Cynulliad Cenedlaethol Cymru

I'm a Commissioner with responsibility for budgets. It's perhaps worth noting.

Just noting it. I don't think it has a bearing on this, but thanks for noting that. There we are. Thank you.

If we can move to item 2, papers to note. We've got one paper to note. Is everybody happy to note that paper? Lovely. Diolch yn fawr.

We'll move then on to our substantive item this morning, which is the financial implications of the Homelessness and Social Housing Allocation (Wales) Bill evidence session. We have the Cabinet Secretary and officials with us. Could you introduce yourselves for the record, please?

Diolch, Cadeirydd. Jayne Bryant, Cabinet Secretary for Housing and Local Government. I'll ask my officials to introduce themselves.

Bore da. Sarah Rhodes, deputy director of homes and people in Welsh Government.

Bore da. Hannah Fisher, head of homelessness prevention legislation.

Thank you very much. We've got quite a lot of ground to cover, because it's quite a hefty Bill.

It is indeed.

If we don't get to ask all the questions that we want to in the time that we've got, then we might write to you and ask some supplementary questions.

Absolutely, and we'll do our best to get those to you as quickly as possible—the responses—as well, Cadeirydd.

Brilliant. Thank you very much. There we are. So, if we can start this morning—. I'd like to just explore the reasons for the introduction of the Bill and the overall costs and benefits. So, to start off with, your estimate over the 10 years is the Bill will cost £325.8 million to implement and generate around £1.2 billion-worth of benefits. Can you set out the key elements of the Bill and provide an idea of where those significant costs and benefits fall, please? 

Diolch, Cadeirydd, and thank you for the opportunity to come before you this morning. The Bill contains a comprehensive package of reforms, as you will know, and that supports our long-term aim to end homelessness in Wales. It does seek to address two main challenges, in managing the increase in demand at the front of the system and, obviously, improving the flow out of the system as well.

So, the cost regulatory impact assessment shows that initial investment is obviously necessary to address the challenges within the system. It does set out significant areas of additional cost in relation to strengthening the prevention work, which is important for our Bill—so, better addressing the individual need and to support people in sustaining their tenancies. Benefit over time is the focus on prevention and that flow through the system, so that we're not pushing people, as the current system is, to crisis. We want to have that early intervention and prevention, so that flow through the system, and that will enable the redistribution of resources, resulting in higher levels of prevention, more efficient relief of homelessness, alongside that reduction in reliance on temporary accommodation, which we know is very expensive, not just financially, but obviously as a human cost.

I can go into some more detail, but the reforms are expected to cost £247 million in present value terms up to 2035-36, and if they achieve the improved prevention and relief outcomes of the core scenario, the implementation costs will be outweighed by that financial benefit, which is projected to have a present value of £481 million.

Thank you.

Just for clarity, you mean 'net present value' when you say 'present value', or do you mean something different?

I think I mean 'net present value'.

Thank you.

But we can—. I think I mean 'net present value'. Yes, I do.

Great. Talking about those net savings and noticing the net savings, or the costs, effectively, in the first year, 2026-27, are £2.5 million, then £12.5 million in the second year, and then it goes into savings after that, rather than costs. Is it going to happen that quickly, because it's such a big Bill? Are the benefits going to be seen that quickly with the implementation of the Bill?

After an initial transition period, the provisions in the Bill are expected to result in that financial benefit to public bodies across Wales, because, obviously, the Bill intends to bring a more public service approach to tackling homelessness. So, the success of the reforms does rely on increasing the rate of prevention in homelessness to reduce the number of people entering the homelessness system to start with, and, therefore, all the associated homelessness costs.

As I mentioned in answer to your first question, increased prevention will mean tackling that access to temporary accommodation and the reliance that we have at the moment on temporary accommodation provision. It'll also rely on case management by local authorities, as well as mitigating the wider negative impacts on individuals.

Again, there is some uncertainty around how much the provisions in the Bill will increase the prevention rates. The central estimate in the modelling is for a 10 percentage point improvement on each rate. So, the estimated financial benefits to local authorities, the wider public sector and registered social landlords is estimated to be £475 million, £167 million and £3.7 million respectively.

So, in addition to the financial benefits, we do estimate £549.6 million in social benefits associated with improved homelessness prevention, and these benefits are expected to accrue to the individual household in the form of those health benefits and earnings benefits. There are some things that'll happen over time, obviously, when we're moving a system from a crisis point that we have at the moment to prevention. That does take time, but we feel that the benefits will be there for people to see.

Yes, I see that. My question is more about is it going to happen that quickly, to be able to turn something that's in crisis around to start feeling those benefits that quickly in the financial space.

We feel that the time that that will take—. We understand—. We've put things in place already, we've had some more investment in the build-up to this Bill around local authorities as well, in terms of realigning their homelessness system in preparation for the Bill. We do think that there will be these benefits in terms of focusing very much on prevention at an early stage and that we will see the benefits within that in the time that we've set out through the RIA. I'm sure we'll come on to this around data later on, but we've done our best with the data that we have. I don't know if Sarah or Hannah want to say anything more, specifically on that.

Thanks, Cabinet Secretary. I think what the RIA does as well is take account of the implementation period before the Bill comes into force and then some time before those benefits in terms of prevention come online. So, we think it's a reasonable estimation in terms of time. What the core scenario shows is that even a modest increase in prevention rates will reap the benefits because the system is so focused on that crisis point at the moment, and it is the most expensive system that we could have. So, the modelling does show that, as I say, even small increases in prevention—. Where the prevention rate is at the moment, in the last year's worth of data, is around 58 per cent, 59 per cent. A 10 per cent increase will only take us back to where we were pre COVID, so, actually, we think that's entirely achievable with additional resource in prevention. We know local authorities can achieve those prevention rates because they've done it before, so we think the time scales set out in the RIA are reasonable.

Reasonable and realistic.

Yes.

Yes. Okay. The benefits are calculated on an estimated increase in the costs of the current system. If the change is not implemented, the RIA then estimates that the number of homeless people will rise by an average of 3 per cent per year, and the related costs by 17 per cent in real terms between 2023-24 and 2033-34. So, what evidence have you used to get to that point?

In terms of estimating both costs and financial benefits, obviously that is challenging. A key challenge with the modelling of the benefits of reforms is that there's little evidence on the potential impact of the individual reforms or examples of similar interventions, which a benefits analysis would typically be based on. As a result, it's not been possible to model the benefits resulting from each individual reform. Instead, the benefits are based on the overall package of reform as a whole, focusing on high-level challenges to the key parameters that lead to the benefits outcome. 

But we've looked at the stock-and-flow model, used as a basis for costing the Bill, and that demonstrates the state of homelessness in Wales and how it's projected to change over time, assuming that there's no change to policy or legislation. So, the analysis is based on Stats Wales outcomes data, and that assumes that the state of the world, as reflected in the data, remains unchanged. Looking at the Stats Wales data and taking that five-year average, as you say, that suggests a roughly 3 per cent increase per year. 

But, as I said, again, around the data bit, it is important to note that the data is erratic over the last five years in particular, but this forecast is really broadly in wider research that suggests that homelessness is likely to increase over the next decade unless we shift the policy to address that. Because, as Sarah said, the system that we have at the moment is the most expensive system that we could have, and we know that the costs will increase in that way. So, we have to do something now, and I believe that this Bill will change the system in that way. But as I said, it is important to note that the stats that we have are a bit erratic over the last five years.

So, it's effectively looking at—. Putting it in real, simple terms, you're looking at sorting out the low-hanging fruit first, and that's the most expensive, and effectively, that's where you get the benefits to come through quickly.

Well, I think the aim is to focus on the early intervention and preventing homelessness. So, some of the aims of our Bill, which is moving from the amount of time you have in terms of prevention, it's going up to six months, and that'll mean that local authorities will have more opportunity to help prevent homelessness. Doing some things at an earlier stage, we all know that that's better than going to the crisis point. So, there are many aspects of our Bill that are preventative in terms of the focus on ask and act for public bodies in Wales, to a number of other things that we're looking at in terms of that prevention.

So, we do think, starting with the preventative side of things, dealing with people at an earlier stage will really help prevent them, say, from going into that crisis point, which is happening now, because we go from the 56 days, which we've got at the moment, which means that local authorities are often helping them at the very last minute, to our proposals around the six months, which mean they have a lot longer to work with individuals.

So, you talked earlier about the 10 per cent, that the basis of the costings in the RIA accounts for a 10 percentage point increase in the share of preventing people becoming homeless. What was the evidence behind that, and how robust is the data that sits behind it? Because you said on one hand that you're looking at Stats Wales data, and on the other hand you're saying, 'Well, it's hard to do it because the data's not great.' So, how do you find that those figures actually tally up? 

Yes, I think that's an important question, and perhaps it'd be helpful to just clarify the purpose of the core scenario as well.

So, as you said, there are limits within the evidence that we can't specify a financial benefit to each provision, as I said, within the Bill. So, that was why we've taken the approach using what's called the 'switching analysis', in which we switch the cost and benefit parameters to understand the break-even point. So, based on the last full two years of data, the current homelessness prevention rate, as Sarah said, is the 58 per cent, and the current homelessness relief rate is 59 per cent. So, we've used those figures as our baseline.

So, if the reforms are able to increase the share of people prevented from becoming homeless by 10 per cent, and we think that is very possible, and the share of homeless people successfully relieved by homelessness is 10 per cent, the financial benefits are set out in the RIA. As I said, we do believe that the 10 per cent is achievable. We've done it before. It would mark a return to rates, as Sarah mentioned, prior to the COVID pandemic, when the prevention rate was around 67 per cent, and so that's an average taken between 2017-18 and 2019-20. So, I think, it's realistic to expect the challenges in implementation, which may result in changes to when and how the reforms take effect, and we've considered the impact this may have on our estimated cost savings.

So, our analysis suggests that the reforms can achieve a financial break-even and value based for many on a 6.1 per cent improvement on the prevention rate and a 0 per cent improvement in relief. So, I think that highlights where we're going with how we feel that we can get to this prevention figure. And when we include the societal benefits, the necessary increase in the prevention rates drops to only a 3.2 per cent increase. So, I don't know if there's anything you want to add on that.

Yes. It's probably just worth saying, as you've noted, that prevention is by far the most cost-effective way of ending homelessness. It avoids—. About the half of the cost of homelessness to Wales at the moment is in terms of TA—temporary accommodation, sorry—so reducing reliance on that as part of our rapid rehousing policy is really important. So, if we can prevent homelessness and avoid people going into temporary accommodation, we make a significant cost saving quite quickly. There are lower cost savings to be made in terms of relief, because you inevitably need to use services, then, if somebody is homeless. You need to make sure that they have appropriate accommodation. But our aims are around freeing up the other end of the system. So, the allocations provisions and the provisions around increasing discharge of the duty mean we're trying to free up that flow, so that people move through the system quicker. And if we do, we aren't using those services for as long and we reduce the savings, but it is a lower end saving, then. Preventing homelessness, as with many other things, is by far the best way of achieving the cost savings we need.

Thank you very much. Mike.

I'll just make a statement, and I don't expect you to answer it, but we have no more houses. So, you're putting some people in houses or flats, then some people will not be getting them. I'll save that up for the Stage 1 debate.

Okay, thank you. 

But we've got lots of benefits given: £1.2 billion benefits, £432.9 million of health benefits, £116.7 of earnings benefits over the appraisal period, and we've got £475 million of financial benefits and we've got £3.7 million for registered social landlords. Peredur knows what I'm going to say. Can you show us the workings? Can you provide us with the workings? Those numbers appear to be your estimates, but there doesn't seem to be any calculations behind them. Because you've obviously made assumptions. If you produce your calculations with your assumptions in them, we can then look at it and look at what other assumptions would give us. So, just through you, Chair, can we have the workings out?

Shall I talk about Alma Economics, because that might be helpful to talk to you around—. Alma Economics developed that cost-benefit analysis, which qualified the economic and societal cost reforms in our White Paper on ending homelessness. We also asked Alma to update estimates of costs and benefits related to homelessness using those existing models developed pre pandemic. So, that provided us with that baseline. They also used that work to develop, as I mentioned, that stock and flow model, which used aggregation analysis of stakeholder data to estimate annual average unit cost related to homelessness in Wales. So, to do those workings out, Alma worked very closely with officials and officials from UK Government departments and with data from local authorities and RSLs. 

In terms of the work that Alma have done, it's not really as simple to be able to publish that in the way that we perhaps would normally do that, because of the modelling tool that was used. But if it would be helpful, Cadeirydd, I'd be happy to ask officials to show you that and go through that with you at a later stage, if that's helpful.

I'd find it very—. Sorry.

Sorry, I was just going to say, what it is is it's a modelling tool—it's sort of a big workbook with lots of different formulas and different things in it, so it doesn't lend itself to publication or anything, but we could take the committee through it, if that would be helpful.

I once earned my living by writing computer models. One of them, which is coming into discussions now, was the blast furnace model in Port Talbot. The one thing I do know about computer models is they're all wrong, it's just how wrong they're going to be. So, I'm just interested in whether we can actually see the model. The model will be based on a number of assumptions, and those assumptions themselves will be the difference between it saving lots of money or not.

Yes, and I think Hannah can perhaps go into a bit more about those assumptions.

Yes. We worked with Alma to, first, cost the homelessness system as it stands. So, using survey data and engagement with local authorities and other partners, we've tried to cost how each element of the homelessness system works. People enter the system at different times, we've tried to cost how different stages work, and we've tried to cost the journey of a person through the homelessness system. That has given us some business-as-usual costs, which we can share with you as part of the model. We've then costed our interventions, as you've seen in the RA, and fed those into the model.

What we asked Alma to help us with was creating a model that we could adjust over time, depending on how this Bill develops through scrutiny, how our thinking might change on particular costings. So, it's a reactive model that we can adjust over time. We've used the best evidence we could to form that stock and flow model. We have Stats Wales data around outcomes, but outcomes data doesn't necessarily tell you about the flow through the system. You can't always pick out where somebody might have got stuck in the system, where backlogs have occurred, and people, as I say, don't always enter at the same stage. There are limits to that, but we do know about outcomes at each stage of the homeless system, and that's helped us.

And then we've worked by gathering data from local authorities to assess what those stages cost them and the kind of work that they do. We've tried to, through the model, cost what a TA unit would be, what a prevention case generally costs a local authority, the difference between general needs, accommodation and supported accommodation, so that we can try to adjust as much as we can for the very different experiences of homelessness incurred by different individuals with different needs.

The other thing that we have tried—. There is a lot of—well, not a lot, there is evidence in the field around what homelessness costs already. Most of that is done outside of Wales, and all of it was done pre COVID. So, another thing we've tried to do is update those costs for Wales. But in doing that, we've been able to utilise some of the existing research by PricewaterhouseCoopers or Crisis on the cost of homelessness to public services and base some of our assumptions on that.

Sorry, Mike—. So, what would be useful for us, going back to Mike's request, is maybe a technical note rather than a briefing on some of these aspects. I think, having done systems engineering in university—

I feel like we're surrounded by experts.

—if everybody is putting in their tuppence worth of what they've done, any model has weightings—

I'm glad I'm not the only one. [Laughter.]

Any model has weightings of different aspects of it. I think maybe an indication of where those weightings ended up, to see where you're putting your priorities or where the tensions are within the model, just to be able to understand—. I think that would probably go some way to be able to understand what we're looking for.

Absolutely, if that's helpful. As I said, if there are any further questions, we could share the latest version with you, if you needed anything after that. But just to say as well—I don't know if Hannah mentioned—the work relates to the White Paper as well, rather than the Bill, and that was due to time constrains. So, when Hannah was talking about it, it was used as a model that we could add in and take out costings as well—just to make sure that you are aware of that.

Just one statement and a question: economic models have always been very good at predicting the past. Gaps in data, which you talk about—how has that affected the accuracy of the estimates of the financial implications?

So, in terms of the gaps in data, obviously, as I mentioned, it has been tricky and we have used the best-quality evidence available to us at the time. We have prioritised official statistics and, again, high-quality published evidence, wherever we possibly could. Again, we have encountered areas where evidence is not available. So, when that has been the case, we've tried to triangulate the data with a range of sources. We've produced appropriate assumptions, which have been tested with key stakeholders, but the estimates that we have presented in the RIA are the most accurate estimates that we could produce, bearing in mind the data available to us. Again, it is important to acknowledge some of the challenges. I think we mentioned earlier on, our homelessness data is fairly high level, and we are taking steps to improve that data, but, for now, it's data based on the outcomes, which, again, has caused some issues for us, particularly in relation to mapping that data on the temporary accommodation to statutory outcome data.

I think, as Hannah said, really, the data captures the state of homelessness at one point in time, and so data used from recent years where possible. So, that data, again, is quite erratic, but we have done as much as we can with the data that is available. I think it is important that that should be noted by the committee. But we will, however, include the financial implications as part of the post-implementation review of the legislation as well. So, we are aware of the data, but we have done our very best using those official statistics to triangulate.

There are also—and this is going into policy, before you stop me at the end of this sentence—the hidden homeless, those people who are sofa surfing. If there is a means of finding housing, they will immediately become a cost on the system. I'll just leave it at that.

The Bill includes powers to make subordinate legislation. How important is that subordinate legislation likely to be?

Thank you, Mike. There are a number of regulation-making powers contained within the current draft of the Bill, and we do have a firm rationale for each of those, as you would expect me to say. For example, there are a number of regulation-making powers concerning the definition of local connection. That's setting exemptions to local connection referrals and entitlements to the main homelessness duty for those exempt from a referral. There are currently no plans to make changes to existing practice at this time, but if there should be, obviously, that would result in a net-zero cost to Wales as a whole, with changes limited to the financial burden on individual local authorities within the wider Wales envelope.

There are also regulation-making powers related to adding, amending or deleting public bodies from the list expected to support the ask and act provision within the Bill, and the duty to co-operate. Again, any changes to that list in the future will require future scrutiny from the Senedd and an understanding of any additional financial costs.

Then, in terms of allocations, there's a number of regulation-making powers in relation to establishing qualifying persons for the purpose of allocating social housing. There are regulation-making powers for further determining how a common housing register and an accessible housing register is kept and maintained, and that's, really, to provide flexibility to respond to potential differing approaches with the establishment of those registers. But just to say, additional financial burdens would be very likely to be limited, very limited, and would be administrative in nature. 

Just in terms of other powers, the Bill contains a number of guidance-making powers and a power to issue direction to social landlords, and that's linked to the duty on a social landlord to support a local authority, and those guidance costs are in the RIA.

Thank you very much. Sam Kurtz.

Diolch yn fawr, Cadeirydd. Good morning, Cabinet Secretary. You're broadening the definition of ‘threatened by homelessness’ and therefore are anticipating an increase in cases of around 15 per cent at a cost of £2.3 million per year from 2027-28. What evidence is that 15 per cent increase based on? 

Diolch, Sam. The proposed changes, alongside other prevention measures, ensure that our homelessness system is placed in a much stronger position with a stronger emphasis on that prevention of homelessness.

The greater emphasis on early intervention will increase the overall rate of prevention that is required to achieve the net financial benefits offered through the core scenario in the RIA. To achieve this long-term benefit, initial investment will be required to ensure that local authorities are able to refocus services on prevention. As I say, some of those local authorities have already done that, and that's to make sure that there are effective measures to keep people in their homes.

In terms of the 15 per cent increase in the RIA, we considered a range of potential increases in presentation of people threatened with homelessness within the RIA, ranging from a possible 5 per cent to 15 per cent and 30 per cent. Following feedback on our original RIA, we've used a central estimate of 15 per cent as the basis for our assessment. We think it's a reasonable assumption—

So is that a worst case or is that a best case, or is that in the middle?

It was in the middle, so we went up to 30 per cent, but it is based on forecasts that show a moderate increase based on current practice.

Sorry, I cut you off there.

No, not a problem. And combined with our policy intention, because obviously we'll be asking people to do that asking and acting before, so we know that that will perhaps increase there. But I don't know if Hannah wants to say a little bit more about how we came to that exact figure.

Yes, of course. We've already noted the forecast moderate increase to homelessness we expect just from the business as usual scenario. We're also mindful that in introducing things like 'ask and act', we are trying to pull people into the system at an earlier stage and we wanted to account for that. If our assumptions on 'ask and act' came to fruition, and it's hard to say exactly what those numbers would look like, it could be a 20 per cent increase at that early stage.

But we wouldn't know how many of those referrals would be completely new. We should assume that some of those people would have found their way to the homelessness system eventually, but at a later stage. And we don't know what stage we might identify those people. So some people identified through public service early intervention may well be in the later stages of homelessness and need to go to a different stage of the system.

So we felt that 15 per cent was a reasonable estimate between those two parameters. We think it's at the upper end of the parameter to try to account for that. As we've said, really, the entire aim of this legislation is, if people need homelessness assistance, that we help them as early as possible. So that's what that 15 per cent accounts for and that's the evidence to support it.

Just quickly, if it was 30 per cent, what was that top figure? Do you remember?

It's in the RIA. I wouldn't know off the top of my head. 

That's fine—

It would be 11,500 cases compared to the just over 10,000 that we've estimated for. 

And then in terms of local authorities, changes related to the prevention, support and accommodation plan will require 64.7 full-time equivalent staff, costing £3.1 million per year for 2027-28. Given this and the other changes that require more resource, becoming resource heavy, how are you expecting local authorities to deal with the increased case load? 

I think that's an important question. The RIA does recognise the additional costs and the capacity that's needed for this Bill, but I'm clear it's not about make do and mend. I don't expect local authorities to deliver the necessary changes without that additional resource. I'm clear that initial investment is very much required to refocus the homelessness system, as we have been doing over the last few years as well. And I think that is very much reflected in the RIA.

The prevention, support and accommodation plan—I'm getting used to knowing that as the PSAP; it's something that I'm aware of that I realise is just coming into the language of everybody else—is very much a new tool. It's delivering person-centred homelessness services and it will form the basis of service tailored to individual needs. Within the RIA, we have sought to cost a process that reflects the level of time, detailed thinking and ongoing review necessary to ensure that plans are provided for those individuals. But I do very much recognise that there is investment required initially, and, as I said, that's very much referred to in the RIA.

Are you convinced that that funding will come solely from the Welsh Government, or are you expecting local authorities to have to burden some of that increase through further council tax rises?

As I said, in terms of the investment that we're putting in, we've already put some initial investment in. It will need some more initial investment. Obviously, you'd understand that future budgets for this Welsh Government are—

In the future. 

They are for future Governments. I can't commit to that. But the RIA does recognise that initial investment would be needed, and I would obviously want to say that future Governments will need to look at that. 

So at this point of time, it's solely funded by the Welsh Government. 

That is where our investment is coming in, yes.

Just on that as well, if you're looking at 64.7 full-time equivalents, that's obviously, on average, three full-time equivalents in every local authority. Are you expecting those to be extra people found, rather than from within the pool of people that are already there, and, obviously, backfilling those roles if they move? I don't know if you can talk a little bit about that.

I don't know if you, Hannah or Sarah, want to say a bit about the work that's been going on with some local authorities at the moment, and how you've seen that in practice.

I'll say a little bit about PSAP, and maybe I'll hand to Sarah. We have costed it as an additional cost. That was quite important to us, because what we're trying to reflect here is our policy intention—that the homelessness system becomes more individualised. As you might have seen in other committees and in the supporting documentation, we've worked with people with lived experience to write this Bill, and all of the proposals are rooted in that lived experience of homelessness.

One of the themes from that engagement was that the homelessness system can feel very technical, it can feel like professionals involved in it are focused on delivering the technical aspects of the legislation and delivering the process, and, sometimes, the person's needs get a bit lost in that. The causes and consequences of homelessness are very broad, they relate to a lot of different types of life experience, they can relate to a lot of different support needs. So what we're trying to do through the PSAP is make sure that that service reflects those support needs in a better way so we aren't just focused on getting the roof over the person's head, we're focused on meeting the person's needs, so when that roof is put over their head, it is a sustainable roof, in the right place, at the right time, in the right community.

We've costed this additional need to reflect that it is quite a different way of working for homelessness services, focused on people, focused on their individual needs, and thinking in a much more holistic way around how homelessness equates to wider support needs. The quick answer is it is an additional ask, we think it is more people. But I just wanted to explain why we're trying to shift the system in that way.

Thank you. Only quickly, just to add that, as the Cabinet Secretary mentions, there are some local authorities in Wales who are already restructuring in anticipation of the Bill and the shift in focus on prevention—so, already looking at the structures. But the RIA doesn't assume local authorities will do that, as the Cabinet Secretary says. We've been really clear we've costed the additional ask through the Bill, and we've tried to capture everything in terms of the RIA that we think will be that additional requirement as a result of the changes to the system that will be required through the Bill.

Just following on from that, if there are certain local authorities that are already doing some of this, they wouldn't be disproportionately affected—. Because they've done it already, they wouldn't get the extra funding. Listening to the answers you've given, effectively local government are going to be fully funded for these roles, so if the roles already exist they'd still get the money to support those roles.

As this will change their statutory responsibilities and statutory duties, obviously everything that Welsh Government funds in terms of the RSG, the local government settlement, the funding is distributed on a formula basis at the moment, so any additional funding that might be forthcoming from a future Government would obviously be—. The intention would be to distribute on the same formula basis, which looks at the differentials in need in different areas, and there are lots of different components. I'm sure you have lots of discussions about local government formulas.

I'm sure you're aware of the local government funding formula. Just to say in terms of—just sort of a side point, though, but an important one—local authority engagement that we've had throughout this whole process in the making of the Bill, the creating of the White Paper, and then, obviously, the decisions that are made around how the Bill is. We have had really good engagement with local authority colleagues. They've been very forthright in their views on the White Paper as well. They do very much understand the aims of the Bill. Continuing to work with local authorities on this, as we have done through the creation of this, is really, really important.

I know Mike wants to come in quickly, and then I'll bring Sam back in.

Have you made any distinction between those local authorities that are depopulating and those local authorities that are having huge population growth—Powys at one end, which is depopulating, and Cardiff, which is having a huge population growth? Are you distinguishing between those, not just in percentage of population? Demand is driven by population, isn't it?

Yes. You make an important point on our demography in Wales and how that's changing. We are already very much engaged with all local authorities at the moment with the system that we have. In terms of the future, as Sarah says, the funding is distributed on that in terms of the local government funding formula, which has a number—as you know, Mike—of indicators, many indicators within it.

I think, when we look at homelessness, where it's appearing at the moment, where we know the trends of people, obviously there are cities where we know there are urban issues, but we have to be clear that homelessness is something that affects all populations within Wales. Sometimes even within cities it's not in a city centre, it's actually hidden homelessness that you can find. I think it is important, as I do at the moment, going around Wales, talking to local authorities about homelessness services and the difficulties they have in terms of temporary accommodation brings their own challenges sometimes, because in some parts of Wales, they don't have the supply of housing as well.

So, it is quite a complex picture, and we know that people become homeless, or are homeless, for a number of complex issues, and it's not simple. I don't know if anybody else wants to add anything to that.

I think we'll move on.

I'm conscious of time. We had a long discussion there, Chair, so I'm happy to hand back, and if the following questions are followed up by letter, I'm content.

Fine. Okay. Hefin. 

You're done, are you?

Yes.

Okay. There's been a lot of discussion today about the assumptions in the regulatory impact assessment and where the outcome may be not what was expected. The RIA says it's got a likely overassumption that all households currently excluded from homelessness assistance would enter either temporary accommodation or supported accommodation, resulting from the abolition of differences in entitlement relating to priority need and intentional homelessness. That's been estimated at £5.4 million per annum. What happens if that overassumption actually comes in as an underassumption, and how will there be capacity to meet that?

Thank you. With all Bills, there's often a real focus on—. They are important aspects of the Bill, the removal of the intentionality test and priority need tests. As I say, there's a lot more to the Bill as well, but the RIA does recognise the additional costs and capacity required to deliver those proposed reforms.

The data in Wales does show the low numbers in Wales in terms of those priority needs. The last full year of data that we have access to, it's around, just under, 300. And in terms of intentionality and how many that's used for, it's 100 across Wales. So, we don't envisage any significant increase in the numbers of people experiencing homelessness, and we're not aware of any data to the contrary of that, but we very much do—. As I say, we had really good engagement with local authorities throughout this whole process and they do require time for those proposed changes coming through in the Bill, and that's something that we've also said in terms of the phased approach that we want to have to this Bill—so, the preventative services at one end, and, obviously, the removal of those tests more likely to be at the end of the Bill. So, while incurring additional short-term costs, increasing that access to homelessness services will likely result in longer term cost benefits.

It's also important to note that if the estimate is correct and people do not need temporary accommodation—. The wider Welsh public service at the moment is incurring a significant cost, associated with the detrimental impact on them, and, in addition, the severely negative impacts of street homelessness on that group of people. So, there are real costs—not just financial, but human as well. But we don't envisage—. When I'm talking about the numbers that we have, you're talking about just under 300 in terms of priority need, 100 in terms of intentionality, but we're aware that there's a shift within local authorities to actually change to this. So, I recognise that it is an important issue for local authorities, but, on the data that we have, that doesn't bear out, I think—. We saw increases from how we dealt with the homelessness system around COVID, where we had the 'no-one left out' approach, and that's where we saw the increase in the numbers. And, you know, we are here.

Yes. And you can only go on the data you've got at the moment.

Yes.

Let's look at some of the incremental issues that might occur as a result of some of the assumptions in the RIA. So, there's a section where viewing accommodation and protection property for prisoners is estimated at zero, because it affects so few people. That's fine if it's just a couple of cases, but could there be a cumulative effect, where a number of different scenarios occur and it actually becomes a more significant figure—from that zero estimate?

Yes. I don't know if there's anything—. Yes, Hannah perhaps can talk a little bit about that. But it is an important aspect of the Bill. As I say, we're looking at the data very closely, that we have, but, Hannah, do you want to—?

Yes, of course. So, I think we could have added more detail in the RIA, on reflection, on this. The thinking hasn't just been, 'This isn't that very many people', it's also about using the Bill to formalise practice that is very well under way and improving practice at no additional cost. So, to take the example of viewing, viewing is standard practice across the sector; there's no suggestion in this Bill that it isn't. But, for particular groups, people in prison, people in hospital, it can be quite difficult to view properties, and in the lived-experience feedback, we heard the occasional example of, 'I was offered a property, I couldn't view it and I was told, if I refused it, that was it.' So, we just wanted to just pull that together, really, and make sure that that's not something that is—. That should be on offer to people experiencing homelessness.

But what the Bill actually requires is that, either in person or otherwise, somebody is given the opportunity to see the property. And from what we can establish, if that was done via film or video, from what we can establish, most housing workers have the tech already to do that and they will already be inspecting properties in order to ensure that they're suitable for offer. So, we haven't been able to identify where costs would fall in delivering this fully, and similar across some of the other proposals. We will absolutely reconsider that if—

Is that why you've attached a zero cost to it?

Yes, that's why—because we couldn't identify any costs.

But there could be a cost nonetheless. 

We haven't been able to identify that. If we are, we will of course reconsider the figure.

Okay.

Just on that, you were talking about prisoners there and people in particular. What about localised issues for local authorities, for example Wrexham with Berwyn prison, people potentially coming out of prison and being homeless in Wrexham but maybe housing elsewhere, is there a cost related to that that's been modelled through to see how that flow works within RSGs and being able to support local authorities that might have a spike in that sort of aspect?

So—. Sorry, Cabinet Secretary.

No, no. You go. I was going to say about the local connection test. 

Yes. And the system and the reforms try to avoid that scenario. So, Hannah, I don't know if you want to talk through the changes specifically in relation to prisoners, because we're changing some of the timing of the duties for prisoners so that the prevention duty kicks in at an earlier stage and they're able to establish the local connection at an earlier stage. So, we know that that scenario that you've outlined happens at the moment, and the reforms are aimed at avoiding that, so that an individual—. We don't want them coming out on a Friday and then staying in the local authority where the prison is whilst all of that wrangling to establish a local connection takes place.

So, from the point of view of this committee, I suppose, it's the financial implications of that. Has that been modelled through well, and do you understand that and are able to reflect that in any funding that's needed in any local authority?

So, one of the things—. As I said, the RIA—. The work we're doing with Alma Economics, and the work for the Bill, was done alongside each other. So, one of the differences in the Bill that's different from the White Paper is that we'll be introducing a local connection test. So, the outline that you've said around Berwyn, for example, that would be quite challenging if we had stuck with what was in the White Paper, so we've developed this local connection test. As Sarah says, it'll mean that local authorities will be working with prison services and prisoners a long time before that, and that's something that will really help to identify where that local connection is. One of the larger aims of the local connection test being introduced in this Bill is entirely to do what we know is important, which is to use the money we have in Wales, the envelope of funding that we have, to make sure that we are serving the people in Wales in this system. Because I think that's a really important aspect of it, that we're dealing with what we can with our own population, and, if others are owed a duty in other parts of the UK, that is something that this Bill seeks to make sure, that they're referred into other services in other parts of the UK. 

Just to answer your specific question, yes, we've costed the changes we're making to the elements that impact on prisoners, and the scenario you're talking about is kind of how the system works at the moment. So, we're trying to improve the way the system works at the moment. So, we don't anticipate that scenario, because we're trying to change the system to avoid that particular scenario.

Thank you. Thank you. Back to you, Hefin. Diolch.

Do you want to go on to Part 2?

No, you carry on.

Part 2 of the Bill?

Where are we?

The financial implications and provisions in Part 2 of the Bill.

Yes, move on to that, so that we cover that in the last bit. Thank you.

Okay. So, that's it, the final bit, really, the home stretch, you'll be glad to know. [Laughter.]

Regarding the three local authorities that don't have a common housing register, and they've got to set one up, how's that going to be done, and is there going to be a burdensome cost on local authorities with that?

Like you say, 19 of the 22 local authorities already do have that common housing register, and, as was said earlier, it is important to remember that this Bill was designed with 350 people with lived experience, and one of the things that they told us was around how this process can be really cumbersome, can be really complicated, and it's really difficult to understand. So, the purpose of our common housing register is to ensure that there's that one-stop shop, really, and there's one route and it's easy to understand. So, it will bring clarity and fairness into the system. It doesn't mandate exactly what a common housing register should look like. There has to be that flexibility for local authorities to understand their population. So, they'll be establishing that and delivering it in a way that's best for them.

So, does that mean—

So, in terms of the cost, a lot of them already—

The current common housing registers are fit for purpose then, are they? And they would fit the requirements of the Bill.

I think that's for each local authority to decide. They have got an understanding of how we want to work in this system. I think it should—. We see the common housing register as actually a cost saving, because it means that there's—. Some of the registers that they do hold are—

It wouldn't be a cost saving for the three local authorities that don't have one now.

There's an initial cost to set those up, which is accounted for in the RIA, but, as the Cabinet Secretary says, we're not seeking to over-engineer common housing registers, really. We have two big policy aims. The first policy aim is that, for the applicant, there is a single point of access. So, if you live in an area without a common housing register, you might need to apply to multiple social housing providers in order to get your home—different systems, different levels of transparency, a lot of confusion. And again, as the Minister said, people with lived experience say that's quite an overwhelming experience. So, we want to make sure there is one access point. And we also want to make sure that all social housing is managed on one register so that it goes to those in the highest need.

So, you're saying that deliberately not being too prescriptive gives local authorities the chance to budget accordingly as well.

So, across Wales, there are—

I know what you're talking about, but it's the Finance Committee, so—

[Inaudible.]

The way that we have drafted the Bill is to try to account for the different types of common housing register that already exist. So, there are fully integrated, partially integrated, choice-based lettings, nominations. They could all continue, we think, under this drafting, but it pulls together those two big policy aims that help the applicant at one end of the system and gives clarity to all partners about what accommodation is being allocated.

And the RIA also assumes two full-time equivalent staff are required to deliver the aims. You've mentioned already the flexibility of the revenue support grant. Is this going to be sufficient then to deliver those two FTEs? Or would you need to additionally fund that?

I think that anything that we're doing at the moment is—. We'll be looking at that. It'd be for something in the future as well, for future budgets. But, at the moment, we believe, within the RIA, that's—.

It is additional funding. The aim of those transitionary posts is to prepare and manage the transition from the legislation we have now to the legislation of the future.

So, if the leader of Caerphilly council came to you, Cabinet Secretary, and said, 'I need some more money', would you give it to him to fund those posts? Just say, 'yes', and I'll let him know. [Laughter.]

I get asked by many leaders of local authorities. I think the important point is—. I think it's easy to get lost sometimes in this, because we're at this stage with the Bill now, but there has been a long development of this Bill, and so the engagement and some of the funding that's come through has helped local authorities perhaps prepare for what they have seen, the changing of the system, and I think it's not just a, 'We've done nothing, and we're starting now.' Local authorities have actually been able to put some things in place already, and we've got some really, really good examples of that, where it's actually making a real difference in anticipation of this Bill.

I think we've done the question—Mike did the question—on how you make an allocation to local authorities where resources are different and needs are different. Is there anything else you want to say on that, and how you might use some kind of system to allocate according to need through the RSG? 

I don't think there's—

If you feel you've covered it, I'm happy to move on.

Is there anything, Hannah?

Nothing from the point of view of the RSG. It might be worth saying that, from a policy point of view, and through the impact assessment process we've run, we've really tried to understand how the differing demography of different local authorities influences homelessness and tried to account for that in the policy-making process, so considering things like denser, Welsh-speaking communities or rurality or things like that.

I just wondered how the system is going to account for that, how the mechanism of delivering funding is going to account for that.

Sarah.

Yes, thank you. I guess it’s going back to the point earlier, in terms of the—. You will be aware as a committee that there’s constant consideration in terms of all the different metrics that make up the local government formula, and there’s a finance sub-committee, obviously, that oversees that. So, housing aspects are a key consideration of that group as well and feed into that formula. So, it would be through that mechanism that any changes would be taken forward.

Okay. I think that's helpful. 

Before we finish, just going back to the registers, are you missing a trick, especially in talking about the gaps in data that you’ve got? Are you missing a trick in not standardising a little bit of the data that’s collected through those registers, to be able to actually review? And I suppose it comes, then, to the post-implementation of this legislation—would that sort of thing be considered and looked at as you’re reviewing how well or badly the legislation has gone once it’s implemented?

So, our working understanding of how common housing registers work now is that they do hold quite a lot of data. The regs-making power we’ve included in the Bill gives Ministers powers to call in some of that information but also make requirements around what information is covered. So, we think they hold—

You've got some mechanisms there to be able to collect the data that you need to actually be able to make sensible decisions, going forward. 

Absolutely. 

And then, as part of the post-implementation review, can you just talk a little bit about that before we come to a close, really, just to see what your plans are about when you're going to review, and how?

Yes. So, just about the data, again, I think it’s a really important point, being clear what the data is. It can be erratic in places, and there have been those gaps. So, we’ve obviously got a long-term strategy to replace that aggregate data collection with that individual-led data collection. And, obviously, when that individualised system is in place, it will really help us move towards more timely data on that individual’s journey through the system. And it will help us with monitoring and evaluating the impact of reforms, particularly on the rates of prevention and relief of homelessness. At the moment, we collect data from local authorities to monitor homelessness in Wales through our monthly and biannual exercises. But we’re also going to consider the actual costs and benefits as part of the post-implementation review and our ongoing evaluation work, which I think will be really key to achieving the aims and ambitions that we have within this Bill.

Okay. Thank you very much for that. That brings us to time this morning. Thank you. Unless you were going to carry on with anything else—

No. Diolch, Cadeirydd, and thank you, and if there are any—

I didn't want to cut you off there. If you want to say something else, carry on. 

No, sorry. But if there are other questions that you'd like to ask, as I say, we will get them to you in as timely a way as we can, and we'll make sure we provide that note for particularly you and Mike, by the sounds of it, with your historical interest and experience in data sets.

Thank you very much for your time this morning. 

Thank you. Diolch yn fawr. 

Obviously, there'll be a transcript available for you to check for accuracy.