Cynulliad Cenedlaethol Cymru

Thank you, Chair—bore da. I'm Mike Usher. I'm the Wales Audit Office's sector lead for health and central government, and I'm the auditor general's observer on the programme board that the Welsh Government has for the implementation of the Welsh Revenue Authority.

Morning, Chair. I'm Richard Harries. I'm the financial audit director, responsible for the work in front of us and the report today.

Morning, Chair. Matthew Coe. I was the audit manager in charge of the project.

Certainly—thank you, Chair. It's just over 12 months since we produced our first report on preparedness for fiscal devolution, and I think it's fair to say that considerable progress has been made during that period by both the Welsh Government and what is now formally the Welsh Revenue Authority. It has come into formal being.

In our original report, we made some recommendations that further details were needed around action plans. I'm pleased to see that the Welsh Government have acted very fully, I think, in response to our recommendation, which they accepted at the time. In my work as an observer on the programme board, I've seen those plans come through, and I think the fact that in our latest report we've made actually no recommendations probably itself speaks volumes and is something of a statement of confidence that the programme is in a pretty good place at the moment. We see no merit in making recommendations to simply say you should continue to do things, and so our report sets out where they've got to. But it's fair to say that a lot of work has been achieved.

There is much for them still to do before 'go live' on 1 April. Our work provides a snapshot as at the end of November, and clearly things will have moved on even in the last few weeks since then. But it's fair to say, in opening summary, that the programme, I think, is in a good place, and our report really sets out how they are getting on on a number of fronts around the various workstreams.

Thank you, Chair. No, there's not, really, and I'd echo Mike's comments. This report was done or the work was done in November and the report was cleared through December, so December has come and gone and we're into January now. I suppose it's a question you need to pose to the Minister and the officials after this session. But, no, I think the plans we saw in place to take the necessary action on the items we identified are things they were still working on. They were all in place. We did raise some issues in terms of some of the digital work, which I dare say we'll touch on later. But, again, it was pleasing to see that the plans are now there to get to 1 April and before.

We haven't gone back in since the report was done. There's a programme board next week, I'm aware of. What we have done in this report is put a sort of marker down that, if things don't progress as they need to, there's an opportunity to come back in February and go in there, and report back to you as a committee, to see where things are not working. But, as it stands at the moment, I think things are going to plan and are looking fairly good for the 1 April go live.

When are you next likely to have a sense of whether things are going to plan? You mentioned a programme meeting or whatever—that'll be the next opportunity, yes?

There's another meeting of the implementation programme board next week, which I'll be attending as an observer, and I think we will get a sense from the papers and the discussions at that meeting as to where things currently are. There is certainly a lot of work going on this month around the final beta testing of systems and developing guidance. I'm not expecting any nasty surprises next week, but, obviously, I'll wait and see. But, as Richard says, we've put a marker down that if we do feel that there are concerns, the auditor general is prepared to issue a further report before 1 April, but only if we feel it's necessary.

We have also put a marker of our intention to produce a further preparedness review in the autumn next year, which, in part, will look back over what hopefully will be a successful launch on 1 April, but also, then, will look at the preparedness for building up to April 2019 with Welsh income tax coming through. Obviously, the sums there are much bigger, but that's really around HMRC rather than the revenue authority.

Indeed. Okay, thank you. Jane Hutt.

Yes. That leads me nicely on to questions about preparing for the Welsh rate of income tax in terms of tax forecasting particularly, because you have acknowledged in your report that Welsh Government has prepared timely forecasts of revenue for the newly devolved taxes and is developing its approach to the Welsh rates of income tax. 

Obviously, there is an issue about data. How suitable do you feel HMRC data are for modelling the Welsh rates of income tax forecasts?

That's actually a very good question indeed, because it's fair to say that, obviously, this has not had to be done before for Wales, but HMRC have had to grapple with those issues for Scotland, and, again, Wales is in the helpful position of being able to build on the experience from both HMRC and the Scottish Government around developing the Scottish rates of income tax.

With that process, it is well documented that there were some issues initially around the identification of Scottish taxpayers and around postcode completeness and accuracy, and further work was needed in quite short order to address that. HMRC, clearly, have reflected on that as they've gone along.

In terms of forecasting quantum of revenue, whilst the tax rates remain the same between the nations, that is rather more straightforward. As soon as you move to any divergence of rate, that starts to raise complications around predictions around forecasts of revenue, and, indeed, starts to influence behaviours, evasion risks, et cetera, et cetera. So, it is problematic, but I think the fact that they can learn from the experience over the last two or three years with HMRC in Scotland, I think certainly puts them in a better place as a starting point.

Yes. I mean, obviously—

Could I just go to Mike specifically on this, I think, if that's okay?

Yes.

Scotland is different to Wales in many respects. One of which is that there's very little daily work travel across the border in Scotland, whereas there's substantial daily work travel in both directions, especially in north-east Wales. It went substantially wrong in Scotland, with some very high-profile cases of people who were not identified easily in the right place. That, I can understand. I can also understand the problems with people who are working in England or working in Scotland and whose main home is in one or the other. These are problems that Wales will have as well, but there's a huge amount of cross-border movement there. I mean, just take two of the big employers—Ellesmere Port car factory and people going there from Wales, and the Airbus factory and people coming from England. I mean, you've got lots of major employers. I don't see any problem with the major employers—they'll know where people live—but you've got a little garage in Wrexham and a little garage in Chester, which may not really have much interest in where their staff live, although they collect pay as you earn. What work is being done to try and ensure that we get down to those much smaller employers employing perhaps one, two, three, four or five people, so they actually know where their staff live and where they live at any one time, because a lot of people move across from Wrexham to Chester and back again. Because, as anybody who's driven up to north Wales knows, you travel in and out of Wales and England several times along the border journey.

Ineed, yes. Yes, absolutely. And I think there is a deal of work here to be done, primarily by HMRC, but there are things here around engagement with the large employers but also, as you say, Mike, with smaller employers. You're right, it is more of a problem in terms of the volume of cross-border travel along the England-Wales border than the England-Scotland border. But the definition of a Welsh taxpayer is defined in terms of residence rather than where people are working, and so it's getting that accuracy around home addresses that will be absolutely key.

Yes. I'm just saying that some of the smaller employers don't necessarily—. They might have the address, but they might not know whether it's in—. They might not actually have it with the postcode, because they might not need it for any general reasons.

HMRC will need it now. That's their clear responsibility.

Yes. But it's making sure that they get that in to HMRC including postcodes.

Yes, it needs to be a validated postcode. At the end of the day, the onus is on an individual to comply with tax, obviously, but, clearly, the employers play a major part here with PAYE. So, it's really around working in partnership, communication and engagement.

And I think, just following that up, it would be—I'm sure it probably would be helpful for us to hear from you again in terms of preparedness, as you mentioned earlier on.

Our work in the autumn report will get into those issues certainly, I'm sure.

This is a key point for Welsh Government as well in terms of their discussions with the HMRC, because they've got to learn about the new responsibilities they're taking on in terms of devolution now for Wales, as well as Scotland.

So, I think it might be worth going on to the next issue, which is, really, looking at partnerships and joint working, particularly between Welsh Revenue Authority and its external stakeholders. You do acknowledge good relationships are being developed, but how effective are they, now it's got to move on to the effective working relationships between Welsh Treasury, Welsh Revenue Authority and their partners?

Thank you for that. I think, as we say in the report, what we've seen to date has been very strong on those relationships. The engagement has been widespread. The key stakeholders are involved in a lot of the discussions and the processes are being developed. I think the key for the Welsh Treasury and the revenue authority is the framework document that's being developed at the moment. Early this year, that's going to be finalised, and that's going to set out a clear understanding of how the two bodies are going to work together. So, that's a key, integral part of this. 

But I think some of these things need to be explained constantly, probably, to the wider stakeholders as well because the independence issue with the WRA is vital, the perception of how that's working is vital. I think a lot of work's gone on and I think we comment that good progress has been made in that area. It is the first non-ministerial department in Wales, and that needs to be explained constantly, probably, as things develop. But, as a fundamental process, through this year, what we've seen this time and last time is the engagement has been very, very strong, and the plans going forward are there to take that forward as well. So, there are no worries coming up from this point of view at this point in time. We will go back in the autumn next year and we'll have another look in terms of the look back for some of the development of the WRA, and, hopefully, that will carry on then.

Yes. Clearly, the Welsh Government have worked very closely with setting up the tax advisory group and the tax forum, the sort of precursors to developing these working relationships. There are many bodies that they have to work closely with, local government in particular. They are represented, I know, on both the advisory group and the forum. How have you assessed the developing of the relationship between WRA and local government bodies?

I think we've seen that evolving over time. They recognised from the outset that there was much that they could learn from local government, particularly around revenue collection, non-domestic rating, council tax and things, and they have looked to draw on some of their expertise. You mentioned participation in the advisory group and the tax forum. Also the lead official within Cardiff Council on revenue collection is also a member of the implementation programme board. Again, I'm seeing that individual making a very helpful contribution, bringing that local government expertise and perspective around some of the real practicalities around collection rates and things. So, I think they're certainly drawing on recognising the expertise that's out there, and, I think, certainly, once the WRA is up and running and into a settled state there's certainly, perhaps, a useful area for them to explore around moving into working very closely with local government around the whole general concept around revenue collection from Welsh taxpayers who are paying both local and national taxes and there's scope for cooperation and coordination there to help tackle non-payment, evasion rates, compliance, and those kind of issues. So, there is a lot in local government that they can work with.

Obviously, strong partnerships already exist with Welsh local government, but some of the other agencies will struggle—England and Wales border particularly, the Land Registry and Valuation Office Agency. I think, again, hopefully, those working relationships are robust. 

Yes.

They have done quite a lot of work to identify the individual properties along the border and to understand exactly where they sit then in terms of the various tax regimes that will apply from 1 April. So, they have gone down to the nitty-gritty, if you like, in that respect. 

Okay. Nick Ramsay, then. Thank you.

Diolch, Chair. Good morning—I think it's morning. Yes, it's still well and truly in the morning.

You've said in the past that you think the Welsh Government need to do more to engage the wider public in terms of getting the message out there about taxes. I was discussing the issue over Christmas and the new year—different meetings—and I was horrified, really, by how little awareness there is of the devolution of—particularly of income tax, I mean; stamp duty's been in the news a little bit more recently. How do you feel about the Welsh Government's performance in terms of getting the message across? Okay, maybe the WRA, that's a trickier sell because people are used to HMRC, but, in terms of the actual fact that taxes are going to be paid here increasingly, how do you feel they're doing with that? 

I think, with public engagement, they recognised from the outset that clearly it was very important, but they've had to prioritise. So, with the two taxes coming on stream immediately, they engaged very promptly with the various landfill operators and with the players in that particular industry. With stamp duty and the property tax, they were engaging then with the profession, with solicitors, the Law Society, estate agents and things. 

In terms of the general public, I think, for these two initial taxes, the level of personal public engagement with paying the taxes is relatively limited, because it's at third hand; it's your solicitor conveyancing who handles the stamp duty payments for you. The Welsh rate of income tax is where it's really going to hit home that there is something that is new and different and people will need to respond differently to it. So, that's where the public engagement is really going to need to ramp up over this next 12 to 15 months. There is almost a danger of starting too soon with that process in that you can put an announcement out and that loses currency quite quickly. So, actually, timing that phased communication and building up of awareness, understanding, realisation of what it means for them has got to be accompanied by clear guidance, a user-friendly website. The Welsh Government have plans in place for those things, but I think to do too much too soon in terms of engagement with every single taxpayer in Wales might actually prove counterproductive. So, I think they're pacing themselves. They've got some quite sophisticated communications expertise, I think, within the programme, which they're drawing on with their plans. 

I think that's a very fair assessment, actually, in terms of where we are. I suppose, with landfill tax and with stamp duty and the earlier taxes, with those stakeholders, (1) they're professional bodies, and it's important that they're aware of the obligations in terms of paying it, but, with regard to income tax, that, simply, apart from self employed, is going to be taken out at source by WRA or HMRC—I don't know quite how it's going to happen.

I'm concerned more, I suppose, rather than the mechanics of it happening, with the public acceptance, certainly over the first few years, of the way it works—the political acceptance, I suppose. Do you think in terms of that the Government are—? You've said that you think they're pacing themselves. Do you think that they're aware of the—I was going to say 'backlash, that's probably the wrong term, if taxes go up, but aware of the public acceptance of the new regime? 

Yes. It is an interesting question, and I think, the last Westminster budget round, with the changes in stamp duty rates was actually quite helpful in just bringing to the forefront the fact that differences in tax treatment in the different parts of the UK are starting to emerge. So, that's probably helping around general realisation. There are things that can be done around social media and those kinds of things, as well as formal communications. There's more, certainly, that needs to be done, but trying to use as many different channels of engagement as possible, I think, will also be important for them.

Okay, and on the wider issue of the Welsh Revenue Authority and the Welsh Government, do you think the relationship between those two has been understood clearly by stakeholders, or do you think more work needs to be done there?

For professional stakeholders, I would say so, yes. Certainly, for the public, I think that public awareness campaign is probably yet to really happen. I think the Welsh Revenue Authority has just launched its website, but it's relatively rudimentary at the moment, and will be built up over the next few months, I'm sure. But, for those who need to know at the moment, I think there is awareness and understanding, but, as Richard says, the point around being a non-ministerial department and assurances around the confidentiality of taxpayer information, those very important constitutional points, cannot be re-emphasised too often, really, to ensure that public confidence is maintained right from the outset.

And do you think the independence of the Welsh Revenue Authority has been effectively guaranteed and understood? I remember when Jane Hutt was the finance Secretary, I was very concerned about people being unhappy with the WRA and simply sending their tax returns to the Minister, which would have caused many, many sleepless nights, I'm sure. So, do you think people are aware, as they would deal with HMRC, that they are dealing with the Welsh Revenue Authority and not directly with Assembly Members and Welsh Government?

I'm sure they will very rapidly become aware. You're right, there is certainly a conceptual risk there, but I think the constitutional separation of the legal creation of the entity, a non-ministerial department with its own board, the very clear separation protocols around use of information, data transfer and things—those safeguards are all in place. So, the mechanisms are in place to deliver something that should work effectively and independently. The key thing, then, is to communicate and reassure the general public that it's in place and it operates effectively, which is partly where audit comes in, of course.

Great. And, still on the WRA, can you elaborate on the potential risks to effective governance, concerning the dual role of the chief exec as the Welsh Revenue Authority implementation programme implementation director?

I think that is a risk that was probably brought to our attention as part of the review, in a way, given the chief exec was covering both roles. And I think they've identified it as a potential risk to them, in terms of the perception, again—in terms of the work he's doing to set the new organisation up, and then running the new organisation. I think he wanted that extra scrutiny, that extra transparency, in the two roles he was performing. They've put a governance arrangement in place now with the permanent secretary, the chair of the board, himself as the chief executive, and the director from the Welsh Treasury, to actually challenge some of the decision making over and above the implementation board. And I think they've put that in place to make sure that, when they get to 1 April, it can't come back to bite them. Because that was the worry—he's performing two roles, a service to himself. Having this in place, over and above, helps him get through that and gives the view to the external world that it has been done properly as well. So, there was a potential there; I think the work he's put in place has addressed some of that.

Just before you come in—. The finance Secretary invited me to one of the early meetings in establishing the embryonic—not you; I think it was Mark Drakeford, actually—Welsh Revenue Authority, and I remember some big discussions about the balance between the chairman and the chief exec, and whether it would be a hands-on chairman, whether it would be a long-term chairman, or a chairman there purely to set it up. Do you think all those issues have been adequately dealt with? 

Yes, I believe so. There is now a chair and board in place. The board have met. The WRA has its own offices with a plate on the wall. The role of the chief executive also being the senior responsible officer for implementation has been—. I think it has been interesting to observe how the programme board governance has evolved over time, and, in fairness to those involved, I think they've been aware of the governance risks and they have managed those risks really very sensibly. I think it's pragmatic risk management through the process. The governance and ownership of the WRA itself has evolved, if you like, because it began with the Welsh Government building an abstract thing, which was going to be called the WRA, being built by the Welsh Government. There comes a point where, having appointed a chair and a board and a chief executive, the thing now building for them becomes actually the WRA building itself, and that migration of responsibility and ownership. They're moving through that process at the moment, so, quite sensibly I think, rather than saying there's been a firm cut-off transfer and at this point everything moves across, they're handling it over time, as things evolve. I think 'pragmatic' is how I'd describe it, but it's actually been quite effective from what I've seen of it in operation around the governance. So, the dual role challenge was recognised as a risk, and I think they're managing it very sensibly. 

Just for Members to be aware as well, you all will soon receive an invitation to a WRA event here in the Assembly, which is for all Assembly Members, and hopefully we'll start to—I'm sponsoring it obviously, so that's how I know—share some of this information. I think it'll be one of the first times that the WRA has come as truly independent to the Assembly to share that with Assembly Members. So, in turn, that starts the information flowing, and we, as Assembly Members, are also part of sharing with our constituents, of course, that this is going to happen and how it's going to happen. Just for you to be aware that that's in hand. So, we'll go on with David Rees now. 

Diolch, Cadeirydd. Actually, highlighting that point goes on to my first area of questioning, because in your earlier reports you identified four tranches of milestones and deliverables. We are now in the fourth, and we've got communications being clearly one of those points. I'm not talking about the communications regarding the Welsh rate of income tax; this is more about what goes on as of 1 April 2018. You've also identified that you think there's expertise now in the communications within it all. Have they done enough to satisfy you that sufficient communication will be issued prior to 1 April 2018? 

Asking us to crystal ball gaze I'm always very nervous about, in terms of making predictions. It's fair to say I think that as at November as we completed our work we were happy with where they had got to. We have looked at the plans they have in place for the many activities through January, February and March to get to go-live. Beta testing of systems—   

I'll come back to that, don't worry. 

The communication plans are in place, guidance is being developed and drafted and that's being itself tested with stakeholder groups for understandability. They have all the pieces in place in the jigsaw and the plan to get themselves safely to 1 April. Whether that will happen, I don't think we would want to commit, clearly.  

But your belief is that at this point in time you see no reason why it shouldn't.

It is well set, I think, to deliver as planned. 

Just to add to that as well, if you look back at the plans before the November review, they've achieved those milestones as well. So, the performance up to that point has been strong, but as Mike said, there's still a lot to do now before we get to April. In trying to badge this report, I'd say it's good progress, but there are some critical months coming with lots to do, which they're aware of, and there are things that can go wrong in that period, unfortunately. That's why we've got to make sure that the plans are there, and we can comment on the plans. Delivery now is the next stage of that process. 

Okay. Let's go on to the beta testing of systems now, because one of your concerns was the digital side of things. 

The area of highest risk is the digital services work stream, certainly. 

I suppose the first question I want to ask is: are you disappointed that, having the history of systems development and always being late and over budget, in fact we are in a situation where there seems to be a failure to completely identify the specifications for the design, which delayed it, and no agreements on how you would manage changes during that process? Because that seems to be fundamental to the development of these systems. Are you, therefore, disappointed that those things have caused problems? 

I don't think I'd want to characterise it as disappointment. I think this was always going to be potentially the most problematic area of the programme. You've referenced difficulties with other IT systems developed in the past. It's well understood that any new system you develop from scratch for a new function involving extensive IT, bespoke, is going to be complicated, it's going to be risky. They recognised that from the outset. In our report last year, we flagged that they were behind where they felt they needed to be at that time, and I recall the auditor general giving evidence to this committee and emphasising this was the one area of particular concern that he had. Since then, they have clearly done a lot to address those concerns. The contingency time that they had within their programme has effectively been utilised as part of managing that process. So, it's probably safe to say that there is very little contingency slack left now in the timetable to get to 1 April.

Having said that, they have made some very considerable strides over the last two to three months in terms of getting beta testing of the various components of the system—the functionality and things—working together. We've reported the position as at the end of November. I suspect that we may well want to get the very latest position from Welsh Government later, in terms of what's happened since then, but, 'disappointment' is probably not the word. We have not been surprised, perhaps, but it's fair to say that they have had a particularly complex set of challenges to address here, against the backstop that it has to be ready by 1 April. Ordinarily with a large IT project, it is possible to slip the completion date and the implementation date. You are up, here, against a hard deadline, so the contingency has been in terms of timing in the timetable, and then bringing in additional resource capacity and expertise in at short order to keep things on track. But they have had a difficult management challenge.

Well, perhaps I would question that, and I will question the Minister on that. But, in a sense, now, we are basically in a just-in-time situation. It has got to be delivered on time now. 

Yes.

Do you believe that the progress they have made will now allow them to deliver on time? Subsequent to that, what are your views on—? Would they be better off delivering something on 1 April, or actually having to delay something to make sure they get it right? You can actually deliver something on 1 April that has still got faults in it.

In our report, looking at paragraph 2.31, we talked around the areas of focus for the digital work stream. One of those was around contingency plans. I think we are aware that those contingency plans are in place. Obviously, with a contingency plan, their hope is not necessarily to call on it, but those plans are there. But, the work that they have done around end-to-end testing over the last few weeks is nearing completion. As I say, you will need to get the very latest position from Welsh Government. Again, I don't think we are in a position to make predictions as to what will happen between now and 1 April. I don't think, as auditors, we can sensibly do that. What I think we can say is that it's clear to us that they have robust plans in place to get them to that finish line, or, indeed, to activate contingency should it be necessary. So, they are in about as good a place as we could reasonably expect, given where they were a year ago.

We'll have to ask the Minister some questions.

Indeed. But from your point of view, what is a contingency in this regard? Have you looked at that contingency?

Contingency plans will be things like what is the absolute baseline minimum defensible specification that absolutely has to be delivered, rather than some of the things that are very nice to have but not as mission critical—

So, we are not looking at an alternative to their own system in that regard, when you say 'contingency'. They are not looking to default to HMRC or something like that.

That's not my understanding, no. 

So, your understanding is that it is the minimum you need to have to operate on 1 April.

Yes, and bear in mind that a key milestone next month will be that final readiness sign-off, which involves both the revenue authority, the Treasury and HMRC. There's a formal process there involving Ministers, which will be the final go/no-go on a lot of this stuff.

[Inaudible.]—the question, really. If we did this review early like we planned this time last year, in the June time when the gateway was happening, we would probably have more concerns. But, with the gateway and the way that they have used that—the findings and the actions—by the September time we'd seen a step-change, to be honest. They've got the expertise in to take that forward. They have got the plans in place to get there. Whether they get there—I think that's the challenge to the Minister next. I think that, in June, we would have had more concerns. By September, we could see the progress being made. I think a lot of the contingency was probably lost right at the start of the process. I think they had five or six months built into the original plans to get there. Putting the supplier in place: that was four months later than they planned, given some of the learning that they had and some of the developments and complexities. So, as of doing this report, it looks like the plans are now there to get there. There's very little contingency, if any, left. 

Okay. Neil Hamilton.

I'd like to ask some questions about recruitment and retention of expertise. The revenue authority is projected to have 75 staff members. Nine of those will be for a transitional period ending in 2018-19. Nineteen have already been appointed. Fifteen have taken up their posts. Of the 19, 15, including senior posts, are on loan from the Welsh Government or from HMRC. We have pointed out in the past on this committee the need to retain tax-related expertise in Wales. The revenue authority, obviously, would be very keen to do that, or it won't be able to operate properly. Do you feel that its recruitment process so far is robust enough to achieve that aim?

Again, coming back to the session last year, I think that was one of the auditor general's concerns, and the committee shared those concerns. I think the recruitment process so far, from what we've seen, has been very good and the calibre of staff coming in has been very, very high. You are right; the issue is how we maintain that staff given they're on secondment. It's an issue they're aware of as a Welsh Revenue Authority. They've got some learning and development plans being developed to share some of that knowledge, but it is a risk to them. As you say, 15 of the 19 are on secondment. That's a big proportion.

They've got some recruitment now over the next couple of months where they're getting another, I think, 44 staff in for 1 April. They've got plans to do it and they're achieving those plans from what we saw, but they've got to do that as well. I think the challenge they see and we see in the report is, going forward, how you maintain that enthusiasm of a new organisation, the secondments, and sharing those skills. That will be a challenge, of which they're aware. I think it's getting the right people in to start the process off is what they've achieved by doing it this way. How they build on that going forward, I think, will be something they need to consider.

Yes. So, is there anything more that the WRA could do, do you think, to improve retention of key skills?

We've referenced that in our report—I'm looking at paragraph 2.17—recognising the extensive use of secondees and getting robust succession planning in place, whether that is through internal promotion or further recruitment or indeed extension of secondments. But then it's the knowledge transfer, as people leave, that they're not simply walking out of the door with the knowledge in their heads. So, they have recognised the need for that to happen. They are, I think, putting in place plans to ensure that that happens, but clearly we've flagged in our report that there is more they need to do in managing that risk. You're absolutely right; with so many secondees, particularly in senior posts, succession planning and knowledge transfer will be key.

Well, you've anticipated my second question there, because knowledge transfer is not something that is easily carried out, especially in highly technical areas of work such as complicated statutes on taxation. Although I imagine that everybody who's being recruited has tax expertise to begin with, they're not simply being recruited straight from university or something like that. So, there is a basic expertise already in everybody who is going to be employed by the WRA. So, what are the processes that they've put in place to ensure this knowledge transfer at a sufficiently high level for everything to proceed seamlessly after the transitional staff have returned, presuming they're going to return, to their original places of employment?

I think I'd characterise that largely in the 'work in progress' box. It's something I think we will want to return to in our preparedness report later this year. But, with knowledge transfer, there are issues here around tax policy and things, but there's also then development of case precedents, and actually the broader context of knowledge management within the organisation is perhaps equally important here as well. But, they're alert to the risk and they need to ensure that they've got learning and development plans in place, which are pretty much bespoke to individual members of staff. There aren't that many staff and, while there's an element of sort of sheep-dip training and things, I think a lot of this is going to be bespoke to specific functions within the organisation, as it evolves. 

So, this is something to come back to.

I believe so, yes.

I think the only other point I would make is that, because the WRA is developing its own guidance, it has seconded staff and permanent staff that are working together to develop that guidance for themselves. So, whilst the seconded staff are bringing in useful expertise from places like HMRC and UK Treasury, what they're actually building is something unique to the WRA. So, in terms of that knowledge transfer, that's happening as they go along as well.

When you say 'guidance', are you referring to the manuals that HMRC produces—?

Yes—and the internal processes and procedures that they have within the organisation itself.

Okay, thank you very much.

'Sheep-dip training' is a new one on me. So, that's one for 2018, by the looks of it. Mike Hedges, please.

It's got nothing to do with sheep. On a WRA dedicted finance system, one of the things you've said quite regularly, and I've agreed with you, is that the sharing of systems saves money. You also say, I think in your report, that it’s going to cost about £0.5 million to set up and that there will be additional ongoing costs by not using the Welsh Government one. Are you convinced of the rationale for having a dedicated system rather than sharing an existing one?

Looking at paragraph 2.34 of our report, I think we pick up on this in some detail, Mike, in terms of the rationale.

I think this was a change in stance. The intention at the outset was that the WRA would be piggybacked on the Welsh Government’s own financial systems—in effect, a separate company set up within the Welsh Government’s ledgers.

I think they have thought long and hard around the pros and cons this, and it is a fairly finely balanced judgment as to which way to go. They’ve opted to go for the stand-alone system. In our report, we’ve set out some of the rationale for that, and it has included looking at the experience of Scotland, but also, I think, recognising that the WRA would be, in relative terms, against the size of the Welsh Government’s finance function, a relatively small player, and to have the influence to get its own precise specifications exactly met by the Welsh Government will be not as easy as if it took control of it itself. It has the benefit of emphasising, again, the separation of the WRA from the Welsh Government, and this will be something more around the perception than the reality, because, clearly, a separate ledger would work.

But there were some challenges there. It has resulted in additional cost, and again, our report brings that out, but it does mean then that, as new taxes come on stream, they are within the control of their own destiny in developing their finance systems to keep pace with that. So, I can see why they’ve taken the decision they have.

Yes. I was going to say, of course, that's the same rationale for smaller authorities not to work with larger authorities. And you've said on a number of occasions that the sharing of back-office functions is a good thing, and now we are saying, 'Well, yes, it's a good thing for local authorities, and it might be a good thing for other bodies, but when we come to the Welsh Government and Welsh Government bodies it's not necessarily a good thing.'

I think, on this one—

I think Mr Hedges has read your previous reports. [Laughter.]

I don't want to add to that; I just wanted to make that point for the record. The other thing I was going to say is: is the finance system to be created in parallel with the collection and management system, or will it happen afterwards?

Sorry, I didn't catch that.

Is it going to be done in parallel? Is it being done at the same time as—

Yes, it is happening in parallel.

And they'll both come to fruition at exactly the same time and, as such, you'll be confident they can talk to each other, if they need to.

Yes. Part of the key to this is the electronic interface between the tax collection management system and the core finance system. That's one of the fundamentals. So, yes, that is taking place.

But you'll only know if they can talk to each other when they're both completed, won't you?

Some of the functionality around communication, that is happening as part of the beta testing. So, that is working through. It won't be a case of turning it on 1 April and hoping it works; it is being developed in parallel.

If I can just complete this: it's developed in parallel, but you've got two parallel systems, and you've got testing of how they are working with each other, but you won't finally know whether they'll work completely with each other until they're both completed, will you? Or am I missing something?

That's what the testing is doing.

Yes, but—. You use the beta testing and you see how they're working at different stages, but until both are complete, how will you know that everything is going to work? Because there will be other bits being added to both as they make their way to completion.

The two systems are being developed in parallel. The key linkages between them are being built and tested. Other developments we've made to both systems wouldn't affect those linkages. So, I'm not quite sure I see where you're going with that.

If I can, another way of asking this question is: is this something that is being monitored by the implementation board? Is that something that you're keeping an active eye on to make sure that this interoperability is—

Yes, the digital services work stream is particularly focused on the interoperability of the systems, and they are providing assurances to the board that those linkages are being tested, and it's developing nicely.

And there'll be no linkages taking place after a certain time? The bits that are added after a certain time will not involve linkages. Is that what you're saying?

That's a level of detail that I simply wouldn't have here today, Mike.

But it is a matter of great importance.

I think it's probably a question to the—

A question to the later witnesses.

I've got one question. You mentioned the governance and management of changes to specifications. Are you comfortable that the governance and management of changes to specifications to the collection system would be considering also the possible implications upon the financial system? If more comes in, are you comfortable that the process is in place so that, as they assess those changes, they will also look at the implications for the financial system?

Yes, they absolutely have to do that because they can't look in isolation at one or the other. The interoperability of all of this is key to its success.

We are just at the time, but there is one final question, if I may, because I think it's quite important. We do know that the implementation costs for the WRA have been weighted in a different way than was originally supposed, but it's also clear from the reports that we've received that the operation costs now will be substantially higher as well. I just wondered if you could give us a flavour of why you think that's happened. Obviously, this is something that we can test at the next session as well, but is that something that's due to insufficient evidence at the start in estimating that or is that something that's flown out of just experience and the way that the programme has developed?

I think that in essence what's happened here is that the original estimates for building a thing called 'the WRA' were made at the outset. The thing that is being created now looks rather different to that original thing and, accordingly, the estimations for costs for that WRA have evolved. It's not, I think, that the original estimates were inaccurate; it's that what is being built is different: it is bigger, it is more complicated and it has greater functionality. That is inevitably more expensive. I think it's more that, as a reason for the changing costs, rather than necessarily inaccuracies in the original estimates, although those were fairly broad-brush estimates, obviously, at the start.

Okay. If we've got any further questions, we can write to you obviously in that regard. So, thank you very much.

Can I move to the land transaction tax? I'm sure you would have preferred the Chancellor not to have changed stamp duty land tax in the autumn budget, but he has done, and he's made changes. How will this affect the Welsh Government's tax forecasts for next year?

Well, Chair, I well recall the debates we had here in this committee when we were establishing the rules for setting rates and bands. I think the system that the committee agreed has stood up to the test of actual experience, because the possibility that the Chancellor would change rates and bands halfway through our own budget process was in our minds when that final system was agreed. So, yes, our lives would have been more straightforward had that not happened, but I think the system we agreed has been able to cope with it and we've avoided the situation of having set rates and bands that we'd have to unset and reset, because although we published our intentions, we're not going to have brought forward the regulations that will set them in Wales until the end of this month. 

What the Chancellor did, as Members here will know, was introduce a new relief in LTT for first-time buyers, and set the threshold for that at £300,000. That did bring a consequential sum of money to Wales, and I think my job, in a way, was to think about whether we wanted to use that sum of money for these purposes, and if we did, how best to deploy it. You'll know that what we will propose to the National Assembly in the regulations we'll bring forward at the end of January is that we should raise the starting threshold for paying land transaction tax to £180,000 for all purchasers in Wales. I think this shows the merit of the system and of devolving this particular tax. I understand that when the Chancellor is setting his arrangements, he has to try and find a figure that will work for Rochdale and will work for Rochester, and the property market in the south-east of England will be very different to the property market in the north of England. Now, what we've been able to do is to, I believe, find a use for the money that reflects the needs of the housing market here in Wales, and on £180,000, 80 per cent of first-time buyers will pay no tax at all. To get 80 per cent of first-time buyers in the same position in England, you have to set the threshold at £300,000. And we've been able to use the consequential money that we got to help not just first-time buyers—there would have been around 4,000 or so of those—but to help 24,000 people who are purchasing property in that relatively modest end of the market. I think there are good equity reasons why we've decided to use the money in the way that we have. 

I was going to say that of course it is a great advantage that you're not putting the rates on the face of the Bill in the first place—

Yes—

—which some people thought was a good idea. I think perhaps they've probably now reconsidered. The other question I want to ask, really, is on learning from Scotland's experience. There was a shortfall of £11 million after the land and buildings transaction tax was introduced compared to the year before, despite a higher number of transactions; this was due to forestalling. Are you concerned about that? And, really, the question that I think is on everyone's lips is: will we get the consequential for it if we are affected?

Well, Chair, I am concerned about forestalling, of course. I think forestalling is inevitable when you move from one system to another, almost regardless of whether the figures are any different, because the system will be different. So, there will be solicitors and conveyancers in Wales who are very familiar with the way the current system operates, and some of them are going to think to themselves in March, 'Well, if I bring those transactions forward into March, I'll be dealing with them according to the system with which I'm very familiar. If I leave them into April, there will be a new system that I've got to be familiar with.' And even if the new system is very straightforward and very recognisable, it would still be new. So, I think there will be some behavioural effects inevitably, and there will be a very small number, I believe—slightly different to the position in Scotland—of transaction that will be brought forward into March in order to avoid a higher rate of tax that would otherwise be paid in April.

Now, that is a one-off windfall to the Treasury, and you'll have seen that the Office for Budget Responsibility forecasts that at around £6 million. And I think that Mr Chote said to you, 'Six million—that's small change'. Well, I suppose when you're the OBR and you're looking at many, many, many billions of pounds, £6 million does look like a small amount, but it's not a small amount for us in Wales. So, I have had a discussion with the Chief Secretary to the Treasury. It was followed up by discussions with officials that I know Andrew was involved in yesterday, in which—. My clear position to the Chief Secretary to the Treasury is that the fiscal framework allows for behavioural effects, second-order effects, to be taken into account. It doesn't guarantee what the outcome is, but it does say that they do have to be considered and that we expect the forestalling effect to be considered within the rules of a fiscal framework, and then, hopefully, we will get the money back. Scotland did get some of the money back—not all—but Scotland did get some of the money back that they lost through forestalling, and my position is that the same outcome should be secured for Wales.

I don't think anybody in this room would disagree with you on that. If 'no detriment' means anything, then we should get that money.

Can I just finally move on to readiness for the transition to devolved taxes? Are you confident that we are ready for the transition, that, on day 1, the new taxes will be able to come in efficiently, effectively and without any problems?

Well, this is a major undertaking, Chair, and there are some risks involved in it, undoubtedly. There's a lot of work still to be done in the weeks between now and 1 April. But we passed a significant landmark, or milestone anyway, yesterday, and maybe Andrew will tell you a bit more about it, because he was at the meeting. So, part of the meeting that took place between our senior officials and senior officials at the Treasury yesterday was to carry out what we had already agreed, which would be a joint readiness assessment between the Governments to see whether that assessment confirmed that things were in a sufficient state of readiness for the Treasury to take the actions it needs to take to switch off stamp duty land tax and go into a landfill disposal tax as it currently is, so that we can take them on on 1 April. So, they have to take action to do that. There has to be a Treasury Order made and laid, hopefully now in February, to disapply UK taxes in Wales from April. There were 24 different tests that that joint committee had set out that would need to be satisfied in order to advise Ministers that the system was in a sufficient state of readiness to do that, and the good news is that after, I think, a pretty thorough exploration of it all yesterday, those senior officials have concluded that things are sufficiently in place, and sufficiently securely in place, that they will recommend to me and to the Chief Secretary to the Treasury that we can proceed to turn those taxes off. I think that's a pretty good independent sign that things are sufficiently on track to take that next big step on the journey.

So, I don't want to sound as if I think everything is easy and straightforward, because it's not. There are complex matters here, and still a lot of detailed work to be done, but I think we can take some confidence from yesterday's outcome.

We might want to explore some of those more complex matters now, so we'll move on with Nick Ramsay.

Diolch. Morning, Cabinet Secretary. In terms of the set-up of the WRA, how it's progressing and the costs involved, the Wales Audit Office, which we've just taken evidence from, stated that there's a very high risk that the digital tax collection and management system may not be delivered in time for when the new taxes are switched on. What's the WRA doing to mitigate this risk, and what contingency plans do you have in place?

Thank you, Nick. Claire, who leads in the WRA on lots of these things, can give you much more hands-on detail than I can. But obviously we keep a very close watch on all of these things. I meet monthly with Claire and with the chief executive of the WRA to be assured about the way in which things are on track and the digital side of things is a major part of the way in which we want the WRA to be able to function in future. It's one of the big advantages of having established the WRA and given it the responsibilities that we have; rather than relying on HMRC, we've been able to design the system for the digital age in a way that HMRC, with all its legacy systems, doesn't have the same blank sheet to work from.

So, we have a WRA digital tax system supplier. It's a firm called Kainos, and they have done a great deal of work over the last two months, and the WRA has a digital tax system ready for registration and is in the final stages of developing its system for filing a tax return. They're on target to deliver a completed digital tax system by the and of February, and testing of the bits of the system that are completed has already started with landfill site operators, for example. In many ways, registration is the key thing to get right in landfill disposals tax. If you've got digital registration right, then many of the other things will follow very smoothly from that. On land transaction tax, you have a need for registration, but we have a very clear ambition for having a full digital tax system for filing a tax return in LTT. That is near the end of its development; there are plans in place to test it with solicitors, agents and so on at the end of January.

So, we remain—our ambition is, and we think we've got good reasons for some confidence, that we will have all of that in place by the end of March and, therefore, ready to go live when the WRA comes live. However, we recognise—

To put it another way, do you think the Wales Audit Office, then, are being overcautious in expressing their concerns about the possibility of this not happening, or do you think they're basing their assertions on the way the system was before, maybe, you stepped in and you tried to address the issue?

I don't want to be critical at all of the WAO's report. I think their reports have been very helpful. The one they did previously—. They're right to say that this is a high-risk part of the business, because getting the WRA to where we want it to be—

There's no wriggle room, is there?

No, and in the last few weeks—we are into weeks now, aren't we, leading into 1 April—they are right to say that there isn't a great deal of time to make things good if things don't work out the way that we hope they will do. So, I'm not quarrelling with their assessment. This is a risky end of things. I think if you look back a year to their previous report, they signalled other things that were a risk but where we've been able to demonstrate that those things have been successfully completed, and I think we will be able to demonstrate that we can successfully complete this too.

But the second part of their report and the second part of Nick Ramsay's original question was: what things are in place to be able to deal with an outcome in which things don't work out in that way? So, the WRA, working with HMRC, have been putting into place a series of fallback measures so that if things don't work out in the way that we hope—so, for example, if the digital tax system were to fail on 1 April—then there is, ready and capable of being immediately deployed, an online form that taxpayers and agents could use to complete their tax return in LTT. That's not there because we think we will need it—certainly, we don't hope to need it—but it's there and it's ready. It could be deployed immediately if the high risk turns out to be more of a real risk than we anticipate.

We accept the high risk. IT in any organisation's implementation runs at risk, but this has been—I suppose 'interesting' is the word, because we've had to change scope halfway through because of the additional rate. So, that has meant different discussions with different people. But we've worked hard on that, and we are assured that when people want to put their tax returns online in April, they will be there. The risk really lies behind the back-office systems that the Welsh Government will have, because we've implemented a finance system in a very short period of time, and the integration between those two systems. So, the customer will see a form to complete, and they will complete it. It's the amount of work we need to do behind. We've taken the risk to try and integrate those systems, because the benefit of that is of huge benefit to the WRA going forward.

So, the taxpayer, very likely, in any outcome, won't notice a difference upfront, but could there then be subsequent delays in terms of the processing of that, or have I misunderstood?

No, because we've got a—not to bore you with a load of amount of detail, but behind everything is an enormous amount of processes, and so we've been working hard to get those processes absolutely right. And behind those processes, you have a number of contingency plans. The number of contingency plans I could bring in today is about this much of a wad thick, to make sure that we actually can do that on that day and subsequently, after that. We will know in February where those systems can actually talk to each other. We've done some initial findings and that's gone very positively, and we've just done an IT health check and that's gone positively. So, we are working hard and our supplier is working hard with us, because we both want to achieve that same outcome.

Great. Turning to the expertise—I think Neil Hamilton asked a similar question to the Wales Audit Office—on the importance of retaining specialist skills and tax expertise required by the WRA within Wales, how has this influenced the recruitment process?

Well, Chair, I think, a year ago the WRA were—and, again, we accepted that—saying that recruitment was one of the major challenges that the WRA would face over that period. And it was a fair warning, because we've never had to recruit people with these skills in Wales before. They are not always in huge supply and you might have thought that attracting people to come and work for a new organisation would have its challenges.

I think I've probably said to committee before that the scale of the WRA—. There are around 40 [correction: 70] or so posts involved in it and we've now got 32 people appointed. The whole of the executive team is in place. We've attracted around eight applicants for every post that's been advertised and have recruited from 14 different UK Government departments to attract those people. So, we've been able to attract people who're working in Wales already for the DVLA, for example, but we've also been able to attract those people who I think we may have talked about before; people who've made their careers outside Wales but have reached that point in their careers where they would like to be able to come back to Wales and settle back here. We've got a number of examples where we've been able to do that successfully. I think 13 of the 32 appointments are of people who are fluent in the Welsh language. So, another part of our staff recruitment ambition was that we would be able to have a fully bilingual service in the WRA and I think having over a third of the staff appointed who are equally fluent in Welsh and English is a good sign too.

So, there are posts that we're still recruiting to, and that effort will need to go on, but, given some of the anxieties we had in the beginning, I think we've probably done as well as or a bit better than we had expected 12 months ago.

Just to add to that, I think one area that we were particularly concerned about a year ago was getting people in who had practical experience of running tax systems, and not necessarily our ones—exactly the taxes we're introducing—but serious operational experience in another tax authority. And we've been really pleased that we've been able to attract a number of very, very able people with that kind of background into the WRA. They're valuable for various reasons. One is to make sure that we hit the ground running on day one, but also, crucially, to transfer that knowledge to another cadre of people—people, perhaps, who haven't got that experience at the moment but who need to gain and develop it over the next couple of years—the sort of future of the WRA, I suppose. So, we've made some really good progress with that, we think.

And, of course, it'll be about retention as well as recruitment, won't it, which is the flip side of the coin.

Chair, that's a very important point, I think, because an organisation with only 40 [correction: 70] or so people working in it, you're not going to see somebody joining it at the age of 21 and still be working their way up the hierarchy when they retire. So, to attract the sort of people we need to attract, what we have to be able to do is to demonstrate to those people that there are other opportunities where their skills will be valuable, and other opportunities for them to develop their careers here in south Wales. It was one of the factors that played its part in the location decision, which people here will remember. But the fact, for example, that we've got the Office for National Statistics on our doorstep in Newport as well, where lots of the same analytical skills will be important there, as here, then, those opportunities become really important in the retention issue, because people come here knowing that the WRA isn't a sort of cul-de-sac where, once they're in it, there'll be nowhere else that they can go. There are other opportunities in Wales where those people know that their skills will be needed and there will be opportunities for them as well. We think that's a strength in terms of being able to attract people, but also to retain them in the WRA and in south Wales.

Can I just ask you about that, if I may? We did have evidence previously from the audit office, however, that a lot of these are secondments rather than long-term appointments. Is that a concern for you in terms of what you've just described?

No, Chair, it's not, because it was a deliberate part of our strategy, really. It's a new organisation, so I think you can understand that someone who might be willing to come to work in it might also want the safety net of knowing that if it doesn't work out and it isn't what they were expecting, they haven't given up the job that they have come from. So, we do have a number of people who have come, for example, from HMRC and who are currently coming to us on a secondment basis.

I've always been very comfortable with the idea that people move around the public service on a secondment basis because I think we benefit and they benefit from that. If they end up going back to HMRC and we have other people who come to us on a secondment basis, I think that will be good in itself. But if, having come to the WRA, they decide that this is where they want to stay, then they'll be able to make that decision as well. So, I think, from the very beginning we anticipated and wanted to make a virtue of the fact that a certain number—and you've got to be careful that it's not too large a proportion—of the people we would recruit would come to the WRA on a secondment basis [correction: on a loan or a secondment basis]. I think we're comfortable with the proportion of people we've got on that basis so far.

Some of those numbers are also Welsh Government staff who are seconded into the WRA, and some of this is just about the initial stages of a new organisation. As the Minister said, it's much easier to attract a wide field if people have the ability to hedge their bets and go on secondment or loan rather than commit permanently to a new organisation where they don't really know much about how it's going to develop. So, to widen the field, you definitely want to be able to offer jobs on secondment and loan, but that doesn't mean those people will necessarily be going back to the previous place after two or three years.

So, most of them are going into the new body with the intention of staying if they like it. It's not simply a loan with the intention of going back to another job.

Different individuals are in different positions, of course, but, yes—the people I've spoken to anyway—a lot of them are very excited about a longer term opportunity rather than just a couple of years. A number of the appointments have been made on a permanent basis now as well, so you've got a mix in the organisation.

Perhaps I could also follow that up. It's good to hear about the high levels already in terms of bilingual skills, and diversity hopefully. I mean, this is going to be a new, modern organisation for Wales and I hope that is also reflected in terms of gender and disability particularly, and ethnicity.

Andrew will know better than me at a staffing level. Chair, you know that the chair of the WRA is a woman, whom we were very pleased to be able to appoint. There's a good gender balance in relation to the board of the WRA. We do struggle sometimes to recruit in people from minority ethnic backgrounds into some of these posts, but it will be our ambition, definitely, that the WRA is a properly diverse organisation that reflects the—

If you don't have anything now, could you share any firgures with the committee later?

Of course.

It would just be interesting to keep an eye that. If you  have, as I understand, quite a high proportion of secondments at the moment, then we'll need to monitor this as it's implemented, if you like, as it moves on. Okay. Back to Nick Ramsay, then.

Could you provide an update on discussions relating to data collection and sharing arrangements at HMRC and Revenue Scotland?

Thank you. Well, all the legal gateways are now in place, and we've talked about this here before, and I've brought regulations to the floor of the Assembly to create the legal gateways that allow data to be shared between the three players there. And, actually, there are two other significant players where data sharing is important as well: in the VOA and also the Land Registry, where we need to make sure that information can flow properly between all of those who have a part to play in land transaction tax in particular.

HMRC will have put the necessary legislation in place to share data with the WRA from 1 April. We are developing memorandums of understanding and information-sharing agreements with those other bodies. We'll have a near-final draft of the MOU with HMRC in place, and we hope to be able to share that formally with HMRC by the end of this month.

So, these are absolutely important building blocks in being able to make sure that information flows through the new system in the right way, and I think that the necessary building blocks are either there already or will be there by the time we get to 1 April.

That was one of the main focuses of the joint readiness assessment that we completed yesterday. To a certain extent, HMRC are not so bothered about us being able to collect taxes in Wales; they want to make sure that they're getting the data and that we're not collecting tax that should be due to them. And a lot of that is about being able to share data between the organisations. So, a good number of the criteria we ticked through yesterday were about the ability to share data between the organisations, and we're well on track with all of that.

We have regular discussions. Even though we're working on the information agreement with Revenue Scotland, we have regular monthly discussions with them around—you know, they're actually operating now; what does that actually mean? And actually they're learning stuff from us at the moment as well because they're procuring their contract for their IT. So, there's that knowledge sharing already progressing.

Good. Can you elaborate on the extra costs for the collection and management of higher rates for additional residential property transactions?

Thank you, Chair. So, this is the single biggest change in the scope of responsibilities to be discharged by the Welsh Revenue Authority between the costs that were known or estimated at the outset and the costs that we now know the WRA will need to incur. So, there are, we anticipate, 7,500 transactions in a 12-month period where the additional residential property rate will apply. Net, we expect that to bring in about £65 million—sorry, gross, we expect £65 million to come in through the tax, but one of the complexities of the tax and one of the reasons it costs money to administer is that you hand back quite a lot of this tax to people too. So, we expect about £15 million of the £65 million to be repaid to people as they dispose of one of the properties that they have within the three-year period that they have to be able to do so.

So, it's a relatively administratively intensive tax to get right, but, nevertheless, it is bringing in quite a lot more than was expected when it was first introduced. And that's why we decided—you remember the debates in front of the committee—to retain that additional rate in Wales, because otherwise we'd be £50 million worse off as far as public services are concerned. But, in order to collect the £50 million, we have to spend about £2 million—just under £2 million, actually.

So, before this, it was the transaction tax. Actually, you didn't really care who the buyer was; you just wanted to know about the transaction. Because we have to refund, we need to know who we're refunding to, and make sure we're refunding to the correct person. Also as well, for compliance purposes, we need to understand who is doing this amount of transactions and if one particular person is doing an enormous amount of these transactions. So, that's where the complexity with the IT system comes in then, and we're working hard on that.

Okay, thanks.

David Rees, please.

Diolch, Cadeirydd. You've already mentioned quite a bit this morning that you seem to have now quite a reasonable working relationship with HMRC. Well, in relation to the Welsh rates of income tax, which will be next year, Cabinet Secretary, you've set five priorities, four of which I can see actually are involving your relationship with HMRC. So, what progress is being made with regard to the Welsh rates of income tax and the discussions with HMRC?

Well, Chair, this is a very important part of our discussions with HMRC. We've got lots of discussions with them, as David Rees has said, about the taxes that are already devolved to Wales, but we have Welsh rates of income tax coming our way before very long and a successful implementation of all of that really will depend upon HMRC being properly prepared for that. So, we continue to have detailed discussions with them. I believe that Jim Harra, who is the most senior person with responsibility for these things in HMRC, is to come in front of this committee later this month. So, I hope you'll be asking him how—

We're having HMRC in, yes.

—well prepared they are. I set out the priorities of the Welsh Government as far as Welsh rates of income tax are concerned, and I included that senior officers of HMRC should make themselves available to this committee. So, I am pleased to see that happening, and I wrote to the Chief Secretary to the Treasury to set out our requirements. We are working very closely with HMRC. We do rely on them to discharge their part of this bargain and we will continue to have those discussions with them so that they can provide us, and you, with the assurances we need that they are putting the necessary resources into this, that they're prioritising it in the way that they need to, and that they can leave us all assured that they will be delivering on the things that they have to deliver in order that Welsh rates of income tax can be implemented on time. I agreed, as part of the fiscal framework with the Chief Secretary to the Treasury, to introduce Welsh rates of income tax at the earliest point in the timetable that they could be introduced, and the UK Government was keen that we should do that. For that to be practically successful, HMRC has to deliver its side of the bargain.

Okay. Obviously I appreciate that HMRC is going to have a lot of responsibilities and they've already given you an estimated cost for their work, between £5 million and £10 million. Have you had a chance to assess those estimated costs? And do you think they're fair, particularly in relation to the fact that the initial costs for Scotland were double that, effectively?

Well, I think we have to take a properly sceptical attitude towards these costs and I'm additionally sceptical because of what has happened so recently in relation to the costs that HMRC are seeking to extract from us in relation to land transaction tax where the bill they are presenting to us went up very significantly in comparison with the original estimates that they provided to us. So, we are having to pay considerably more to HMRC for them to put themselves in a position where our land transaction tax can be successfully implemented, and I will certainly be asking Mr Harra questions about how that came about when I meet him, I think, on the same day that you will be meeting him. The fact that they got that so badly wrong inevitably casts some scepticism into our minds in relation to the figures that they've provided for Welsh rates of income tax.

Because that was just about turning off a tax. That's just the cost of turning off a tax.

Yes. I mean, most of the costs are to do with them getting access to our data and ensuring that that data flows through HMRC systems in the way that it needs to. So, you kind of think, 'all you're doing is disapplying it in Wales, that can't cost very much', but it's all the continuing interest that HMRC have in transactions in Wales that is really where the costs arise.

It sounds a bit gilded to me, that.

Well, I will certainly be asking Mr Harra how this came about, and whether the fact that they got their sums so wrong in the beginning—that the answer to that is that the Welsh Government should pay the costs of their mistake. But there we are. That's a discussion we will have. 

We may also have that discussion.

Much bigger sums of money—. Sorry, David. 

It's a bit of a concern, in one sense, because both with land transaction tax and with the income tax, Scotland has gone first, and you would think they would be learning from examples as a consequence of that, and therefore their estimates would be closer.

Well, I think in LTT, what they will say to you is that HMRC had moved from one system to a completely different system between the time that they dealt with turning off the Scottish equivalent and now dealing with us. And it's because there is a completely different system that they are dealing with that their costs turned out to be higher than anticipated. However, on the Welsh rates of income tax, exactly the point that David Rees has made ought to be true. And the estimates that we have been offered from HMRC, of between £5 million and £10 million, are much more encouraging to us, and they are where they are, lower than the costs in Scotland, we are told by HMRC, because they have learnt from what happened in Scotland and we are able to get the benefit of that experience by going second. So, the costs to us will be lower than they were in Scotland, because they put things right and learnt things when doing it with Scotland, and we're going to get the benefit of that. Chair, I suppose it would be fair for me to say that the prize of getting much lower costs on Welsh rates of income tax would significantly outweigh the costs we are incurring in LTT. In other words, if you wanted one of these to go your way and one of them not to, it's much better that we are getting the benefit in Welsh rates of income tax, where it will be many several millions more than the extra money we're having to pay on LTT. So, our job is to hold them to these estimates. I certainly don't expect the same problem to arise a second time.

Does that mean it also looks like the same problem with—? Because they actually seem to have lost 400,000-odd Scottish taxpayers as well.

Well, that is a way in which we are working closely with them, to make sure that the difficulties that they experienced in identifying individuals who were to be held responsible for Scottish rates of income tax—. As you know, the legislation sets out who will be liable for Welsh rates of income tax, and I think we have to have some optimism that the glitches that they're experiencing correctly identifying individuals in the Scottish context—that they will have learnt from that. They tell us they have learnt from it—we have a representative on the project board with them, preparing for exactly these sorts of issues—and that we won't expect the same problems to reoccur in Wales.

But as Mike Hedges pointed out in an earlier session, the cross-border flow between Wales and England is far greater, percentage-wise, than the cross-border flow between Scotland and England. And therefore, that does give us some concerns that they may not be ready for that, and therefore the costs could be higher as a consequence.

All I can say, Chair, is that we share the same anxieties, and we're right to. They are the things that we should be concerned about, and we will work away with HMRC on those things. But these are things for which they are responsible, and that's why them coming here and being accountable to the Welsh Government as well, and to the UK Government, given what I said to you about how keen they were that we did this at the earliest opportunity—that's why they too have an interest in making sure that it doesn't go wrong because HMRC doesn't prioritise it or invest in it or give it the significance that we know that it has for us.

Mike Hedges, do you want to come in, just on this?

Quoting other people, if Steffan Lewis was here now, he would say that this is not abnormal in Europe and that perhaps HMRC could be learning from what happens in other parts of Europe where people work in one area but their tax has to be paid in a different one.  

I think we talked here in the committee about the city in Holland where property transactions are the responsibility of the Belgian authorities, for example. So, yes, in a European context this would not be unusual at all. 

We'll move on with Neil Hamilton.

I'd like to ask a question about the potential new devolved taxes. Some of these taxes' success will be measured by how little they raise, I suppose, like the disposable plastics tax on the one hand, and others like—just to take one at random—the tourist tax will probably be the opposite, by how much they raise. What work have you done, or what thinking have you given, so far, to the criteria for selection from the list that's already been published about how suitable these taxes might be and, on the other hand, the potential impact upon the people of Wales? What's going to be uppermost in your mind here when you form a view as to (a) the short list and (b) the final choice? 

Well, Chair, first of all, just to agree with the point that Mr Hamilton made at the beginning, and it is a really important point to try and get into the debate, that taxes have more than one potential purpose. Some of them are primarily about raising revenue, some of them are about changing behaviour, and not all taxes have the identical underlying motivation behind them. So, I think that's an important part in the debate that people don't always appreciate at the outset. 

I think it's important just to say again that what we are doing in the very first potential tax that we will select is to test the new machinery. So, that's also in my mind as to whether this is the sort of tax that is suitable for that machinery-testing purpose. So, that's one of the criteria that I will be thinking about against the four different taxes that were on the short list. I'll then be thinking about the criteria that we set out in our tax policy document, the tests that we set out for what taxes should be in Wales, and I have to pay some attention to the criteria set out in the Wales Bill itself, of 2014, because the command paper there set out a series of tests that would need to be applied to any new proposal.

So, that's what I will be doing, and have been doing. I've met all my Cabinet colleagues who have an interest in this already. We are moving into the final few weeks of the work that we are doing to assess the different possibilities on the short list. My ambition is to be able to come to the Assembly before half term and to say, 'This is the one that I've now decided on, for now. That doesn't mean to say that other things on the list are just forgotten, but we've got to test the system for the first time. This is one of the four that I think will be most suitable for doing that.'

We've had a series of discussions with the UK Government about the part that they will then play. We talked about this on the floor of the Assembly yesterday. I'm not sure whether we will get to a completely final agreement with the UK Government in advance of testing the machinery for the first time, but we will be pretty close, I hope, to an agreed protocol with the UK Government that will set out how they will respond. I will send the tax up the line, the first action then lies with them, and I hope that we will have an agreed document with them that will set out how they will then go about responding. So, a lot of that work is also going on over the next few weeks to try and get that in place by the time I identify the one of the four that we will use in the first instance. 

If I may just ask, specifically on that protocol, which is also what I was asking yesterday on the floor: would that include some expectation of a reasonable timetable? Because one of the difficulties of the LCO system, of course, which this is worryingly similar to, is that things got lost in Whitehall for many months. 

Well, Chair, part of my ambition in negotiating the protocol is to avoid some of what we would now see as the downsides of the LCO process. I'm not certain that we will be able to agree a timetable for this first one, because everybody will be in the position of not knowing quite how long any part of the process will take. I'm afraid we are also getting a bit of a feeling that we are going to be meeting the Brexit legislative load here too, that we will be asking the Treasury to do this new piece of work at the same time as they are having to deal with a very considerable degree of legislative activity that flows from the Brexit decision. That might have an impact on how quickly the Treasury will be able to take forward the work that they will need to do on this proposal. But we will try and get a sense—. But it shouldn't take more than 12 months for them to complete everything that we would expect them to complete. After we've done it the first time and we've all learned from that, I would hope that, after that, the process would be considerably shorter.

Sorry, go on. Thank you.

Okay, fine. Good, well, thank you very much for that. Time's marching on, so perhaps I better not explore that point any further. But I'd just like to ask, now: what consideration has been given to developing a legislative budget process? As you know, the Wales Act enables Wales to develop and introduce its own financial framework legislation. We've talked about this before, of course, and I wondered if you could give us an update on the thinking there.

Sure. I don't think, Chair, that I'm very much further forward than the last time we discussed it. I think we've agreed, between the Government and the committee, that, the process we used in the current round, we should replicate it next year, and that, in the current state of devolved taxes, it more or less worked in the way that we hoped that it would do. So, we know what we're going to do next year. This committee has committed itself to doing a piece of work, in holding an inquiry into legislative budget processes. I'm looking forward to following that inquiry to see where it takes you, and then, hopefully, to continue to do things in the way that we've done up until now, which is to work co-operatively, and to see if your inquiry leads you to a conclusion that a legislative budget process is proportionate to the level of responsibilities that we have and that it would provide advantages over and above the system that we currently have. Then, I will be very open to discussing that with you. 

Diolch yn fawr.

We'll certainly take up those offers, I think. Can we conclude with Jane Hutt? Thank you.

A couple of questions about forecasting and your work with the OBR. When Robert Chote came to speak and gave evidence in December—well, you would have seen what he said. He actually did—. The issue about timing, he said, was a challenge for them, not just in terms of our draft budget, the November budget. He also talked about their other responsibilities as well, in terms of the national accounts of the UK Government, et cetera. So, could you just say something about how you're working with them in terms of those comprehensive forecasts that you need, but also this continuing question about Wales-specific data as well?

Well, Chair, I heard what Mr Chote said to you, and I met him later in the day and had a further conversation with him about it all. The good news is that the Chief Secretary to the Treasury I think has been helpful in this regard and has made it clear to her officials that she wants us to have a proper conversation with the OBR about what it could do if we were to ask it to undertake this role for us, and what it would be able to do, by when, in the annual timetable for these things. That, I think, is what Robert Chote was most concerned about with you and with me—not that they couldn't do the work, not that they didn't want to do the work, but that the work they might want to do for us, we might be asking them to do it at the very point in their cycle where they are most busy in discharging responsibilities they already have. Now, I don't believe at the moment that those difficulties are insurmountable, and a detailed discussion is going on with the OBR about that. But if, at the end of it, there was on offer we had to weigh up then I'm certainly happy to share that with the committee, because you will be as interested in that work as I will be.

If we were to conclude that what the OBR is able to do in that practical sense doesn't match what we need, then we've always had other possibilities for this. The good news here is that the Bangor work that we had this year has turned out to be, I think, of good calibre and the Treasury are less anxious about us continuing to use that route for the time being while we set up something else. So, I hope the OBR will still be the route we can go down; I think there are significant advantages in doing it. If the practicalities defeat us, because they just can't do what we need them to do, then we will go to some of the other possibilities we've talked about in front of the committee.

There will be, next year, a lot more Welsh-specific information in the tax view [correction: a lot more Welsh-specific information] because the WRA will be producing Welsh-specific information that we've never had before. In landfill disposal tax, we rehearsed this here before. The way the OBR forecasts moved around was because they've never had or needed Welsh-specific information before. The WRA will be a very important source of that information. If we're talking about Welsh-specific data in relation to more macro-economic analysis then what I'd ask Mr Chote, and I think he would think about it, is: is there data that, if we were to make what could be quite a considerable investment in obtaining that data—did he believe that it would be likely to make a material difference to the sort of forecasts that they could make? Now, if the answer to that is 'yes', that, if we make that investment and produce that data then the forecasts would be better, more reliable, would be different for Wales, then there will be a case for making that investment. If the answer is, 'You could produce all that data, but actually it might make very little difference to your macro-economic forecasts because they are dominated by not just what happens in the Welsh economy but what happens in the wider UK economy', and we could produce all that data and it would make very little difference, well, I'd be much less attracted to finding money to produce data that didn't lead us very far and I think the jury is still out, really, in that debate.

Okay. Fine. I think—. Just a final point: we've talked about block grant adjustment, but of course it is OBR who will be calculating in terms of their fiscal and economic outlook from November in terms of block grant adjustment. We've heard already about how we should be hearing by the end of the year, but it would be helpful if you've got any predictions about the impacts of this in terms of the outcome.

Briefly, Chair, I think there are three aspects that are worth reporting this morning. So, over the three-year period 2018-19 to 2020-21, we expect the impact of the fiscal framework, on what we know so far, that that will bring about £70 million to Wales that otherwise would not have come to Wales. So, we can see that 105 per cent multiplier beginning to make a difference there. The way we have chosen to use our tax responsibilities I think adds another £30 million into that over a three-year period. So, there's £100 million now available in revenue for Welsh public services as a result of us taking on these responsibilities that otherwise wouldn't have been there, and, in tough times, £100 million—I know it's over three years, but it all counts and it all makes a real difference. And, because we are taking on fiscal responsibilities, we have new borrowing powers as well. As from 2019-20, we will be able to borrow £150 million a year on the capital side of the budget as well. So, there, I think, you begin to see the fact that there is a genuine impact, and a genuinely beneficial impact, on both the revenue and the capital sides of our budgets as a result of the new responsibilities we are taking on and the way we are choosing to exercise them.

Thank you.

Just on that, to note, as we close this session, that we did of course invite Alun Cairns to give evidence on precisely this aspect of the block grant adjustment. He has, unfortunately, declined the committee’s invitation once again. We’ll discuss that in a second as a committee, but just so that that’s on the record as part of this discussion.

Disappointing, I would say.

Disappointing, yes.

Very disappointing, because there is a good news story as well in terms of the impact of fiscal devolution.

Fiscal devolution seems to be on track, it seems to be timetabled, the risks are acknowledged but are being taken care of, and there's a lot of substantial work being done to make this a success, by Westminster as well as by the Welsh Government and independent bodies such as the OBR. So, to me, it is disappointing that the Secretary of State does not want to discuss this in public session—the Secretary of State for Wales.

He is visiting Wales today.

Well, it's quite clear that it's not diary commitments that are preventing his appearance before this committee; that is a deceit.

He is actually in Wales today, Chair.

He's in Wales today, is he?

He's actually in my constituency this morning.

Well, thank you for that local information. We would have been delighted to have even visited your constituency to talk to him, if necessary, David. We'll note that for now, but—

Could we make an appointment to visit his constituency surgery, possibly?

Well, we could possibly do that; I think that's—. We'd prefer as a committee that he came to us, I think. Obviously, when we report to the Assembly on these matters, we'll note our disappointment in that regard, if that’s acceptable. In which case—.

So, we'll break until 11:30 now and we'll be in private session after that. Diolch yn fawr iawn.