Y Pwyllgor Cyllid - Y Bumed Senedd

Finance Committee - Fifth Senedd

11/10/2018

Aelodau'r Pwyllgor a oedd yn bresennol

Committee Members in Attendance

David Rees
Jane Hutt
Llyr Gruffydd Cadeirydd y Pwyllgor dros dro
Temporary Committee Chair
Mike Hedges
Nick Ramsay

Y rhai eraill a oedd yn bresennol

Others in Attendance

Chris Vinestock Prif Swyddog Gweithredu a Chyfarwyddwr Ymchwiliadau, Swyddfa Ombwdsmon Gwasanaethau Cyhoeddus Cymru
Chief Operating Officer and Director of Investigations, Public Services Ombudsman for Wales Office
David Meaden Cyfrifydd Ariannol, Swyddfa Ombwdsmon Gwasanaethau Cyhoeddus Cymru
Financial Accountant, Public Services Ombudsman for Wales Office
David Phillips Cyfarwyddwr Cyswllt, y Sefydliad Astudiaethau Cyllid
Associate Director, Institute for Fiscal Studies
Dr Ed Poole Pennaeth Rhaglen, Rhaglen Dadansoddiad Cyllidol Cymru, Canolfan Llywodraethiant Cymru, Prifysgol Caerdydd
Head of Programme, Wales Fiscal Analysis Programme, Cardiff University's Wales Governance Centre
Guto Ifan Ymchwilydd Arweiniol, Rhaglen Dadansoddiad Cyllidol Cymru, Canolfan Llywodraethiant Cymru, Prifysgol Caerdydd
Lead Researcher, Wales Fiscal Analysis Programme, Cardiff University's Wales Governance Centre
Martin Jennings Gwasanaeth Ymchwil, Comisiwn y Cynulliad
Research Service, Assembly Commission
Nick Bennett Ombwdsmon Gwasanaethau Cyhoeddus Cymru
Public Services Ombudsman for Wales
Paul Davies Yr Aelod sy’n gyfrifol am y Bil Awtistiaeth (Cymru)
Member in charge of the Autism (Wales) Bill
Stephen Boyce Gwasanaeth Ymchwil, Comisiwn y Cynulliad
Research Service, Assembly Commission

Swyddogion y Senedd a oedd yn bresennol

Senedd Officials in Attendance

Alex Hadley Dirprwy Glerc
Deputy Clerk
Bethan Davies Clerc
Clerk
Christian Tipples Ymchwilydd
Researcher
Georgina Owen Dirprwy Glerc
Deputy Clerk
Joanne McCarthy Ymchwilydd
Researcher
Meriel Singleton Ail Glerc
Second Clerk

Cofnodir y trafodion yn yr iaith y llefarwyd hwy ynddi yn y pwyllgor. Yn ogystal, cynhwysir trawsgrifiad o’r cyfieithu ar y pryd. Lle mae cyfranwyr wedi darparu cywiriadau i’w tystiolaeth, nodir y rheini yn y trawsgrifiad.

The proceedings are reported in the language in which they were spoken in the committee. In addition, a transcription of the simultaneous interpretation is included. Where contributors have supplied corrections to their evidence, these are noted in the transcript.

Dechreuodd y cyfarfod am 9:02.

The meeting began at 9:02.

1. Cyflwyniad, ymddiheuriadau, dirprwyon a datgan buddiannau
1. Introductions, apologies, substitutions and declarations of interest

Bore da, bawb, a chroeso i gyfarfod Pwyllgor Cyllid y Cynulliad Cenedlaethol y bore yma. A gaf fi nodi bod yna glustffonau ar gael ar gyfer y cyfieithu neu i'w defnyddio ar gyfer addasu lefel y sain? Mae'r cyfieithu ar sianel 1 ac fe allwch chi addasu lefel y sain ar sianel 0. A gaf fi hefyd atgoffa Aelodau i ddiffodd y sain ar unrhyw ddyfeisiau electronig sydd gennych chi os gwelwch yn dda? A gaf fi hefyd ofyn a oes gan unrhyw Aelod unrhyw fuddiannau i'w datgan? Na, dim byd, diolch yn fawr.

Rydym ni hefyd wedi derbyn ymddiheuriadau gan Steffan Lewis a gan Neil Hamilton.

Good morning, everyone, and welcome to the meeting of the Finance Committee at the National Assembly this morning. Could I note that there are headsets available for interpretation or to be used for amplification? Interpretation is on channel 1 and you can amplify the sound on channel 0. Could I also remind Members to ensure that any electronic devices are on silent, please? Could I also ask Members whether they have any interests to declare? No. Thank you very much.

We've had apologies from Steffan Lewis and from Neil Hamilton.

2. Papurau i'w nodi
2. Papers to note

Ymlaen â ni, felly, i'r eitem nesaf ar yr agenda—yr ail eitem—sef papurau i'w nodi. Fe welwch chi'r papur cyntaf i'w nodi yw llythyr oddi wrth Ysgrifennydd y Cabinet dros Gyllid ynglŷn â newidiadau cyllideb amlinellol blwyddyn ar ôl blwyddyn, a hefyd mae'r cofnodion o'r cyfarfod a gynhaliwyd yr wythnos diwethaf i'w nodi hefyd. A ydy pawb yn hapus? Diolch yn fawr.

We'll move on, therefore, to the next item on the agenda, papers to note. You will see that the first paper is a letter from the Cabinet Secretary for Finance about outline budget changes year on year, and also the minutes of the meeting held last week—those are to be noted as well. Is everyone content with that? Yes. Thank you very much.

3. Bil Awtistiaeth (Cymru): Sesiwn Dystiolaeth
3. Autism (Wales) Bill: Evidence Session

Y trydydd eitem, felly, yw Bil Awtistiaeth (Cymru) a mae'n bleser gen i groesawu Paul Davies, Aelod y Cynulliad a'r Aelod sydd yn gyfrifol am y Bil. Gydag ef heddiw y mae Martin Jennings o Wasanaeth Ymchwil Comisiwn y Cynulliad a Stephen Boyce, hefyd o Wasanaeth Ymchwil Comisiwn y Cynulliad. Croeso i chi'ch tri. 

The third item, therefore, is the Autism (Wales) Bill and it's my pleasure to welcome Paul Davies, AM, the Member in charge of the Bill. With him today is Martin Jennings from the Research Service of the Assembly Commission and Stephen Boyce, also from the Research Service of the Assembly Commission. Welcome to the three of you.

Mi wnaf fi ddechrau, os caf fi, drwy ofyn: yn amlwg mae yna lawer o ansicrwydd ynglŷn â llawer o gostau mewn rhywbeth fel hyn a rŷch chi wedi defnyddio pa bynnag dystiolaeth sydd gyda chi a nifer o ragdybiaethau sydd ar gael, ond pa mor hyderus ŷch chi bod yr amcangyfrif o gostau yn adlewyrchu'r opsiynau'n gywir?

I'll start, if I may, by asking: evidently there is a lot of uncertainty around some of the costs in something like this and you have used whatever evidence you have and a number of assumptions that are available, but how confident are you that the estimated costs are an accurate reflection of the costs?

Rwy'n hyderus iawn fod y costau mor gywir ag sydd yn bosib. Amcangyfrif cost y Bil hwn yw £7.5 miliwn dros bum mlynedd. Mae'r asesiad effaith rheoleiddiol yn nodi'r amcangyfrifon gorau o effaith cost yr opsiynau gyda'r wybodaeth sydd ar gael i ni. Nawr, roedd yna ddiffyg gwybodaeth ar gael ar y nifer o bobl ag awtistiaeth. Roedd yna ddiffyg gwybodaeth ynglŷn ag amserau aros ac roedd yna ddiffyg gwybodaeth ynglŷn â gwasanaethau a ddarperir ledled Cymru. Wrth gwrs, mae hyn wedi peri pryder mawr i mi yn ystod y broses yma. Ond, er mwyn helpu llywio costau manwl, fe wnaethom ni berfformio ein hymchwil ein hunain: fe gysylltom ni â byrddau iechyd lleol yn uniongyrchol ar gyfer data amseroedd aros; fe gysylltom ni â Dr Dawn Wimpory ar gyfer trafodaethau manwl ar sut y gall y system gasglu data weithio a beth y gallai hi gostio. Fe gysylltom ni â nifer o academyddion blaenllaw sy'n ymwneud â chostau awtistiaeth ac fe gysylltom ni â Llywodraeth Cymru am amcangyfrifon o'r costau gweinyddol hefyd.

O ran costau Llywodraeth Cymru o ran darparu gwasanaethau cyfredol ac effaith newidiadau o fewn y Bil, dywedodd Llywodraeth Cymru wrthym ni mewn e-bost—ac rydw i'n dyfynnu yn y fan hyn:

Dim ond i gadarnhau ein sgwrs yn dilyn y cyfnewid llythyrau diweddar rhwng Ysgrifennydd y Cabinet dros Iechyd a Gwasanaethau Cymdeithasol a Paul Davies nad yw Ysgrifennydd y Cabinet o blaid darparu unrhyw wybodaeth bellach ar hyn o bryd. Fel y soniais yn fy e-bost cynharach, gellir dod o hyd i'r costau sy'n gysylltiedig â chyflawni gweithredoedd tebyg yn y memorandwm esboniadol a osodwyd ochr yn ochr â Biliau'r Llywodraeth yn y gorffennol.

Felly, roedd hi yn siomedig iawn fod y Llywodraeth wedi methu neu wedi gwrthod rhoi rhagor o wybodaeth i ni, felly rydym ni wedi cyfeirio at femoranda esboniadol blaenorol Llywodraeth Cymru, a gwybodaeth arall y gallem ei ganfod, er mwyn llywio ein hamcangyfrifon ar y sefyllfa bresennol a chostau'r Bil hwn.

Cysyllton ni hefyd gyda ASDinfoWales am wybodaeth mewn perthynas ag adnoddau hyfforddi a hyfforddiant a ddarparwyd ar gyfer gweithwyr proffesiynol addysg, iechyd a gofal cymdeithasol, ac mae ein costau ni yn seiliedig ar eu data nhw hefyd. Mae'r diffyg gwybodaeth ynghylch nifer y plant ac oedolion sydd ag awtistiaeth, y gwasanaethau sydd ar gael, a chanlyniadau'r gwasanaethau hyn yn dangos, rydw i'n credu, angen am y Bil hwn, ond mae hefyd yn arwain at heriau o ran costio effaith y Bil, oherwydd y diffyg gwybodaeth rŷm ni wedi ei gael. Ond rydw i'n hyderus iawn ein bod ni wedi cael llawer o wybodaeth a bod y costau yma yn realistig.

I am very confident that the costs are as accurate as possible. The estimated cost of this Bill is £7.5 million over five years. The regulatory impact assessment does note the best estimates of the cost impact of the options with the information available to us. There was a lack of information available on the number of people with autism. There was a lack of information about waiting times and there was a lack of information about services provided across Wales. Of course, this has been a cause of great concern to me during this process. But to help to shape the detailed costs, we performed our own research: we got in touch with local health boards directly for data regarding waiting times; we got in touch with Dr Dawn Wimpory for detailed discussions on how the system for gathering data can work and how much it would cost. We got in touch with a number of prominent academics involved with autism costs, and we got in touch with the Welsh Government for estimates of the administrative costs as well.

In terms of the Welsh Government's costs in providing current services and changes within the Bill, the Welsh Government did say to us in an e-mail—and I am quoting here: 

'Just to confirm our conversation that, following the recent exchange of letters between the Cabinet Secretary for Health and Social Services and Paul Davies AM, the Cabinet Secretary is not minded to provide any further information at this time. As I mentioned in my earlier email, costs associated with delivering similar actions may be found in Explanatory Memorandum laid alongside government Bills in the recent past'.

So, it was disappointing to know that the Government has failed or has refused to give us any further information, so we have referred to the previous explanatory memoranda by the Welsh Government, and other information that can be found, in order to shape our estimates on the current situation and the costs of this Bill.

We also got in touch with ASDinfoWales for information in relation to training resources and the training provided for professionals in education, health and social care, and our costs are based on their information too. There's a lack of information also about the number of children and adults who do have autism, and the services that are available and the outcomes of those services. That shows, I think, the need for this Bill, but it also leads to challenges in terms of costing the impact of the Bill, because of the lack of information that we have received. But I am very confident that we have had a lot of information and that these costs are realistic.

09:05

O ran ymateb y Llywodraeth, o safbwynt gwybodaeth, a ydych chi'n credu bod y wybodaeth rydych chi ei hangen gyda nhw, neu a ydych chi'n credu mai beth maen nhw'n ei ddweud mewn gwirionedd yw nad ydynt yn gwybod yr ateb i rai o'r cwestiynau yna?

In terms of the Government's response, in terms of information, do you think that that information that you need, do you think they have it, or do you think that they don't know the answer to some of those questions?

Wel, nid wyf yn siŵr. Nid ydyn nhw wedi egluro hynny. I gyd maen nhw wedi ei ddweud wrthym ni yw nad ydyn nhw'n rhoi'r wybodaeth inni. Felly, fe fydd yn rhaid i chi ofyn i'r Llywodraeth y cwestiwn yna, rydw i'n credu.

Well, I'm not sure. They haven't explained that to us. All they've told us is that they're not providing information to us. So, you'll have to ask the Government that question, I think.

Ocê, diolch yn fawr iawn. Mike Hedges.

Okay, thank you very much. Mike Hedges.

We do have to ask the Government that question. For us to be able to undertake our role of scrutinising this Bill, then we need to have accurate information.

Can I preface any remarks that I make here, or questions I raise here? This is solely about the financial aspects of the Bill, so the questions I ask have nothing to do with whether I actually support or do not support the Bill, although I have supported it up until now, but there are questions that will need to be asked about—. 

No—. Sorry. Although it's been said in here, I know a number of people in the autism community will be listening to this and will see transcripts of this, and I just want them to realise that this is not about whether it's a good idea or not; it's just about the financial side of it. I know you know that, as the Member responsible, but there are several thousand people out there who don't, and I think sometimes we talk amongst ourselves too often and not to the public.

And that sentiment, I'm sure, is reflected across all Members of the committee.

I know, but I think—. The first question I've got is: how do you decide that the Bill is value for money? I've seen costs, certainly for health boards and for other public sector organisations, in indirect costs. How do you work out that their spending of this money is value for money?

I believe it is. This Bill does provide value for money. Obviously, the Bill is intended to provide better diagnosis, targeted and more consistent services leading to better outcomes, and will therefore, I think, provide value for money on the basis of an improved system for individuals with autism.

Now, it also places and enshrines this regime in legislation, obliging the Welsh Government to undertake these activities and guaranteeing a higher quality service for people with autism. However, due to the lack of data available in Wales in terms of the prevalence of autism, we haven't estimated any savings in the regulatory impact assessment, which no doubt you've noticed. But in response to the health committee's Stage 1 consultation, I do note that the Association of Educational Psychologists stated that, and I quote,

'Overall, we would anticipate that improving the way in which mental health needs are met, not just meeting the needs of those diagnoses with autism, would represent a significant saving.'

So, they obviously believe that this Bill is value for money, and I think there is strong evidence of many benefits with this Bill. For example, the early reduction in numbers of people waiting for a first assessment meeting and better information in relation to people diagnosed with autism will actually lead to more impact from scarce resources, less likelihood that services are provided to undiagnosed or unrecognised autism, and better information to allow more detailed understanding of autism in Wales. Enhanced training will improve care services and diagnosis at all levels as well. Now, as I said, whilst no savings are quantified, the regulatory impact assessment actually makes reference to the National Audit Office work, which suggests that, actually, substantial savings could be achieved with modest increases in the identification rates. So, I believe that this Bill is certainly value for money.

09:10

Yes, it shows a £4.5 million cost to health boards over a five-year period, although, interestingly, unless I'm misreading it, it only actually shows an increase of £1.7 million over five years if we carry on as we are now, which is not, I would suggest, a very substantial figure. I know it seems anathema to health boards the length and breadth of Wales, but if they're spending this, how about making some savings further down the line?

You half answered this question on my last one, but is there any work able to be done by someone to actually see if we pick up a person at three years of age, how much we are going to save within the health system compared to if we wait until they're 11 years of age? What are those eight years of health visits going to cost? I think that an estimate of that by someone who would be considered to be an expert and unbiased would perhaps be helpful to us in looking at the overall costs. It always amazes me that, every time we find costs, the savings somehow seem to disappear into the ether.

Yes, and we have done some work around this, but obviously the lack of information available does make it difficult to actually estimate savings. However, a 1 per cent reduction in the annual direct and indirect costs associated with autism would actually result in a saving of approximately £1 million. But it should be noted, of course, that every effort was made to obtain reliable data from the Welsh Government, and those figures I've just quoted to you have actually been based on a recent comprehensive study called 'The Microsegmentation of the Autism Spectrum: Economic and research implications for Scotland'. So, what we've done is we've looked at that and we've applied that to Wales, and that's how we've actually come to that figure, but I don't know whether Martin wants to add something to that at all. 

Yes. It's difficult to put exact costs and savings, because we don't know the services people are provided or how they're apportioned around Wales, but we know there is inconsistency. There were examples of benefits of timely diagnostic assessments for children and adults, and things like supported employment schemes; they can provide individualised training and workplace support through job coaches. So, it's really about early intervention. We often talk about preventative investment. There are many approaches to early intervention. Not that many have been evaluated, but still they could benefit children with autism both with and without intellectual disability. Parent training and support programmes—. There was a lot of evidence in the microsegmentation report, which talks about the fact that, currently, there are sort of good costs and bad costs, so this Bill is about directing to identification, early intervention and stopping a lot of the bad costs, which are services that really aren't relevant and, potentially, could do, in some cases, more harm than good.

Welsh Government has recently updated its autistic spectrum disorder delivery plan. Have you determined the impact this has on your cost for option 1?

Again, the unavailability of information regarding the level of services currently being provided and people in receipt of those services would mean that, without information from the Welsh Government, I'd be unable to determine the impact on costs for option 1. I did ask the Cabinet Secretary what the cost of his code of practice would be during Plenary, and he responded, and I quote:

'The code is already planned in to the work that we have, so we have budgeted for it and expect not only to go through the process of consultation but to deliver services.'

However, it is worth noting that many of the issues that the Welsh Government is planning to address in this particular code are, of course, the same as those in my Bill. For example, assessment and diagnosis, accessing care and support, staff training, planning and stakeholder engagement and service planning and delivery. So, surely introducing the code would also use resources and cost money, because obviously the Welsh Government at the moment are saying that they don't want to see legislation being passed because they feel that resources will be diverted in order to introduce this legislation—

09:15

But they're already spending the money then on that code.

Well, it seems that way, but, of course, we don't know how much this code will cost because, obviously, when I asked that question, I didn't get that answer. And given that the Welsh Government is looking to introduce this code, surely it will cost money, it will use resources, but the difference between this legislation and the Government's code is that I've set out clearly the cost of this piece of legislation; the Welsh Government has not yet set out the cost of introducing this code.

I think that's answered my next question about the RIA.

Thank you, Chair. It is important that we understand the costings, and you've just identified that one of the big problems we've got is that we are unsure as to the costings of the current system, the code being proposed and the delivery plan. Which also means, therefore, perhaps that the costings for training and other aspects are not there as a consequence of that. Do you anticipate your Bill to add to the costings of training that the current plan would be—? Because if the Government goes ahead with its current plan, that's going to be spent anyway. The question is: will this Bill add to the cost of training as far as you are aware?

We have costed for that, and I can certainly provide more detailed information, and perhaps Martin could just very quickly—

The question I want to know is: is that in addition to what you anticipate to be delivered in training through the delivery plan and the code or is it in addition to that?

When we've been working with Paul and stakeholders around providing some costings, we didn't want to assume that the Government was going to continue providing training or provide more training. So, what we've done is estimate a range of costs—so a high range of cost being that nobody's got any training yet and a low range of cost looking at the amount of training that's been done by professionals, applying rates of staff turnover and potentially having more training. So, that gave us a range of just under £2 million to just over £3 million, and then we took a midpoint there. We didn't put in costs around training in option 1, because we didn't think—not knowing what the Welsh Government were doing and what they were planning on doing going forward, we didn't think we could net those costs off. So, in a way, it's sort of an overestimation to be on the safe side, really, as we don't want to be making claims that the costs aren't really as high as they potentially could be.

So, the possibility is that those training costs actually could be subsumed already in option 1 anyway.

Yes, and the Government does plan to provide training and continue to provide training, but not knowing what their plans are or what the costs are, we've costed what Paul's option is and we haven't taken the costs off of what we don't know is being delivered yet.

And you mentioned other things, you talked about some in the RIA, you used the Scottish example. Can you go through the process of the Scottish example? Are there any major differences between Scotland and Wales? Do they have better data than we have to be able to just apply some of the figures and therefore our figures may not be able to be equivalently justified because we haven't got the data to match up? Could you just go through that process for us, just saying why you chose Scotland, how you did comparisons between Scotland and Wales and the issues that appeared to be in Scotland—are they similar issues that we're seeing in Wales?

Because of, obviously, the lack of information that was available, obviously this particular report, which I referred to earlier, was actually appropriate, and that's why, obviously, we've looked at the Scottish example. Again, perhaps Martin can just quickly go through the process as far as that's concerned.

Yes. We wanted to look at the background costs of autism services that were delivered in Wales and also look at the indirect costs around employment, both people with ASD as they're young and in the future, and also parents and carers who were forgoing opportunity of employment or not able to have as many hours of employment. In I think it would have been the first consultation that Paul ran on the Bill, one of the witnesses—I can't remember whether it was the association of psychologists—pointed us in the direction of work that Dr Martin Knapp had done at the London School of Economics. So, we had a look to check to see how balanced and in-depth that work was, and it seemed to be the best work that was available on costings, so we got in touch with him. He then put us in touch with the authors of the Scottish report, if I can call it that, which at the time was being drafted, and they were very, very helpful and put us in touch with their lead researcher who had a number of discussions with me about the work they'd done, how that could be applied in Wales and any differences between local service provision in Wales and Scotland and policy. The Scottish report is based on the best international evidence and work within Scotland and the wider UK. And from discussion with her, it didn't seem that there would be significant differences at all. The prevalence rates will be the same, the support of issues and support needed would be the same, and the levels of support provided won't be massively different. So, it definitely was the best information that is available and could be used.

09:20

So, you managed to actually do a rough assessment of comparative—to compare Scotland and Wales and you came to the conclusion that it was appropriate and the analysis was able to be applied to Wales.

It's definitely the best information there.

Can I ask one further question? Clearly, we've talked about this, Paul—the ability to extend this Bill to other conditions, because this is specifically on ASD, but there are many other, as has been mentioned in debates, conditions that this could be applied to. Have you done an assessment as to the costs that might be incurred, because it will be by regulation? So, have you done an assessment of the costs that might be incurred if it is extended to a wider audience?

No, we haven't looked at those costs because obviously they would be future costs, of course—

I understand. I just wondered whether that had been done or not.

Sure. No, we haven't actually looked at those costs, no.

Still on the issue of additional and data collection costs, option 2 would incur £7.4 million of additional costs on top of the current provision. Did you consider undertaking a sensitivity analysis to provide a range of costs based on different scenarios?

Yes, there are two areas where there is more uncertainty in relation to the costings, so we have applied sensitivity analysis. The approach we've taken is to identify a range of costs, from which we've taken the midpoint for our cost estimate. The two areas are costs around reducing waiting times and indirect opportunity costs in relation to training, which has just been referred to. In terms of the numbers of people requiring training, we costed: all staff; all staff minus those who have already received training or would be expected to have received training in the next two years, based on annual training figures from ASDinfoWales and staff turnover rates. We've taken the midpoint between these two figures. The range of costs we identified is between £1,998,032 and £3,039,385 over five years—

—and the midpoint we've taken is £2,518,708. And for estimating the costs of reducing waiting times, we did contact local health boards to find both the number of people on waiting lists over 13 weeks and over 26 weeks, and this Bill does relate to people waiting 13 to 26 weeks. So, our low estimate considers the cost of assessing these people; our high estimate is for all people waiting over 13 weeks; and the midpoint reflects an assumption that the Welsh Government investment in multidisciplinary teams to reduce waiting times to their 26-week waiting time target will have had some impact on reducing the number of people waiting over 26 weeks by the time of the introduction of this Bill. And the range of costs we identified is between £3,148,500 and £5,576,625 over three years, and the midpoint we've taken is £4,362,625.

You may have touched on this, but why did the RIA not include additional costs for meeting increased demand for ASD services as a consequence of the Bill?

Sorry. Why did—my voice is going. Why did the RIA not include additional costs for meeting increased demand for ASD services? So, in other words, if the Bill itself then generates a greater demand, did you factor that in or is it based on the current—?

It's based on—. What I was talking about in option 1 and option 2 is an estimate of the costs of autism, which includes, I think, without looking at it—it includes, as I was saying, education, housing, employment support and lost cost income and a number of other costs. It's very, very detailed. So, that's covered in both areas and it's assumed that the Bill will—by identifying people earlier, there'll be different services required. So, rather than—I was talking about the good and the bad costs: it will be services that won't be provided because preventative interventions have happened, and there'll be additional services provided that relate to more tailored services.

09:25

It's kind of the reverse, actually, isn't it? If, as you say, preventative measures are in place, then you won't have costs, so it's almost consequential non-costs. Okay. A bit different from the normal situation.

Just on that point, I think it's important for us to explore this because with the Social Services and Well-being (Wales) Act 2014, it was always argued that if services became available, the possible demand of take-up would increase. I just wondered whether you've done any work to see if that has happened on that Bill, and if, therefore, there's a likelihood it will happen on this Bill. I understand you talk about preventative—and you might be able to—but will the uptake increase as a consequence? Because I know, and you know, many families are out there, desperate to have some support, so there is a large proportion out there looking to have these services. Have you done analysis or had any information from the social services and well-being Bill to see whether, as a consequence of that Bill, uptake increased, and could that be translated into this Bill?

I don't think that kind of data is available yet for the social services Act. So, we haven't really had that resource available to us. But also, I think there is a point to be made about—well, another point has been made about the preventative effects of earlier intervention. One example of that, for example, from Dawn Wimpory, was that, if those people being referred for diagnosis can be screened earlier, then some of those people who are inappropriately being referred can be diverted to other services. So, there are pluses and minuses, I think, on this, and it's quite a complex picture; it's quite difficult to quantify.

I appreciate there are pluses and minuses, but it's nice to know those pluses and minuses—what are the pluses and what are the minuses? I just wondered whether that was available—to look at one of the consequences that might happen.

Because of, obviously, the lack of information that has been available, it's been, obviously, difficult to even look at some of these issues. As Stephen said, we don't have data yet with regard to the social services and well-being Act, so it would have been difficult for us to look at those costings.

There is also evidence of different prevalence rates and assessments of ASD rates that suggests that it's not been done consistently. There is evidence that sometimes if people have waited a long time, they might feel they're owed a positive assessment; that came through in Dawn Wimpory's evidence. So, I think by improving the training and reducing waiting times, that would have another positive impact.

Thank you. The RIA mentions that NHS bodies will be able to use existing systems to collect data for children, but how confident are you that these bodies have the IT infrastructure in place to collect and share consistent data?

Paragraph 484 of the explanatory memorandum summarises the current child autism data collection system in Wales. The community child health 2000 database is in the process of transitioning to the new children and young persons integrated system database. These databases contain a module that allows for autism data collection, and this module has been successfully piloted, I understand, within the Betsi Cadwaladr University Local Health Board area since 2012. I've been advised by the Wales Autism Research Centre that it will be possible to capture adult data via the same system with little difficulty. Further data sharing via a cross-agency system to actually promote integrated services, including autism services, may also be achieved via the planned Welsh community care information system. So, I'm quite confident that, obviously, the current systems will be able to capture further data.

And, obviously, everyone's looking now at the impact of the general data protection regulation—GDPR. Have you considered whether this could have an impact on data collection and sharing and the related costs?

09:30

Yes, we've given careful consideration to both the impact of the GDPR on human rights obligations in actually developing the Bill. Paragraph 492 of the explanatory memorandum actually deals in detail with the relevant GDPR considerations, including setting out those bases on which sensitive health data may be processed. Also, paragraphs 493 to 495 actually set out the various safeguards employed within the NHS when processing sensitive person data. And it should be noted that these are both stringent and compliant with all necessary GDPR and human rights requirements.

Thank you. The other issue, going back to the capacity of NHS bodies, your cost model for data collection does assume one LHB will host the databases across Wales. Has this approach been discussed with the health service—with LHBs?

Yes, we've taken a number of views on board in consideration of the best approach to actually take to data collection. There are a number of projects that have been undertaken to attempt to collect information in relation to autism diagnoses. We've based our costings on the most proven successful model—that which has been successfully run, as I said, in north Wales by Betsi Cadwaladr university health board and Bangor University. 

Staff who've developed and administered the database for the last six years were very generous, I think, in terms of taking their time to help us to understand how the systems work, and how such a system could actually be compatible and rolled out across all local health boards, and the level in staff resources that would actually be needed to get such a system up and running, and then maintained. So, yes, we have spoken to the NHS about this.

Thank you. You spoke to us a bit earlier about the costs for the waiting times for diagnosis, and getting those down—clearly, this is a key aspect of the proposed Bill. But you've also referred regularly to the lack of data—begging, stealing and borrowing data, and reports and costings. We've also heard reference to the possibility that more people might present themselves for diagnosis, given that it might be more readily accessible. So, are you really confident that that particular aspect has been costed sufficiently? Because I think you've gone for a midpoint, as you explained earlier. If you get that wrong, then it's quite a big flaw, isn't it?

Yes, I am confident that the figures that we've presented are pretty accurate. Obviously, it is challenging, given the lack of information available, but I am absolutely confident that the figures that we have presented will be accurate and will make a huge difference if this piece of legislation is passed.

Okay. And just to reiterate the point, as well, that was touched on earlier: there is a risk, is there not, that the capacity of ASD support services would be undermined as a result of extra resources being required for diagnosis? That's a risk, isn't it?

There's always a risk, of course, but the transitional costs identified in the Bill would be additional to current funding, and it's not intended to replace or put pressure on existing services. This Bill is intended as investment in preventative services, which will, in time, reduce expenditure on reactive or acute services. But, of course, there will be costs in terms of implementing this Bill—we've made that absolutely clear, of course—and we've received considerable evidence that giving children and adults an earlier diagnosis, and having trained staff to deliver services appropriate to their needs, will actually result in better outcomes.

I thought about that Welsh proverb: there's no uphill without a downhill, and there's no early expenditure without later savings. I think that that's something that we need to address at some stage. I'll tell you what I know happens, and Paul can clarify if I'm right or wrong—people keep on going along, keep on going through assessments, nothing much happens, so they keep on going along and having assessments and, eventually, they get assessed as having autism after several assessments. Now, each of those assessments have a cost in terms of expenditure by the health boards and others. So, surely, if you get an early assessment, then you save the next five or six assessments. Every one of us, I'm sure, has had constituents say, 'I've been, I've been assessed, I've appealed, I've then gone back to be reassessed, I've then appealed, I've had my fifth assessment, and now they've actually realised that my son/daughter has autism.' If we're going to cut out all those, surely there must be savings somewhere along the line. I don't believe my constituency is unique in some of these thing. Sometimes, I say things and people seem to think it is, but there's lots—. The same person being seen several times has a cost to health. Do we know how many—and I know you're going to say the health Minister hasn't told you yet—but do you know how many assessments, on average, and what the range is for people going in to be assessed for autism before they agree they have autism?

09:35

Can I first of all say, Mike, that you've been a long-term advocate of spend to save, and this is what this piece of legislation is all about, in my view? Yes, we will probably have to spend around £7.5 million over a period of five years, but the savings, in my view, will be absolutely huge. And it'll make sure that people with autism will actually receive the appropriate and the right services, and that those services will actually be consistent across Wales; that's the whole point and purpose of this legislation in the first place. 

Can I ask one more question, particularly on that point? 

Mike has highlighted the possibility of assessment, appeal, reassessment. Will this Bill stop that?

Well, obviously, you know, I'd like to think that this Bill will ensure that services will be consistent and then people will be able to access those services quickly, and that those services will be available.

Nice answer, Paul, but the reality is, you can't be sure that it will stop that happening. Those costs could still be at play.

No, and I accept that. I accept that, of course. No-one can be certain, but I am confident that if this legislation is passed, obviously it will plug huge gaps in services and it'll make sure that services are consistent across Wales.

Nick wants to pick up on this as well. Sorry, Mike, I'll come to you.

Following on from Dave's point, I'm coming from the same sort of angle—I'm playing devil's advocate. You know I support the legislation, I think it's a good idea, but, of course, playing devil's advocate, the Government will say, 'This will cost money.' They've got the code, they're taking these steps, doing these actions. What Dave and what we're trying to ascertain is: what is the difference between having the legislation and simply having the code? Now, it'll obviously underpin it, but that's the key question that you're going to come up against, I think, in progressing this.

Yes, or what the additional cost of the Bill, your Bill, will deliver, extra to what the Government's is delivering.

The answer to that question is, we don't know how much the Welsh Government's code is going to cost because they haven't told us how much the code is going to cost. I've made it absolutely clear in my Bill the costs associated with introducing this piece of legislation, and that's the difference.

Could I ask just one final question? We passed the social services Bill, which had exactly the same, in my opinion—do you agree it had exactly the same open-ended potential increase in demand in that as we've got here?

Well, there's a possibility that, obviously, by passing legislation, you will see an increase in demand, but I'd like to think that the costings that we've put forward with this Bill will, of course, absorb that demand as well.

So, how would your proposed legislation, then, be monitored for effectiveness and value for money?

It's been made clear on the face of the Bill that it would require the Welsh Government to commission an independent review of its progress in achieving the aims of the autism strategy within three years of the publication of the strategy, and actually publish it as well. It'll also be required to publish further independent reviews at intervals of no more than three years after this. The Bill also requires that the Welsh Government publishes and updates reliable data on autism. So, we've made sure, obviously, on the face of the Bill, that it will be monitored and reviewed on a regular basis.

And explicitly so in terms of value for money as well.

Can I just clarify a point? The answer you just gave was—and if I'm wrong, please correct me—that the Bill will require the Welsh Government to make sure its strategy is updated and the cost of its strategy. But will it ask the Welsh Government to include the cost of the Bill as well as the Government's strategy?

09:40

Ocê. Unrhyw gwestiynau pellach?

Okay. Any further questions?

Any further questions?

Na. Iawn. Ocê. Wel, a gaf i ddiolch i chi, felly, am eich tystiolaeth? Mi fyddwch chi'n derbyn, wrth gwrs, copi o'r trawsgrifiad i chi ei gadarnhau e o safbwynt cywirdeb. Felly, diolch yn fawr i chi eto. 

No. Fine. Okay. Well, can I thank you, therefore, for your evidence? You will receive, of course, a copy of the transcript for you to check it in terms of accuracy. So, thank you very much again.

4. Cynnig o dan Reol Sefydlog 17.42 i benderfynu gwahardd y cyhoedd o'r cyfarfod ar gyfer eitemau 5, 7 a 9
4. Motion under Standing Order 17.42 to resolve to exclude the public from items 5, 7, and 9

Cynnig:

bod y pwyllgor yn penderfynu gwahardd y cyhoedd o'r cyfarfod ar gyfer eitemau 5, 7 a 9 yn unol â Rheol Sefydlog 17.42(vi).

Motion:

that the committee resolves to exclude the public from items 5, 7 and 9 of the meeting in accordance with Standing Order 17.42(vi).

Cynigiwyd y cynnig.

Motion moved.

Rydym yn symud nawr at yr eitem nesaf, sef y cynnig i symud i gwrdd yn breifat. Felly cynigiaf yn unol â Rheol Sefydlog 17.42(vi) fod y pwyllgor yn penderfynu cwrdd yn breifat ar gyfer eitem 5, eitem 7 ac eitem 9. A ydy'r Aelodau i gyd yn fodlon â hynny?

We will move now to the next item, which is the motion to go into private session. So, in accordance with Standing Order 17.42(vi), I propose that the committee agrees to meet in private for items 5, 7 and 9. Are all Members content?

Diolch yn fawr iawn. Felly mi symudwn ni i sesiwn breifat.

Thank you very much. So, we will move into private session.

Derbyniwyd y cynnig.

Daeth rhan gyhoeddus y cyfarfod i ben am 09:40.

Motion agreed.

The public part of the meeting ended at 09:40.

10:00

Ailymgynullodd y pwyllgor yn gyhoeddus am 10:02.

The committee reconvened in public at 10:02.

6. Ombwdsmon Gwasanaethau Cyhoeddus Cymru—Amcangyfrifon 2019-20: Sesiwn Dystiolaeth
6. Public Services Ombudsman for Wales Estimates 2019-20: Evidence Session

Croeso cynnes—croeso nôl i gyfarfod y Pwyllgor Cyllid. Rŷm ni'n symud ymlaen at yr eitem nesaf ar ein hagenda, sef croesawu Ombwdsmon Gwasanaethau Cyhoeddus Cymru i drafod yr amcangyfrifon sydd wedi'u cyflwyno ar gyfer 2019-20. A gaf i ofyn yn gyntaf, er budd y Record, i chi gyflwyno'ch hunain a dweud eich rôl, os gwelwch yn dda—pob un ohonoch chi?

Welcome back to the meeting of the Finance Committee. We're moving on to the next item on our agenda. I welcome the Public Services Ombudsman for Wales to discuss the estimates that have been submitted for 2019-20. For the Record, could you please introduce yourselves and outline your roles?

Fy enw i ydy Nick Bennett, a fi yw Ombwdsmon Gwasanaethau Cyhoeddus Cymru.

My name is Nick Bennett, and I'm the Public Services Ombudsman for Wales.

David Meaden, financial accountant.

Chris Vinestock, chief operating officer.

Dyna fe—diolch yn fawr. Bydd hynny o fudd i'r Record. A gaf i ddechrau, felly, drwy ofyn cwpwl o gwestiynau?

Roeddwn i'n gweld yn eich amcangyfrif chi ar gyfer eleni, a gyflwynwyd y llynedd, eich bod chi'n nodi costau o £340,000 a oedd yn gysylltiedig â'r pwerau newydd yn y Bil, wrth gwrs, sydd yn mynd drwy'r Cynulliad ar hyn o bryd, ond rwy'n gweld bod y rheini ddim yn ymddangos yn yr amcangyfrifon ar gyfer 2019-20. Nid wyf yn gwybod a allwch ddweud wrthym ni pam.

Thank you very much. That will be of benefit to the Record. May I just ask a few questions to start?

I saw in your estimate for this year, which was submitted last year, that you set out costs of £340,000 associated with the new powers in the Bill, of course, which is currently passing through the Assembly, but I see that they do not appear in the estimate for 2019-20. Could you please tell us why?

Yn gyntaf, buaswn i'n leicio diolch i'r pwyllgor am y gwaith ffantastig sydd wedi digwydd ar y Ddeddf newydd. Y rheswm pam nad wyf wedi rhoi ffigurau llawn i mewn ar gyfer blwyddyn nesaf ydy oherwydd nad wyf eisiau rhagfarnu pethau. Nid wyf yn siŵr yn union pa bryd fydd y pwerau'n cael eu pasio gan y Senedd, na ychwaith pa bryd yn ystod y flwyddyn bydd y rheini'n dod i rym. Felly, rwy'n meddwl y bydd hi'n well inni roi cyllideb atodol i mewn pan ddaw'n glir i ni a ydym ni'n sôn am gyllideb ar gyfer hanner y flwyddyn nesaf neu dri chwarter neu'r flwyddyn lawn.

First of all, I'd like to thank the committee for the fantastic work that's happened on the new Bill. The reason why I haven't put full figures in for next year is because I don't want to predetermine things, and I don't know when the powers will be passed by the Senedd, or when, during the year, they will come into force. So, I think it's better for us to submit a supplementary budget when it's clear to us whether we're talking about a budget for half of next year, or, perhaps, three quarters or the full year.

Iawn, ocê. Y tebygolrwydd yw, buasai rhywun yn meddwl, ei bod hi'n mynd i ddigwydd yn y flwyddyn dan sylw, ond rŷch chi jest yn teimlo mai dyna ydy'r approach gorau.

Okay. Well, the likelihood is, one would imagine, that it will happen in the year in question, but you just think that that's the best approach.

Rwyf wir yn gobeithio—

Well, I genuinely hope that it will happen—

—gan ei fod e heb ddigwydd y tro diwethaf.

—because it didn't happen last time.

Nid wyf yn hollol glir pa bryd—os yw popeth yn mynd i fod yn ei le erbyn 1 Ebrill y flwyddyn nesaf.

I'm not entirely sure when—whether everything is going to be in place by 1 April next year.

Iawn, digon teg. Ocê, diolch ichi am yr esboniad yna. Rwyf hefyd yn gweld eich bod chi'n gofyn am gynnydd o ran adnoddau o 3.8 y cant, ac mae hynny'n is na'r cynnydd yr ŷch chi'n dweud a fydd yn y bloc o 3.9 y cant. Rwyf jest yn gofyn ar ba sail ŷch chi'n seilio'r dybiaeth yna, o ran y cynnydd yn y bloc, oherwydd pan oedd Comisiwn y Cynulliad yma yr wythnos diwethaf yn siarad â ni, mi oedden nhw'n dweud bod Llywodraeth Cymru yn rhagweld cynnydd bloc o 1.67 y cant.

Okay, fair enough. Thank you for that explanation. I also note that you are requesting an increase in resources of 3.8 per cent, which is lower than the change that you say there will be in the Welsh block of 3.9 per cent. I would ask on what you've based that assumption about the block, because the Assembly Commission were here last week, and they told us last week that the Welsh Government foresees an increase of 1.67 per cent in the block.

Mi wnaf i ofyn i Dave ateb y cwestiwn yna, ond rwy'n meddwl bod eglurhad hanesyddol ynglŷn â'r cwestiwn yna.

I'll ask Dave to answer that question, but there is a historical explanation for that.

10:05

We don't base those figures on the estimate of what the block is going to change. We always calculate it in arrears, so we compare the annual budget motion, for example, in 2018-19, with the annual budget motion of 2017-18, and we've consistently done that over the last 10 years. So, in the last two years, the block moved from £17.3 billion to £18 billion of total managed expenditure, which is where we get the 3.9 per cent. So, we consistently use the December budget motion change of what's actually happened, rather than estimate or guess what the block is going to change. We don't have any access to know what's going to happen to the block next year, so it's based—

Neither does the Welsh Government, I suppose, but there we are.

So, that's where the 3.9 per cent came from.

Thanks, Chair. How does the estimate reflect the projections that your overall case load will increase by between 5 per cent and 12 per cent in 2019-20?

The workload obviously fluctuates quite a lot, and you will know from the submission and from previous reports that there are changes in the balances of enquiries and complaints, as well as changes in the numbers. But I think it's important to say that the budget submission isn't simply a mathematical exercise, so that if there's an 8 per cent increase in complaints, there's an 8 per cent increase in budget. It isn't as simple as that, and I'm not sure that an 8 per cent to 12 per cent increase in budget will be one that the committee would be particularly pleased to support.

So, what we do is look at the projection of workload and the impact of that on our resources and, in particular, on our staff and on our IT systems. So, the budget reflects what we think are the staffing resources that we need to manage that workload, which does increase every year, but that increase does fluctuate. At the moment, it's something like 8 per cent so far this year, compared with the same period last year—an 8 per cent increase—but it does vary. The estimate reflects our best estimate.

What's the maximum increase you could accommodate within your current budgeting?

That's not quite such an easy question, because it depends a little bit on the type of case, but broadly speaking—

And health is more expensive; I remember from Finance Committee.

Exactly. So, if we have a huge increase in enquiries, that is much easier to accommodate than a huge increase in health complaints. But, broadly speaking, the way that the budget has been prepared is on the 5 per cent to 12 per cent increase in case load that we've talked about in the annual report, and also in the context of the new powers.

Can you explain how your submission shows the differential cost of elements of your case load and the related financial pressures, as recommended by this committee in its report previously?

Certainly. I refer back to what I just said, that it isn't a simple mathematical exercise. But what we have done is look at the different types of cases, the balance of enquiries to complaints, the types of complaints that we are typically getting, and the balance of health complaints compared with local government complaints, and tried to come up with a view based on the differential time that is involved in each of those types of cases and, consequently, the difference in costs of dealing with them. So, typically, a health complaint will cost a lot more and take a lot more time to investigate than a local government complaint. It obviously depends a bit on the two complaints, but if it's, for example, a local government complaint about waste collection, it's not surprising that that is rather easier to handle than a complex complaint about healthcare, professional and clinical judgment, sometimes integrated with elements of social care, which then, obviously, becomes a much more substantial piece of work. We have looked at the make-up of our workload, and the budget submission reflects the differential element. If you want, I can say a little bit more about the work that we've done on the specific costs.

I suppose it's worth starting by saying that, actually, once you start working on unit costs, you can do a huge amount of work and go into a huge amount of detail because every time you make an assessment of cost you then can branch out to two or three different options in terms of how far a complaint progresses—whether it's simply an enquiry, whether it's resolved at the first stage at initial assessment, or whether it goes through to final investigation and report with professional advice. So, you can look at how a complaint actually progresses through the system and work out the cost of that. You can also look at the type of complaint it is and identify the typical difference in cost between having a health complaint, say, and a local government complaint. So, there are different ways of looking at costs. We've tried to keep it proportionate and relatively simple, partly because no two cases are the same anyway. So, we do group them but we don't get down to the level of costing every individual case.

But just to give you some of the sort of headlines in terms of unit costs: dealing with an enquiry will typically cost about £114 in terms of unit costs; with complaints, it depends on the type, but typically a complaint will cost £1,600, but that obviously varies a lot from one that is dealt with very quickly, which would be below £1,600, to a long and complex health complaint with professional advice and a full report in the public interest, which might be up to maybe £13,000 or £14,000. So, there is a range. We've done quite a lot of work to try and analyse those costs. They inform the budget submission, but it isn't simply a mathematical calculation of health complaints multiplied by £14,000 or £9,000 equals—. It is based on an assessment of how that case load can be accommodated within our office and what implications that has for our staffing and resources.

10:10

I know I've got the next set of questions, but can I ask something on that particular one? I appreciate what you've just said, but can you therefore tell us—? You say an average of £1,600 up to the odd case in health that is £13,000, which is quite a difference. That means there are a lot of small ones compared to some of the bigger ones. Have you got an analysis of the trend of the more expensive inquiries you have to undertake to see how they are increasing as to whether that's going to be dramatically increasing? 

I'll try and answer that. The elephant on our doorstep is the nature of health complaints that come to the office. They've more than doubled over the past decade. Last year, they increased by, I think it was 7 per cent or 8 per cent, but 11 per cent for health boards themselves. If we go back—there was one Welshman, Sir Idwal Pugh, who was UK health ombudsman in 1976, and he received 560 complaints across the whole of the United Kingdom, and he was alarmed that, by 1978, it had increased to about 780 across the whole of the United Kingdom. Last year, I received 900 in Wales, and his successor in England will receive 27,000, and there'll be corresponding figures for Northern Ireland and Scotland. So, we're talking about a trend in terms of the most expensive complaints that has moved from 500 some 40 years ago to over 30,000 across the United Kingdom this year. So, I think that gives you the bigger picture in terms of the trends we're looking at.

But not all of those will be the expensive ones. The question I asked was whether the trend of the more expensive ones is increasing.

Actually, I would argue that a lot of them would be expensive and also what we're finding—and again it's the reason why I've described it as the elephant on our doorstep, really—is that it's fantastic that we're seeing this expansion, the ageing population, but it does mean that, increasingly, the advice that we have to receive is more complex and multifaceted and that people will perhaps be suffering a number of comorbidities when something goes wrong. So, that means that the cases take longer and are more complex and are more expensive, simply because we need more clinical advice. That's good in a way. The fact is that, back in 1976, when Sir Idwal Pugh was dealing with this, people perhaps only had one thing wrong with them, whereas today perhaps those people will be 15 or 20 years older and there are more issues to investigate.

And there isn't just a cost to you as well; there is a cost to the health boards, for example, isn't there, in dealing with all of this?

Yes, indeed there is.

Okay, thanks. Can I go back to your annex in your report, which gives us the estimates? You've changed the breakdown of the revenue budget in your submission from those reflected in the estimates for 2018-19. Can you just talk us through why it's changed?

10:15

Sure. Dave, do you want to do that?

They're broadly similar, actually. It's more descriptive rather than a wholesale change. I think, if I look back at how we've looked at data over the last few years, we had different headings for Finance Committee, different headings in our annual report and accounts, and different headings in our internal monitoring systems. So, to me, the sensible thing is that we use the same headings everywhere, and these headings that you see here are the headings that we use internally. For example, 'advisory and legal', and we will use those headings in our annual report and accounts so that we're consistent everywhere. But really, it's more cosmetic; there's no wholesale structural change to them and we haven't moved money between headings or anything like that. It's placing 'training and recruitment' next to 'staff costs', because to us it's more linked to it. It's calling 'advisory and legal' what it is rather than just 'professional fees' or 'professional advice'. It covers a lot more things than that.

That's understandable enough. We just needed to know why you are doing it.

Well, I think we've been meaning to do it for a couple of years, but we took that decision this year to change the way it looks.

Okay. Obviously, in doing so, I'm assuming it helps you to demonstrate priorities a bit better as a consequence of that. I notice that the staffing costs are clearly a priority, and your estimate for 2019-20 is basically £0.5 million more than it was for the actual spend in 2017-18. I know you were talking about the case load, but that's quite a dramatic increase. Can you just run us through why you estimate that increase in staffing?

As I said, we have taken money from some budget headings and moved it to the front line. For example, we've reduced our legal costs budget.

So, the actual figure for 2017-18 in your annex, which I've got, says £2.9 million, near enough, and your estimate for 2019-20 says £3.4 million. So, are they all staffing costs? You say you've moved things, but was the actual cost in 2017-18 £2.9 million?

That was the actual cost.

So, your estimate is still £3.4 million, which is £0.5 million more.

Yes, which probably represents something like eight staff more, 2 per cent pay awards in each year, pension increases and increments. For example, the pay award and increment effect in this estimate would be something like £80,000.

I see from my notes here that half of the increase in salaries and related costs for this current financial year that was presented last year by you was to cover a specific set of circumstances; they weren't recurrent costs.

So, are you envisaging a similar situation this time or are these recurrent costs?

Yes. The maternity has continued.

Yes. Last year, we did have specific issues that we did cover in the Public Accounts Committee the other day as well. So, I think the internal and external specific challenges that we had were an 8 per cent increase in health-related complaints, which were, obviously, as we discussed, more expensive and more complex, but then 10 per cent of the workforce took maternity or adoption leave last year. There will be more this year, but not 10 per cent. So, what we've tried to do is really concentrate on risk, and the risks for us are meeting our key performance indicators, particularly given the pressure that we're under with health, and making sure, therefore, that we have sufficient staff. And then, another key risk from our risk register, which, again, we covered the other day in public accounts, is cyber security and ICT given that, obviously, with a growing proportion particularly of health-related issues, we've got a lot of sensitive data that we have to be very, very careful with.

So, the majority of the increase for next year is recurrent costs, but there is an element of special circumstances.

There was a smaller—[Inaudible.]—we had for maternity. Some maternity will—[Inaudible.]—we've got other people going on to maternity next year, so at least half of that will continue. But, we've also moved—. We've taken on a Welsh language translator directly and we've stopped using our—

—well, not stopped, but considerably reduced—

In-house. So, that's shifted money from office costs to staff. 

10:20

Okay, thank you. Sorry, David, do you want to continue?

No, it's okay. It's just Mr Bennett was saying about, obviously, the ICT, the cyber security issues. Are you therefore talking about that you're going to employ specialists, or is that going to be more of the IT spend?  

We, again, outsource a lot of the expertise that we need for different ICT aspects.

Okay. And I suppose—. I also see the 4.1 per cent net cash requirement from the Welsh consolidated fund. How are you taking on board the recommendations of the committee that you should keep in line with the increase in the Welsh block grant? 

We've got some data that I think Dave has prepared and can present on that, but I can assure you that we have certainly kept well within—the 0.03 per cent of the Welsh block that we were at one point, we are below that. 

Yes, on the resource basis, we kept within the Welsh block, because there are a little bit of depreciation figures around that that don't quite work their way through into cash. When we agreed to stay within the movement of the Welsh block, that was on a resource basis, and the cash should almost be exactly the same, just slightly different this year, so we have kept within 3.8 per cent, compared to 3.9 per cent on resource.   

Yes, but, as a proportion of the Welsh block, I think the question is. 

As a proportion of the Welsh block, it's 0.025 per cent, which is another one of our indicators. We set ourselves a maximum target many years ago of 0.03 per cent, and we're hovering around now the 0.024 per cent, 0.025 per cent as a percentage of the block. So, we're within both indicators. 

Can I ask you about improvement? Firstly, can you tell us a bit more about the role of the improvement officers and how they're driving improvement across the public sector, and, secondly, how you plan to make the improvement activity more of a mainstream part of the whole process?  

Yes, thank you. I think, first of all, in terms of managing expectations—wonderful job that they've all done, the improvement officers, at the outset. I don't think it would be fair to expect them to drive improvement across the whole of the public sector. But, you know, I'm not stopping them—we want to incentivise that as far as we possibly can. The five improvement officers that work with health boards are probably the best example that I can give. Their role is one day a week. What we've got is a cadre of improvement officers, the most experienced investigators that we have. Most of their jobs are still about investigating complaints, and our role is still very much about investigating complaints about public services, and we're very clear about that. 

However, particularly given the figures that I gave you in terms of what's happened in health over the last 30, 40 years, we need to try and turn that curve. I think there's a lot more learning that could happen in the health boards across Wales, so their role is to really try and develop that culture of better complaint handling in those health boards through ensuring that the health boards, their staff, understand what good complaint handling looks like, the way in which front-line staff should be empowered to deal with complaints effectively and soon, the way in which boards and leaders of health boards should be looking at complaints. In a non-market-based system, how do you know that you're responding to the citizen? Where's the contestability? So, I think it's really important that we see that there is adequate accountability and attention given to complaints, and due prominence within the different corporate cultures of those health boards as well.

I think there's still an awful lot more that we can do. As I say, it's been difficult for our improvement officers, given that they've been under a lot more pressure, with a lot more health complaints coming in, for them to devote perhaps 20 per cent of their time on the improvement function. It's been piecemeal in many ways. However, our proposal is to have a full-time learning and development officer to really ratchet up our activity in this area. I think that that could provide a number of benefits, not least, given that we're analysing our health complaints so much more now—. Twenty-five per cent of the health complaints that we receive are about complaint handling, which I find very, very surprising. Less than 5 per cent are about surgery. Less than 10 per cent are about waiting times. So, we could do an awful lot, I think, if we could increase the performance of health boards, because this is in nobody's interest—it's not in the health board's interest, it's not in the complainant's interest. And I think, for a very modest investment, we could do a lot more to reduce not just the complaints coming to us but to improve the experience of taxpayers and health service users across Wales.

10:25

And the role of the learning and development officer, can you explain a bit about the—

Chris, do you want to—?

Yes, thanks. The thinking is that we will have somebody as a learning and improvement officer who will co-ordinate the work of the improvement officers across the ombudsman's office, but will also come at things with perhaps a different approach and different skill set. The people we've got as improvement officers are primarily investigators. They have got great skills in investigating and analysing data. They haven't necessarily got the same level of skills and experience in terms of garnering support for change, for actually encouraging organisations to embrace change. Also, I think there's scope for improving the consistency of improvement activity across health boards. The learning and improvement officer will come at it not from an investigation stance but from looking at the scope for learning and improvement across bodies and will actually identify best practice, will help apply some of the improvement actions that relate to complaint handling and learning from complaints that are identified or progressed in one health board or one public body into others. So, it's actually about trying to get maximum benefit from the improvement work that we do, and also making sure that as well as blinding people with data we've actually got some of the skills to work with, particularly, health boards to garner their support for improving complaints processes but also complaints culture. I think that's why we think we need to have a separate role rather than have it as one fifth of five investigation and improvement officers. We think it needs to have greater focus and greater attention.

And is it your intention to grow this resource, resources allowing, in future years? Because, clearly, you're investing upfront in order to potentially save issues down the line. 

Well, the Act—. The draft legislation does include provision for a complaints standards authority, which is best practice from Scotland and now on the statute book in Northern Ireland, so it would go with the grain of that type of work to try and ensure better complaint handling amongst different sectors of the public sector. 

Yes. Just following that up, would that include the possibility of you extending the work of improvement officers to other bodies within your jurisdiction? I think that was a committee recommendation, but I don't know whether this would be possible in terms of the work that you're already undertaking and whether you've looked at the benefits of extending it in that way. 

Yes, I think there's—. Well, there's always more that we can do. As I said earlier, the elephant on our doorstep is really health, the fact that we're getting more and more health complaints. I'm pleased to say that we've seen a 10 per cent reduction in local government complaints—second year on the run. An example of where I think improvement activity has been particularly effective with a public body was Ceredigion council. We designated them for improvement about three or four years ago. Ceredigion—what, 2 per cent of the Welsh population—was at the time accounting for about 35 per cent of upheld local government complaints in Wales. Our improvement officers worked with them—and in fairness to Ceredigion they've embraced this approach—and their figures are much, much better. They no longer account for anything like one third of the upheld complaints in Wales. So, that's really good.

I think it's important as well that our improvement ambition, if you like, isn't just about specific organisations and complaint handling, but I hope about improving public services. The best example I can give of that would be the thematic work that we've done. We published a thematic report on out-of-hours services in Wales a couple of years ago. I know that the Government were not happy and the NHS were not happy about us publishing that report but it drew out the lessons from the out-of-hours themed health complaints that we'd received and made a number of recommendations about having a broader review in Wales, about making sure that there was effective supervision of junior doctors—because we'd seen a horrible case in Betsi Cadwaladr area where a patient had passed away from sepsis and no consultant coverage over the weekend at all—and that there should be electronic handover notes and other issues. Despite the pushback at the time, I think both health Ministers were magnanimous in the way in which they took that thematic work on. They have instigated a health board by health board review of our services.

I went up to Ysbyty Glan Clwyd in, I think, April or May of this year, and the staff there were able to show me what they were doing in response to that and some other public interest reports that we'd specifically published. It was brilliant, because all I usually deal with is complaints, but for them to be able to tell me, 'From now on, there will be no junior doctors in this hospital or any other hospital in north Wales who will not have at least access to a middle-ranking or senior doctor every weekend and every bank holiday'—well, you know, that is a real improvement and something that can make a huge difference, given that we do still see these very unfortunate sepsis cases that can occur.

So, I think, if for nothing else, it was worth doing just for that thematic report. We need to see improvement as a pan-Wales theme, as well as something that is a geographic intervention for certain local authorities or health boards that might be struggling.

10:30

Of course, that means that you are helping to develop best practice for public services, so how are you engaging and consulting with the other bodies that are actually developing—? Obviously, the auditor general and the Wales Audit Office are crucial in terms of their role, so how—? Are you actually liaising together? Presumably, your learning and improvement officer might be able to help, but, I mean, we don't want it in silos, do we?

No, no. Absolutely true. We were very concerned from the beginning, going down this route, Chris and I—we didn't want to be creating any confusion or duplication. So, our improvement officers, where they, let's say, were working with Betsi Cadwaladr or another health board, we made sure, as well, that they were also linking in with the lead contact officer from the Wales Audit Office or from Healthcare Inspectorate Wales, because we want them to be able to triangulate. If there's something in the complaints data that we have that informs them, then that's great, so we make sure that there's that joint learning.

I certainly meet with the auditor general and some members of his staff on a regular basis. We've even done joint seminars on good practice and complaint handling in north and south Wales. I think we plan to do some more—we had a member of staff from the Wales Audit Office come to see us with a further proposal to do that, recently. That's good—where we can do that jointly, I think it does have more impact.

Can I ask: what would the impact be on your office if you weren't to get the increase that's outlined in your submission?

Well, we wouldn't be able to employ the learning and improvement officer. Clearly, we would be under pressure when it comes to meeting our key performance indicators. Chris, are there other issues?

I think that that's pretty much it. It would be about not having the learning and improvement officer, but also, perhaps, looking to scale back on the improvement work that we do with the investigation and improvement officers. It would have an impact on case load.

I guess the other thing—and we've always been reluctant to do this—is to take a view and actually apply a higher bar for complaints that actually are accepted for consideration and investigation by the office. We're reluctant to do that, because, by the time people reach the ombudsman's office, they generally do feel pretty strongly about the complaint they've had. They've had to go through the complaints process of the body concerned, which is sometimes more than one stage; they've got to show persistence. They tend to have strong views that there is an issue there to be investigated. But, if we didn't have the resources, I think we would have to look at saying, 'Well, we understand your concern, but we don't think it's significantly serious for us even to look at.' We're not really keen to go there, but, equally, I don't think we want to be in the position, which I think is the alternative, of having a constantly growing backlog because the cases are coming in much faster than we are able to deal with them.

Okay, thank you for that. I just want to ask about pensions, as well, because I'm wondering how you're managing the risks relating to the pension fund. Specifically, of course, there was a recommendation of ours previously around not investing—or, investing in a diverse portfolio, which clearly isn't the situation that you're in at the moment.

Just to be clear, everybody's in the civil service pension scheme in our office, bar one person. Back in 2006, we inherited staff from the local government who were in the local government pension scheme, who had a choice of transferring into the civil service pension or remaining. Over the years, that is now down to one active member. It may be worth just going back on the history a little bit. Following the financial crisis of 2008-09 and the actuarial valuations that took place of the local government pension funds, our element of the Cardiff and Vale pension fund was identified as a £1.6 million deficit, which we've been paying off over the last six to seven years, and we actually paid it off a year early and gave nearly £300,000 back last year to the consolidated fund.

So, I have regular meetings. There's a minimum of an annual meeting with the Cardiff pension fund manager in county hall across the road, and the professional scheme actuary, called Aon Hewitt. But I had a special meeting with them when I suddenly found we're now back in surplus. And I suppose my tack was to say, 'How can we stop what happened happening again?' And it's clear what happened—it happened because there's a diverse portfolio that the pension fund had. There were two things, really: the equity value falling and a very pessimistic view from the actuaries.

The recommendation by the actuary—not our recommendation, but his independent recommendation—was to move into a low-risk Government bond portfolio. The current value of that fund is £1.5 million—£1.5 million surplus—and it's not exposed to stock-market fluctuations. One of the reasons for doing that is, when we exit this scheme, when the last person employed by PSOW leaves the scheme, any surplus will be given back to us. So, if that person left today, we'd have £1.5 million cash given back to us, which we would probably give back to the Welsh consolidated fund, as we can't keep bank balances at the end of the year anyway. We would obviously, seriously, give that back. So, over the next few years, if that person leaves the scheme, whatever value is left in that scheme will be returned.

I guess the answer to, 'What risk is there?'—well, there's none. There is no risk to that pot of money, because it's invested in Government bonds. Okay, the Government bonds may not do as well as the stock market or vice versa, but if, say, in two years' time, that person left the scheme and there was a stock market blip, and that put that value as halved, we'd only get half returned. So, there's no risk, really.

The figures quoted, which you might see in the accounts, were of a much lower value because the international accounting standards on pensions do not allow an organisation to show the whole value of their surplus in the accounts. So, they carry out some complex calculations that are beyond me and bring that value down. But the underlying figure, the higher figure, is in our annex of the accounts. And the reason that that's done is for the private sector, really, because they don't want the private sector borrowing money on a pension fund asset. 

10:35

Okay. That's a comprehensive answer. That's excellent, thank you very much. Mike.

Can I carry on with pensions, then, for a few seconds?

Really, you've got the choice between low risk, low returns, or high risk and potential high returns, and it's not abnormal when you know that you may well be coming to a time when you're going to have to meet the payment to make sure that the risk is reduced. I mean, for the stock market to lose 10, 15 per cent of its value in a three or four-day period is not abnormal when we go into economic crises, which we're quite good at doing. So, I would say it's fairly sensible to do that. You will get criticised by people who say, 'It should all be in the stock market. The stock market's doing very well.' And I've heard that criticism here from other Members regarding the pension fund we've got here. But you know you have to pay at certain times and it's a balance, and I tend to agree with you on balance, but you think it's a balance between risk and reward and that it's much better to have a guarantee of being able to meet your liabilities and have some money over, as opposed to knowing you might not meet your liabilities and have to find some money, but you could have a windfall as well. Isn't it safer to do what you've done?

10:40

'Yes' I think is the one-word answer, isn't it? Okay, I think we've exhausted the pensions.

Okay, on to IT. We asked last year for a three to five-year programme. Have you got one?

How do the costs under the new ICT contracts compare with those previously incurred? And have you achieved any efficiencies through the joint procurement activities? Are there savings to be made in joint procurement, working with others, especially in terms of activities that everybody has to do but will not actually be one full-time equivalent, such as ICT security et cetera? Are there savings to be made by actually having somebody working between a series of public sector organisations like yours as opposed to having to buy in very expensive people?

Well, certainly we can see the value in joint working when it comes to things like governance, so joint working and training when it comes to ensuring that the right type of audit and risk arrangements are available when it comes to cyber security. Fundamentally, with our expenditure here, though, I have to say we are not looking for efficiencies in that way. What I'm looking for is a cyber security level that means there are no payments requested from anywhere in bitcoin and also no fines that emanate from the Information Commissioner given GDPR and other developments there, so it's safety first rather than looking for further savings, to be honest, Mike.

On data security, sometimes organisations will pull the internet plug out; that gives you absolute data security. Have you thought of splitting your systems so that you have one dealing with external and one working in internal? It might involve some data transfer, but it would mean that your internal system would be safe from external attack.

Do you want to answer that, Chris?

I'd love to. So, to an extent, our system is already split in that way. For example, one of the main routes for complaints into the organisation is through the website, which is independently hosted, and then there's a channel for information to come through to our systems. It is one of the things that is a trade-off, because a true internal system makes it quite difficult to work remotely, for example, and once you start having remote access and encouraging working at home or working off-site, you start opening up gaps and ways into the system, which have the potential to be exploited. What we've done is we've changed our IT support arrangements; we've had a full review of our IT systems by them and by an independent expert, and we are in the process of implementing a complete change to our infrastructure that tries to boost that security to reflect the risk profile, to try and make our systems as robust against cyber security issues as they can be. So, I think our plans reflect the broad spirit of what you're asking, but we're not having completely separate systems, but there's a lot of emphasis on making sure there are robust firewalls and filters and access controls.

I went to a seminar once on this and one person said, 'There's only one guaranteed way of cyber security', and he pulled the plug out. 

No, but I think that is a difficulty, but can I just test it with one more question? There are things that we should not be accessing from outside, such as personnel records, payroll and things like that, the company's accounts—not the company's, but the ombudsman's accounts. Is there any way of keeping them separate from the publicly accessed parts of the system, or do you?

They are already separate in the sense that they are separate parts of the system with separate access rights and separate access controls. We are looking at how we can make sure that our systems are resilient and so they've got mirroring of servers, and we're also looking at how we can get things into position where they can be restored from the cloud if needs be. So, there are security issues there that we need to address. I think we're giving a lot of attention to the architecture of our systems to make sure that the systems are separate. However, if they're all on the same network, it comes back to the only absolutely guaranteed safe way—to pull the plug. So, there will be links, but we're making sure that the architecture actually separates out the different systems as far as possible and of course builds on the existing access controls that we've got within the system in terms of who, even within the organisation, can access what.

10:45

Just for clarification purposes, I assume—you said separate systems, but I talk about separate servers, and there was other data on separate servers. Are there firewalls between those separate servers and is that the type of mechanism that you've put into place to ensure some form of safety? I do have concerns about the cloud because the cloud is what it says—it's a cloud and you have to access it and so does everybody else. But, in a sense, you're talking about different servers for your back-up purposes, for your fail-safe, but are you talking about different servers for your different storage, so you're separating the personnel system from other systems, and do you have firewalls in between them?

Each part of the system will be protected by its own security and its own firewall. I can't honestly tell you whether the way the system will be designed will be that all of the personnel records will be only on one server, but we are differentiating between the case-management information, which is handled in a particular way with particular servers, and the more internal information. The whole purpose of the design of the infrastructure is to try and make sure that those controls are as robust as they can be, and you'll have seen from the estimates submission that we've identified IT security as a really high potential risk and therefore a very high priority.

And that's our focus, isn't it, really, in terms of making sure that the resource is there in order to meet those concerns that have been expressed? Okay, thank you very much for those answers.

Can I just ask as well: in relation to the advisory and legal fees that you list, I note that there's a reduction in the estimate that you propose for next year? Can you explain the basis for that reduction, particularly given, of course, the concerns that you've outlined about increasing workloads and the volume and complexity of a number of those complaints?  

That heading encompasses clinical advice, professional advice on casework, but also things like human resources advice, audit work, building advice, and general legal costs too. So, only part of that is casework-related. So, although that's fallen in total, from, I think, £281,000 to £260,000—I haven't got it in front of me—the actual clinical and other casework advice element is the same as what our forecast is for this year, which is £165,000. It rose from last year, it's levelled off this year, and we're keeping that value the same for next year. We haven't reduced the front end—

So, how are you realising that difference, then? If it's not in terms of complaints and support around that, where are you making those savings?

We're making savings on the legal costs.

The provision for additional legal or corporate-type costs have been reduced but there has been no reduction in clinical or front-line services-related advice.

We've traditionally kept a reserve in for legal costs, just in case. There comes a point in time where, I think, you have to say, 'Well, we actually need that at the front end.' And should we get a challenge, hopefully, we can deal with it. If we had a massive challenge, we'd have to come back to this committee. But we'd probably have to do that anyway, if it was big enough. So, that's where we made the reduction.

Of course, if you had a massive challenge, you'd probably drift into the following year anyway.

What I was about to say was that, if there's a big challenge, you don't tend to deal with it in one 12-month period.

Just one question that I've kept to the end. Health is said to have always been a big issue and it's increasing in demand. Now, I know that Keith Evans did a review on the complaints processes in health a few years back. Have you seen that working within the health boards or are there still issues with the complaints processes in the health boards? Is it causing you to, actually, take on more detailed work? 

That's a great question, David, and you remind me—I do remember the Keith Evans review. Was it 'The gift of a complaint'? It was published as I started back in 2014. I hope it's made a difference. Certainly, I'm pleased to see a reduction in complaints in some health boards. Of the ones that we've highlighted as improvement, three came down, but we did see a 30 per cent increase in one, a 34 per cent increase in the volumes from another health board. But let's put it this way: we wouldn't be asking you for additional resource to do more on improvement if that report had the impact that, perhaps, Keith Evans would have hoped it would have had.

What I can say as well, though, that is quite positive, is that we have seen less maladministration and service failure during the course of that year. So, that is something that is positive. So, only four public interest reports that year. Sometimes, we have eight, nine, ten very, very serious issues. That didn't come up. And the overall level of maladministration, I think, was down to about 56 per cent of our investigations, compared to 60 something per cent the year before.

But we're talking about culture change and that takes time. I doesn't happen in a year—it could take seven.

10:50

Excellent. Diolch yn fawr iawn. Thank you very much for being with us this morning, We appreciate your presence very much and, clearly, we'll be deliberating the evidence and the estimates, and reporting in due course.

Felly, diolch yn fawr iawn i chi.

So, thank you very much.

Diolch yn fawr iawn i chi i gyd.

Thank you very much to you all.

Daeth rhan gyhoeddus y cyfarfod i ben am 10:51.

The public part of the meeting ended at 10:51.

11:25

Ailymgynullodd y pwyllgor yn gyhoeddus am 11:26.

The committee reconvened in public at 11:26.

8. Cyllideb Ddrafft Llywodraeth Cymru 2019-20: Sesiwn Dystiolaeth 2
8. Welsh Government Draft Budget 2019-20: Evidence Session 2

Iawn. Croeso nôl i Bwyllgor Cyllid y Cynulliad. Rydym ni ar eitem 8, sef craffu ar gyllideb ddrafft Llywodraeth Cymru 2019-20. A gaf i estyn croeso i'r tystion sydd ger ein bron ni am y cyfnod nesaf yma? Croeso i Dr Ed Poole, pennaeth rhaglen dadansoddi cyllidol Cymru, Canolfan Llywodraethiant Cymru, Prifysgol Caerdydd; Guto Ifan, ymchwilydd arweiniol gyda'r rhaglen dadansoddi cyllidol Cymru yn y ganolfan llywodraethiant, Prifysgol Caerdydd; a David Phillips, cyfarwyddwr cyswllt y Sefydliad Astudiaethau Cyllid. Croeso i'r tri ohonoch chi.

Mae'n werth nodi hefyd fod Dr Ed Poole, wrth gwrs, yn ymgynghorydd arbennig i'r pwyllgor, ond mae e yma i roi tystiolaeth inni y bore yma. Felly, diolch yn fawr.

Mi gychwynnaf i gyda chwpl o gwestiynau, os caf i. A gaf i ofyn i ddechrau os gallwch chi grynhoi pa mor effeithiol ŷch chi'n credu mae polisïau cyllidol Llywodraeth Cymru wedi bod dros y flwyddyn gyntaf o ddatganoli trethi?

Welcome back to the Finance Committee of the Assembly. We're on item 8, which is scrutiny of the Welsh Government draft budget 2019-20. Can I welcome the witnesses before us for this next period? I welcome Dr Ed Poole, head of the Wales fiscal analysis programme, Wales Governance Centre, Cardiff University; Guto Ifan, lead researcher with the Wales fiscal analysis programme at the Wales Governance Centre; and David Phillips, associate director, Institute for Fiscal Studies. I welcome you all.

It's worth noting also that Dr Ed Poole is a special adviser to this committee but is giving us evidence this morning. So, thank you very much.

So I'll start with a couple of questions, if I may. Could I ask you to summarise how effective you think the Welsh Government's fiscal policies have been over the first year of tax devolution?

Wel, diolch yn fawr iawn, Cadeirydd, am y cyfle i roi tystiolaeth i'r pwyllgor y bore yma.

Thank you very much, Chair, for the opportunity to give evidence to the committee this morning. 

I think, with regard to the effectiveness of the tax policies so far, we've seen—. It's the first new year, so it's very early days, but what we've seen, perhaps, is there's been a slightly different approach taken. In Scotland, for example, we've seen the Scottish Government using tax powers quite early, changing the rate, changing the allowances. There's a slightly different approach that the Cabinet Secretary's outlined in this budget, which is to test the system first, steady as she goes, make sure that Her Majesty's Revenue and Customs is doing its job, collecting taxes in the right way, testing the system. So, there's no right way and no wrong way, but it's interesting that we've seen a difference in approach to the use of the powers.

What we have seen, of course, is that, on the land transaction tax side, a different set of rates have been applied to mirror in some ways the Scottish land and buildings transaction tax, making the rates a little bit more progressive—lower at the bottom, higher at the top. So, we have seen some use of the powers on that side. But I think the Cabinet Secretary is right in that we want to make sure that the new powers are bedded in so that we can see that there are no surprises down the road.

Okay. Anyone want to add anything, or are we okay with that?

I would just say that it's too early to detect any impacts on economic performance, and, of course, as Ed was saying, in the first year, whilst they changed the structure of the land transaction tax, they didn't change the revenues raised, or there wasn't a plan to change the revenues raised. There was a revenue-neutral change, and the difference, looking into the second year, is that while the Scottish Government decided to actually use its income tax powers to raise some revenues, the Welsh Government has not decided to do that. It was in, actually, I think, its manifesto, or the plan for government, that there would be no increases in income tax. So, quite a different discussion there. If you look at the debate, actually, back in the Assembly election in 2016, the debate in Wales was for income tax cuts largely, whereas the debate in Scotland has been around increases, which is maybe not what you'd expect given the politics of the two counties. But I think there has been a difference in what people are talking about with income tax in the two countries. In Scotland, it has been seen as a way to increase revenues, to make things more progressive. In Wales, it's been seen as kind of a way to make the country more attractive, potentially, to top earners—not across all the parties, but there's been more of that discussion in Wales than in Scotland.

11:30

Okay. Thank you. Diolch yn fawr iawn. So, in your experience, then, what are the key elements to implementing an effective tax strategy, and do you think that the Welsh Government is developing suitable policy levers to accomplish that?

What we argued in our report, which you might have seen, which we released in June, was that these taxes should be taken together, and especially council tax should be seen as a potential policy lever to use alongside the new tax powers that have come on board. In this budget, there's a good example of that in the fact that we haven't seen an increase in income taxes, though the cost and demand pressures on local government will make it likely that council tax will increase and increase the financial burden, not through income tax, but through council tax. There's a good case for considering these two taxes in the round, together.

Of course, implementing something different would be a long-term project and it would probably require a manifesto commitment and a wide debate about how best to raise revenues from the Welsh tax base. Again, it might be too early to assess that.

But that integrated approach that you referred to in your report, to what extent do you think that is happening?

I think there's viable work being done in terms of looking at the council tax reduction scheme and an increase in take-up of that; introducing a softer approach to the management of council tax arrears. But that's treating the consequences of a regressive tax system and not, maybe, addressing the fact, or the underlying fact, that it is quite a regressive tax and it's ripe for reform. Maybe that is a longer term policy.

Sure. Yes, because it's still early days, in fairness. Nick, you wanted to come in. 

Yes. I might be treading on someone's toes here, but I don't think I am. You're right, it's a fascinating discussion. I know the Cabinet Secretary has said that he wants it not just to be about raising or lowering taxes but to engage the public, and yet—we talk about this all the time, and it's fascinating—when I go out and I speak to people out there, I can't name one person I've met yet who actually realises that income tax is devolved to us. And I'm just wondering at what point they will get it, if they will, and how on earth do you bridge that gap? Because the Welsh Government, to be fair, and we have all been banging on about it now for ages, but people just don't seem to have registered it. Those are my constituents, anyway. Maybe it's different for others.

Well, they'll be getting this soon, although, this, in itself, isn't going to answer the question.

And I think it's going to be a heck of a shock to them. They'll actually fall from the doorstep.

It won't be a shock because nothing's going to happen. But, just to make the point, Scotland had tax-varying powers from the first referendum in 1997 and did not do anything with them until they had a fiscal framework. And, obviously, there's a lot of education that we need to do to understand more and the implications, even, of what they did in terms of raising tax, and what that means for the budget.

Can I take that? I think the issue with Scotland and why it took them so long was less around the fiscal framework and more around the ability to vary rates independently. So, the powers that Scotland had until 2016 were just moving rates up and down in a lockstep. It was politically as attractive [correction: politically not as attractive] to do it, frankly. Politically, it was more challenging to say, 'We're going to make taxes higher for 70 per cent of the population' rather than saying that, 'Actually, a large fraction of the population are gaining.' We saw that in Scotland recently where they introduced a 19 per cent, a 20 per cent and a 21 per cent rate, where the maximum gain from the 19 per cent may have been £20 a year—not really economically meaningful, but politically quite important. So, maybe it's around the politics of it, of having the kind of powers to vary rates differently.

Maybe that's why things play out differently in Scotland and Wales. In Scotland, they really wanted to show, 'Look, we are using these powers to do things differently; we a different country.' In Wales, it's been more of a steady-as-she-goes approach, not wanting to rattle the cage given concerns about whether Wales would be perceived as a high tax country if the first thing it did with the tax powers was to increase it for higher earners.

And I think the big issue—those are important issues as well—because HMRC are going to continue to collect the income tax, which is the right approach, because most people will pay through PAYE, is how many of us will look at our pay slips in detail and the different particular lines on a month-to-month basis. So, people might well not notice if they're paying through PAYE or actually even through self-assessment. And, of course, we go back to the Holtham commission, the Silk commission—the whole point of tax devolution was this accountability link between Welsh Government spending decisions on the one hand and the revenue collecting on the other. And so that information is really important. I know that Welsh Government are trying to do this, and taxes are not everybody's favourite subject to talk about, the committee being different, of course, but it is going to be a challenge. I think that, once rate decisions do start being taken, I think then we'll see—we have seen, in Scotland, quite a lot of discussion about the Scottish Government's fiscal decisions. So, I think that, after the powers come online, there will be some more uptick. Whether it will be the uptick that people would perhaps have hoped at the time of the commission is an open question at the moment.

11:35

In terms of general fiscal policy, I think it's important to note that, this year, there's been a slight shift in UK Government policy so that the added consequentials from the NHS announcement equal about £365 million, and had the Welsh Government taken its income tax powers and raised that revenue itself, it would have required about a 2p increase in the basic rates, which is quite a significant increase in the basic rates. So, that may be an important point to make, in that spending decisions in England will still be the main determinant of changes to the annual block grant, and we shouldn't over-egg the fact that—the income tax powers do change the debate and change the budget process, but the UK Government's decisions in England will still be really important.

Yes, that's an important point, thank you. I just want to ask about Brexit as well, because it wouldn't be an Assembly committee without somebody referring to Brexit. [Laughter.] But, in all seriousness, of course, I'm just wondering how Brexit could affect fiscal policy in Wales and what maybe could be done to mitigate some of the potential risks around that.

Certainly on the programme side, on the budget programming side, we're going to obviously continue to take part in EU programmings after 29 March next year for the rest of the multi-annual financial framework through to 2020. That, I think, includes provision for Wales to continue to participate in funding programmes until their closure in 2023. Of course, the successor arrangements are still yet to be determined, so the absolutely key ones being the shared prosperity fund and the replacement for the common agricultural policy. So, those are—the Cabinet Secretary has raised this as well, and that obviously has major financial consequences potentially for Wales and the future shape of those programmes. 

In terms of the devolved taxes, the relative effect if Wales has a particularly—if the shape of the eventual deal disproportionately benefits or disadvantages Welsh businesses compared with the rest of the UK, we might see some changes in the devolved tax collections on the income tax side. For example, goods are relatively more important to the Welsh economy than is the case in the rest of the UK, so if there is a goods-friendly deal, that could potentially be an upside relative to the rest of the UK. A more financial-friendly deal would perhaps be the other way round. Of course, the biggest impact will be the impact on the UK budget as a whole, because, as Guto just mentioned, still the largest component of the budget will be determined by the block grant, and if there are fewer resources or more resources in the UK Government budget, that will have the biggest impact, but obviously there's a lot to watch for on the spending programme side as well.

Yes, if I could add a couple of points on that. I think that, in the short term, if there is a short, sharp, shock Brexit, a disorderly Brexit, the immediate effect could be to have more money for the Welsh Government budget. I think that, in that circumstance, it's likely that there would be fiscal easing by the UK Government. It would want to prime the pumps by increasing potentially, say, capital spending, investment spending and some other areas of spending to ease the shock, but that wouldn't mean that it was a Brexit dividend—what it would mean is you're increasing spending in the short term to help ameliorate the problems, but it'd be another slow, long slog to get the public finances back on track if you think of this as a kind of permanent hit to the economy, rather than a temporary hit to the economy. So, you might see, in the short term, actually an increase in spending. It might be labelled a Brexit dividend—that's unlikely to be true.

Another point I'd raise is around the shared prosperity fund. I think, here, there's really a couple of issues that will rear their head. One is that there's been some discussion in Scotland that, actually, we should be avoiding the use of complex systems—just put this into the Barnett formula. I think putting this kind of funding into the Barnett formula would severely disadvantage Wales. Even if you started off with the current levels of funding, the Barnett formula has a squeeze in it. If you start off with a higher level of funding, it squeezes it down, so, over time, Wales would lose out from putting this funding into the Barnett formula. I think you would need to think about some kind of different allocation method.

I think the other issue to have in mind is that if this follows the approach that's been taken within England for allocation of regional funding or development funding, it could be deal based. So, what you've seen in England is that rather than having formulae or a more mechanical system to allocate funding, which is how the EU funding works, funding for redevelopment and for city deals has been based on a kind of case-by-case-basis negotiations. That could pose issues if certain areas have more political power in the UK or have stronger links to the party that is governing at a particular time.

So, I think there's something to be said about having a system that is mechanical, to some extent. That doesn't mean it can't have any performance-based measures in there. It doesn't have to just be based on how poor you are. It could be estimates of what impacts the programmes could have and estimates of the impact previous programmes have had—some kind of performance-based system. But, actually making that as objective and as transparent as possible would be beneficial, rather than having it based on a deal-based approach.

11:40

Okay, thank you. We've deliberated on many of those issues in a recent report of ours, so thank you for that. Jane.

Yes, moving on to the Barnett formula, as you can imagine, I'm very interested in it—well, we all are. We're particularly interested in the adjustments that have been made now to the Barnett formula, incorporating a needs-based factor. Can you comment on the significance of the changes to the Barnett formula for Wales?

As you'll know, more than anybody else in the room, of course, the Barnett formula automatically allocates a population share of changes on spending on devolved programmes in England to the budget. Obviously, the major upside of the fiscal framework for this—the fiscal framework deal—was the additional 5 per cent multiplier. So, we keep all of the existing calculation of the Barnett formula, but add what's been termed a needs factor, where you multiply that formula by 105 per cent at the end, giving an uptick. I understand that it's added about £70 million to the Welsh budget so far, and about £271 million over the five years covered by the UK Government's NHS funding announcement. So, it's a relatively significant change in terms of the finances, and particularly because there's been the NHS consequential, so we're seeing that 5 per cent applied to that.

What's interesting, of course, is that there's an additional layer to the 5 per cent, which is at the point—. David mentioned this, that the Barnett formula, as you'll, of course, be well aware, has a squeeze mechanism in it, just as a quirk of the formula, so that, particularly when there's spending growth or an uptick in the spending environment at the UK level, we see funding levels converging towards the English levels. At the point at which we hit 115 per cent, the needs-based multiplier switches from 5 per cent to 15 per cent. So, actually, putting the existing European budget programmes in the Barnett formula, there might be a mechanism in there—. Now, whether or not there's an additional change to the fiscal framework to account for these big changes, there might be a mechanism to stop further convergence over time. But, there are relatively consequential changes to the formula.

You said that convergence was a quirk in the system. Wasn't it, initially, the aim of the system?

Well, by making it a percentage of the cash base change, then, yes, over time, it would result, particularly in Scotland and Northern Ireland, in a convergence down to England. 

11:45

I'm not sure it was the aim. I think the idea behind it was it was a temporary stopgap measure to avoid, kind of, issues around budget negotiation just after the International Monetary Fund bail-out. 

Mike's on to something: it wasn't designed to be beneficial. 

It was designed as a temporary kind of measure. So, I'm not sure anyone ever thought—. I'm not sure if the person who put it in ever thought that convergence would become a significant issue, because I think they thought it would be replaced by a new system because, actually, in the late 1970s they were debating a needs-based system. I think that was maybe the first assessment of what relative needs were at that time.

It was supposed to cover the gap until there was going to be a Welsh Assembly, wasn't it, in the late 1970s? They assumed that devolution would happen, but, of course, it never happened.

If I could comment on this as well, so I think this kind of switch from 5 per cent to 15 per cent is quite interesting. 

Well, the work that we did with Ed and Guto suggested that it could take quite a long time. So, even if you had growth in spending of 5 per cent a year in nominal terms in England, it could take around 18 years to reach this 115 per cent level, given the differences in relative population growth between England and Wales, and given that we're starting at a level of about 120 per cent of spending compared to England. 

But I think one of the interesting things is that once you reach that, and it switches from 5 per cent to 15 per cent, what you get is rather than the Barnett squeeze, a Barnett kind of divergence, because once you start giving 115 per cent comparability factors, what will happen is that Wales will start to move up from 115 to 116, 117, 118, 119, 120 per cent, again, because slow population growth in Wales means that you give Wales 115 per cent of the increase in England, per person that's more than the 115 per cent increase in England, because the population of Wales is going up less quickly. So, I question whether that is actually a sustainable part of the system. Will the UK Government actually want to keep that in place, given that, actually, that would lead to Barnett divergence rather than Barnett convergence? I'm not sure whether that is actually a politically sustainable thing in the context of the debates within the union, but if it goes ahead it's a very, very beneficial thing to Wales.  

I mean, obviously, we've called and this committee, I think, has called for a reform of Barnett, basically, particularly post Brexit. And, you know, that's a key point in terms of reform is needed, but that's very helpful to that analysis. 

In terms of the block grant adjustment model, what are the strengths and weaknesses of the comparable model used to adjust the Welsh block grant? 

Some of the good things about it, compared to other systems internationally, the first is that it does protect the Welsh Government budget from UK-wide economic shocks. I think we've seen this in this budget; there's been a downward revision in the land transaction tax revenue, but because that reflects UK-wide trends, the Welsh Government budget is protected from that because the block grant adjustment will also fall.

I guess the other upside, the other strength, is that it fully incentivises the Welsh Government to grow the tax base to get extra revenues, to grow faster than the block grant adjustment and comparable revenues in the rest of the UK, and it also incentivises the Welsh Government to increase economic growth to avoid a shortfall in the budget. 

In terms of, perhaps, a weakness, for the Welsh Government it would be that it doesn't adjust for Wales's slower growing population on the revenue side of the budget. So, the chief economist report that accompanied the budget announced that if that slower growth in population translates directly into slower growth in revenues, then the negative effect on the Welsh budget could be around £150 million a year by the end of the next decade. Of course, on the expenditure side of the budget, that slower population growth, as David alluded to earlier, does benefit the Welsh Government in terms of its per head spending power. There was a recent example in Luke Sibieta's work on education spending, comparing education spending in Wales and England—the fact that per pupil spending had fallen faster in England than in Wales, not because of trends in the total amount spent by the two Governments, but the fact that the pupil population in England had grown by 10 per cent and it had stood still in Wales since 2009. So, that gives an indication that the budget the Welsh Government has goes further when it has less of a population to spend that money on. 

11:50

So, certainly the reason why the UK Government was arguing for the comparable model both in Scotland and in Wales was it said that was symmetric with the Barnett formula. The Barnett formula meant that Wales was benefiting from having its low population growth, because it didn't adjust for population growth. On the same side, actually, if you're going to benefit on the spending side, you should bear the risk on the revenue side. So, I think that's the argument the UK Government made on that point. In Scotland, they didn't buy that. If you look at why this has happened in Wales, I think it was part of a political bargaining process that other people in the room would know more about. 

I think that was my next question answered as well.

Yes, it was on the population growth, which I think you've just answered.

Yes. Land transaction tax—would you accept that land transaction tax in Wales is now more progressive than stamp duty land tax was?

Well, as I mentioned at the very start of this session, land transaction tax is one of the taxes that we've seen relatively significant rate changes for. I think, particularly at the lower end, we have seen a reduction in the rates. Then, at the higher end, we've seen an increase in the rates. So, up to transaction values of £250,000, Welsh rates are lower than they are in England; above £400,000, the opposite is the case. Of course, that's different for first-time buyers, where there's a zero rate for properties values of £300,000 or less in England, and that hasn't been carried over in Wales, as we know.

So, there is the third comparison, of course, in the UK at the moment, which is the land and buildings transaction tax in Scotland, where we've seen the rates also have that kind of more progressive tilt. So, Wales has cheaper rates than Scotland for anything from £146,000 to £226,000, then Scotland takes over and has cheaper rates up to about £333,000, and then the situation reverses again. So, both of them have adopted a slightly more progressive slope on the rates schedule, if you like, than is the case in England, excluding first-time buyers of course, which is a quite different treatment. 

I would say that I would agree with all that analysis. I think it's important to bear in mind, when thinking about tax policy, that whilst distribution—progressivity—is going to be one of the factors to bear in mind, another factor to bear in mind is also, if you like, efficiency. Do you think that this tax will have a bigger or smaller negative impact on the economy? I think, in thinking about that, in general, economists think that stamp duty, that landfill transaction tax, is a bad tax. It's a taxation of transactions that means that the people who value properties the most don't end up living in them, and it taxes mobility. People move to get jobs and get taxed more on that, which can reduce the flexibility of our labour market and can reduce people's job opportunities.

What this tax is doing in Wales is the fact it's gone to a more progressive system means that for the majority of transactions actually that efficiency cost is smaller. It's less grit in the wheels of the economy and the housing market moving, but at the top it's a bigger bit of grit. I think that depends on where do you think this tax would actually have more of a negative impact—on the lower part of the distribution, or the top of the distribution, where the tax increase can be fairly substantial. I'm not saying there's a right or a wrong answer there, but I would say that I would hope—and I kind of know it is doing this—that the Welsh Government is thinking about progressivity but also thinking about broader impacts on how the economy is performing as well as the distributional impacts.

Just looking at property prices, you have London and the south-east and you have the rest of Britain. Wales is part of Britain, as is Scotland, as is Northern Ireland, as are the midlands and the north of England and the south-west. I once read that the borough of Kensington and Chelsea had more £1 million properties than Wales. That may not be absolutely true, but it's close enough to the truth to be acceptable. With property prices in Wales growing more slowly, and, in terms of absolute money, if our properties all went up by 10 per cent and the ones in Kensington and Chelsea went up by 10 per cent, our average house price is over £200,000 and Kensington and Chelsea's house prices are well over £1 million. So, you're talking about an £80,000 difference per property, even if they were all going up by 10 per cent. The question I've got is: what is the effect it'll have on Wales even if we have the same rate of growth in terms of percentage, compared to what England will be getting from some of the very affluent parts of the south? And Kensington and Chelsea is not the only very wealthy London borough.

11:55

I think I'd make two or three points on that question. The first is that, because we're using the comparable method, which adjusts for the initial size of the tax base, what actually matters is not the absolute increase in the property prices, it's the percentage increase in property prices that matters for the block grant adjustment. So, if property prices go up by more in percentage terms in Wales, in general, that will have a bigger impact on revenues than it would in England, although you do need to think about actually the rate structure as well as the relative property value increases. It's actually quite complicated to work out the full effects, but it's not as simple as to say, 'a 10 per cent increase is £80,000 in Chelsea and it's £10,000 in the Valleys, and, therefore, they'll benefit more in England.' You need to think about where you're starting off from and what the exact distribution of property prices looks like in those two areas.

The second thing I'd add is that, actually, whilst, certainly over the period between 2010 to 2016, say, 2017, property prices were growing faster in London and the south-east than the rest of the UK and certainly in Wales, I think, actually, that has started to reverse a little bit recently and property prices in the north and Wales have been holding up a bit better than they have in the south-east of England and inner London in particular. Whether that continues or not, I'm not sure, but that trend has reversed slightly in recent years and actually could be impacting a little bit at the moment on stamp duty revenues in Wales. Because I think—and this is what Ed was saying earlier to us, or to me—the stamp duty revenues in Wales or land transaction tax revenues in Wales have come in lower than forecast—they reduced the forecasts—but the forecasts have come down even more in the rest of the UK, so Wales actually benefited slightly from the fact that the slowdown has been slower, or less of a slowdown, than in the rest of the UK.

In terms of the relative growth in Wales compared with the rest of the UK, if you look at past decades of the house price data, it does tend to be quite cyclical. You have periods of quite rapid convergence and then periods of divergence. And, as David said, stamp duty has been devolved right at the top of that wave, if you like, in terms of that the differential between Wales and the rest of the UK has never been greater than at the point of devolution. So, if it follows trends that we have seen in the past that property prices in Wales then converge again, that could be of benefit to the Welsh budget.

Isn't there a difficulty with that, however, that London is an international city? Very few Russian oligarchs and very rich Americans and Arabs buy property in Wales; they buy up bits of central London by the street. And with the pound having collapsed following—. The absolute cost in their home currency has reduced, so isn't there a danger, because of that, of even more money being sucked into London in terms of pounds from abroad, and that increasing divergence?

Well, certainly, a few years ago, when these proposals were being discussed by the Silk commission and the UK Government, we were very concerned about that issue and the lock, the tie-in, in that Wales's block grant adjustment would be linked to a market that, as you say, is entirely international in parts of London. But what we've seen since then—and Guto was mentioning this—is we've seen a house price drop in London and so what we've seen in terms of the revenue so far is that, although land transaction tax revenues have been a little off forecast, the block grant adjustment has been even worse hit in England, so we've actually had an advantage from that relative differential. So, while it's completely true that London is operating in a different market entirely, the historical trend is quite interesting, where we see divergence and then economic circumstances leading to rapid convergence in growth rates, and so, if we're following on from those historical trends, it's a relatively good time for this tax to have been devolved. We won't know the future, of course, with Brexit, with the internationalisation of the property market, and so on. We don't know if history's repeating itself, but, if it does, it was the right time to devolve the tax.

12:00

I think on that last point I would just add that whilst of course there's a potential upside from international transactions in London, there's also a potential downside from that as well. So, actually, because a large fraction of the properties are owned by international investors, by people that have used London as a base—because, actually, a fair part of the population working in, say, financial services and associated industries are from, say, the EU or other countries—. If that continues, and London continues to grow as an international centre, that could be an upside to London property prices, but if, actually, post Brexit, or post-clampdown on things with Russia, that declines, that could have a bigger impact. So, I think I would see it not necessarily as that there's—. There's more scope for upside in London, but there's also more scope for downside, and what that implies, actually, is some uncertainty about how this will all play out for Wales, although, as Ed says, it does look like we were at the top of a peak in divergence if things follow what they did in the past.

Yes. Thanks. The devolution of the Welsh rate means it's important for the Welsh Government to grow the tax base. What strategies do you think the Government could employ to do that?

I think we touched on it in our report, that it needs a cross-departmental approach in the Welsh Government, recognising that policies relating to the economy, skills, education and housing will have an effect on decisions of taxpayers and on the income of taxpayers in Wales. So, for example, one policy in higher education, a policy that incentivised more graduates to stay in Wales after leaving university, or of attracting graduates from the rest of the UK, or attracting Welsh graduates back into the Welsh economy, for them to become Welsh taxpayers, and increasing the skills base of the Welsh economy, might be a way to go. Is that fair?

Yes. I think the other thing that we noted in the report that we issued about six months ago now was the importance of high-paid jobs and closing the productivity gap. So, it's not that the Government is not interested in doing either of those things, but it becomes even more important in the context where the budget is linked in with the tax base—it's not necessarily the raw number of jobs in Wales, but high-skilled jobs that will help close the productivity gap that we have at the moment.

I guess it's important that, even though we have lower income distribution in Wales, the top 10 per cent of the highest earners, taxpayers, will pay around 70 per cent—or, no, will pay as much as the bottom 70 per cent, the lowest-earning 70 per cent, of income tax payers. So, even though we're less reliant on those with very high incomes, any changes to those types of jobs—

Can you say that again? The top—? I'm taking a while to get my head around this.

Sorry, yes. I'm trying to remember off the top of my head as well. So, the top 10 per cent of income tax payers will be paying the same amount as the bottom, or the lowest-earning, 70 per cent, because of our progressive income tax structure, and the fact that the personal allowance takes more of a chunk of people's earnings. So, in terms of the incentives for the Welsh Government, and in terms specifically of the income tax base, the incentive is to increase those higher-earning jobs, and the incomes of higher earners.

Of course, there are obviously other considerations for the rest of the tax base, but specifically in terms of the incentives, maybe, Welsh Government faces, those are the sort of jobs that—.

I was going to ask you about the behavioural change, and we talk a lot about migration. How real an issue is that? I think it was Gerry Holtham who said that you can move the basic rate around, and you probably wouldn't have that much impact, but the upper rate tends to be the more mobile jobs. I think he said that, if memory serves. Is that still the case? 

12:05

Yes, it's among the higher earners that you'd expect more of a difference, or more of a migration response. There's a good discussion on that in the Welsh tax policy document that was published alongside the budget, in terms of looking at international comparisons of the migration responses to the income tax rate changes. What it concludes is that it's very difficult to use international evidence, because Wales will be quite a specific case. The border is quite porous, there are big centres of population on both sides of the border—  

If Steffan Lewis was here, he would now say, 'It's a border like any other', and I would say in response, 'Yes, you've got a point but it is a very—'. A lot of people within Wales live close to the border, whereas in Scotland most people live in the central belt, don't they, away from it? 

Yes, I think it was about 6 per cent of the Welsh tax base—. Well, actually, no—6 per cent of Welsh taxpayers will be working in England, and therefore will pay Welsh taxes if they're domiciled in Wales, and because they tend to earn slightly more, about 8 per cent, give or take, of the tax base will actually be earned in England. So, the commuting will remain important.

Just on the behavioural effects, because we only have a share of each of the three bands of income tax—it's 10p in each band, but as a proportion of the 40p and 45p rates, the Welsh part of it is a relatively small percentage, so a quarter of the 40p rate to two sevenths of the 45p rate. So, Welsh policy changes only have a small effect on the overall decision of that taxpayer, so even if they change it by a couple of rates, most of the tax rate payable will still be going to the UK Government. So, the biggest potential behavioural effect would be on the basic rate, because that's where the biggest share of the tax is being collected by the Welsh Government, but these are the taxpayers who are least likely to move in response to a rate change.

So, unlike Scotland, where all of the non-savings and non-dividend income tax is devolved, and there's potentially more scope for that kind of cross-border migration, because of the way that this partial income tax system is working in Wales, I think it's less likely.      

Yes, I would concur with that. So, you might expect that in Scotland, because of the distribution of population, to some extent, there would be less migration response, but migration response would be more important for the Scottish tax base because the Scottish Government bears all of the impact, because it has the 20p, the 40p and the 45p. In Wales, you might get more migration response, but it will have less impact on the Welsh revenues because Wales only bears that on a final 10p—. It gets to keep all of the impact of the higher rate—say you increase the rate at the top, Wales gets all of the mechanical, direct effect of the higher increase in tax, but it only bears a small part of the behavioural effect, whereas in Scotland, it gets all of the benefit but bears all of the cost as well. So, I think there will be different incentives offered in the two countries. So, actually, it kind of makes, if you like—. All I'm saying is that probably tax increases make more financial sense in Wales than they do in Scotland, despite the fact that they may even have a bigger migration effect in Wales, because of the operation of the fiscal framework.   

Specifically on the migration point, because there's very little comparable—. There is no case study that we can use to gauge what the migration response might be to rate change in Wales. What we did in the report was to look at certain tax policies and changes in the tax rates, and work out how many taxpayers would need to shift or move from Wales to England, or from England to Wales, to reverse the policy. So, for example, the 5p increase in the additional rate—so, increasing the additional rate to 50p in Wales—would raise around, I think—I can't remember—it's about £28 million, but it would take around 850 average additional rate taxpayers to move from Wales to England. That's still quite a significant migration response, considering that the additional rate population is only around 5,000, so you'd have to expect a significant chunk of them to migrate from Wales as a response to income tax rises.  

On the other side, if you cut the additional rate by 5p, it would take more than 1,100 taxpayers to move in from England to Wales to make up the cost of that tax cut. I don't know how likely that is. There is a big pool of additional taxpayers. There are 360,000 living in the rest of the UK. There are roughly 40,000 to 50,000 additional rate taxpayers living within 50 miles of the Welsh border. So, then it becomes a question of—it's highly uncertain what that effect could be on the cost or the increase in revenues that you get from the tax change.

12:10

We have some experience of tax differences in terms of council tax, and the Americans have got exactly the same, haven't they? In terms of council tax, there is about a £600 difference between Newport and Blaenau Gwent, but we have not seen a flood of people. America has got lots of examples—none of which I can remember at the moment, but I'm sure you know far more about it than I do—where you've got states with substantially different—. Connecticut was relatively low and whichever was next to it was relatively high, and yet you had very little movement.

I would say on that, first of all, I think you need to distinguish between different kinds of taxes and the kind of responses we expect. So, the property tax, you'd actually expects a lot of the impact of that tax difference to be effected in property prices rather than in migration responses. So, there's a higher tax in Blaenau Gwent. People know that and therefore probably prices are lower in Blaenau Gwent than they would be if the property tax was lower. We just published an article by researchers in Denmark that looked at a tax change in 2007 where they changed property taxes—the Government imposed property tax changes across different local authorities, some went up, some went down, and within three to four years property prices were just completely 100 per cent changed to reflect that. So, I think for property tax, I wouldn't necessarily expect migration responses, I would expect it to be reflected in the property prices.

I think, for something like income tax, or if there was a local corporate tax or sales tax, that's where you'd expect there to be a bit more migration response because that wouldn't be reflected in property prices in quite the same way, I'd imagine. There have been studies in Switzerland that show that, actually, there are migration responses to differences in the taxes in cantons, although, of course, they are much smaller areas than Wales, or than the US states. 

Perhaps we can talk again about the US states because I'm going to get told off because that's outside what I was meant to be talking about.

We were very pleased that the fiscal framework agreement notes that the block grant adjustment will be applied separately to each band of income tax, rather than to the whole. Were we right to be pleased?

Yes, in short. When we published work before the fiscal framework negotiations, we had almost raised the alarm about the potential downside budget if income tax were given one block grant adjustment. The simple reason is the composition of the Welsh income tax base is very significantly different from the UK as a whole—far more than is the case with the difference between Scotland and the rest of the UK. So, what the different block grant adjustments for each band do is that they mean that the impact of UK Government policy decisions at different parts of the income tax distribution will be treated in a similar fair way across the various parts of the UK.

So, the example that we had raised was that increase in the personal allowance, which is pulling far greater numbers of Welsh taxpayers out of paying taxes at all, which, obviously, if you're looking for a healthy tax base, is something that you'd want to have a look at to make sure of a fair balance of people paying taxes and people being drawn out of the tax base. So, what this will do is it will treat—because there's one block grant adjustment for the lower rate, it means that decisions on things like the personal allowance are treated fairly between England, Wales and Northern Ireland, that it'll have a similar impact on basic rate payers all across the UK, and just, in the same way, if there are policy decisions that significantly affect the additional rate, which is much less important in Wales, but if there are additional benefits, it makes sure that we're treated on the basis of our very different tax base. 

I would agree with that. I'll just very quickly add, what this does, as Ed says, is it means the block grant adjustments move in ways that a bit more closely reflect what the Welsh tax base looks like, but it doesn't necessarily mean that Wales will always gain from that. So, there could be circumstances where, actually, Wales would have benefited from a single block grant adjustment. So, an example of that might be, for instance, if the UK Government said not to increase the personal allowance, but, say, freeze it, and then—'We need to raise more revenues to pay for NHS, et cetera. We've increased the personal allowance massively in the last few years, so we're going to freeze it at £12,500 for five years.' In that circumstance, Wales actually might have benefited from having—. That will actually increase the Welsh tax base more than it would increase the English tax base, and Wales would have benefited more from having a single adjustment, in that case. So, whilst this system actually seems to provide the chosen separate adjustments, a more closely matched tax base provides better insurance for the Welsh Government against policy changes that differentially affect the tax base—you can benefit from insurance, should that risk be the case, or you could end up paying the insurance and actually you would have been better off not paying it, if the risk doesn't go that way. You can think of it as a bet, to some extent.

12:15

Importantly, this is about giving the Welsh Government the incentive to increase the tax base, but assuming that the Welsh Government has powers over that area. So, in the absence of this kind of adjustment, the Welsh tax base is at the mercy of forces that are outside its control, and that doesn't seem to be really linked to the kind of accountability mechanisms for tax devolution in the first place. When we calculated the potential impact—had income tax devolution been in place since about the start of the decade, and if we'd had just one block grant adjustment, we calculated the hit to the Welsh budget, within about three or four years, to be in excess of £150 million, whereas this would reduce it way down to about £30 million. And at that time, of course, we didn't know about the trade-off on the expenditure side with the 5 per cent needs-based factor. So, it has significantly reduced the risk, as David said. 

Can I just say that I don't think the Welsh Government has needed any incentive to try and grow the Welsh economy? I don't think there's any political party here, who, if they were in power, would not want to grow the Welsh economy, whether—[Inaudible.]—more income tax or not. We don't agree on many things, we don't agree how it ought to be done, but we all want to see the Welsh economy growing. 

The question I've got regarding what you said about increased personal allowance changes is: we may be disadvantaged by an increase in personal allowance, which takes people out of income tax and reduces income, and that could well affect Wales more than it affects the UK as a whole, depending how many people are close to and around paying income tax. Would no detriment come in at that stage if movement in personal allowances had that effect of disadvantaging Wales? 

The three block grant adjustments for income tax are going to take care of that problem, to most extents. So, in the very first year, we have no detriment, effectively, so there'll be a full reconciliation. Then, going forward, an increase in the personal allowance would, all things being equal, change the effect on the English tax base at the basic rate, as it would in the Welsh tax base at the basic rate. So, the point of the three block grant adjustments was to do lots of things: one of them was to take care of the personal allowance problem, which could potentially, in the absence of that, be quite severe. 

It won't do it completely perfectly, because even within the basic rate range, you might see more concentration at the lower end of that than further up, so it won't be perfect. I don't think that the UK Government would invoke this as a no-detriment-type scenario. They haven't in Scotland, so far, where we have the same kind of effects going on. The no-detriment clauses, if they are similar to the ones in the Scottish fiscal framework—I should have looked this up before I came—

If they're similar to that, what they will basically say is there needs to be agreement on both sides on the concept of detriment. The UK Government, in the past, has indicated that it sees these as clear, demonstrable, large things that are not just part of the design of the fiscal framework. So, whilst I think an increase in the personal allowance would likely see a slight negative impact—a lot smaller than it would be if it was a single block grant—on the Welsh tax base, I think it's rather unlikely that that would lead to a situation where those principles are invoked, because that would also give a situation where, if there was a freeze in the personal allowance, we gain a little bit more from that, relative to England in terms of its revenues. Would the UK Government invoke that in reverse? Probably not. I think these small things—no detriment wouldn't apply. And, actually, when you look more generally, the no-detriment clauses are really quite unusual internationally. Other countries tend to, kind of, tolerate a degree that there will be interaction between tax policies in the different levels of government and they have not formalised this in quite the same way. 

12:20

Yes, okay. That's an interesting point. I just want to ask about new taxes or potential new taxes. We know that the vacant land tax is something that the Government is actively looking at, but there's also work happening around the disposable plastics, social care levy and tourism as well. I'm just wondering what potential effect you think that might have on the Welsh tax base.

Maybe in terms of the vacant land tax, if that's successful in securing development, bringing more empty land into development, it might impact the NDR—the non-domestic rate tax base, the land transaction tax tax base, and, then, if it builds more homes, that's more council tax revenue for local authorities as well. Of course, they're going to be quite minor in terms of—. Except for, maybe, the social care levy, they're likely to be quite minor in terms of revenue items in the budget. The tourist tax, if implemented at a local authority level, may—. Perhaps, given more sources of different revenues for local authorities, it might ease pressures on council tax increases. So, if they can diversify their sources of income, that may lead to less of a need to increase council tax levels. But, as I say, in terms of over the overall Welsh budget, these new taxes are going to be relatively minor except for, perhaps, the social care levy. 

Yes, and on the social care levy, which is obviously the bigger potential revenue boost from the various taxes we've discussed, I suppose it's the way that that's implemented that will have an impact on the tax base. So, if it follows the Gerry Holtham kind of model—that contributory model of different payments at different ages—although it's relatively complex, it would perhaps be more favourable on taxpayers of different ages than a straightforward income tax increase to pay for social care. So, a few pennies on the basic higher rates to pay for social care might have a disincentive on younger workers and perhaps an incentive on older workers to take advantage of the new social care package, if you like. So, the implementation of any such social care levy would potentially have an impact on the tax base as well. 

Yes, I would concur with that. Looking at the social care levy again, the kind of system that Gerry proposes is quite different to what we had in the UK. It's actually quite similar to what many people think national insurance is. They think that national insurance is actually contributory in a way that it isn't, but it is quite a departure from what we've done in the UK. But contributory systems do have one beneficial effect in the sense that the kind of incentive effects of them are normally not quite as negative, if you like, as general tax rises, because, 'I get something for paying something in; I'll have less incentive to cut my earnings or move away than I would if I was just paying into a general tax pot that didn't lead to any particular benefits going forward.'

On that social care levy, though, rather than just thinking about what this means for overall funding levels and how this could affect existing tax bases in Wales, I think the bigger question with the social care levy is how that fits in with our system of local government in Wales. Currently, social care is provided at a local authority level, whilst the national standards and national provisions for social care—actually decisions are made at a local level about certain parts of the eligibility criteria in terms of care needs requirements. If you move to a social care levy, would that be a national care levy? In that case, would people expect the same eligibility across the country? They're all paying the same levy. Would that mean centralising social care, taking one of the big areas off local government? On the other hand, if you had local social care levies, you might need quite a complicated—some equalisation between areas, or quite different tax rates in different areas depending on how big their tax base is. Put simply, Monmouthshire would have a lower tax rate than Blaenau Gwent, potentially, for a social care levy. Cardiff would have a lower one again because, actually, it's got quite a young population. So, I think there are bigger questions that open up about that social care levy, rather than just the current taxes. It's about how Wales is organised as a country.

12:25

Okay, thank you, and, again, that's an area that we've grappled with as well. So, just to conclude, from me—I'm not sure whether other Members might want to add any further questions—I'm just wondering whether you feel that the Welsh Government publishes appropriate fiscal data to enable effective scrutiny of its policies or should additional information be made available?

On an initial point about that, we welcome the fact that the chief economist report this year is particularly useful, looking at some of the medium-term projections of the Welsh budgets. I think that's quite a useful new addition to the budget document that's been released—the fact that we can look at the medium-term outlook for the Welsh budget.

It raises the point about the NHS consequentials coming to the Welsh budget and whether that will be funded by tax increases, more borrowing or by cutting departments that are also devolved, and the net consequential might be different for Wales. Those sorts of discussions, especially in the context of setting future income tax rates and having a sense of where the budget will be going over the next couple of years, I thought were really useful. 

It has significantly improved in recent years. It needed to, especially in the context of the new fiscal framework. I think, coming back to the accountability point, one of the things that's important is that the Welsh public and civil society can work out—. Okay, say the block grant has fallen, from Westminster, is that because, actually, Westminster's cut the budget it is giving to Wales, or is it because the block grant adjustments have gone up? Because, when working out what is driving changes in budgets, a risk could be that income tax revenues go up in Wales and they go up faster in England. If you just show the block grant and the income tax revenues, there's a risk that the Government could say, 'Look our revenues have gone up, but what's happened—the reason we're having to cut—is that they've cut our block grant.' Well, actually, what is happening is that your revenues have gone up but they've not kept pace with the rest of the UK, so, relatively, you've fallen behind. And that's quite a different circumstance to, 'They've cut our budget', versus, 'Our tax revenues haven't kept pace with the rest of the UK.' And, I think, the fact that Welsh Government is putting that information out is useful, because it helps people make those decisions and scrutinise the Welsh Government as they should be able to. So, I think, that's a really important thing.

One thing that Ed reminded me of earlier was that one of the difficulties we've had when we looked at what's happened to the Welsh Government's spending on different areas in the past is the constant changing in the departmental categories. What the Scottish Government does in its budget is that it has a table where it puts everything consistent, back six, seven, eight years, on the new categories. That hasn't been done in the Welsh budget in the past. I think it's been done for one year now, but not back further. And I think that, actually, providing those figures on a consistent basis back until 2010, as far back as possible, would allow scrutiny, to say, 'Look, can we compare what's happening to the Welsh Government in terms of health, education, et cetera, to the decisions being made in London, decisions being made in Scotland?' That's one area where I think still more work is needed, because, at the moment, I can look back one year. Looking back further, I need to do a lot of complicated calculations myself. 

And just very briefly on that. It's, obviously, not a ministerial priority to sort out the complex recategorising of expenditure from five, six years ago, so we understand why that wouldn't normally happen. But, actually, almost every week in Prime Minister's questions in Westminster, you'll have back and forth arguments about statistics, about how much Welsh health spending or education spending has varied compared with England. So, actually, it's really important for the public to see that historical record in the new budget lines as well.

I was going to say that the other way of doing it would be to break everything down into functional areas, so that it doesn't matter if you move a function. If you spend £20 million on sport, it doesn't matter where it sits—when it moves, the £20 million moves with it, and then you could decide whether, within that, an increase or decrease had taken place. 

Yes, and that's what the public expenditure statistical analyses—the Treasury's analyses—try to do. The issue is that that is only backwards looking. What the Welsh Government's budget has going for it is that it's forward looking as well. So, as well as looking at what's happened in the past, you can look forward as well. That's why I think, actually—. Whilst I can agree that it doesn't seem a priority in the first instance, putting it in the budget is a useful thing, because you can see how the Welsh Government has prioritised different areas in the past but also what its priorities are for the future on a consistent basis. That's what's been hard to do, given that—. Unless you go line by line yourself and try to reallocate things, that kind of thing is not as easy as it should be.

12:30

Thank you very much for your evidence—really valuable stuff. We appreciate it very much. You will be sent a copy of the transcript to check for correctness. With that, can I thank you again? The committee will now move into private session. Thank you.

Daeth rhan gyhoeddus y cyfarfod i ben am 12:30.

The public part of the meeting ended at 12:30.