Finance Committee

Felly, fe wnawn ni symud ymlaen at ein heitem fawr gyntaf y bore yma, sef eitem 3, goblygiadau ariannol Bil y Gymraeg ac Addysg (Cymru), a sesiwn dystiolaeth efo’r Ysgrifennydd Cabinet sydd gyda ni y bore yma. Ydych chi’n fodlon cyflwyno eich hun a’ch swyddogion, os gwelwch yn dda?

Bore da, Cadeirydd. Diolch yn fawr. Mark Drakeford ydw i, yr Ysgrifennydd Cabinet sy’n gyfrifol am y Bil. Gyda fi y bore yma mae Bethan Webb, sydd wedi arwain ar ochr y Llywodraeth ar y polisi, a Joanne Corke, sydd wedi’n helpu ni drwy’r analytical services ac sydd wedi bod yn gyfrifol am bron popeth rydyn ni’n mynd i siarad amdano y bore yma.

Gwych, diolch yn fawr. Wrth gwrs, bydd y Cofnod yn cael ei yrru atoch chi ar y diwedd i wneud yn siŵr bod pob dim yn gywir.

Felly, fe wnawn ni symud yn ein blaenau. Fe wnawn ni gychwyn efo edrych ar gefndir y Bil yn ogystal â faint o randdeiliaid allweddol sydd wedi bod yn rhan o greu'r costau ac yn rhoi tystiolaeth i bwyllgorau eraill hefyd.

Er bod rhanddeiliaid wedi croesawu nodau’r Bil, cododd nifer o randdeiliaid bryderon ynghylch y weithred ymarferol a pheidio â chael yr adnoddau i gyflawni ei gynigion. Gwnaeth pobl fel Cymdeithas Genedlaethol Prifathrawon Cymru nodi bod cyllid i ysgolion yng Nghymru ar bwynt argyfwng. O ystyried y pryderon yma, pam mae’r Bil yn cael ei gynnig rŵan a sut fyddwch chi’n sicrhau bod adnoddau digonol yn cael eu cyflawni i allu cyrraedd y nodau?

Chair, on the question ‘Why now?’, this Bill has been long in preparation. You can date its origins as far back as 2017. So, the Bill is not something that could have taken anybody by surprise. It’s been in the Government’s legislative programme from the start of this Senedd term. It was part of the co-operation agreement between the Labour Government and Plaid Cymru. So, I don’t think anybody could fairly say that the Bill has been rushed on them.

I followed the evidence that other committees have taken closely, and one of the things that strikes me about some of the evidence that’s been given is that it hasn’t really given adequate weight to the fact that this is a Bill, the impact of which will be felt, not in the next half term but over the next 10 years. So, at the school level—and some of the evidence you referred to from the trade unions talking about the work impact for schools—it’s year five of this Bill before any impact at all happens at school level. So, the Bill is designed to be incremental in the way in which it is implemented, and flexible as well.

The actual costs, the new costs, are accounted for in the regulatory impact assessment, but the Bill is essentially about making the very best use of resources that are already there in the system, because you know that the purpose—the overriding purpose—of the Bill is to make sure that the education system, and all the efforts that it currently makes, are aligned with the ambition of creating a million speakers of the Welsh language by 2050. And there are extensive resources in the system already; Welsh has been a compulsory part of the curriculum in Wales since the Education Reform Act 1988. So, this Bill is about using the resources that are already there, and adding to them where we need to make extra effort to maximise the contribution that the system can make. It is not about a day-zero approach to things, and it is certainly not about costs that are going to rapidly come down the line either at local education authorities or at schools themselves.

Felly, wrth ichi rampio i fyny dros y pum mlynedd, saith mlynedd, degawd nesaf, sut mae'r ariannu a rhoi'r adnoddau i mewn yn mynd i gael eu monitro, a sut mae e'n cael ei adlewyrchu yn yr RIA i ddangos bod yr adnoddau—? Achos wrth ichi rampio i fyny unrhyw beth, dŷch chi'n gwario mwy o amser a mwy o adnoddau'n ei wneud o, felly mae'n mynd law yn llaw bod angen rhoi rhywbeth i mewn—fedrwch chi ddim jest cadw pethau'r un peth a gobeithio am rywbeth gwahanol.

The starting point is that the Bill does assume that the investments that are currently being made will be sustained into the future. So, that's the foundational principle of the Bill: that the significant resources that are there now will continue to be available. The RIA then does set out, at the different milestones of the Bill, where additional costs might be incurred—very often they are opportunity costs rather than actual financial costs, although they are monetised for the purpose of the RIA.

One of the things that becomes clear when you read the RIA is that providing very specific figures about investments in the future has to take into account the very, very wide nature of the sector itself—schools that go from a very small primary school in an English-speaking village in Monmouthshire to a large secondary school in the west of Wales serving a largely Welsh-speaking population. So, the size of schools varies, the linguistic character of schools varies, the geography that they serve varies, the buildings that they are in and the resources that they have vary.

So, the RIA offers the best estimates we can, but says regularly, as you will have seen, that we will need to return to these figures, because one of the other things that the Bill will do will be to improve the data that we have about the nature of Welsh language teaching, particularly in mainly English schools in the future, and that those figures will need to be returned to as that data improves. But the RIA provides the best estimates that we are currently able to provide, using the data sources that are already available.

You said there that you have been very precise with the estimates and that you have given precise figures and not a range of things. So, what's the rationale behind that? Because, as you say, as you move forward from today to five years, things are going to flex. Why not provide a range of, 'This is the minimum to the maximum'? Why give one precise figure? What's the rationale behind that?

Ministers are advised by our experts in knowledge and analytical services on the best methodology to deploy when providing an RIA. In this case, the advice was clear that had ranges been the preferred method for describing and capturing the costs involved in the Bill, the ranges would have been so wide that they would have effectively been meaningless, because you would have had to have captured, as I say, costs from that end of the spectrum where you're talking about a very small school where very little Welsh is currently taught, to a very large school where Welsh is the medium of instruction, and everything in between. So, the preferred method, the best method for conveying to the reader the costs that we believe will be associated with the Bill in the state of knowledge that we currently have is to provide a midpoint figure.

Ffigur pendant, nid ffigurau sydd ddim yn dangos y ffaith bod pethau'n fwy amrywiol ar lawr gwlad.

So, there are many instances when those costs are known and can be quantified and we can be confident about them. And the advice that Ministers have—and I certainly myself endorse it—is that the way in which figures have been assembled for this RIA is the best way of reflecting the nature of the sector and the nature of the costs that are likely to be incurred.

Achos fel Pwyllgor Cyllid, dŷn ni'n gweld Biliau yn rheolaidd yn dod gerbron, ac mae rhai yn rhoi amrediad, ac mae rhai yn rhoi—. Beth ydy'r maen prawf? Ydy hwnna'n ddibynnol ar y Gweinidog, neu'n ddibynnol ar yr advice rydych chi'n ei gael? Pam mae hwn yn wahanol i rai o'r Biliau eraill dŷn wedi'u gweld, lle mae amrediad yn cael ei roi?

Of course.

Well, essentially Ministers do rely on the advice that you're given, because we have an analytical knowledge services unit in the Welsh Government who do this all the time, and whose job is to make an assessment of the best method that can be used to give the most useful description of the costs that a Bill incurs. I am, I guess I would have to admit, a slightly unusual Minister in this way, in that I have taught research methods at the university here, so I have a view of my own as well as the advice that we are given.

I think the method that will be used will vary from Bill to Bill because the subject matter of the Bill will lend itself either towards a range, because that range is narrow enough to tell you something meaningful, or as in this case, where the range would be so wide that it would essentially be an arbitrary set of figures that the committee would have in front of it, so that it would be very difficult to derive anything meaningful. Now, in this case, the advice we had, and my own assessment as well, is that offering these midpoint figures gives the committee and those who are interested in the Bill the most insight into the likely costs and the future trajectory of those costs.

Before I bring Peter in—.

Rydych chi'n sôn yn fanna am y midpoint. Ydy hwnna hanner ffordd rhwng y top a'r gwaelod, neu ydy o'n agosach at y gwaelod neu'n agosach at y top? Sut mae ffeindio hwnna? Sut mae'r average point yn digwydd? Jest o ran y fethodoleg.

No, it isn't a midpoint in a range. It is taking the specific costs and giving you the actual figures that we think best reflect the current state of play. Now, to give you just one example, we provide a cost that we say a local authority will incur in analysing the school-level delivery plans, and for that we make an assessment through our contact with local authorities about how much time that would be, what sort of person you would be looking for, and then we give you a cost for that. Now, I recognise a small local authority with a small number of schools may not need that amount of time, and a very big local authority may need a bit more than we've offered, but we have given you what we think would be the standard cost, recognising that it will vary up and down.

Diolch am hwnna. Peter.

I think you've answered my point, actually, Cabinet Secretary, because the overall figure—I can see exactly what you're saying. The range of things to accommodate—it's very difficult to separate those down. So, we can be confident that the total cost or the element relating to schools has built in those assumptions from each school, each local authority, to come to as near a figure as we can. Obviously, for a small, generally English-speaking school, perhaps in a border county, there would be additional costs that may have to be incurred in staffing, for instance, I suppose, if there weren't the skills to—. So, I can see the flexibility, but I suppose what we'll need to try to ascertain is how do those assumptions come about and how they are created in this global picture, I suppose.

Maybe the methodology of—. Is it similar to what Peter was explaining, having been through the numbers, and how did you get to that point?

In terms of the local authority cost, in terms of looking at the—. Sorry. In terms of the local authority costs, we had a range of different situations described in the responses that we had from local authorities. And as the Minister said, there were certain job titles and we were looking at the job titles and what kind of a range they would be on, the scale for local authorities, and then we were making an estimate of the—. I think we came up with 0.4 full-time equivalent. That would be from year 6, I think, on that one. Instead of just year 3 and year 8 when the delivery plans would be approved, we recognised that there was ongoing engagement with each individual school within the area, so it's a recurrent cost that goes across to 12 months from year 6 for that one. So, that's how we came to the estimation for that one.

In terms of the schools and the cost for workforce development, what we recognised was a range, or the different types of experience of different schools were so wide that what we recognised was that the money within the system to support workforce development was a key driver. That was the key information that we had back from all of the stakeholders that we spoke to in all of the schools that responded, that that workforce development was going to be key. Hence why the RIA sets out that the money within the system needed to be retained throughout the period of the RIA assessment in order to continue to support that.

Now, some schools would have little requirement in that regard and some schools would have a higher requirement in that regard. So, in the first instance, we've said, until the new Welsh in education strategic plans are developed with more detail, and until the national framework is set in terms of workforce targets, et cetera, then at that point it would be appropriate to use that information to provide, perhaps, more detailed estimates that we would have at that point. So, that's how we went about looking at both those different types of circumstance.

Mike.

Do you expect to follow normal distribution, and have you got any confidence levels?

We didn't extrapolate from base data; because the base data was so wide, we just felt that that would be arbitrary, as the Minister set out, hence why there wasn't a range. What we've done for the school workforce activity is to say that the money currently within the system to support schools in doing that should remain, and then perhaps when we get to the stage of having that more detail within the new WESPs, et cetera, we would be able to undertake an exercise like that. So, we didn't undertake a modelling exercise for this, which would have, perhaps, been able to come up with some ranges, et cetera, but the base data wasn't there to be able to support that.

I was talking about that in another place, but the Welsh sabbatical, is that going to continue, because that has been the big thing for getting people who speak some Welsh to be able to teach Welsh, especially in primary schools? That's made a huge difference in the primary schools I know in Swansea. Is that continuing, and that's not within these costs because it's a current cost?

Those costs are reflected in the RIA, but it is assumed—it's £3.3 million for the sabbatical.

It's £3.8 million.

It's £3.8 million, I beg your pardon—£3.8 million. And the figures that you see in the RIA of investment that is required to continue through all the period include that figure.

But it's just a continuation of what's currently being spent—it's not new money. I know I'm boring on this, but I've tried to differentiate between new money and cash and opportunity costs that are different, and the big one is additional cash.

The Bill assumes that that level of investment continues into the future. Given that those are decisions that will be made by future Governments, in that sense, they will be additional, but then, they—. The Bill assumes that those investments will need to continue.

But in the same way, are we going to assume that Morriston Hospital is still going to be funded, whatever Government is in power after 2026?

Yes, absolutely.

So, it's just a continuation?

Yes.

Rhianon wanted to come in on this as well. Rhianon.

Diolch, Chair. Just very briefly, it's been very clear in terms of the emphasis on workforce development, and the reasoning around the wide cohort of schools' ecology, however, just a quick question, if I may, and if I've missed it, apologies: you said the schools that responded; what is the actual percentage or scale of schools that responded in terms of this?

I think it was—. Yes, was it 23 out of 295 that responded?

Ydy hwnna'n—? Sori.

In that element of the research and the evidence for the Bill, there were 23 schools: 11 primary schools, 11 secondary schools, one all-through school; there were six Welsh-medium schools, seven dual language schools, and 10 English-medium schools. To be clear, though, the information collected from the schools is not quantitative data, it's qualitative data. And 23 schools with that sort of range within it does allow you, at a qualitative level, to get quite a lot of insight into what schools believe the impacts of the Bill to be. There are other sources of data, including some quantitative data, but that wasn't the purpose of the schools survey, to produce percentages, and so on.

Ac mae hwnna'n rhoi digon o hygrededd i'r—. Oherwydd ei fod o'n quantitative nid qualitative, ei fod o wedyn yn rhoi blas i chi o beth mae pobl yn ei feddwl, yn fwy na beth mae'r ffigurau yn ei ddweud wrthych chi—ai dyna'r math o beth?

Wel, mae e yn rhoi blas o beth mae pobl yn ei feddwl, yr heriau maen nhw'n eu hwynebu, beth maen nhw'n ei feddwl fydd yn mynd i ddod atyn nhw yn y dyfodol. Wrth gwrs, maen nhw yn gallu rhoi i chi rhai ffigurau am beth sy'n mynd ymlaen yn yr ysgol benodol yna, ac mae hwnna'n help, jest i gael rhyw fath o insight i beth maen nhw'n ei ddweud wrthym ni. Ond nid dyna ydy pwrpas gwneud yr ymchwil yna. Y pwrpas oedd cael y qualitative data—

—insight into, not just figures, but how the system works and how it looks to people who are at the chalk face of it all.

Ac rydych chi'n hapus bod hwnna'n rhoi digon o amrediad. Un o'r pethau roeddech chi'n sôn amdano oedd y mathau o ysgolion a gwahanol ysgolion. Oedd o'n edrych ar amrediad daearyddol hefyd? Roeddem ni'n siarad yn gynharach am lefydd ar y bordor, neu mewn ardaloedd lle mae'r Gymraeg yn cael ei defnyddio ychydig bach yn fwy rheolaidd. Oedd o'n rhoi hwnna i chi hefyd?

Yes. We commissioned Arad Research to carry out this work for the Welsh Government because of their expertise and their independence. The information that could be derived from the 23 schools did cover a very wide range of the waterfront of schools in Wales, and sufficient for us to be able to get the sort of information we were looking for. Had we been looking for quantitative data, 23 would not have been enough. But for the type of data we were looking to that exercise to produce, I think it gave us a pretty wide set of insights.

And did it give you anything that you weren't expecting, or—?

I think the thing that maybe I was most struck by was just the sheer variety of experience at that individual school level. Schools reported very different things, in all sorts of different ways. Now, I don't say that was a surprise, but I think it was even more pronounced in the way that different schools described their own circumstances: how they went about it, how well prepared they were, what the gap would be for them, what sorts of resources they would need for the future. I think it underpinned the decision to go about the research for the RIA purpose in the way we did it, because the range would have been very, very hard to capture.

Okay. Diolch yn fawr. Mike Hedges.

Diolch, Cadeirydd. The RIA sets out transitional costs in Parts 1 to 5 of the Bill, which relate to Welsh Government staff time. Is it going to be new staff or is it opportunity costs of existing staff doing something different?

These are opportunity costs.

Okay. Thank you very much. The RIA sets out alternative options to achieve the policy intent of some proposals in the Bill without legislating, and I'm sure you remember that comprehensive education was brought in by a Department of Education and Science circular by Anthony Crosland.

Indeed.

Would there be any savings to do it via the equivalent of a DES circular—a sort of Welsh Government educational circular? Would that actually save money or would it make no difference in cost?

I think the calculation that the Welsh Government made in making that choice was less about cost and more about effectiveness of the policy. So, yes, there would have been other ways in which the policy could have been taken forward, including non-legislative routes, but the Government decided that, to be effective in marshalling the education system behind our ambitions for the Welsh language, you did need the force of statute behind it. And having made that decision, I think, if I've understood Mike's point rightly, the RIA does not provide costs for the course of action that the Welsh Government had decided not to take. It provides costs for the course of action we had decided to take and the reason for it was not financial, it was policy.

Okay. The cost for each school to prepare and publish the first and second iteration of the Welsh language education delivery plan is described as an opportunity cost. That means it's a non-cash cost, I hope. Are there any savings in terms of opportunity costs by schools working together with alike schools in the same area?

Well, the Bill does provide for schools to come together in that way. I myself believe that the case for doing that is pretty strong, particularly, maybe, in the south-east of Wales, where this Bill is concerned. We don't cost it for the Bill, because that is a choice that schools themselves would make. But the Bill does make provision for schools to be able to act in that way, and I think Mike Hedges is likely to be right, Chair, that schools getting together to do some of this work will be a more cost-effective way for them to do it.

Well, I only know Swansea in any detail, so, apologies for that, but Swansea has got a number of school clusters based around the secondary schools. I assume that other councils do it the same way, and I would expect the primary schools, within a secondary school cluster, to work with the secondary school to produce this. So, would you see that as an advantage financially?

I think that is potentially an advantage financially. To give an example that I'm more familiar with—

Cardiff.

—Fitzalan High School in my constituency serves seven different primary schools. They have lots of things in common because of the nature of the communities that they serve. Therefore, the sorts of challenges that you may need to address in a school-level implementation plan, there will be lots to learn from one another, and that would be effective in a cost sense as well as a getting-a-better-plan sense.

I'm not allowed to talk about the plan; I'm only allowed to talk about the cost. But I have views on the plan and on primary school clusters working together, but I just hope that they do it in terms of—as far as this committee is concerned—reducing the costs and producing better outcomes.

The RIA doesn't factor into its estimates the potential for some schools to incur additional costs if their delivery plan is rejected by the local authority—I would think that would be very unlikely—or changed by a school. Now, my experience of school governors is: we get a policy from the local authority, we change the name of the school at the top, and we change the name of the headteacher some way through, and then we approve it at an annual meeting. So, it doesn't actually cost anything to do that, apart from a bit of typing. I'm not sure why this would be any different, in terms of cost, by being done by the local authority sending it out and schools just putting their mark on it.

Well, Chair, the Bill does introduce a new system in which there will, for the first time, be a national framework, which will set targets. The WESPs will have to respond to those targets and show how they are to be delivered. And schools will, for the first time, have to produce individual-level delivery plans showing how they will translate the local authority's ambitions into activity on the ground. That is new.

Now, my own view is that this will not be an onerous burden on schools. They already produce school delivery plans. We do not expect those delivery plans to be lengthy documents. But, nevertheless, because it is a new thing, although it's an opportunity cost, we do monetise it in the RIA, and we monetise it at the cost of the most expensive person in the system—we assume that this will be the headteacher doing it, not somebody who's less expensive. And then we provide the cost for that in the RIA. I think those are costs that will be absorbed by schools, and they have five years before they have to do this. So, they have plenty of time to think about it and to prepare the ground through the school development plan work that they already have to do.

Diolch yn fawr. I'm happy, Chair.

With the headteacher costs there, in the RIA, you've put a cost of £7,100 on it. A headteacher’s salary is more than 10 per cent, and it's 0.1 full time, so, something doesn't—. I know my maths is basic, but £7,100 is 10 per cent of £71,000. Some headteachers are, obviously, on more than that. So, could you give us an idea—

And some are on less.

And some are on less. So, how's that come about?

We took the average headteacher salary in Wales, as set out in the school workforce annual census, divided it by 12, added 30 per cent on for on-costs, so, that may be some of the difference. And then, we've looked at year 5 and year 8 as being slightly different: so, the first time that you do a plan, it being six months at 0.1 per month for year 5, and then half that for year 8.

Because you've done it once, it's going to take you less time next time.

Yes, it's going to take you less time to do the basic—. Because you'll have done the education workforce side of it for your school and the plan, and then it will need to be updated rather than completely refreshed. So, when you do the calculations on that amount, it comes out as £7,116, so, we rounded to £7,100, as we have with all the figures, and then multiplied by all the maintained schools. So, it's a bit convoluted—apologies. But that's where the data came from—we used the base in the schools workforce annual costs.

Peter.

Have you made provision in here, or what's your expectation around school improvement services and the shared services for driving improvement, and then regulation through Estyn? I doubt there's much of an extra cost for Estyn, but what's your expectation on local authorities on improvement services?

So, Chair, there is a cost for Estyn, which is identified in the RIA. These are costs that Estyn itself produced. It assumes that they will have to review, in a thoroughgoing way, one of the WESPs each year, and that they will be able to continue to, in a more monitoring way, look at the implementation of other WESPs. And, temporarily, the figure's gone out of my head, but somebody might—. But it's a relatively modest cost.

It's £23,500.

£23,500 a year, and it's what Estyn itself tells us that they would need.

Peter Fox asked a very good question about the middle tier, as it's sometimes called, and, of course, the Government is concluding its review of those middle-tier services. So, the RIA will be, I anticipate, reproduced during the passage of the Bill, as it normally is, after Stage 2, and if that work has concluded by then, and there's more that we can say, then we will update the RIA to take account of the conclusions of the review.

When it comes to workforce planning—and you’ve made the assumption that the headteacher will be doing this—and when it comes to additional support within the implementation of this, can you talk a little bit about how you’ve costed that into the RIA? And is most of it going to come from existing staff, or are you looking at creating new roles that will help support, or help with that workforce pressure?

Well, the workforce pressure at the school level, Chair—we assume, as I said, that all the help that is currently available within the system will continue. And there are a whole series of programmes that the Welsh Government already provides, primarily through my colleague Lynne Neagle. So, we have costed into the RIA the cynllun pontio. We have included the Welsh-medium workforce capacity development grant. We have included our plans—or the Cabinet Secretary’s plans—to introduce a teacher retention bursary in the Welsh language field. We’ve assumed that the Open University’s work—which is another way in which we are able to widen the pool of people who can come into the profession—continues.

The sabbatical scheme has been, as Mike Hedges said, the primary driver for helping people who are already in the workforce, and who’ve got a Welsh language capacity already, but want it to be increased. That’s been the major driver—that’s included in the RIA.

That's been hugely successful in Swansea. The main thing is the hyder, isn’t it?

Yes, it is. It’s people's confidence to use the language they already have in the context of the classroom. Last year’s figures from the annual survey of the workforce said that there was over 6 per cent of teachers already in Wales—over 1,500—who reported that they were able to speak Welsh but didn’t use it in the classroom. Now, there are very good reasons why some of those people don’t choose to do that, but there will be people within that group who, given the necessary help and support, would be able to become part of the workforce of the future.

So, all of those things are assumed in this Bill. Those things will continue and there’ll be new possibilities. There’ll be new innovations through the work of the athrofa, for example, which is a new part of the landscape and which will be charged with making sure that there are new forms of support to the profession so that we can look confidently to create that workforce that we will need.

Great. Before I bring Peter in, there were three scenarios that you’ve talked about within the Bill. If it becomes law, you haven’t quantified scenario 1 and scenario 3, but have said that scenario 3 would lead to considerable impacts and costs for stakeholders. Why have you chosen scenario 2, and can you quantify the costs if scenario 2 does not deliver the outcomes and scenario 3 occurs?

Well, the three scenarios, Chair: scenario 1 is business as usual, and our assessment is, and the reason why we have a Bill is, that if we just let things continue as they are, then we would not be extracting the maximum impact we can from the education system for the 2050 ambition. So, we don’t cost it because we don’t think that it does the business.

The Bill is designed to be ambitious but achievable. It’s trying to find that spot where you are stretching, you are making better use of the resource that is there already, but that it is not so far beyond the capacity of the system that it becomes self-defeating. That’s option 2, and that’s why option 2 is costed, because that is the option that the Welsh Government believes offers us the best opportunity to achieve our ambitions.

Scenario 3 is a scenario where the efforts that the Bill makes don’t succeed. Well, we believe they will succeed. We believe the mitigations that the Bill produces will be sufficient to see to it that scenario 3 does not materialise. And that’s why we don’t cost it, because, in our view, it’s not the likely scenario. The middle scenario is what the Bill is designed to support and to promote, and we provide the costs for that.

But you've said that it would be a considerable impact and costs for stakeholders if scenario 3 did come about. So, you've got to have a basis to make that statement, I suppose.

Well, if things go wrong, then you'll have to make a greater effort because you won't have captured the contribution that we think is there to be captured. Now, in those circumstances, I don't think that you can say confidently that the way a Government would respond to that would simply be to use what this Bill says, only to try to make it work faster.

A 'once more with feeling' type of thing, or—.

Yes. There may be a whole other range of actions you would need to take if scenario 3 were to materialise. We don't believe it will. We believe that scenario 2 and the Bill itself and other things will give us that pathway to 2050.

Okay. Thank you. Peter.

Thank you. We may have covered this second question of mine. The RIA assumes a cost of £4.37 million per year for schools to constantly improve or increase their Welsh language provision, which it says is based on requiring funding levels similar to 2024-25. Can we get an explanation of the source of that funding and why it has been based on this? And how have you estimated that cost? I think you probably have covered it already somewhere.

Well, the costs are not so much estimates, they're actuals. This is what we spend now, and that £4.37 million is made up of a number of elements. The biggest one is the sabbatical scheme at £3.8 million, but it also includes things, for example, like the canolfan dysgu Cymraeg—the provision that is made through the centre. That is £0.5 million that is provided for that. There are staff costs, then, associated with managing that, which are on top of that. So, the headline figure, the combined figure, comes to £4.37 million, and then the RIA breaks down in one of the tables, which is probably at a level of detail that the committee hasn't needed to go to—. But in the tables, you will see the breakdown, and it's how it's actually spent today and then it projects that for the future.

Okay. Thanks for that. You say that the duty to specify the amount of Welsh language education for schools in each language category, including the minimum provision, will be set out in regulations and is likely to lead to financial implications for stakeholders. Can you advise on the costs of this, and are there any other areas of secondary legislation that will result in costs?

I think the answer to the final part of the question, Chair, is 'yes'. There are pieces of secondary legislation that flow from the Bill. I don't want to be too tedious for Mr Hedges, who had to hear all of this on Monday from me as well, but all the significant uses of subordinate legislation are subject to the affirmative procedure. So, when they appear before the Senedd, they will have RIAs of their own. So, the costs will be available to the Senedd in coming to its own conclusions.

As to the minimum provision matter, there is a 10 per cent minimum of the compulsory school day that will be expected to be devoted to the purposes of Welsh language education in schools that are primarily English in their medium of instruction. And there are costs that we provide in the Bill of doing that. By regulation, in future, following consultation—all the things you'd expect—the Bill has a regulation-making power for raising the 10 per cent. And at that point, the costs would need to be identified. I don't think I can identify them sensibly for you today, because, for some schools, the 10 per cent won't be reached until well into the next decade, because the Bill allows two periods of three years beyond the starting point for schools that are the furthest away from this to be able to reach that. So, we are talking some way into the future. But the Bill provides the scaffolding, which means that the costs will always be known to the Senedd in making its decision as to whether or not to support a proposal, for example, to go from 10 per cent to a higher figure.

Yes, okay. Thank you. The RIA suggests that a more efficient use of existing resources might absorb some of the additional costs for schools to improve or increase their Welsh language provision and move to a higher language category. However, it recognises the situation will need to be reviewed regularly to ensure this funding is sufficient to achieve the Bill's aims. Given the potential variation, why haven't you provided a range of costs? And I think we've already answered that as well.

It's the same answer, Chair, indeed, in relation to the range issue.

On the issue of effective use of existing resources, like I said earlier, Chair, primarily English-medium schools have had an obligation to teach Welsh since the Kenneth Baker Act. These are both Acts of Conservative politicians, Peter, so you'll be pleased to know this—[Laughter.]—and both Welsh Conservative politicians. So, the Kenneth Baker Act, a native of Newport, in 1988, and then the Wyn Roberts Act of 1966, which extended the age from 14 to 16 in which Welsh was a compulsory part of the curriculum. Schools have been doing this for a very long time.

Do we believe that all that effort is turning out young people who are confident speakers of the Welsh language? Well, I think the answer in any number of investigations, including investigations by Senedd committees, is it doesn't. And when we talk about more effective use of existing resources, I'd point him to the fact that a huge amount of time is spent now, because it is obligatory for schools to do it, and yet the results are not what they want to be. So, we think that this Bill will allow for better use of all that effort to more effective outcomes by using the new framework and the new approach that the Act sets out. So, that's what I mean by more effective resources—resources that are there already and are not doing the job we think they could.

I'm not sure what Bill was in place when I went to secondary school in Carmarthen in 1976. I was required to learn Welsh or have compulsory Welsh language at secondary level even then, all those years ago, and it didn't do me too much good, because I sort of left the Welsh involvement. But, clearly, the focus was on it at that point. Anyway, I digress; sorry, Chair.

My last question: the cost of late-immersion education is estimated to be around £16 million over the appraisal period, which represents seven years of costs. The RIA has noted invariably that the costs will fall between years 3 to 10, 2 to 10, and 4 to 10, which could amount to a difference of between £2 million to £4 million in additional costs. Can you clarify which period the costs will fall in?

I'm going to ask Joanna, if you don't mind.

Yes. Just looking at that part in the RIA, I think there is definitely a sort of inconsistency there. So, I think there's a typo on table 8, but, looking at table 8, just looking at my overall—

If it helps, if you want to come back to us with that, rather than—

It looks like the overall amount is correct in table 8 on the main cost of the Bill, just looking at the amounts there, but I would like to come back in writing, if that's okay.

Okay. That's fine. Mike.

I'm just going to go outside schools, because you have youth provision. I mean, at Twrw Tawe, in Swansea, which covers the Swansea valley and east Swansea, that provides youth service through the medium of Welsh. Now, at the next stage, it may not be in this Bill, but are you looking at that—at any expenditure on that to support this youth provision so that people can see Welsh as something outside of schools, rather than something to be used in schools?

That's Bethan, because it's more the policy side of it.

And I'm just conscious of time as well.

Can you write to us?

Unless it's—

Os wyt ti eisiau bod yn sydyn, ie.

So, the WESPs do include—I think it's question 5 of the WESPs—language use, and every local authority has a Welsh in education forum, and there are representatives from partner organisations like the Urdd and the mentrau iaith around the table to ensure that use is looked at in the wider sense, alongside school education.

Diolch. So, I'll bring in Rhianon now. She's sitting there patiently. I'll bring Rhianon in. Diolch, Rhianon. Hang on. Yes, that's it, you're on.

Diolch. Thank you, Chair. I'm going to move to the financial implications, if I may, of preparing the national framework and the amendment of the WESPs. So, the RIA says that there is a considerable variation in how the resources to administer and support WESPs are organised. How, then, have you reflected that evidence in the cost estimates, and on what basis have you estimated a cost of £4.5 million per year?

Thank you, Chair. So, the basis of that cost is the same basis as costs that we've already discussed, because it is derived from the grants that the Welsh Government currently provides to support local authorities to develop WESPs. So, that's the money that's already in the system. We assume, for the Bill, that those costs will remain into the future.

In our evidence from local authorities, as Rhianon Passmore said in her question, Chair, local authorities did reflect quite a wide variety of costs to them in preparing WESPs, but local authorities were very positive about WESPs and about their wish for them to continue. And that's one of the changes between the White Paper and the Bill, where we had proposed some changes to WESPs, but local authorities asked us not to do that, because they feel they've been a success story. So, we assume that we will continue to invest in them and that the costs that are there now will be needed for the future.

Thank you very much, Cabinet Secretary. The evidence from local authorities has also suggested it was challenging to quantify additional costs of analysing workforce data to plan the teaching workforce, which we've been discussing previously, and that those estimates varied widely. Can you explain on what basis this was determined—that it would cost precisely £11,600 per year per local authority and why there is no variation factored in on that, bearing in mind our early commentary around the very different ecology across Wales?

Well, the figure, Chair, is an estimate that a 0.2 full-time equivalent, the equivalent of a day a week, would be needed in a local authority in order to carry out that work. We assume that that is needed continuously, so, it isn't like when you're looking at a WESP and then it's over. These are workforce matters, so the workforce is required continuously, and when you take the level of local authority officer that we think will be needed for this purpose and you look against the Soulbury scales, then that comes out at £11,600 a year for each local authority. It will vary, as we've said throughout the last hour. What Ceredigion may need at one end of the spectrum and what Cardiff might need at the other will be different, but if you look at it in the round and you multiply that by 22, that's what we think that obligation will cost in cash terms.

Do you foresee that a person doing that role one day a week would then be potentially shared by local authorities, or is that adding to somebody else's role within a local authority?

I think we think it's more likely to be adding to the role of somebody you already have, but, as I said earlier, I think there are parts of the Bill where a shared resource might be best in some parts of Wales. It might not be everywhere, but, in some parts of Wales, I think it will.

Diolch. Rhianon.

Thank you. Diolch. On what basis, then, did Estyn provide an estimate of £23,500 a year to conduct the rapid review of one Welsh in education strategic plan, and what assurance have you been given that that estimate is reasonable? And what is the threshold for Estyn initiating such reviews?

So, again, Chair, just to go over what I said in an answer to Peter Fox, that figure is designed to cover two sorts of activities by Estyn. One is a rapid review, where something is thought to be going wrong and Estyn needs to go in and do a rapid review of what is happening. But Estyn will also be doing routine monitoring of WESPs that are in steady state and where they don’t need that sort of intervention. So, the estimate of £23,500 is based on the resources that Estyn itself has provided to us. So, these are the actual costs that it would assume. We rely on their expertise in that to give us what it would cost them. There's certainly no incentive for them to give us a figure less than it would cost them. So, that's the level of confidence we have.

I think the circumstances—the final part of Rhianon's question, how Estyn would decide how to trigger such a rapid review—probably that is a matter for them as the independent inspectorate, and I don't think I myself have any insight into how they would do that.

Okay. Thank you. I understand that. I'm going to move on to the financial implications around the National Institute for Learning Welsh and the cross-over to that. So, you're replacing the current National Centre for Learning Welsh with a new National Institute for Learning Welsh. How will that transition be organised so that it delivers the objectives of the Bill, and what evidence do you have to estimate the resources required by the national institute?

I thank Rhianon Passmore for that, Chair. So, in a way to just restate what Rhianon said, this is not one organisation becoming another. The current organisation, the national centre, will end, and a new organisation will begin. So, the transition arrangements are different in that circumstance than if it was one organisation simply morphing into another. And, of course, the Welsh Government has extensive experience over the years of that sort of institutional change on a much larger scale. Just this summer, the Higher Education Funding Council Wales came to an end on 31 July, and Medr came into existence on 1 August. So, there's a lot of experience in the Welsh Government of assisting in that sort of transition.

As far as the costs are concerned, the sources we have drawn on to provide the figures in the RIA for this Part 5 of the Bill include the current national centre's grant, the staffing structure of the current arrangements. We've looked at the annual accounts of the national centre. We've taken advice from the centre itself, its own expertise, as to what it thinks the costs of the athrofa will be in the future, and, of course, we have a considerable body of expertise within the Welsh Government in those civil servants who have been the sponsors of the national centre, who know how it works, who know what the costs are, and have used that expertise in providing the costs that the RIA does for what the athrofa will cost in future.

Diolch. Thank you very much. And finally from me, the Welsh Language Commissioner noted the costs of the national institute have not included programme costs, and she goes on to say that she describes this as crucial for the providers of Welsh language training courses for adults. So, is there any rationale as to why those programme costs have not been included, please?

Chair, first of all, I agree of course with the commissioner that there's no point in setting up a new organisation and it having no money to do what you've asked it to do. What the RIA, however, covers is the costs of the body, not the costs of its programme. So, the current programme is around £13.6 million, I think, that the centre has. And we would expect that funding to be available to it in future. It's not in the RIA because it's not the cost of the institution itself; it's the cost of what the institution will do. Of course, it will have to have funding to do that, otherwise there would be no point in having it. So, I completely agree with the commissioner—it's central—but it's not in the RIA, because the RIA doesn't have that as its focus.

In your letter to—. In your evidence, I think, to the Children, Young People and Education Committee, the cost of the new institute is a best estimate and you expect to provide a more accurate analysis of the costs once the second phase of the review of the school improvement is concluded in early 2025. Given that more accurate analysis of costs will only be available after the conclusion of the review, how confident are you in the accuracy of the current cost estimate, and what steps will you take when it becomes clearer?

Well, Chair, we are confident about the estimates that we use in the Bill. This is not one of those areas we discussed this morning where there's a wide range of costs involved. We have a clear idea of the staffing levels we would expect the athrofa to have. We know the running costs of the current centre and so on. So, this is one of those areas where the costs that you've got in the RIA are pretty dependable. Now, if the landscape changes, then those things may change as well. That's why that review is important and what I said earlier, that if it's concluded while the Bill is still in front of the Senedd, we'll reflect that in an updated RIA. But to give some comfort to the committee as well, the RIA is very clear at a number of points that this is a 10-year programme of work and that updated costs will be needed at various points as better evidence and actual experience becomes available to us.

Ocê, diolch. Mae gen i ddau gwestiwn, gobeithio, sydyn i orffen. Dwi'n ymwybodol o'r amser ac yn gobeithio y gwnewch chi faddau i mi am ofyn dau gwestiwn arall. Pryd fydd y gwerthusiad economaidd ôl-weithredol wedi ei gwblhau, sut ydych chi'n bwriadu ei gyflawni, a sut fyddwch chi yn monitro llwyddiant y Bil? Dyna ydy'r cwestiwn cyntaf.

Mae pennod 11 yn yr RIA yn rhoi mas y cynllun i fonitro'r Bil a sut y gallwn ni fynd ati i gael mwy o wybodaeth yn y dyfodol. Mae'n rhannu'r gwaith dros gyfnod y Bil i gyd. Mae'n dechrau ym mlwyddyn 1. Yn y canol, rŷm ni'n mynd i wneud ymchwil i weld a yw bwriadau'r Bil yn cael yr effaith rŷm ni'n gobeithio y bydd. Ar ddiwedd y cyfnod, ym mlwyddyn 9 a 10, bydd mwy o waith i'w wneud, fel mae'n dweud, mewn summative evaluation, a bydd yr effaith economaidd, rŷm ni'n meddwl, yn fwy tebygol yn y cyfnod yna, y cyfnod tuag at ddiwedd y cyfnod. Mae'r costau yn yr RIA yn adlewyrchu'r gwaith rŷm ni'n mynd i'w wneud. Fel gyda phob Bil sydd o flaen y Senedd, mae cynllun gyda ni i fonitro effaith y Bil a'r costau, ac i gymharu'r costau sy'n mynd i fod yna gyda beth sydd yn yr RIA yn bresennol.

Ocê, gwych, diolch yn fawr. Dŷch chi'n cydnabod y gallai fod angen am gyllid ychwanegol ar gyfer costau staff ysgolion a throchi yn yr iaith, a fydd yn cael ei ystyried yn ystod gweithrediad y Bil. A allwch chi esbonio'r sail ar gyfer hyn a pham nad ydy hyn wedi cael ei ystyried fel rhan o'r RIA?

Well, immersion is a relatively new part of the landscape and a very successful one, and, as a result, has grown significantly in recent times. The Bill assumes that everything that we are currently investing in late immersion centres will continue to be there, but the Bill also has a real ambition to grow late immersion as part of the work to increase the number of confident Welsh speakers. So, it's a hard area to give those sorts of future costs, because it's the most dynamic part, probably, of the whole current landscape. But we provide the costs of the current system. We've set out the ambition to do more, and I think trying to capture what that future might look like, and then put a cost alongside it—we wouldn't be giving anything that the committee could rely upon. It would be a best guess, but it would be at the guessing end of the spectrum, compared, for example, to the athrofa costs, which we've just described, which are much more—

Mwy pendant.

Ie, pendant.

Okay. There we are.

Gwych. Diolch yn fawr iawn i chi am eich amser y bore yma. Rydyn ni fymryn bach dros yr amser, ond ddim yn rhy ddrwg. Diolch i'r Aelodau am y cwestiynau, a diolch yn fawr i chi am ddod. Mi fydd yna dransgript ar gael i chi allu ei siecio, a dwi'n meddwl eich bod chi'n mynd i yrru nodyn atom ni ynglŷn â'r pwynt yna, i wneud yn siŵr ein bod ni'n cael y ffigurau yn iawn. Felly, diolch yn fawr.

Roedd un pwynt gyda Mike Hedges dŷn ni'n mynd i ysgrifennu amdano—youth services. Dŷn ni'n mynd i wneud hwnna.

O, ie, wrth gwrs. Gwych. Dyna fo. Diolch yn fawr.

I meant outside the Urdd and the formal ones—the sorts of informal ones, where young people who aren't into the Urdd type of thing get an opportunity to use Welsh in normal circumstances.

Informal settings.

Mentrau iaith and all that sort of—. All the other—. The things that go in the community more than in an educational setting.

Yes.

Wrth gwrs.

Gwych.

Grêt. Diolch yn fawr.

Yn unol â Rheol Sefydlog 17.42(vi), dwi'n cynnig ein bod ni fel pwyllgor yn penderfynu gwahardd y cyhoedd o weddill y cyfarfod. Ydy pawb yn gytûn? Dwi'n gweld nodio. Diolch yn fawr. Felly, fe awn ni'n breifat.