risk of something uncertain happening, asbestos, radon or whatever it may be, but we are reasonably confident.

There has now been a change in the NPPF to recognise the importance of sites for major infrastructure. Is that going to be beneficial? Have you looked at what it might mean for your projects going forward?

We are going to do a slightly different thing. Jim will be able to explain this, but that is not, in itself, what we are relying on. We are relying on a different Crown development route, which will be faster for us.

That is exactly right. We have had really good support from colleagues in MHCLG around all of those intended reforms to the planning regime. We considered in detail whether we should bid to be included in the changes that were announced last week around nationally critical infrastructure. We decided that, on balance, that was not the right way to try to accelerate future planning permissions for new prisons, on the basis that the way that that is designed is more around things such as the Thames tunnel, where there is an extensive public consultation period, which just does not apply in prisons and we would not want it to. Instead, we are relying on the planned changes to Crown development routes that will reduce and, potentially, remove the risk that we get stuck in such a prolonged cycle of local planning decisions being remitted to the centre and going back out to the local authority. Utilisation of the proposed changes to Crown development routes, which I understand are due to come into force over the next couple of months, is the thing that we see de-risking us ending up in a future Garth, Grendon or Gartree when we are bidding for the next planning.

That bypasses local planning committees.

As I understand it, it expedites the process to bring that to central decision making. It is Government policy that is being worked through.

Does it bypass local planning committees?

It removes some of the powers that currently sit with local planning committees.

Local planning committees no longer have any say in whether a prison is built.

To be clear, we are not responsible for planning policy. We are just taking advantage of the route.

I am just asking for a factual description of what the situation is.

You are right that, in the main, it brings that decision into central Government.

Can I perhaps wade in here as a chartered surveyor who knows something about this? Part of the problem with procuring asylum home sites was that the local communities were not consulted. If you do go for this Crown procedure, which, as Mr Betts says, does bypass local planning, it is still important that local communities are involved in the decision making.

Yes, absolutely, and one thing that we have learned off the back of the experience at Garth, Gartree and Grendon is that, while we did put effort into engaging local communities, we should have done more. It is not in our interests to build a prison in a community that is opposed to that prison being built. A challenge that I am not sure we have properly worked our way through yet is that there is incredible benefit to communities of having prisons on their doorstep in the sense of the jobs that are created, the stability of those jobs, and the money that goes into the local community. Over £100 million of the moneys that we spent at Millsike have been spent on local companies within 25 miles or thereabouts of the sites, so there are huge benefits. I absolutely understand why some individuals within communities might have questions about what it means to have a prison on their doorstep.

I concur with those remarks. My father’s farm had a prison converted from an old RAF site in the 1960s, and it has been very beneficial for the local economy.

How many prison places have been delayed due to nutrient neutrality rules?

I do not have that number, but, off the top of my head, we have at least half a dozen projects that have had significant delays due to nutrient neutrality.

Perhaps you can write to the Committee.

Yes, absolutely. I will try to do some maths while we get through the remainder as well.

You are suggesting that hundreds of prison places have been delayed due to nutrient neutrality rules.

Yes, 100%.

This is a very intractable problem if you are somewhere like the Wye Valley, where nutrients are a really important problem. If you could let us have a note on this, I would be very grateful.

It is definitely hundreds. I am just pausing on whether it is a thousand or more.

I would like to move us on to look at some of the effects of the prison capacity crisis. In particular, I would like to start by considering the impact on prisoners and prisoner outcomes. The Report lays bare—and we have also received quite a lot of evidence on this matter—that, in particular, overcrowding results in people often being confined for longer periods to their cells, restricting purposeful activity, which, in itself, then leads to higher rates of violence. First of all, Permanent Secretary, from the Ministry’s side, what have you done to understand the impact of this current prison capacity crisis on prisoners and prisoner outcomes?

The first thing to say is that we, as the Ministry and as HMPPS, have been doing absolutely everything possible in our power not to allow the whole criminal justice system to go into gridlock. That has involved a massive amount of work in HMPPS and HMCTS, and, indeed, with our colleagues such as the police, in order to prevent that from happening and to do that work. Nobody wants to run the system anywhere near the level at which it has been run. We have been running the system at over 99% capacity. That is a testament, by the way, to the exceptional people who work in those prisons and, indeed, probation officers and others, who have allowed that to continue. That is not a situation that anybody would have wanted, so may I, first of all, reassure the Committee that we have done absolutely everything that we can both to stop getting into that position and then to ensure that we got out of it as soon as we could? You have already identified that there are increased workload pressures on OMU staff, but also prisoners being held overnight in police cells puts pressures on the police, as well as on prison officers. It also affects PECS and makes moving those prisoners around much more complicated and costly. It is not good for the prisoners. Although Amy and her colleagues do everything possible not to allow it to affect regime, the fact of a prison operating at capacity inevitably impacts on regime. Amy has to very carefully guard her operational red lines to ensure the safety and security of the prison. The safety and security of staff and prisoners is absolutely paramount. We have been putting that as the No. 1 priority, but that, therefore, affects rehabilitation outcomes. I should also note that it significantly impacts on victims, because schemes such as SDS40, or ECSL under the last Government, are not what victims or we would want.

I will come back to staff and, indeed, as you raise, the public impacts of this capacity crisis. If I may turn to you, Ms Rees, you are running this hot and there is this level of overcrowding. It seems almost that this has become the norm for how you are having to operate. What are you doing to try to manage some of the harms that we see, whether that is self-harm or prisoner-on-prisoner assault, in terms of keeping prisoners safe? We will come on to some of the rehabilitation and purposeful activity stuff in a moment.

There is lots I would like to say about this. The first is just to reiterate what the Perm Sec said: our staff have been exceptional right across HMPPS. That is everyone from a prison officer to an OSG on the gate, to probation officers, to senior leadership. They have been absolutely outstanding in what has been too pressurised for too long. That is the first point. The second is that we introduced red lines, which we have stuck by very rigidly. For the benefit of the Committee, those three red lines are, first, that we would keep out at least 1,500 places to do the required fire maintenance, so we did not go off schedule on that; secondly, that any crowding would be dictated by us operationally in the normal way, which means we cannot just do it the night before, as it takes months to decide where it is safe, and we might come back to how we decide where it is safe to crowd; thirdly, that we would safeguard operational stability. If we could not staff the place to 75%, or, indeed, we were worried about the operational stability, as we were at times in Swaleside and Woodhill, regardless of the position with prison places, we would take those places out of use, as we did do. They are some of the first things that we did to ensure that there were red lines and we understood exactly what we were doing. I might just put some flesh on the bone about what it means to run at this level of efficiency. We have something called an operating margin. That is an analytical exercise of the number of cells in our huge estate that you will not be able to access on any given day. That is because they are in the wrong geography, because we have cat C places, or because they are specialist units where you can only put high-security prisoners, or long-term high security estate. At the start of this, that margin was 1,850. Through the brilliant work of teams right across HMPPS and MoJ, we got that down to 1,190. That means, in a huge estate of our size, 88,000, there are only 1,190 places on average that we are not accessing every day. That represents huge efficiency. What that means when we are running this late is that prisoners arrive too late with us, having been in court for too long. They often arrive in the wrong part of the estate in the wrong part of the country, which affects not just how late they arrive that night, but how early we can get them back to court in the next day, which in turn has a knock-on to things like Crown court backlog, which I know that you have discussed a lot in this Committee. It is just a totally inefficient way to run the estate. We spend so much time just managing what is coming very late at night, and then very early in the morning, trying to get them back out to court for longer journeys, meaning that the time to do resettlement activity and the time to help people with the needs they have is reduced. There is just no getting away from that. In terms of what we do to try to improve safety metrics all of the time, we have a huge variety of things, everything from what we do to try to prevent things coming in—I know that drones have been the subject of a lot of discussion—what we do in terms of airport-style security to stop prisoners, visitors and others bringing things in, how we try to support both staff and prisoners who are struggling, what we have done on drugs and on accommodation. We try to do all of these things, but all of these things were a challenge before trying to run the estate at 99% occupancy.

On the one hand, you are saying that you are highly efficient because you are running at 99% and you have only this small number of spare beds, but you are also admitting that trying to manage at that level with that tiny margin of cells and places means that, effectively, you have to do things that are very inefficient, which impact the whole system and make it inefficient, such as transporting people to the other end of the country and back. They certainly have very poor outcomes on prisoners’ ability to get back into the community on release, have relationships and so on. I just want to push a bit further on what you think would be an efficient operating level, given that that is almost too hot, that would allow you to actually do things?

I never want to be above 95%. Ideally you would get down to 90%. There is no question about that.

What would 90% allow you to do in terms of improved prisoner outcomes?

It would allow more staff time. It would allow more efficient operating in terms of how we move around the country. It would allow us longer to do assessments. It would mean that people could be more efficiently placed to geographical parts in the country. It would just allow a greater efficiency of outcome. If I may, I just want to touch on crowding, because you mentioned that twice as well. In terms of how we assess crowding, that is an operational assessment done by my prison group directors. That is the equivalent of an area manager. It is an operational assessment. Most cells could physically take two people, but we do not just assess that. We assess whether the ancillary services and the staffing, etc, will safely allow us to do that. Crowding levels at the moment are at 23.6%, so that means about a quarter of the estate are living in a cell designed for one person with two people. To be clear to this Committee, that means that someone has to defecate where someone else eats their meal. They were designed for one person. There is an open toilet in that cell.

A quarter of prisoners at the moment are living in those conditions.

Yes, 23.6%.

Is it your policy to move towards single cells everywhere?

It has actually been our policy in new prisons to build places more designed for two—they have a separate bathroom and a separate shower—so that you can get good efficiency from the estate, but it does not have the problem that I just so graphically alluded to.

I hear you on your new estate, but clearly you are doing maintenance, refurbishments and one thing or another. Is it your stated policy that you are trying to move across your estate to single‑cell occupation?

We would always try to eliminate crowding. It is not necessarily through single-cell occupation, because there is quite a lot of evidence that sharing for some people is a very good protective factor for violence, self-harm, etc, but we definitely try to reduce crowding.

Can I just ask a follow-up in terms of the projections and assumptions? What assumptions did they make about the 95%, 90% in terms of occupancy, and what assumptions did they make in terms of crowding and that 25%? When you did your projections about the point at which supply and demand do not meet, have they built into them assumptions that you will continue to run as hot as the current estate, at more than 95%? Do they continue to assume that you will use crowding of a level of 25% of prisoners in unsuitable conditions?

Ross is the world’s leading expert on projections. I am going to let him answer this, but I will just say that the supply line would include the operational margin and the buffer above where Amy says, within that supply, she cannot hold people. That will be built into the supply line. The demand line is actually about what is coming in. It would not be built into the demand projections in that way.

The supply line sets out the number of safely available places, which then goes through the process that Amy has described. That would include the provision of crowding. That does not, of course, mean that it commits us to using said crowding in the future, but that is what the projection does.

What would you consider a safe level of crowding for your projections?

There are two things. My colleague has already talked about one of them. HMPPS will not crowd where it regards it as unsafe. There are very strict operational red lines on that and a process for scrutinising it. Beyond that, there is a policy decision for Ministers about what levels of crowding the Government regard as acceptable in the estate. That has been quite longstanding.

At what level is it currently set?

It is the figures you just heard.

That is acceptable.

Yes. That 23.6% is safe; otherwise I would not sign it off. It is not optimal, though.

It is important to say that these figures go quite a long way back. I do not think anyone is saying that the level of crowding in the estate is desirable, but it is not something that has happened recently. There have been modest changes in the projections, but it is included in the supply line that we are talking about.

I would like to come on just to the impact of this in terms of rehabilitation. Back in 2020, Dame Carol Black produced a report about drug use specifically. We have evidence from both His Majesty’s inspectorate and Camurus Ltd that set out the scale of drug use in the prison population. Random drug testing at the time of Dame Carol’s report was 15% of the prison population. There does seem to be a link between this lack of purposeful activity, the hopelessness that that engenders and the use of drugs. We also received evidence that one in four prisoners are being detained because of an offence relating to their drug use—so it could be theft—and they tend to cycle in and out of prison, rarely getting the long-term recovery or treatment that they need. Could you say what you are doing to address drug use in prisons and ensure that there is access to effective drug treatment to try to reduce, on your demand side, the reoffending rates that are obviously so prevalent and the public harm that that causes?

We have done two main things since that time. One is to focus on continuity of treatment between what they get in prison and then what happens as they go out, because there are actually quite good levels of treatment available in prisons. We work really closely with our NHS partners, but we often find that it is very difficult to keep that continuity of treatment as they go out into the community, so we tried to do a lot. For example, we have a system in Wales, Dyfodol, where it is seamless, so you can access the same information before they were in custody, while they are in custody and actually in the police custody suites, and all the way through. That is something we have really been trying to work on. We have also tried to work on drug-free wings, so incentivised substance-free living units. We have a variety of them. I recently spent a good chunk of time on one in Cardiff with the men who are on there. We think they really are a good, effective way, particularly when they are in small, specialist wings, where people can access the treatment they need, etc. As we talked about earlier, we are also trying to prevent the drugs from getting in. We have had quite considerable investment in what we call airport-style security to try to prevent the drugs from getting in, but I do want to be candid in this Committee that all of those efforts are not made easier by running at this level of population.

In terms of policy, are the drug rehabilitation requirements in terms of the sentencing review being looked at as a cost-effective alternative to these short custodial sentences for people who are known to have drug use and be part of this cycling in and out?

Everything is in scope.

As the Committee will be aware, the scope of the sentencing review is quite broad, so they have been given the remit to look at a whole range of options. We cannot comment on whether they are looking at that specific thing; you would have to ask David Gauke, but the remit is wide.

What progress has been made on implementing the Carol Black recommendations from 2020 in regard to drug use in prisons?

Did you have particular recommendations in mind?

Well, basically making sure that there was drug treatment available. We have not had any evidence that these numbers of 15% are going down.

We have focused on two areas since that report: the continuity of treatment and trying to do the ISFLs. Without getting into loads of detail, random mandatory drug testing is a very small sample of any single prison population, so you can easily have a skewed figure. We have been trying, as well as our MDT, to look at other things we can do, such as waste water testing, which does not give you an individual view of the prisoner but it does tell you about drug use overall. It also tells you about trends in drug use. The change of drug use has been considerable in the last few years. That is where drones have become a real issue, because it used to be that the product itself was highly valuable. Since we have moved to new substances that can be synthetically produced, it means that the product itself is less valuable and they are prepared to take much riskier methods to get it into prison, because if you lose the product it does not have the same impact.

Are we talking about spice?

Yes, spice. Anything synthetically produced is just so much cheaper.

I have one final question on this matter, which comes to education. There seems to be evidence of links between getting education in prison, as part of prisoner rehabilitation, and reoffending, but also employment on release. We received quite a lot of evidence that those educational facilities were in a state of decrepitude, if that is appropriate language to use, and very physically substandard. What are you doing to try to secure the physical facilities as well as releasing people to give them time to go to education? What proportion of your prisoners would be able to access education at the moment?

There are a few things. First of all, for education or anything else in purposeful activity, you need to run a really good regime, which means some of the stuff we have talked about. You have to be able to unlock everyone on time and prisoners have to be in the appropriate establishment. You then have to have good staffing levels. I am really pleased to say we now have 99.5% of our required operational staffing in place. That is the first platform. We then need to have really good relationships with our education suppliers, so once you have the prison there you have high-quality education, and ensure that those places are filled. We have been doing loads of work. You might be aware, but two years ago we started employing specialists with an education background, directly employed by the prison, to come in and look at what they could do. I spent a good time with one in Brixton recently. I spoke to Charlie Taylor, the chief inspector of prisons. He really thinks those individuals have added a lot in terms of what we do in terms of education and the quality of education on offer for prisoners. But, as you can appreciate, not all prisoners want to go into education. We would normally have a proportion of places available that would be somewhere between a fifth and a quarter of prisoners who want to access it, and not all full time. We often struggle to fill those education places, so we need to try to make them exciting for prisoners and something that they want to attend and they want to go to. We are starting to get improved outcomes from those regimes in terms of maths and basic English standards that they are achieving. From when I recently met with Charlie Taylor, he thinks our purposeful activity scores are now beginning to go in the right direction and showing some improvements.

The bottom line on all of this is that the current crisis in capacity is having a negative impact on the ability to get into drug rehab, the ability to educate and the ability to reduce reoffending rates by giving people purposeful activity.

In straightforward terms it means all of our resources are stretched over more people with less time.

This is a really important question. When recidivism is 50% in some cases, if you could even make a dent in that you would reduce the amount of crime and the number of prison spaces required. It is not just education; it is giving prisoners a purposeful use of their time, rather than locking them up in cells for 22 hours a day as the Report makes clear. Is this something that you are seriously looking at in prison strategy? It seems to me that there would be a lot to be gained from improvements in these aspects.

Yes. Reducing reoffending is, alongside protecting the public and swift access to justice, the No. 1 raison d’être for the Department. In SR21 we received £550 million over three years to support reducing reoffending, looking at things like getting into a job, getting accommodation and substance misuse. It is a combination. As Amy has said, where we are at the moment in terms of capacity has definitely had negative repercussions for our ability to rehabilitate, but it is an area where funding matters. We are keen to work out how we can maximise funding available to help rehabilitate prisoners, because it is so essential and it is the only way you break the cycle.

The further ahead we can plan for these things, the more likely we are to have programmes that will succeed. The approach that we are taking to setting out a long-term strategy that you heard about earlier should help as well.

It is really nice to see you again, Dame Antonia. I wanted to move on and talk a little bit about the capacity shortfall numbers. I was hoping you could help me to understand the numbers in the first instance, because that would be really helpful for the Committee. I can see from paragraph 3.5 in the NAO Report that by the end of 2027 there is a projected shortfall—and there are three versions of this—of approximately 2,100 places in the low-population scenario, approximately 12,400 places in the central scenario, and approximately 21,200 places in the high scenario. My understanding from the assumptions around that is that that excludes the impact of SDS40, which is obviously the early release scheme that was launched in September last year. In terms of the numbers that are included in the Ministry of Justice annual statement, obviously there are different projections, but that is because there are different assumptions. My understanding is it that is because SDS40 is probably one of those differences, so there is a number of 5,400 projected as the capacity shortfall. Is that in line with your understanding? Is the 12,400 excluding or including SDS40?

Yes, roughly. There might be some other changes as well. When we publish data, we obviously have to publish the most up-to-date and accurate data as we have it. When we published the annual capacity statement in December, those numbers were based on the latest, most accurate data, so they included the projected line, which included SDS40, for example. At the time the NAO produced its Report, also in December, it wanted to keep working with the original data that it had. We all discussed this at the time, and the NAO understood that it was not therefore up to date and accurate, and would not match what we were publishing, but was based on the data that it had previously been working with, which is why its number says 12,000 for central and we project it as being about half that. Ours are based on the latest, most accurate data with all the implemented policies at the time. With Gareth’s agreement, I would go with capacity for the latest numbers on this.

Yes, the 5,400. That is really, really helpful. Thank you. Effectively, the SDS40 and some of the other measures reduce the shortfall in capacity by roughly 7,000.

Roughly, which is one of the principal reasons for doing it. The reason the Government took the decision to do it was to stop the lines crossing when they were about to cross before, but it also does help in the future.

No, absolutely. Do you have any thoughts on how long SDS40 is going to need to continue in order to allow you to manage the very challenging situation that you have at the moment?

As I already set out, the line includes it in perpetuity, because in line with best practice you work on the basis of the policy that is in place. Government could decide to unwind SDS40 following the outcome of the sentencing review if it puts other measures in place. That is a policy decision that could be made. It has not yet been made, which is why the numbers include it at the moment. I would not want to prejudge what Ministers would decide on that policy. It will depend on a number of things, not least, no doubt, what comes out of the sentencing review and therefore what policy decisions the Government make to reduce demand.

I totally agree with that. In addition to that, as you will be aware, the Lord Chancellor has agreed there will be a review no later than 18 months after implementation to see how the policy is developing.

That is helpful. Thank you. Just linking back to the previous hearing we had where we talked about the significant remand population, the number you gave us at that hearing was 17,600, so it is actually 20% of the prison population. I am not going to go over old ground that we went over before, but I am interested in some of the comments that were made by the Lady Chief Justice, Baroness Carr of Walton-on-the-Hill, around the capacity in the courts to sit. Capacity is about 113,000 days, and the decision was made by Government to move to only 106,500, leaving 6,500 days unfilled. This is going to have a drastic effect and put pressure into the system. Have you considered what the impact would be on the capacity shortfall numbers if there were actually more sitting court days? Have you thought through that process in terms of what it would mean for the capacity shortfall if you did get your maximum? Would that relieve the pressure on temporary schemes such as SDS40?

It is a very complex system. The short answer is yes. Just to be clear, the agreement reached between the LCJ and the previous Lord Chancellor was for 106,000 sitting days to be sat. You referred to a reduction. The agreement reached was for 106,000. The Lord Chancellor has now increased it to 108,500, so there was not a reduction.

It went up, but just not to the maximum. It did not go up to 113,000, which is the maximum.

Yes. One reason for that—not wanting to get diverted on to my other favourite topic, which is Crown court backlogs—is that there is the maximum that could be sat by the judiciary and then the maximum that could be sat with other constraints on the system, as we discussed last time I was here. To your main point, Ms Paul, you are absolutely right. The whole thing is a system. It is difficult to bring down the outstanding caseload in the Crown court if you do not have space in the prisons in order to allow that churn through. Equally, as we discussed last time, an increasing backlog in the Crown court creates a much bigger problem for Amy and her colleagues, because it has increased numbers on remand. It is now 17,000, up from 9,000 in 2019. That makes it much harder for Amy to manage the prison population. You are right that there is a virtuous cycle one could get into. We are on our way to getting there, because the first thing you have to do is create headroom in the prison population, as we have now done with SDS40, in order to bring down the Crown court backlog, which we are seeking to do, through magistrate sentencing powers, for example, because you need that headroom to allow you to do that. Overall, as you bring down that backlog, that should then help you with remand. Over time, you move towards a more sustainable criminal justice system as a whole. To a long way of answering your question, do we consider the impact of one on the other? Yes, absolutely. On almost all the governance we have in place, such as the custodial options taskforce and the criminal justice action group, it includes both HMPPS and HMCTS, but also the police, so every part of the system. We are looking at the whole thing end to end, and working out what is optimal for the whole system.

Do you mind if I just add a word on remand? You raise a very important point. Pretty much all the way through this hearing so far we have talked about total prison numbers; I wish it was only that straightforward for me, because total prison numbers are not the only thing I have to manage. I also have to manage a geographical constraint. If I have a prisoner who was sentenced in Exeter Crown court and I have a place in Leeds, that is really no use to me, if you see what I mean. Equally, to the point you are referring to on remand, if they are on remand the difficulty I have in efficiently accessing the estate is that they need to stay next to the local court where they are going to be heard. These things provide very particular pressures. It is that bit of the estate, our reception estate, that has been under the most acute pressures.

I fully appreciate the challenge you are under with that, because, as I said at the beginning, 20% of the prison population are on remand. That is a really, really difficult thing to manage. I appreciate you obviously did not make the decisions around the court sitting days, but I am interested in what the capacity shortfall would be if the courts were sitting for 113,000 days.

I am not keen to go too far down this line, because it is outside of the scope of this inquiry.

It is relevant in terms of understanding the capacity shortfall.

All right, but just on this one.

I obviously do not expect you to have that number to hand.

I am not sure I have understood your question. When the system is running at more than 99% capacity, increasing sitting days at that moment would mean the CJS would go into gridlock. In a rationed system, you cannot necessarily bring one number down and assume it is going to bring the other down, because if it is full you actually cannot allow the increase in demand for that to happen.

You would not want the court sitting days to go up to 113,000. Is that what you are saying? It would not work.

That is not quite right. I also want to be very careful, because when the former Lord Chancellor was doing the concordat letter a year ago with the Lady Chief Justice the agreement for 106,000 sitting days was what at the time was thought to be the maximum that could be sat. Now, as it turns out, things changed during the course of the year. If you have understood that bringing down the backlog automatically leads to bringing down the prison population, that is not quite right when you are operating in a full capacity.

No, I am trying to understand whether, if you moved to the optimal amount of sitting days, it is more than what we are currently sitting. That is the question, because you said at the time it was 106,000.

That was what was agreed in the concordat letter between the Lady Chief Justice and the Lord Chancellor, so I am just setting out the agreement.

In addition to what the Perm Sec has said, there is no single answer to that question of optimal. Obviously that is not the reason we do court sitting days, and it is separately constitutionally governed anyway, but it will depend on the balance between the remand population and the sentence population, and the balance between Crown courts and magistrates courts. You would find a different answer and lots of different dates if you asked that question anyway.

It is interesting to get a different perspective, because there has been evidence given recently at another Select Committee on this. I will not labour the point and I will move on. I appreciate that you have had to use various emergency measures. You brought in Operation Safeguard and Operation Early Dawn in order to manage the challenging situation that you have. How much extra has it cost for you to put that in place in light of the expansion scheme running late?

I will say a few things and then Amy will want to comment. All these different operations are different types of things. Operation Safeguard in particular, which is the use of police cells, has cost £70 million. At every stage we have authorised both the spend and the volume of cells used. There is a very specific process that we go through to authorise use of Operation Safeguard. That is signed off by me and by Ministers, and that has been the total amount there. We are very conscious. We go through a number of gateways and, every time we want to increase the number of cells or we want to reactivate, we go through that process again.

It is worth me just adding something operationally on Operation Safeguard. First of all, I am extremely grateful to police colleagues who have made Operation Safeguard available. It has been an absolutely critical part of our contingency planning. Secondly, Operation Safeguard does not provide any cells; it provides liquidity in the system, meaning that on the night where—an example I have used—someone gets sentenced in Exeter and I might have spaces but they are in Leeds, I simply cannot get that person from Exeter to Leeds overnight, or whatever the case may be. Operation Safeguard has been a very important one-night contingency. It is not really additional places, because the following morning we have to get them out of Operation Safeguard and back into prison custody, but it has allowed us to move the pieces around the chessboard. Once you get into 99% occupancy it is very difficult to move it very efficiently.

How many of these rapid deployment cells from tranches 1 and 2 have been delivered?

770.

That has taken roughly four years since the inception of the idea.

From the very start of the programme in the sense of conversation about rapid deployment cells to today, yes, it is four years, but each project has not taken four years.

My thesis is that, in totality, four years to get to 700 does not sound very rapid. It sounds more like really slow deployment cells, and Project Speed sounds like Project Slow. In this overall operating environment, Permanent Secretary, were you coming up with other contingency plans that you could present to Ministers, knowing that the rapid deployment cells programme was actually the very opposite?

Yes, as evidenced by the fact that the criminal justice system did not go bust. The period leading up to October 2023 when the previous Government took the decision about end-of-custody supervised licence was a series of interventions in order to prevent the system reaching gridlock, and then again up to the point where this Government took the decision on SDS40. There have been a huge number of things, both policy changes and supply things. On the rapid deployment cells, they were entirely innovative when first identified. I completely accept your perhaps slightly tongue-in-cheek suggestion that they are not that rapid. When we were first talking to the market about them, one of the reasons they ended up being delayed or coming in later than we wanted was that we were not doing intrusive site assessments at the time we were doing the design. We were literally in the process of working out what the design would look like and doing testing as to what category they could hold. I really believe that at some point they will be potentially game changing, because they will end up being able to be brought on rapidly. As Amy says, it is really good-quality accommodation. I feel like sometimes you have to take a bit longer to get the thing that will then give you significant flexibility.

Do you think maybe with the benefit of hindsight there was too much emphasis on them being a silver bullet earlier on, in 2021 or 2022?

Possibly. I myself was quite excited about them when I arrived in this job in January 2021, because they seemed to be something significantly different from what we had previously done. We are always looking at ways to do something that is innovative. Even at the time we knew that there was uncertainty and there was risk. We did not know what the cohort would be. For example, would we be able to get something that felt like a rapid deployment cell that could hold cat B? All this stuff had to be tested. Where could they be? How quickly could they be brought in? Would you stack them? Would you not stack them? They were ambitious at the time.

A bit too ambitious.

Can you be too ambitious? Maybe. Do not forget that they will end up delivering nearly 1,800 places and they have already delivered 770 places. In the future they could easily end up delivering more. They are the biggest change in terms of accommodation we have brought on and been able to provide that has happened in the last five years. Amy is now going to tell me I am wrong about that.

No, you are not wrong at all, but it is not just the sum total of nearly 1,800 places. It is that they provide places in corners of the estate and for certain prisoners that it would otherwise be very difficult to access. It seems to me that the basis of both your first questions and these questions is implementation times, because, in the end, whether it is new supply, ECSL or anything, I have to implement it. All of these things have different implementation times, with a brand new prison probably being at the extreme end and ECSL being at the extreme opposite end. That is why both the Perm Sec and Ross’s comments are so important. The thing we need to do is try to project outwards, so you can work out exactly which one of these things you want to implement with what implementation window. They do all provide something slightly different.

Can I shift gear in my next set of questions from the supply side to the demand side and the sentencing review? I am slightly worried that we are going to have the sentencing review and it is going to support some of the projected shortfalls, wherever the ministerial decisions land. Could you help set out some of the risks that it may be almost too late, because of course there will be a lag and a lead time into some of the benefits you get in managing that demand side? Could you talk through some of the risks that we might not get there quickly enough?

We are planning very carefully for that. There is a reason why the sentencing review is happening on the timescale it is: it is to make sure that the decisions are available to Ministers and then, of course, if legislation follows, that will need to come to Parliament. We have been considering that very carefully. Of course, it is not the only thing that we are doing. As was set out in the annual statement, Ministers have also authorised—and, indeed, Parliament has now granted—some additional measures on HDC and on RARR. It is going to need a portfolio of different things. Going back to your last question, the same is true with RDCs. We did not just rely on RDCs; the previous Government did demand-side measures as well. I totally agree there will need to be a whole set of different measures. I do not think it is the case that the timeline is impossible in the sentencing review.

One of my colleagues might come on to some of this in a moment. If I just end my little cluster of questions wrapping together supply and demand, they have interactive effects. Ms Rees, you have given some really illuminating answers so far. Perhaps you could help unpack some of this. When prisoners have time in custody, they are hopefully undertaking meaningful rehabilitation and so on. Where you are reducing time in prison, are you doing the analysis as a Department to understand that that might lead to increased pressure and increased demand through heightened reoffending, because of the very consequence that they have had a shortened period in prison? Could you give some reassurance that you are almost coming up with that systems‑level interactive thinking that reduced time in prison may actually yield increased demand?

I would start at the opposite end of the telescope. What we are absolutely clear about—and there is loads of evidence—is that, when you disrupt people for a very short prison sentence and you lose their ties to accommodation, to employment and to family, that is the worst outcome of all. It is certainly true that, when we have prisoners for much longer periods, we can do more effective work with them, and the outcomes are better when we have them for longer periods. But, if you are looking at anything under 12 months, the disruption effect in terms of the links for resettlement back to what people need, so jobs, family relationships and accommodation, leads to very, very poor outcomes, because you disrupt all of that.

If one of the points behind your question is, “Do we understand that having fewer people in prison will lead to an additional burden elsewhere in the system?”, the answer is absolutely yes. We talk about this a lot. Amy never lets me forget that, if anything, the impact on probation is significant but also rising, and will significantly increase if we attempt to drive down the prison population through decisions that Ministers may or may not make in the future.

It is not just probation; it is all the other services. It is health and all the other people who need to pick those up. But it is certainly true that, with all of those other community services, had we kept them in the community and had they kept the link to those services in the first place, it would be much better than trying to manage their move into custody and then, in a very short period, back out of custody into the community.

I am slightly worried that we will have the sentencing review but then we actually will not necessarily think about some of the resource requirements for the Probation Service as well.

A key part of our planning, not just for the sentencing review, but for any demand measure and any supply measure, is to look ahead for the staffing requirements of that. It is part of the modelling that we do. There is a governance structure, which the NAO commented on, that helps check all that is in place.

On probation, it is also true, to be clear and candid, that, should the outcome of the sentencing review be that we are going to put more volume into probation, we will need to take some decisions about what they are going to prioritise, because it is impossible for them to manage an increased workload beyond where they are at the moment.

We are getting the message that the system is dynamic. We got that at the last hearing on the courts backlog as well. You are saying that the concrete measures such as increasing the number of prison cells will not resolve the issue on its own. Other issues such as reducing recidivism are long-term strategies. Are you not letting the prospective policy changes do quite a lot of heavy lifting here, so the sentencing review and the criminal courts review? Given you said previously that you assume government policy stays the same until it changes, what are you doing to anticipate those reviews, in terms of the effects on prison numbers?

It is not so much that we assume that government policy stays the same; it is more that, as I just explained, when we publish projections we publish on the basis of known variables, so policies that have already been implemented. We project on that basis. Obviously, as the annual capacity document shows, there is a gap projected in the future between supply and demand. One of the reasons that the Government have asked David Gauke to lead this independent review, the sentencing review, is to look at a number of issues in sentencing. One of the reasons to do that would be to want to have a sustainable system. Something is going to have to happen. To your point, we are building what we are building and that is also public. Other things have already been done. We constantly consider a large number of options that we would put to Ministers to solve these problems, as we have done and as you have noted has been done over recent years. Last October, the Government announced an extension of the curfew for HDC from six months to 12 months. There is also the risk‑assessed recall review. There are a number of other policies that are already coming down the track to help focus on that.

There is also SDS40. Yes, there are, but my point is that these things, even in combination, are not resolving the crisis. You presumably are not allowed to put figures in that say, “Well, we hope to reduce recidivism by 10% over the next 10 years, so we will account for that”.

It is not that we would not be allowed to; it is that I would not let that happen. Unless you could guarantee you could deliver it then the projections would not be able to be accurate.

We have been told that both of the major reviews are reporting in the spring, which seems very quick, but there could be legislative implications of them. It will take time to introduce the measures they suggest. You might suspect what measures will come in but you do not know. What planning are you able to do, and where can we see the results of that planning? This seems to be absolutely axiomatic as to whether you are going to succeed in reducing the prison overcrowding situation.

As we were discussing earlier, the capacity gap numbers have already significantly come down from those that were included in the NAO Report. The current Government have taken action that has substantially reduced that. We do look ahead and we plan very carefully for measures that come in. Obviously we cannot plan for the measures of the sentencing review until they are finalised and it is independent. There is a mix of secondary and primary levers. As you will be aware, some of the secondary legislation measures can come in more quickly than primary legislation would, and there might be other policy options that the Government wish to consider as well. As the Lord Chancellor makes clear in the introduction to the capacity strategy, all these measures will come together in totality. We are confident that our planning systems can support that. We will need to wait for the policy decisions from Ministers after the sentencing review.

What planning are you doing for additional costs?

We set out the cost whenever we put forward a policy measure. It is very often done in the impact assessment, as was the case with HDC recently. It will be part of the planning. We look at the operational impacts, the staffing impacts and the impacts on our CJS partners. There is a thorough and extensive process. I am not suggesting any of that is easy, but we have very well‑developed processes in place to do that.

When do we get to hear about that? Presumably you are in a fairly constant narrative with the Gauke review, for example, because you would not want it to come up with recommendations that would increase the prison population. If that dialogue is going on, will anything be published prior to the sentencing review reporting?

It is an independent review. We will have to wait to see what the review says, and then Ministers will have to decide what they want to do about it. If it is legislation, Parliament will need to decide whether it grants those powers. There are a whole series of steps to go through. The planning will be taking place. It will be very thorough and we will check all the things you just described.

I have one last question on this issue. I will use the example of Dartmoor, which is closed as we speak. I do not know whether you have a prospective reopening date. Do you have any other situations like that, or potential situations, where substantial numbers—not game-changing perhaps, but substantial numbers—of prison places could go offline or come back online in the next 12 to 18 months?

It can always happen.

There are a variety of operational reasons why these things might happen, including RAAC, which we did not experience there, but we still have 183 places off due to RAAC at the moment in two establishments in the estate. A boiler, a leak or a fire—touch wood—could bring off substantial numbers. In terms of Dartmoor, as you say, they are quite significant numbers, but we would need to find a technical solution that means we are safe to reopen them.

What is the position with Dartmoor?

The position at the moment is that we are still looking with HSE at what we could do and what mechanical ventilation, etc, we could introduce that would allow us to safely reopen it.

You are paying a lot of money for it. Is it to the Duchy of Cornwall?

Yes, we certainly pay rent to the Duchy of Cornwall.

There are a lot of prisons in crisis at the moment, not just in terms of their physical state. One solution to that might be to close a wing or to redistribute prisoners in some way. Again, that has big operational effects and could temporarily at least reduce the number of available places. Is anything like that going on?

Yes, there absolutely is. Going back to my red lines, we took places off where we thought we were unsafe or we needed to, even when we were at our most capacity constrained. A more recent example is Wandsworth, where we have taken places offline to try to reduce the operational burdens.

How many?

We have done 150 so far.

Is that 150 cells?

Places.

Of how many places in the whole prison?

The op cap varies, but it is circa 1,500 in Wandsworth. We have tried to reduce those.

So 10%. When did that happen? When were those 10% taken away?

We started taking them off in late autumn last year.

Right, and there are no plans to reintroduce them. It is running at 90% of its previous capacity.

To the conversations we had earlier, we have tried to reduce crowding in Wandsworth to relieve the pressure there, given they have operational difficulties. In answer to your question, yes, for staffing reasons and for stability reasons we will take places offline if we need to.

I have some questions on probation. Shall I keep them for later?

No, please do them now if you wish to.

I raised it now because you just alluded to the implications for other parts of the system. I went with the Justice Committee up to Manchester to look at Buckley Hall prison—which I am sure you know; I am sure nothing gets past the officials—and to talk to Greater Manchester Probation Service. Although prison officer numbers have undoubtedly increased, there are still a lot of gaps in the probation system and there are a lot of additional pressures. For example, they will have to cope with pressures due to rioting and other pressures as we go along. There is a shortage of officers. When they start, they will not be experienced officers.

Sorry, can we clarify who you are talking about?

Probation officers.

Probation officers rather than prison officers.

Yes. In some ways the situation seems to be more acute with probation officers in terms of that recruitment process. We know there are a lot of risks. There are serious further offence risks that the Probation Service is understandably cautious about. What is your overall feeling about the Probation Service’s ability to cope now, but also if you are going to put all these additional pressures on it, whether that is by home release or whether it is by non-custodial sentences?

I am going to ask Amy, who is also responsible for the Probation Service, to talk to this, but I would say that, first of all, it is under a huge amount of pressure at the moment. It is doing an absolutely excellent job. We are conscious that a lot of the changes that we make when we are diverting, be it HDC or the risk-assessed recall review, lead to increased pressure on probation, which is why we have a big programme of recruitment in place, which Amy is overseeing. I will ask her to say something about that.

If I may, before I get to your question, I would like to give some positives. You started with prison staff. We are now at 99.5% in terms of prison staff. That has been a huge effort and we are really proud of everyone’s successes. It is worth saying, though, that it now means that nearly 40% of our band 3 to band 5 prison officers have less than three years’ experience, which is something we will need to manage through the estate. In terms of probation, we have increased staffing by 20% since we took it over and unified it. We have recruited over 4,500 PQiPs, so trainee probation officers. I absolutely accept the premise of your question. To be clear, we cannot put any additional pressure on probation without an offsetting measure and recruitment alone will not be sufficient. While we have been successful in recruitment, it will not be sufficient. What we are doing and have already started to do, as the Perm Sec said, as these new measures come in—things like the changes we have already made to HDC and SDS40, meaning people are spending a longer balance of time in probation—is a programme called probation reset. What we are attempting to do is to say, “First of all, focus your efforts on medium and high-risk prisoners, as opposed to the lower-risk offenders. Secondly, all the evidence says that the first two-thirds of the licence are the time when we will have the most impact on the risk reduction. Focus the effort on the first two-thirds”. I am pleased to say we think that has been successful. On the lower tiers we have managed to have a really significant percentage reduction. On the higher tiers we have had a 10% increase of the time that probation professionals are spending with them. That is exactly the sort of change we were seeking to pull off.

You are constantly robbing Peter to pay Paul here. What is your equivalent to the 99% figure for probation? If you have a 99% filled establishment for prison officers, what is it for probation?

It varies by the different staff grades in probation. It is quite complicated, because some cases can be overseen by PSOs.

Probation officers and PSOs, yes.

We are probably at about 102% overall in terms of the workload that they are managing at the moment. That will be a wide variation. Some probation officers will be way under that and some will be over it. In terms of robbing Peter to pay Paul, that genuinely is not quite the way I would characterise it. In reset, we are saying that there are some things that the evidence says it is not worth us doing any more. There are actually some things we are attempting to stop. It is not robbing Peter to pay Paul; it is saying, “Peter will not do some things and Paul will do some other things”. We are trying to say there has to be a principle, to your point, that we cannot put any additional work on to probation without trying to establish what probation will not be doing going forward and where we would like to focus its efforts.

Those less serious, less dangerous offenders are exactly the type of offender who, if the Gauke review recommends alternatives to custody, there is going to be a flood more of coming out. The expectation among everyone, from the public to the sentencers, is going to be that they have a high degree of supervision and they have what are testing and meaningful alternatives to custody. You are saying, “Well, we actually do not have time to look after them”.

No, there are two things about that that are not quite accurate. The first is that the lowest rung of probation caseload are people who would never go into prison. These are people who receive only community sentences. They might only receive a standalone unpaid work order. This is at the very bottom of the caseload.

That is my point, or it is another side of the same coin. We hear from magistrates that they are reluctant to sentence for community sentences for the very reason that they think people are going to have a soft ride, not be supervised and get away with murder—well, not murder, obviously.

I have a couple of points on that. First of all, the evidence base would say that those people at the very lower end need more support, but there is no evidence that lots of intensive supervision is very helpful for those people. Those people often need support from the services we were talking about before, such as accommodation. Yes, it is a level of supervision, but the question is, “For how long and for what duration is actually helpful, and what support services are you going to move them on to?” Actually, the people you are talking about shifting from short custody and into probation are not the bottom end of the tier. They are people who will need more support from probation and supervision from qualified probation officers, which is why we must make capacity for them.

I will stop here, because I am very sympathetic to the situation you are in; I just cannot see the way out of it. If I were a magistrate sentencing, if it was on the cusp and I was deciding between a short custodial and a community sentence, which I may not even have the choice of in a short time, and if I asked questions about what the nature of that community punishment would be and was not satisfied that it would be in any way either rigorous or take place at all, and therefore went for a custodial option, would you not be concerned about that?

No, I completely agree with you, but that is what I am saying the remedy is. The remedy is to make capacity for exactly those people who are not at the lowest end of the tier and instead look at the ones who would never be a choice for custody. These are not people who would ever have been referred to custody; they are people who just get a community sentence. The choice is, “What do we do with them as part of that community sentence?” in order to make capacity for the people on the cusp you are talking about.

Can I just briefly add one quick point about technology? Things like tagging can make a significant difference as an option of what one can do when sentencing. Sentencing is a matter for an independent judiciary, but knowing that things like tags work, along with the technological change on things like GPS tags and sobriety tags, has made a difference in terms of the attractiveness as a sentencing option of electronic monitoring as opposed to custody.

Do they require any supervision or monitoring other than purely electronic for the person serving that sentence, which, again, may be either an alternative to or early release from custody? Do you have confidence in the companies that are running those? Serco is the one that springs to mind.

I have opened up a whole can of worms here. The short answer is yes. Jim is the world’s leading expert on this, but the point I am making is that tag as well as whatever would have gone alongside it is at least as good as whatever would have gone alongside it, so it can provide an additional level of confidence.

As an example of that kind of supervision, a couple of years ago we introduced what we call the acquisitive crime pilots in electronic monitoring with 19 police forces across England and Wales. Individuals released from prison having served time for burglary, robbery and a few other offences get a GPS tag on their leg by mandate. We supervise them, but in addition we receive data from those police forces on robberies, burglaries and other linked crimes on their patch. We match that with the GPS data coming off the legs of those offenders. We use that to rule people in and out of being suspects of crime. It is also really powerful because it gives additional information for the probation officer to use in their supervisory conversations with those individuals, so it is exactly the kind of innovative use of technology alongside practitioner judgment that we would want to see more of in the future. It is the kind of thing that absolutely looks, feels and demonstrates the toughness of probation when that is the right response for certain individuals. On Serco and Allied, which is the other company that provides the tags, it is absolutely right there have been transitional challenges as they have moved from the old IT systems that they inherited to the new systems that they are introducing. We are a long way through that now. There has been very close scrutiny of their performance by Amy and others. We are hopeful that they will have completed transition to those new systems in the next couple of months. That then provides a really strong platform upon which to build.

I have just a couple of other points on technology. To answer your question directly, it is not a free good even when these systems are working perfectly, because, even if someone is tagged, you need a responsible officer. You need probation officer capacity in order to do whatever is needed with the data that comes through. We will definitely factor that in, to Ross’s point, to all future projections. More broadly, we joked earlier about AI being the answer to all issues. I am not saying it is as something on its own that will help us, but one of the things we do need to do for probation officers is provide them with the best possible technology to try to reduce the amount of double keying and things that they have to do to mean that their time is most efficiently spent. There is some good work to be done there too.

Thank you for those answers. This Committee did a lot of work in the last Parliament on tagging, so that was a very useful update. Mr Slaughter mentioned HMP Dartmoor. I reckon that the costs of remediating are probably fairly substantial. When you renewed the lease last year, I hope you managed to reduce the cost of it considerably.

We certainly did some good negotiations with the lease and we also got some investment from the Duchy into things that we could do in the prison. It is certainly true, from what we know today—and there is more we need to do to work out exactly what we are going to do on Dartmoor—that it would require significant investment, but it would also be a lot less than building a brand new prison of that size, 640 places. The question for us is whether we can find a technical solution that means it is safe for prisoners and staff to go and work there, to reduce the levels of radon.

Is there a risk that further parts of the prison may have to be closed yet?

There is no part of the prison in use at the moment.

Do you think it is fair to taxpayers that we are still paying rent to the Chancellor of the Duchy of Cornwall for the premises, given the radon issues? Is it fair to my constituents and other constituents?

And that fact that it is unusable now.

The thing is we have not made the determination on whether it is unusable. We are trying to find a technical solution. If you are asking me honestly, straightforwardly, whether I would want to give up Dartmoor as a site right now, the answer is a resolute “no”.

Do you think it is fair that we have paid the Chancellor of the Duchy of Cornwall rent for a radon site?

It is the contract.

I am not asking about contractual fairness. I am asking from the sense of value for money for our taxpayers.

I am still occupying it, am I not?

To be fair to our witnesses, that is outwith their scope at the moment.

Just to follow up, did you not think about inserting a clause into the contract, when you negotiated it, that indicated that there would be a reduced payment if the prison was taken completely or partially out of service because of radon, given that radon was a known problem when the contract was entered into?

To be fair, before we get too far into contracts and payments, first of all, I have been clear that I would not want to give up Dartmoor as a site. It is 640 places, if we can find a solution. The second thing is that we are still in there at the moment. We do not have prisoners in there, but we have staff in there and we have lots of works. We are trying to figure out if we can find a technical solution. I would not want to say it could be used for another purpose, because we are trying to work it out. The site still belongs to us and we are still using it.

With the greatest of respect, Ms Rees, you are not quite answering any of the questions from the three of us. The question was how well you negotiated. You knew the problem at the time, when you renewed the lease last year. Did you renegotiate the lease sufficiently downwards to cover the known costs, at that point, or the fact that you would likely have to vacate a considerable part of the prison? Did you negotiate downwards sufficiently?

They negotiated with us really well and they made some investment into the site. We continue to have negotiations with them as we find technical solutions, but I am still trying to find a technical solution to that site.

It is not unusual to have to pay for such a site. As Amy has been talking about, there are still staff there. There is still work going on and the Committee was asking a question earlier about the balance between new build and remediating places. This is a prime example of that. We need to consider really carefully whether it is going to be cost-effective to bring the prison back into use. That is, in this case, unfortunately, tied into some quite complicated scientific questions about radon and the health risks involved there. There are HMPPS staff on the site. If you were renting a flat and the flat was empty while you were doing some decorating, you would still have to pay for it, would you not?

With respect, your questions only come into play if we determine that we cannot do anything with that site and cannot find a technical solution.

Not quite, Ms Rees, because you know that you have considerable costs. You knew at the time, when you renegotiated. We cannot ask a sensible question, because we do not know how much investment they have put in and we do not know how much you renegotiated the lease. I may ask you, in due course, having considered it, to give us the precise details on confidential terms. To avoid me having to do that, because I know this is a sensitive subject, could you perhaps reassure the Committee that you negotiated to make sure that your costs were well and truly covered, either through investment from the Duchy or by a reduction in the lease?

I can certainly assure you that the renegotiation was a positive experience for HMPPS, in which we got some benefits. The other thing is that you are assuming I knew everything at the time of renegotiation and that is not true. We had much more substantial radon levels and readings than were expected in communal areas after the time that we renegotiated the lease.

We have been doing radon testing and, as we say, those levels were higher than we were anticipating, but we are also trying to work through, with HSE, what exactly would be safe levels. That is not as absolute. There is quite a lot of detail to work through in that sense, because it is different from residential accommodation in terms of prisoners, due to the amount of time they spend in confined spaces. It is not exactly the same as residential property.

Why did you extend the lease for 25 years?

First of all, there were options for how long we extend for, for what period, and then the period afterwards. We are going to have to make significant investment so, frankly, if we make any investment, it has to be over a considerable period of time.

That implies that you are sure that matters can be resolved.

No, we are not sure, as I have said, that we can find a technical solution. At the time of negotiating, we were more sure, though, because we have had higher radon levels readings since.

You could be stuck with a lease that you cannot resolve. Then what happens?

Then, we would go back to negotiating. As I say, my genuine personal experience with the Duchy has been really positive. If that is the situation that we get to, I am confident we get around the table and see if we can find a solution for both parties.

Just to come back to fire safety, which is clearly a really important issue, I understand from the reports we have had that there are currently 23,000 occupied prison places that do not meet current fire safety standards. You are committed to put that right and make them all fire safety compliant by 2027. Are you still on course to do that, Ms Rees?

We have already covered that the collapse of some companies may mean it is impossible to do them all by the end of 2027, but we will take them out of use if they are not remediated by then.

How many do you think you will not be able to do by 2027, which you will have to take out of use?

I genuinely do not know the answer to that, because we have not managed yet to re-let all of the contracts that were originally under the supplier that has now gone into administration. I can tell you, though, that we have averaged 4,000 places a year for the last couple of years. In the second year, we have done a lot more places. If we were able to continue on the level we have been doing for the last year or so, we would get nearly all of those done but, sat in front of you today, I really do not know, because we have to re-let the contracts, which we have not yet done.

You would have to be going at something like 8,000 a year.

Yes, and that is possible.

How long will it be before you are likely to have a clearer understanding of what your future position will be on this?

It will be six months, probably.

In the meantime, you will be looking at contingency measures, if you have to take all those cells out of use.

Yes. We will certainly not have to take them all out of use but, whatever balance that we are left with—let us say, for argument’s sake, it is 3,000 at the end of 2027—then we would definitely look. They would be factored into the projections and then we would decide what we had to do to manage them. That is, in a sense, like all the other things—like boilers, like staffing—where, three times a year, we look at the projections. It is all of these things that are factored in.

Do you have the necessary resources to complete all this work?

Do you mean in terms of funding?

Yes, in terms of funding and capabilities.

We have covered the funding earlier, in terms of maintenance monies, what we would need to do and how we would need to speed up getting something from the spending review. At this point, the answer is no, but we have a lot of negotiations to do. In terms of resources and project management, yes, we do have that. Like I say, what we have averaged in the last year in particular would mean that is possible, but there are two factors that I am still concerned about with maintenance in general. That is supplier capacity and the headroom in order to do the work at this level.

Earlier, we were talking about the impact on prisoners and you did raise the impact on staff themselves. The Committee would also like to recognise the work that the staff are doing under very difficult conditions. You raised the issue of recruitment and that you feel you have done very well on that, but you recognised that the result of that is a dilution in experience. The NAO Report suggests that the number of band 3 to band 5 officers with two or more years of experience had decreased from 81% to 68% between June 2021 and June 2024. In addition to having staff recruited is the need to retain. Clearly, working under these conditions is leading to sickness absence rates that would be higher than the civil service average and also reports of burnout. In order to secure not only the physical capacity, but the workforce capacity that you need to run the service, what are you going to do about workforce retention and wellbeing?

It is a really good set of questions. On recruitment, we have already talked about what we have done and that it has led to a balance of experience and inexperience. We have also done national campaigns for the first time. Some of you might have seen that we now do, “An extraordinary job, done by someone like you”. It has been a really successful campaign. We then have to look at retention rates, which have come down. We are now at about 8.5% in the last set of published figures. That has come down, but there is more to do. We do a lot of retention deep dives to try to look at the issues and what causes that. It is quite hyper-local. The issue as to why people leave or stay in Woodhill might be a different from the issue in Berwyn, for example. We also find it is a combination of what I would call hygiene factors: “Do I have a locker? Did I have a line manager who met me on day one?” Then there are also big factors, such as pay, safety and all of those things. We try to have an action plan for how we deal with any sites that we recognise are struggling with retention. In terms of wellbeing, there are three layers to what we do. First, you have to have a workload that is manageable. We have talked about that, both in prisons and probation. That is the first thing. The second is then the respect that they get for the extraordinary job that they do. You only have to look at last week and the Southport trial to know what an outstanding job people in my organisation do. Not only did they sit through that horrific testimony and statements, they then took him back to the establishment and we will work with him for the 52 years that we have him. It really is outstanding work and they need respect, dare I say it, from Committees like these, in terms of what they do. Pay is another considerable factor. We have increased pay for prison officers quite considerably. We have had a three-year multi-pay deal. Then there are the human things that we do to support people, such as the employee assistance programme. We have actually recruited a dedicated wellbeing support for each region in prisons and probation. It is right through, from the hard data of workforce metrics and what you do to pay people to the human support.

Do you have targeted support for any staff who are victims themselves of assault?

Yes, we do. We have the employee assistance programme, which provides counselling 24/7. We also have a really well-developed and well-regarded system of incident management, Gold, Silver, Bronze. We do incident debriefs after those and I do things such as writing to any staff who are seriously assaulted. There is local support, etc, but there is genuinely always more we can do on this.

Is there any chance of getting any returners, given these experienced people have left you?

Yes. We have been working on returner campaigns. As you will appreciate, being a prison officer in particular is a physically demanding job. There comes a point in your career when you might not want to do that and we have tried to think about those as maybe being trainers, maybe bringing them back to do support. We have talked about the balance of experience and inexperience. We can use returners to come back and do some support work for new starters, so that they can learn from someone with a lot of experience. Yes, we are looking at returning campaigns.

I only say that because, before I came to this place, I did a lot of work on age-friendly employers and how we enable people to work for longer. Some of those suggestions would be very welcome, I am sure. Those were the main points I wanted to ask on workforce. The only thing is your assumptions about the workforce you need are based on the fact that we currently here have a prisoner to prison officer ratio of four to one. Is that safe? It is double the European average.

That is just an average and I am honestly not sure how helpful that is. In terms of my long-term high-security estate, it would be very different and, in the open estate, it is very different, on the opposite end of the scale. It is right that it is governors and PGDs that decide the safe ratio for their prison and it really depends on the context. It also depends on things we talked about earlier, such as lines of sight and the actual building that they are managing. There are really local factors that determine how safe that is.

How do you plug all that into your long-term workforce strategy, to be confident that you have that longer-term supply of workers?

Ross might want to say a little bit about how we do it analytically and how we manage our staffing profile. I would say to you that we have done well. We have not had to delay any opening of new places because of recruitment. We have managed to recruit really well those that we need. We have had to do more work on retention, as we have talked about quite a lot. I can honestly say our issue is not accurately forecasting what we need. It is more about making sure these people feel valued, we train them well and they stay.

Amy has put it very well. It is one of the things we model and, to date, we have done so very successfully.

Can I just say, Ms Rees, your points about your hardworking workforce, in both the Prison Service and Probation Service, are very well taken by this Committee? We really do appreciate what they do. It is a tough job and that is much appreciated. If there are no other questions, I have one sweep-up question for you, Dame Antonia, if I may. Last summer, after the riots, when some very quick sentences were handed out, your margins for vacancies in the prison were very tight indeed. Are you confident that, if some other shock to the system like that happened again, we would not be in a similar position? Through your supply side, demand side and bringing extra places on stream, are you confident that it would not happen again?

That what would not happen again?

You were down to such thin margins at the time when the rioters were sentenced.

All the work that we are doing is to try to ensure that we are not in that position. We were in a very difficult position at that time. The announcement of SDS40 had already been made and, although tranche 1 had not happened, partly because we were in the summer, we were already in a good position, thanks to the herculean efforts right across the CJS. I should also, at this point, thank the judiciary for all the work that they do with us, in partnership. That is partly a reflection on the discussion we were having earlier on sitting days. At the moment, right now, we do have some headroom. We need to avoid getting into a position where we have fewer than 300 places available across the whole estate. At a time when nearly 1,000 people are being sentenced, many of whom will go to custody, if you only have 300 places left, at that point the system has ceased to function. The whole purpose of everything that we are doing at the moment, the whole purpose of the publication, is to get a strategic, advanced look at things. The purpose of all the interventions that we do on the demand side and the supply side is to move away from a position where we are right up against the limits, which would stop us being able to respond. The first job of the whole of the CJS is to be able to provide a space for anybody that the independent judiciary sentences to custody. That happened in that case and will continue to happen.

We did survive and we did manage. That is because people in the CJS were outstanding in this summer period. That was right across the CJS. That is also the reason why tough decisions must be made by all parties, because we all witnessed in the summer, did we not, that you absolutely have to have a functioning CJS? When things get very difficult in society, it is people like me and others across the CJS that we look to and rely on. We must be properly funded and prepared.

I had two very quick factual corrections, if that is okay. I was not able to answer Mr Charters’ question earlier around the number of places that have been significantly delayed due to nutrient neutrality. The answer is about 300 with a really material delay. I misspoke earlier. I mentioned that we had built in additional contingency to the forward plan. We are operating that plan as a P76 contingency, not a P43. Apologies.

I did think 43 sounded quite low.

I transposed two numbers in my mind, so apologies.

Thank you for making those corrections. Can I thank all our witnesses very much indeed? We have had an interesting hearing today and we want to be producing our report with recommendations shortly. The uncorrected transcript of the hearing will be published on the Committee’s website in the coming days. We will consider the evidence that you have kindly produced today and, as I say, produce a report in due course. Thank you again.