National Assembly for Wales

Diolch, Chair. Good morning, committee. Thank you for that question. I would start by recognising that I think that the transition has been smooth. The long period of work that went up to it really was to ensure that the transition was smooth, orderly and eventually successful, and I think that that has been achieved, not least because of the work that was done with a wide range of stakeholders to ensure that the way in which we approached this piece of work was very much inclusive. I think that that has borne fruit and certainly would be the way in which we would intend to go forward in other areas of developing our approach to devolved taxes and setting those particular rates. I think it's also been successful in terms of demonstrating that we can take a specifically Welsh approach to this. So, in terms of the stamp duty land tax, for example, our approach in Wales has meant that a larger number of individuals and businesses have been taken out of the scope of the tax, and that's because we have set different parameters to those that existed previously. Similarly, on LDT, although we've maintained the same rates at the standard level and the reduced level as England to prevent any waste tourism at this point, and we've committed to doing that for the first two years, we have set that higher rate for the unauthorised waste disposals, which I think has been very effective in terms of leading the way, because we are the first nation to do that. So, I know other nations now are intending to follow our step. So, I think that's been very successful as well. We've had good feedback from stakeholders, who I've mentioned previously, and also a really effective tax advisory group as well, which I look forward to engaging with in due course.

Okay. Thank you for that. So, in terms of taking some learning or lessons from your experiences, what lessons do you think you can take from the already devolved taxes that you've learnt that you can maybe apply in future to others?

I think I'll ask officials to reflect on what we can have learned from the steps that we've taken thus far—clearly, they would have had more effective engagement in that up to this point—but I would also say as well and recognise that we have developed very good working relationships at official level with the UK Government. I think it's only fair to recognise that that work has also given dividends.

Yes. Okay, thank you. So, I suppose one thing I would certainly say was an important lesson in the process was that it took quite a long time to do all of the work and we were very glad that we started in—you know, 2013 to 2018 seems like a long time. Obviously, it is a long time in lots of ways, but it wasn't more time than we needed. So, getting the work started in 2013 and having that kind of continuous effort rather than trying to cram it into a much shorter period of time I think was really important. I suppose you don't always have the luxury of having the right amount of time to do bits of work in Government, and I think that made quite a big difference to how we were able to do things because there were three pieces of legislation, creating a new organisation, so a lot of work to do.

I echo a point that the Minister just made, that the relationships that we established through that time were really important with other bits of Government in the UK, Scotland in particular—we learnt a lot from following the experience of Scotland, and when I say 'following', I don't mean copying, I mean going after them; that helped a lot, having them having gone down a similar path a little bit ahead of us—but also, crucially, stakeholders in Wales, tax experts, all kinds of people who have an interest in tax. We didn't have all the expertise in the Welsh Government at the outset that we needed to do the job properly, but we were able to draw on others' expertise and experts in the field. So, those were really important lessons, I think.

Okay. Thank you for that. Now, the Welsh Government published its tax policy work plan for 2017 to prepare for devolved taxes in 2018-19. In your view, how successful were you in delivering the plan?

Thank you for that. Our annual tax plans have been really important in terms of setting the direction for the work that we need to do. So, 2017 was very much part of laying the groundwork for where we are now. Particular areas of interest in that would be the commitment by October 2017 to announce the rates and bands for land transaction tax, including the higher rate surcharge, and the rates for landfill disposals tax. Clearly, that was done, but also, during the 2017 period, there was important work done on consulting on the powers for the Welsh Revenue Authority as part of their enforcement suite of powers, so that was also an important piece of work and they now, I suppose, are looking to explore how they would use those powers alongside Natural Resources Wales and others as well.

There was also a focus in 2017 on improving the evidence base. So, there was a commitment to develop analytical tools to support the development of income tax policy in Wales. Clearly, that was successful. So, 2017 was very much about laying the groundwork, and then, of course, we had our 2018 report published in February and the update, I suppose, provided to committee and others in October, and there were some really significant developments in the 2018 report, not least the agreement with the Office for Budget Responsibility on devolved tax forecasting. The progression, then, of the Welsh rates of income tax—so, again, taking that long view through the various tax programmes—and the development of the new taxes agenda—so, the vacant land tax and social care levy, and we had a debate on that in the Chamber yesterday. Again, we will be reviewing the committee's work on that in the Assembly today, and I look forward to responding to that debate.

And also, in the 2018 plan, we wanted to do some work to articulate and develop our thinking on the long-term strategic challenges around the Welsh tax base—so, looking at population trends into the future that we might expect, looking at levels of income and so on. They do, I think, present us with some challenges that we will be looking to meet in the future as well. So, I think those tax plans are very important in terms of setting out the direction of work. It would be my intention to publish the tax plan for this year early this year; we've done it in February previously, and I would be keen to meet that kind of time of the year with, again, the view to updating in October.

Okay. There was reference as well to reviewing small business rates relief and council tax, and I'm just wondering whether you could give us any update on progress on those particular areas as well, please.

I'll ask the team to perhaps advise on any pieces of work that have been done specifically reviewing those schemes, but, on the small business rate relief scheme, obviously it's providing over £110 million now of funding, supporting more than 70,000 ratepayers across Wales. Most pay—yes, more than half pay no rates at all, and that's fully funded by Welsh Government. So, I think, in that way, it has been very successful, but, in terms of any reviews that have been done into the impact of that, I'll ask officials for their input there.

And, on the review of council tax, that was, again, part of the 2018 tax policy work plan. So, I can confirm that we launched the awareness-raising campaign, because we know that lots of people who are eligible for support with their council tax still aren't taking that up. That's something I've asked officials to do some further work on in the early part of this year, because, obviously, next week in the Chamber, we'll be making announcements further.

We've announced our intention to remove the sanction of imprisonment for enforcing council tax debt from April 2019, and this is very much about increasing the fairness of council tax, because we don't believe that being in debt is a criminal offence, so we're taking action there, and also looking to local authorities to share with us good practice, because some local authorities don't take that kind of enforcement activity at all in terms of threatening people with imprisonment and yet they still have very good rates of tax collection. So, we'll be looking to those, really, to see what the best practice is that we can share.

And, again, we've published a consultation on exempting care leavers from paying council tax up to the age of 25. Again, that's part of our approach to creating a fairer system because we know that care leavers, particularly, are very vulnerable; it can take care leavers a longer time to settle in secure accommodation, just because they simply haven't had the support that others might have had. So, I think this is a really good thing to do. Alongside that, we've got a joint campaign with Money Saving Expert to improve the application of the severely mentally impaired discount—and I cannot stand the name of that discount, but, nonetheless, it's really important that those people who are able to receive that do. So, there are lots of pieces of work going on to ensure that we do have a fairer system.

But, in terms of the small business rates relief in particular, there was an intention to move it to a more permanent arrangement. So, is that work ongoing, or where are we at with that?

So, the permanent scheme is in place, introduced in April last year. And, of course, there was an announcement in December of an enhanced—well, retail relief scheme, really, because it applies to all retailers with a rateable value up to £50,000. So, that was announced in December, but that's not necessarily a permanent scheme at this point, given that we're not sure whether the consequential that we used to fund that will continue into the future.

So, that's still something that's—

Being kept under review.

Okay. And, in terms of council tax, are we looking at more fundamental reforms?

There's a lot of work still going on on that. We've set out a lot of things that have been done already, but, yes, there's still work about how we can continue to develop and enhance—

Do you have an intended time in terms of when you hope to come to some sort of conclusions?

I think to some extent it's an ongoing programme of work, but I think, in the workplan that we publish later this year, or in the early part of this year, that will set out a bit more of what comes next on that.

Okay. Thank you. Mike.

Welcome, Minister. Can I just say two very quick things? The first one is that reliefs are good but only mean that you've got less money for public services. That's got to be balanced. I'll again make the point, which I don't expect you to reply to, that we need to have more higher bands. The way it stops at the moment—whether you've got a £0.5 million house or a £5 million house, you pay exactly the same thing, and that's got to be wrong.

But you're not going to answer either of those, so can I come on to some general questions? Your report mentions you expect the block grant adjustment to take into account the impact of forestalling. Can you clarify how these discussions are progressing?

My understanding is that it will still take some months to understand the impact of forestalling on the stamp duty land tax and land transaction tax—to understand the impact that it's had and for us to get together those particular numbers. But, under the terms of the fiscal framework, we would expect our block grant adjustments to be reduced, because we know that Scotland was afforded that and we would expect the same kind of treatment.

You could expect, just thinking about it, with LTT, there would be some forestalling if people would try to get it done under the system they knew. But, land disposal tax—why would you expect that to have forestalling, because people tend to get rid of things as and when they've got them?

Yes, that's a useful clarification. We don't think there was any forestalling in relation to landfill disposals tax, but just in relation to land transaction tax where, as you say, there are various reasons why people might want to have brought their transactions forward to complete them under SDLT rather than under LTT.

The system you know rather than the system you don't, and you don't want to be one of the first ones testing it out. I can understand that.

Can you confirm when HMRC costs for switching off stamp duty land tax and landfill tax in Wales will be finalised? Have you actually got what the net cost will be now?

I'll answer that. So, I would say it's largely finalised, if that makes sense. I think the last estimate we gave the committee was between £1.75 and £2 million in total. The latest estimate that we've—well, the latest confirmed amount of cost is about £1.6 million so far, but the work is not entirely complete. You would expect it to have been fully completed by now, but there's still a little bit of work still to be done on completing the data feed to the Valuation Office Agency. There's a flow of data that goes from us—well, from the WRA—to HMRC to the VOA, and that's not been finalised yet. There's a workaround operating at the moment to make sure there's data flow, but the final solution, as it were, for that is still being worked on. That's going to cost another £100,000 or something like that, which is our cost. So, we anticipate around about £1.75 million as the final cost, but it's not finally final yet; it will be in the next few months, we think.

And when will you report that?

I would have thought as soon as the work is fully complete and the project closes, which, as I say, we were expecting to have been finished by now, but is not quite there. We would expect to report that to the committee so that you've got the full information.

Orally or by letter?

Whichever you would prefer, I suppose. Probably a letter.

A letter initially, at least, and then we'll come back to you if we need to know anything else. Thank you, Mike. Neil.

I'd just like to consider your experience in the last couple of years of transition to devolved taxes with the WRA. At the start of this process, the Government estimated the costs of setting up the WRA at between £4.9 million and £6.4 million over three years. That was subsequently reprofiled to £6.3 million over two years. The outturn was £6.246 million, which is a remarkable degree of precision, comparing the forecast with the outcome. Can you give us your own view of how the transition to devolved taxes has gone so far and is going now? 

Thank you very much for that question. In terms of developing my own view on it, it would be difficult, in a sense; I've only been in post a short time. So, in that sense, I would look to, for example, the Wales Audit Office report to guide in terms of what the progress has been like, and I think the Wales Audit Office report was very complimentary. I think that we are still in early days very much for the WRA, but, nonetheless, the transition does seem to have been very smooth. And again, I think that's partly about having those good working relationships with stakeholders, engaging widely, listening, and so forth.

I'm aware that the Wales Audit Office report does highlight some areas for further development as well—so, for example, the development of the WRA's digital capacity, focus on staff recruitment, because, again, it's a young organisation. We need to ensure that those staff that are recruited are developed and given the kind of support that they need to be retained in those jobs. I understand the team is a very impressive one, so we would want to support that team. Also, the development of joint governance arrangements needs to take place with the Welsh Government, and also the development of the three-year plan. So, I'll be having some discussions with the WRA on the three-year plan. I know you're taking evidence from the WRA today, so there'll be the chance to reflect on some of the operational challenges they might have seen from the inside at that point.

I know they're also developing performance measures reflecting the organisation's approach and their key objectives, and I look forward to having some further discussions on those later today, actually, after they've given evidence to the committee.    

You mentioned digital capacity. One of the interesting things that we've been able to trial with the WRA is that their systems are all entirely cloud-based, which is charting new territory, in a way, for the UK in this respect. Given the complexity and the volume of information that they have to acquire and use, when you're using new technology, that always comes with a potential risk. It doesn't seem so far to have developed into anything, but are you concerned in any way about that? 

No, it does seem to be working very well. I know it's actually award-winning at the moment in terms of the way that they've developed that. Obviously, it allows people to work remotely, which I think also is a good thing, and allows all the stakeholders that would need to access that to be able to do so in an easy and secure way. So, it seems to have gone very well. I think there are probably lessons for us across Government in terms of what a good system looks like, because we do have digital challenges that we have to address.  

Especially in the NHS. 

Yes, indeed. So, if there is a system that is working well and that people find good and user friendly to use, then certainly we should be looking at that.  

One of the interesting, I suppose, learnings from that is that it's probably easier to start off a new cloud-based service than it is to transfer an existing thing on to a fully cloud basis. I think one of the big challenges public bodies are grappling with now is how you transition from the server-based systems to cloud-based systems in a safe and secure way. And I think if you're going to start something new, it definitely makes sense to start it this way, based on our experience.    

Mike just wants to come in here very briefly. 

Surely a cloud-based system is really—all it is—an overseas server, isn't it? I mean, it hasn't actually been stored in mid air, on clouds. [Laughter.]

You're getting beyond my level of knowledge here, but I think, yes, basically you're hiring space on other people's— 

I think it's probably somewhere around Dublin, or it might be in west Wales. 

Or Newport. 

You might want to check this with Dyfed when he comes in subsequently, but I think the servers we use are UK based. 

I'm just talking in general about—

But they can be anywhere in the world, I think. But I think we were quite keen on ensuring that our data was being held in the UK. 

But it's actually held physically on a server, just not on a server within their centre. 

There might be some potential data protection issues if it's not held in the UK as well, I think. 

Basically, yes. 

As long as it's held in the European Union up until 29 March. 

There we are. That's a whole other debate, Mike. Sorry, Neil. 

Mike quite properly brings us back to earth on that. [Laughter.] But, moving on to the future role of the WRA, how do you envisage the WRA's future role in the analysis and reporting of Welsh-specific tax data? One of the problems that we've had in forecasting the effects of different tax rates, et cetera, in Wales is that we lack the basic ingredients on which we can make these forecasts in a credible way. So, the WRA, in addition to its obvious revenue-raising activities, has an important role to play in providing us with the raw material on the basis of which we can make sensible policy decisions for the future. 

So, I understand the WRA will be reporting, or is reporting, quarterly on LDT and monthly on LTT to give us that information, but then will also be providing annual reports, which do go into a greater level of detail. 

Yes. So, we've already found it extremely useful, obviously, to have Welsh-specific tax data being produced by the WRA on a regular basis, and that's already informing our forecasts for future years. I think, as the Minister says, it's still work in progress in terms of exactly what nature of data is most useful and what kind of additional analysis the WRA can do of the data to help us with the kind of jobs that we need to do in the Welsh Government to develop policy options and forecast, and also combining the data that the WRA have with other data to make useful analysis. So, that's going to carry on as a developing area of work for us. 

Thank you.

Thank you. 

Can I come in?

Yes, go on then. 

Just a quick point on that one. I therefore assume you're having regular discussions with the WRA about the data and how that can be used, and the type of data you would want to seek if you haven't already got it. 

Yes, absolutely. So, we have a data group between the Welsh Treasury and the WRA that meets, I think, monthly, and they talk about all this stuff. People who know what they're talking about on my side talk about this stuff with people in the WRA who are handling the data on a daily basis. 

Okay. Rhun now. 

What about work with the Office for Budget Responsibility to develop more Welsh-specific data, because we discussed with the chair of OBR the fact that there was a gap in Welsh-specific data. He wasn't particularly concerned about it. What are your thoughts?

Well, we're always looking for more useful data. I think a lot of the data that we need to do proper forecasting and analysis of the devolved taxes is already in existence. So, we have lots of data on, for example, waste arisings in Wales—municipal waste, commercial waste, how many landfill sites there are, what their plans are going forward. So, on the landfill side, I think we've got a lot of really good data.

On the LTT side, there's good data on the housing market in Wales, transaction volumes, trends, et cetera. There is very fine, granular detail on a local authority basis. So, a lot of the kind of very specific stuff is already there, and I think, probably, the gap that you're touching on is the kind of more macro-level stuff where there were lots of debates between the committee and the previous Minister on what type of data is possible and feasible and useful. And that discussion is carrying on. And, yes, I think you're right—OBR's view is that there's not necessarily any glaring absence that will prevent them doing the job we want them to do, but that's not to say we can't develop better, more useful data and analysis going forward.

Before we move away from the WRA, can I just ask what the Government's thinking is in terms of potential WRA roles in relation to other new taxes that might be in the pipeline? I'm just thinking of the vacant land tax, for example. Would you foresee the WRA actually delivering some of that?

I haven't had those discussions with them yet, but it's certainly something I could explore in our meeting today. 

And the WRA are very much part of the project work that we're doing on the vacant land tax. It's probably a bit early in the process to be making any kind of real proper decision about who might collect and manage those taxes, but, obviously, we've created a capability in the WRA that has potential uses for other taxes that come along, and maybe other functions too. So, yes.

Okay, thank you.

With vacant land tax, we've been very, very careful, as committee will know, to try and separate out the two points. So, the policy development is one side of it, but then making the case for the devolution of the powers is separate. So, when the UK Government considers the case for the devolution, then it's not considering it within the context of what policy the Welsh Government and the Assembly might decide on. It's a separate—.

Yes, sure. Okay, thank you. Nick.

Diolch, Chair. Congratulations on your appointment—your new position, Trefnydd, and Minister, Gweinidog. Can you provide us an update on how preparations for introducing Welsh rates of income tax are progressing, specifically with regard to governance arrangements?

Thank you for that. I did have the opportunity to meet the head of devolved income tax at the Her Majesty's Revenue and Customs earlier on this week, to have those discussions as to how progress is going, and particularly looking at some of those governance arrangements. I think it's worth reflecting as well that officials have had really excellent engagement with HMRC over the period of the development of this piece of work, and there have been some really important key delivery milestones that have been reached. Some of those relate to governance, some of them are more operational. The successful creation of the WRIT data set has been achieved, so that's required to identify those citizens who are resident in Wales, and who should be paying WRIT. And obviously now, those citizens will have a 'C' in front of their identification on the systems. Notification letters have gone out to around 2 million HMRC customers living in Wales, informing them of the crossover. And a lot of the technology and the development there that needs to be done has been done. There is still some work to go, I understand, on the technology side, but we are meeting the milestones that we had set out for that.

Officials continue to scrutinise the costs of the work, and HMRC does have that contact centre in place, which is providing information and advice in the Welsh language, and they've had some good take-up on that as well, which I think is really positive. The service level agreement will set out some of those governance arrangements, and that is being finalised at the moment; I would expect it to be complete very, very shortly. It probably will mirror a lot of what was in the Scottish service level agreement, because some of the issues are just so similar, but, nonetheless, there will be other parts of it, such as undertaking duties through the Welsh language and so on, which will be specific, obviously, to us.

Did you want to add something?

Yes, just to add: we've had a number of reviews undertaken of WRIT, and in particular the governance. We've had an internal HMRC review, then we've had one that's been done independently by Welsh Government and HMRC, and also the Wales Audit Office, and all three reviews have commented on the strong working relationship between HMRC and Welsh Government, and the strong governance arrangements that we've got in place.

It's probably worth mentioning also that the National Audit Office are in the middle of doing a review of HMRC's work on the Welsh rate of income tax, and I think that's going to be published later this month—Liz, is that right?

Yes.

I think that's what we're expecting. And so, obviously, the committee will have a strong interest in that. So, the WAO have said some things in their recent report about, I suppose, our end of the WRIT work, but, of course, most of the work is being done by HMRC, and that's part of the NAO's remit to look at that. So, that will be very interesting, to see what they come out with.

I'm pleased you mention possible technology issues from the HMRC end. I've received one of these letters within the last week, notifying me that I'm a Welsh taxpayer—clearly—and saying there'll be a 'C' on the tax code. But the letter is not addressed to me—it's addressed to someone that I have no idea who it is. So, I don't know whether you can take that back to HMRC. Because, obviously, I'm aware that there's a big change coming, so I think if I was just a normal member of the public, and I got a letter addressed to somebody that doesn't live at my address—

I hope that isn't suggesting that you don't pay tax, Nick. [Laughter.]

Well, I'm wondering now—I'd better go back through all my previous tax—. [Laughter.] I have had letters from HMRC in the past that are addressed to me, but, in terms of this transition, and no-one previous to me at the address was known by this name, I won't—

I think what that might be is that the majority of the letters went out to taxpayers in November, and what they've been doing recently is looking at people that may have more than one address, so it could be that something has gone slightly wrong there, but I can look into that for you.

The rest of us, I know, because we've asked this previously, have received our letters, and it was back in November, and they were properly addressed, so let's not—

If you share that information with the—

I should have brought it, really, shouldn't I? I didn't think this morning.

We'll certainly look into that and find out what happened.

I imagine there are going to be some teething problems, but as this is a process that's never happened in Wales before, there are clearly going to be a few months, not just at this end, but at the HMRC end as well, where they've got to keep on top of things.

And an ongoing one as well, because people move all the time, don't they?

Yes, that's a really key point. So, the first job is to create the database and there are always, inevitably, some errors and teething problems in getting that right at the start, but then, of course, you've got to maintain it as people change address and move into and out of Wales. Well over 100,000 move into and out of Wales every year. So, it's an ongoing job that HMRC need to do there.

When we met earlier this week, they did talk through the ways in which they are ensuring that they do have as up-to-date information as possible. One of those is looking at addresses held by other sources, including employers, child benefit and tax credit processes, the Department for Work and Pensions data, for example, and also cross-referencing data with third-party sources, such as the electoral register, credit reference agencies and so on to try and make sure that the data that they have is as good as it possibly can be.

Also, we intend to get involved in some of this work as well in terms of encouraging people to update their records when they move. Lots of people still think that HMRC is automatically informed when they move their house, but that's not the case. So, we're encouraging people to update, either through their personal tax account, telephoning HMRC, or completing the paper form. Also, when a customer informs HMRC of a change of address, the date on which they moved is mandatory. And that's important because HMRC then will use that data to look at the number of days that the individual lived in Wales, Scotland, or elsewhere in the UK, to calculate the correct amount of tax that that person should pay. So, there is lots of work going on, but I take the point about your specific case, which we will look into.

That's personal service, isn't it, from the Minister?

My last question you've answered actually—it was about the effectiveness of the bilingual service and that's all on track?

Yes, yes—very positive about that.

They're very committed to providing a Welsh language service and we've seen an increase in the number of people who have signed up for HMRC's Welsh language service on the back of receiving the letter, so that's positive.

Okay, thank you. Rhun.

Thank you. That original poll, as you say, was very much a baseline poll against which we can measure the impact that our communications and HMRC's communications are having in terms of raising awareness, both in terms of employees, but also employers as well. There'll be a further burst of activity in order to, again, raise people's awareness, and, of course, there's ongoing work through websites, social media and so on. But in December, we launched a doopoll survey to ask whether members of the public had received their letter. So, as well as raising awareness of that, also it gives us useful information in terms of whether people had received the letter and whether they'd recognised that they had had that information as well.

Evaluation of the comms campaign will be taking place throughout, but there will be a more formal evaluation of the effectiveness of that, using the Government Communication Service evaluation framework, and that will take place soon after April 2019.

The doopoll work has happened, you say—are there some results from that that are shareable?

We're still working on the analysis, but on the wider survey of awareness, we're going to do another Beaufort set of questions before April to try to see whether there's been—we'd hope to see some movement from 24% to a higher level in that period. 

Tell us why that's important to you, because you could argue that it doesn't matter—if somebody pays their tax. For most people it happens through PAYE and that's that. Why is it so important to you that the public know what's going on?

I think it's important to Welsh Government and the WRA in terms of developing that relationship with the taxpayer, because now we have a different relationship with taxpayers in Wales in terms of that 10p and the way in which Welsh Government potentially in future could change the rates of tax. So, I think it's important that people do have a better understanding of where their tax is going and what it might be doing in future. Clearly, we've got the existing situation where we're not diverging from the tax across the board, but nonetheless, developing that kind of information and relationship will be important in future. 

So, it's more about a good relationship between Government and the people rather than it being important that people understand how it works so that they can engage with it in a way that's more effective for HMRC or Government in terms of collecting the money. 

Yes, that's right. There's no kind of call to action here in relation to income tax. The engagement and work is, to a large extent, certainly the stuff that—HMRC's correspondence has, I suppose, a more specific purpose, to ensure they're trying to get their database as accurate as possible. But a lot of the way that we're doing it is a more general effort to help the people of Wales understand more about who they're paying their taxes to and what that money is being used for. 

So, to sum up, what I'm getting to: it doesn't worry you that perhaps a lack of understanding could affect in any way the tax take or—

No, it shouldn't do.

Okay. On the tax that we hope will be coming in, reflecting on what you said about following Scotland, one mistake that happened in Scotland was quite costly, of course, on understanding how much money would be brought in. What lessons have been learnt from that, and that over-forecasting that cost Scotland dearly, which you hope or you're confident will mean Wales doesn't end up in the same hole?

So, Andrew might give a bit more detail, but we did have a discussion about this with HMRC earlier this week as well. I think part of the issue they had in terms of identifying the correct taxpayers was to do with the systems that they used and the way in which addresses were logged onto those systems. We're taking a different approach in the way we do it. We're doing it by postcode, which will be inevitably more accurate. The fact that 2 million letters were sent out on the basis of that suggests that we have the right cohort of people. So, we wouldn't envisage the same problems that Scotland have experienced would be ones that we would experience, but Andrew or Liz might want to reflect further on that. 

It's certainly the error in the forecast—there are different things going on here, I suppose. Identifying the right number of taxpayers and getting the address database right is really important. But in terms of your understanding of how much revenue is raised from the tax at the outset, the error in those estimates in Scotland is certainly a cause for concern for us. I think I've said this already in committee, but, just to reiterate it: it won't affect how much money is available in year 1, because that's a transitional year and the actual amount that will be deducted from the block grant will be the actual amount of tax paid rather than the forecast. I know Scottish colleagues are very uncomfortable about the size of the gap between what they were expecting tax to raise and what it actually did in year 1, and the understanding of the reasons for that is still imperfect, shall we say. So, more work is needed there to get to the bottom of that. 

When will you be able to breathe easily? When will you know? Will you know halfway through the year? Will you have to wait until the end of the—?

It's well after the fact—that's part of our concern. So, we've got a figure already and the numbers now for what the Welsh rate of income tax will raise in 2019-20, but we won't really know what it actually did raise until well after the end of 2019-20, when we're looking back. I'm sure that will be an issue in the committee if there's any significant difference between those. 

And that's when an adjustment occurs in terms of the block grant. So, if Wales hasn't received enough then that would be— 

Yes, it will be done retrospectively, if that makes sense.

Mike, very briefly. 

Very briefly. I think that there are two things, aren't there? There are people on pay-as-you-earn and they'll move across the border, but that should net out, even if people do it wrongly. The people moving from Wrexham to Chester, Chester to Wrexham, should be very close to netting out and it will not have a particularly meaningful effect. The other one is higher rate taxpayers who are multi-property owners. They pay an awful lot of tax, but they may well own three or four properties. I think that was part of the Scottish problem, wasn't it, the people who were in Scotland but also were in London? There's not a lot of those, but in terms of income, they're fairly substantial. How are we certain that we will actually get the right number of people in Wales in the higher rate actually paying it? And that'll be on self-assessment, almost certainly. So, it won't be your pay-as-you-earn people. 

Yes, a lot of that is related to what kind of compliance work HMRC are going to do once things kick off. An obvious point, but, while the rates are the same, it doesn't make that much difference in aggregate terms, in terms of tax paid, but it clearly makes a difference as to whether we get it or whether the UK Government gets it. So, it's very important to us certainly to get that stuff right and not wait until there are big differences in rates to get the processes set up to establish all that. The rules for whether you're resident in Wales or not, in the legislation, are very clear and straightforward. There will be individuals, I'm sure, whose circumstances are complex and who don't fit neatly into the categories, but they should be a relatively small number.  

But they will be the highest rate taxpayers who will be paying—

Possibly. 

Well, almost certainly. Perhaps you could give me some examples of where they wouldn't be amongst the highest rate taxpayers. But they would be amongst the highest rate taxpayers who, although a small number, will be paying a lot.

Yes, you're almost certainly right. You can imagine some people whose income isn't very high but who are very wealthy, because obviously we're only taxing earned income with WRIT, but I'm sure you're right. 

Thank you.

On that continuous monitoring of who should be paying, the Wales Audit Office said that

'taxpayer identification is not a one-off exercise'.

Are you confident that that in-year monitoring will be solid enough?  

The Minister has already touched on a few of the arrangements that HMRC are going to put in place to help them maintain the database, because, yes, there's no automatic notification to HMRC at the moment when you move house. So, to an extent, the system relies on customers informing HMRC of a change of circumstance, but there are other data sources that can be used to check that. So, we're obviously working very closely with HMRC on all that. I don't know if there's anything you want to add, Liz. 

Just to say that there'll be performance measures built into the service level agreement and obviously we'll be monitoring and working closely with HMRC to develop that.

Very briefly, on forecasting, you've already mentioned, Mr Jeffreys, earlier on, that work is continuing with the Office for Budget Responsibility and HMRC presumably on making sure that data available for specific Welsh forecasting is as good as possible. Do you have anything else to add in terms of specifics as to where you think the gaps might need to be plugged? What extra bits of data might be able to be sourced in future? 

I suppose the key thing that we will have in future that we don't currently have is a not perfectly live but much more timely monthly data on the amount of income tax that's being paid by people with the C code. So, the PAYE system will generate monthly data on how much is being paid in the Welsh rate of income tax. It won't be 100 per cent of the tax base that's covered by that, but it's a pretty high proportion. I haven't got the exact figure in my mind, but it's 80 to 90 per cent of the tax base that would pay through PAYE.  

And what's the lag? To find out how much tax is paid in January 2019, when do you find out?  

I'm not sure exactly, but it's pretty much—. The PAYE system runs every month and so it would be a matter of a few days, I would have thought, rather than—in terms of the raw output of that. But as I said, that's not going to give you 100 per cent because the self-assessment stuff, for example, doesn't run on that cycle and there are all sorts of adjustments that need to be made as well. So, you'll have that regular data flow coming through that will help in terms of providing a dynamic picture of how things are evolving.

In terms of having the full-year outturn, as we touched on earlier, there'll be a bit of a lag with that after the end of the financial year. So, over time, this picture will build into a much clearer picture, I suppose, than we've got at the moment, and I feel like, in two or three years' time, our ability to forecast—not withstanding the obvious uncertainties with any future forecast—will be significantly better than it is now. We're still a bit blind at the moment.

I was just going to say that all the forecasts for 2008-09 were absolutely wrong because we fell off the cliff. That's going to happen at some stage, we'll go into another recession, so the forecast will be great, then all of a sudden—because the OBR never forecast a recession. At some stage we'll go into the next recession and things like land transaction tax will lose about 40 per cent. 

Yes. It is that important distinction between uncertainties in your forecast based on not having very good data and uncertainties based on genuinely impossible to predict things.

Whilst we're on the OBR, I've got a couple of questions I wanted to ask because you are clearly working with them on the independent forecasting stuff. One of the things that they told us when they appeared before us a while ago was that producing comprehensive forecasts would be quite challenging because of timings in relation to the Welsh budget and the UK Treasury autumn budget as well. So, I'm just wondering what kind of impact that would have on their independent forecasting role in your view—is that a concern?  

This has been pretty well rehearsed with the committee, hasn't it, already I suppose, but the timings of the fiscal events are not perfectly well aligned from our perspective. So, we produce our draft budget at the start of October, the UK Government budget is later in the autumn than that and this year was towards the end of October. So, the OBR are producing those macro-level forecasts for the UK fiscal event, not for ours, so that kind of misalignment does create an issue. I think, we didn't see very, very significant revisions in our forecast between October, the draft budget, and December, the final budget, this time around and in normal circumstances I wouldn't expect to see—. In a perfect world, you would have a better alignment between all of these things and it's just a factor that we have to deal with. It doesn't make the job impossible by any means. I think the thing we would be really concerned about is if we had reason to think that our October forecasts were going to be systematically wrong—'wrong' is not the right word, but requiring significant revisions every time at final budget stage. I think we would need to start to rethink the process a bit if that was the case and we need to keep a close eye on that, I think. If we are seeing very significant revisions between October and December every year because of changes in the macro forecasts, then I think, you know, we—and also the committee, I would have thought—would want to look again at the sequence of things. 

Okay. Thank you. And in relation to updating the memorandum of understanding and terms of reference between the OBR and the Welsh Government, is there any update you can give us on that? 

I think we're nearly there with the MOU. It should be completed shortly. 

Yes. There'll be another document that's more like a terms of reference type thing, which I think we would publish. These clearly will need to be finalised before the new arrangements come into force, and obviously all of that will be shared with the committee. 

Okay, thank you. David. 

Thank you, Chair. Welcome, Minister, and congratulations on your appointment. Now, to date we've been talking about taxations that are now in place or are coming into place very shortly, but of course we also have the powers to create our own devolved taxations and your predecessor clearly indicated one of four options he was pursuing after vacant land tax. I suppose what I'm trying to find out is where we are with that. Because I understand that it is now with the UK Government laying down the necessary legislation in place, and I understand the chaos that's in Westminster and whether they actually are giving the time to this, but where are we on the vacant land tax, because you'll be the Minister who will be taking this new taxation through and basically following the process for the first time?

Thank you very much for that. I set out earlier the kind of principle, which you're obviously aware of, in the sense that we're keeping the two sets of work separate—the policy development side and the seeking devolution of the power side. My understanding is that, obviously, timescales aren't within the gift of the Welsh Government or the Assembly completely, but following UK Government agreement to take forward the proposal, parliamentary time will need to be made available to consider the draft Order. However, we do hope that the powers will be devolved this year and a timetable then can be agreed as part of the official discussions that are currently taking place. Is there anything further?

That's all right. I think, just to add, that we have begun discussions with Treasury officials about what a draft Order might look like and how you would define the competence necessary to develop a proposal for a vacant land tax. So, that's ongoing. It is slightly awkward that there's no—. This process has never run before, so there's no very clear expectation or blueprint for how long each of these stages take, and given the other challenges in Parliament, this isn't necessarily the top priority on the list there, I don't suppose. But it's very important from our perspective and we're going to keep on trying to make the process move in a timely fashion, rather than let it meander along too much.

I appreciate the timescales are not within your gift because there are issues in relation to Parliament, but are there timescales that you've identified as to which point you think you can't complete this process in this Assembly term?

As the Minister said, our expectation is that competence will be granted this calendar year. So, that would take us to, obviously, the end of 2019 and then we would turn to the proper policy development part of this, which is a case of, 'The Assembly has the competence for this, what might such tax look like? What are the pros and cons of it?' It's perfectly feasible to be in a position to introduce legislation this side of an Assembly election. I think that kind of timescale. But whether that's what Ministers or what the Assembly are prepared to consider is another matter.

So, if we don't get it by the end of this year, that feasibility becomes less.

We expect by the end of this year and, of course, all of this is very much within the context of Brexit and the additional pressures that there will be on the Assembly's legislative time.

Tell us about it.

Okay, and the other question also linked in to that is the borrowing cap. Now, your predecessor, again, informed us that he was in discussions with the Treasury to extend the borrowing limit. Now I assume that's on capital, but I'm also assuming that discussions are ongoing, so where are we with those discussions? And I'm assuming because the annual cap is 15 per cent of the total cap that you're also looking at changing the annual cap as well, as a consequence of that increase in borrowing.

So, as was set out in the command paper that accompanied the 2014 Wales Bill, any changes in terms of the borrowing cap would come about as part of the spending review. So, I know that the previous Minister made some strong representations to the UK Government on that in terms of ensuring that this is considered as part of the spending review and I certainly would seek to continue those discussions. I'm meeting the Chief Secretary to the Treasury at the finance quadrilateral on 15 February, so there are lots of issues, actually, that we've touched on today that I would be keen to pursue, and this certainly would be one of them. 

This is an issue that's still under discussion then.

Yes.

This committee in the past and this Assembly have supported the prudential borrowing model that exists with local authorities. The fact that Blaenau Gwent can borrow without a cap and the National Assembly for Wales can't appears strange to some of us. Have you continued to make the case for the ability to have prudential borrowing, because all that exists within local authorities is within the total net expenditure of the moneys made available in the Public Works Loan Board? 

Obviously, in an ideal world, Welsh Government would be able to decide its own capital borrowing requirements in order to optimise our investment plans, and, obviously, we think it would be more appropriate for the National Assembly for Wales to vote on our borrowing limit than the current system, so having a similar process, I suppose, as the one you've described for local authorities. I don't know what the latest discussions have been on that with the UK Government. I know that they haven't been open to that previously. However, I think that the fact that there are Members raising this—the committee might have a view on this as well. And I think all of that helps.

The committee's had a view on it several times, and I think it's just important that we do keep on pushing for what are reasonable controls, because we won't be able to do what we'd like, in the sense that if you increase your capital, you're having a serious effect on your revenue.

Anything on that to add?

No, you've answered it very well.

Any further questions?

Diolch yn fawr.

Thank you. I think the first thing to say is we are very proud to be Wales's first non-ministerial Government department, and quite conscious of the responsibilities and the governance requirements that that imposes on us. I think there are really two elements to the question that you've asked us, and the first is about responding, in fact, to a recommendation, which we were well aware of, that the Wales Audit Office made in their previous report in December 2017 that was about the importance of ensuring that the governance arrangements and the structure of a non-ministerial Government department in Wales is widely understood. We've done a considerable amount on that, and this year the Wales Audit Office asked us again about our work on that, and clearly felt we had made progress. In that context, I think it's important to state—we know it; everybody knows it; we operate on this basis—that tax decisions within the organisation are made on the basis of tax data, which is confidential to the organisation and is separate from the political process. At the same time, though, we believe we have—. And indeed, in the previous evidence session, you've heard people say the same thing. We believe we have a very good relationship with Welsh Government, and that's a very important part of our structure. 

In terms of the broader communication and promotion of the authority, we did an enormous amount of work before we went live, in stakeholder engagement, in communicating with key customers, taxpayers, with their agents. We ran a whole collection of webinars, and we have continued that work following our go-live date. It's very important to us to make sure that we sustain that. We talk about partnership a great deal. It's a key part of the thinking we're doing in our corporate planning work, and we want to live up to that in terms of the way in which we engage regularly across Wales. 

Well, I think the first thing to say is just to remind us all that we are a very young organisation. We have been established for just over a year, but we have not even been operating for a year. We've made a good start. We were very pleased to read the Wales Audit Office assessment of that start, but we know we have more to do. We're a small organisation, with a range of very specialised staff and skills. And we think carefully about diversification, and we're doing that thinking as part of our corporate planning work. We are developing a purpose statement that we have consulted on internally and that we are engaging in consultation on more widely. And that will form one of the key pillars of the corporate plan, which we will publish later this year. The reason we are concentrating on the articulation of our purpose—and this is precisely to your question—is that we're simultaneously trying to ensure that we, as a small, specialised organisation, stay focused on what we have been set up to do, but also seize the opportunities to deliver value for the people of Wales in terms of the investment that's already been made in us. And that's a balance. One of the reasons organisations of any kind have a purpose statement is to reduce the likelihood of what I call—forgive me, it's a slightly business-school phrase—'unconsidered diversification'. I don't mean to reduce the likelihood of diversification. I don't mean that. I mean to reduce the likelihood of unconsidered diversification. Because, with a small organisation doing something that's quite visible, it's important that we manage that process well. 

Having said that, the purpose statement that we've developed articulates not just the operational service to customers and taxpayers, designing and delivering Welsh national revenue services, but also talks about the data—and you've had various conversations in the previous session about that—and about the opportunity to maximise the benefit of that data in entirely appropriate and confidential ways for the benefit of tax policy and for Wales. 

I could say a lot more about various other things, but—

That's fine, thank you. And we were taken through the purpose as well, when we were with you. 

Indeed, of course you were.

David. Thank you. 

Diolch, Chair. Since you've mentioned data, it's does seem natural to ask the question. We've raised questions many times on this committee on specific Welsh data. You've heard yourselves the questions to the Minister on the Welsh data. I suppose the question we need to ask is: what lessons are you learning about collecting specific Welsh tax data? It is important if you do your forecasting, to have that information. Are there any lessons you've learnt in the process to date? Are there any lessons you've learnt from perhaps other nations that we might not yet have come across? Where are we on it, and where are we going with it?

Happy to try and answer that. So, on the point about how we learn from others about data collection, it's probably not just about data collection, but how you manage and house the data, and how you make sure that that data is available to the organisation internally in order to be able to interrogate it and to provide meaningful analysis on the back of it, to support forecasting, policy making or whatever it might be. So, yes, I personally went to meet colleagues in Revenue Scotland, and my career background is working on exactly that in Her Majesty's Revenue and Customs, and the Valuation Office Agency,  so in other parts of the UK tax system. So, I'm aware of all sorts of pitfalls and things that can be difficult.

It probably starts with an understanding of what data model you want to have, and that is what then informs the design of your digital systems. So, some of what we do about data goes way, way back to decisions we took very carefully 18 months ago, about saying that we want to make sure that the data model maximises the ability for us to be able to bring data together and be able to interrogate it in a meaningful way. So, that's the first thing we did, and that's implemented into IT systems—the physical infrastructure that allows you to do that. So, I think we've done that from the get-go. I'm not saying it'll be perfect, but we've certainly thought about it and done it in a way that allows us to match data in a way that, perhaps, we wouldn't have done had we not done that thinking to start with.

What have we then learnt? I think one of the most important things that I've learnt, from my experience of this, is actually publishing it as soon as you possibly can, because the public and interested stakeholders will spot stuff—'Hang on a second, what's going on there?'—that you might not do, as well. So, we've been publishing data on a monthly basis from the very beginning of our operations. So, we publish land transaction tax data monthly and have done for—what are we on, now? Nine, are we on now, or something? So, we do that with LTT, and land disposal tax, we publish on a quarterly basis. We provide commentary on a quarterly basis on both of those things. So, we put data out there.

I think the other thing I'd say about what have we learnt and what's important about that is—and this is reflected in the conversations we have with Welsh Treasury—it's less about the actual data themselves; it's more about the skills and managing the people who are dealing with the technology, the systems that are capturing those data in the first place, which means that you really understand what they mean. There's a very big difference between the data that you get out of transactional, operational, digital systems than you get out of, say, something that the Office for National Statistics runs as a survey. Because once you've done a survey, it's fixed, it's not going to change. That's not the case in transactional systems, for all sorts of reasons, people amend things, change things. So, I think what we've really done from the very beginning is make sure that we've built a team, and, having had experience of working with teams like this in the past, I'm really proud of small but very, very competent data analysts in the WRA, and they're made available to Welsh Treasury. This is not about revealing details of the data themselves but an understanding of where they come from and why they're structured in the way that they are and what they actually mean. So, I think that will be the extent of what I would say by way of a response to that.

Okay. I understand the data model, therefore, which sets up your data you want analysed, and analysts work on that basis.

Yes.

I fully understand that, but I suppose in a sense, therefore—. Are you scheduling a review of your data model to see if actually it is delivering what you anticipate it to deliver? And, are you looking at the role of the data analysis in the sense that: will you be expanding it into the Welsh rates of income tax agenda as well? You talk about the data; you can only analyse the data you know you have, so you need to know exactly what data you have, is it working, and, therefore, the technology people—you work on what you're told to work on.

Yes. I suppose, to that point, it's early days. I'm sorry, but that's the reality. Yes, we will want to look at—are there things that we can do to make that better? We've not yet had a year's worth of operations, let alone a year's worth of data to be able to do that properly. But I think you're certainly on to something with—. That's why our purpose statement points out the fact that the WRA could well have a role in making sure that the most is made of the data that is in Wales for taxpayers—Welsh taxpayers' data. So, we would want to do that, but it's a bit early days, I think I would just say. But yes, we'll get there.

And have you a schedule of review for the data model, back to the 12-month period, for example?

I suppose what we have scheduled is a review of how the first year of implementation of the two taxes has gone. That would be one element of it, yes.

Okay. And a final point, in a sense, from me. The forecast for the landfill disposal tax was slightly out, by £16 million—have you understood the reasoning for that and, as a consequence, do you think now that your future forecasts and the data you're collecting will probably be more accurate from that agenda?

The first thing to say is that the forecasting of LDT is Welsh Treasury's job, not our job. I think it's probably still a bit early to be absolutely certain about all of the factors that contributed to that difference because we've only got six months, two quarters' worth of data on LDT. I'm just anxious to make sure that we don't draw any very strong conclusions at this early stage, but I suppose I would refer to the previous answer that having people who really understand the data will be hugely advantageous to understanding what the forecast should be going forward. I think also—I think Welsh Treasury colleagues would agree with this—having access to Welsh-specific data would definitely make it easier to do forecasting in future. As for where we're at at the moment, it's just a bit early to come to a definite view, I think.

Can I just add one thing on that? That answer and the previous answer to your question, I think, underlines the fact that one of the things we're building during this first year of operation is a baseline of data on which—obviously we haven't built it yet, because we're still only nine months into the operation. An obvious point, but I just wanted to emphasise it.

No, I understand that and how much you need to have that baseline for you to work off in the future. But I suppose what we need to be reassured of is that you do have that baseline in view, and that you are looking at reviewing that baseline at some point to see whether it did give you the accurate information you need to use it as a baseline.

Yes, and again I think only time will tell a little bit. I think we've got much better data, much better understanding than we had previously, and I'm confident that the people that we've got will be able to give a robust understanding of that, but not till a bit more time has elapsed, basically.

Thank you. Neil.

Sorry, I didn't realise I was next in line. You started off by creating a corporate plan for your first year of operation. Now you've got the slightly more challenging task of developing a corporate plan for the next three years. Can you give us a bit more detail about the process that you have internally for developing this plan—what the top priorities that you've got are within it?

Yes, certainly. So, I think the first thing to say is that we have a range of internal processes, but we are, referring back to my earlier answer, also very focused on external consultation and stakeholder engagement. We talk about that all the time and we wanted to live it out through the preparation of the plan. So, we have a purpose statement. Those of you who were able to visit the office at the back end of last year will have heard more about that. It isn't finalised yet. We're still going through some consultation processes, but it has been discussed internally with members of staff, with the board, and externally with the previous Minister and with the Permanent Secretary. This forms one of the three pillars of the corporate planning work that we're doing. The other two are a statement of our approach to tax collection and tax compliance, which you will know from your visit is quite innovative, is quite different from the way things have been done elsewhere and in the past, and the third pillar is the set of measures that we build on the back of those. Because of the innovation in our approach, because we are consciously choosing to take a different approach, a much more engaging, partnership approach with taxpayers, the measures that will result from our work will look different from those you will see in HMRC's reports and even in Revenue Scotland's reports, because the measures reflect the range of activities we're trying to do and the way we're trying to do them. We're still working on those. The board has seen some of them, but some of them we're still seeking for something that is both hard enough and informative enough to help us track whether our approach is working.

Well, you've got 10 weeks left, I think, roughly, before you need to submit this report. So, you're on track to meet that deadline.

We are on track. The board has already seen one of the third or fourth drafts. We're in regular engagement on it.

Right. And what indicators have been put in place for 2018-19 to measure the overall performance of the WRA? You referred earlier on to the WAO report, some of which was very complimentary. I think it's fair to say as a committee that we share the WAO's opinion of the work that you've done so far. So, how are you going to monitor your overall performance going forward? How have these indicators that you have in mind been monitored hitherto? Some of these things are not statistically measurable, so how do we make these meaningful?

Let me start and then I'll come to Dyfed. So, some of the measures are exactly those measures you would expect any organisation to have about financial probity, about living within our budget, about size, about the range of skills in the organisation, about the diversity of staff, about staff engagement. We've just had the civil service annual survey results published, and we were delighted to find that we were the second-highest organisation in the United Kingdom on a range of scores, and on four particular scores we were the highest in the UK.

So, those are the things you would absolutely expect any organisation to monitor. We monitor them and we report on them. But, as I said earlier—and I think this is the point of your question—there are some things that are specific to the way in which we're working, and some of those are quite hard to define in ways that are informative. So, for example, if you took the amount of time people spend on the telephone, you might think that it would be good to have that as low as possible. But in a context where we're really trying to assist taxpayers right at the outset to pay the right amount of tax at the right time, actually, the opposite is often true; it's better to have a one-off call that is a bit longer and more substantive. So, that's an example of a measure that we will need to interpret and use differently.

There are others. Dyfed, do you want to pick up some of the work that we're doing now? 

Yes. The idea of this year was to develop a mechanism to work out what sorts of things were the right sorts of things to measure, as Kathryn touched on in the last point, and then have those ready for the corporate plan. The sorts of things that we've been looking at this year—this obviously hasn't been signed off yet, but I would fully expect to see as part of our measures going forward—are things like the number of people who are filing their returns online. Now, obviously, we're not entirely in control of that—we need our partners who are dealing with us to actually file online—but I think it's a fair measure of looking at whether or not our systems are easy to use, because if only half the people file online then, obviously, there's something wrong with your system, you might contend. So, that sort of thing. 

How many people pay by electronic means or pay by cheque? Those sorts of operational transactional things. We've been looking at monitoring this year month by month, as we've got the data along, and seeing what's happening with it, to say, 'Okay, so we think that the baseline is this, and therefore what we should be looking to do is improve by that amount.' So, that work we've been doing. Cheques is a good one, actually, where we've seen a significant reduction already in terms of the number of people who were paying by cheque to the number of returns—sorry, returns rather than people. The number of returns that were paid by cheque has reduced significantly. We'd expect that to reduce more over time. 

So, we have a suite of measures. We've developed 16 draft measures that we think are there or thereabouts. We are in the process now of discussing that with our board to make sure that they're comfortable with that and then, clearly, that will be a means of indicating how well we're doing in delivering our corporate plan. And that's something that we'll need to agree with our Minister before the end of March—as you said, it's 10 weeks to go. 

So, we have a range of them. We have been looking at some of them this year. But I think it's also important to say that given where we're at as an organisation, for example, one of the big things we want to measure is reductions in tax risk. 

Yes, I was coming on to that; you've anticipated my question. 

Sorry. So, that's something that, clearly, we want to be able to measure effectively, but obviously we had no idea what the extent of it was on 1 April 2018. So, it was just impossible to measure in this first year, so we've focused on trying to make sure that we establish what that measure should be and the best way of doing it, really, in the first year.  

Can you flesh that out a little more, and say how you're defining tax risk and what you're looking for? 

Yes. So, what we're attempting to do here is identify areas where we think there is a possible risk of the amount of tax that has been paid not being correct, which might be for a number of different reasons, and clearly then define what the parameters of that are, and therefore define, 'That's the risk.' So, in other words, there are parameters that sit around saying, 'If the following is true, then we think that for this type of transaction, there's a potential risk here.' Then, what we want to do is baseline that at the beginning of the period, so that will be from 1 April of this year, to say what we think the extent of that is, and then re-assess that at six months, 12 months further down the track for that same risk: what's the likelihood of it materialising to be an actual error in the amount of tax that's paid? So, what you're seeking to do is reduce the amount of both opportunity and actuality of people paying the wrong amount of tax.

I would be completely open and say that this is very innovative; this is a really new way of measuring the efficacy of a tax authority. And I think I've tried to explain the shift—it's not a perfect analogy, but, in effect, it's a bit like the fire service going from measuring the number of fires they put out to measuring the number of fires that actually happened in the first place. And they're quite sort of—. It's a difficult transition, but everyone can see now, 'Yes, that makes more sense', because of the way it incentivises the operations of the organisation. So, that's what we're doing at the moment: we're identifying those risk areas, working with HMRC as well to get their input into what they think would be the relevant parameters as well. So, we're relying on their experience too, and using that helpfully.

Does HMRC do this in any way, or is this new ground that you're tilling here?

A bit of both, I think. I don't want to get drawn too much into exactly what HMRC does, where I'm no longer expert. But, certainly, the conversations I've had with colleagues in HMRC, they do this type of thing. I think there's a desire to—there's an understanding of, if you can get it right, this is actually a very helpful way of incentivising the tax authority to act in a way that is genuinely all about making sure that people pay the right amount of tax. But is that universally what HMRC does? No, it isn't.

Good. Okay, thank you.

Thank you. Rhun.

I think that there are a couple of elements to that. And the first of them is about creating the kind of organisation that people want to work for, want to join, and want to stay in. I've already commented on the civil service staff survey results, with our score of 85 per cent. And I think that's a reflection of the effort that has been made, and the atmosphere that has been created by leadership, by the kinds of structures we offer. And we are looking consciously at the kind of culture we have created, and the kind of culture we want to sustain. So, that's part of it.

The other aspect of the question though is about permanent employment and secondment—something that this committee has asked about before. And before we cover the detail of that, it is worth reminding ourselves that it's not necessarily a bad thing to have some staff who are on secondment. It gives the organisation flexibility, it provides the opportunity for refreshment, as posts change, as our operations develop, it makes it much easier to flex. And it has some benefits for them, to be able to come in to a young, new organisation, doing some innovative things, and then take that expertise out elsewhere.

Do you want to say anything about any particular details? So, the only thing I would say is that the board has a sub-committee, the people committee, and one of the things that we focus on is succession planning—succession planning, obviously, at senior levels for the tîm arwain and the chief executive, but actually also looking organisation wide, considering those people who are on secondment and who have a secondment end date and those people who have just joined us permanently. So, we pay attention to that particularly, because we're a small organisation and are very reliant on these capable and competent people. 

I would like to answer half that question by saying that we have been participating with tax authorities from other parts of the British isles in a regular devolved—. It's the British isles tax authorities forum, and one of the pieces of work they are doing—it's top on their list—is the development of an articulation of what it is that a tax professional needs—they use the words 'a tax professional'. So, we're starting to collaborate with them in thinking about those things. 

The question you asked, though, was: does Wales need one? I think part of the answer is that participation in that wider body is useful in itself, aside from the opportunity to collaborate and share the work. Do you want to comment on that too?

I'll help that along. What feedback have you had from stakeholders regarding the guidance you have produced?

Pretty good, I think. I don't think we're going to be up for any Booker prizes for fascinating literature on guidance, but I do think that the use of it by professionals and the feedback that we get has been really good to date.

I think also, I like to think about guidance as a process rather than just as a product. So, we work continuously and openly with stakeholders to understand what we can do to improve the guidance we've got and also to use technology to perhaps facilitate people's understanding of when they need to use it—making it more diagrammatical and that sort of thing. So, we're in the process of doing that at the moment. I'm pretty happy with the feedback we've had so far.

My final question: there were delays in publishing some guidance documents for April 2018; did this have any impact on the effectiveness of the guidance?

I don't think so, because what we tried to do is prioritise making sure that the guidance that was available for day one was what you needed to do the things that you needed to do by then, rather than ensuring that everything was available. So, as far as I'm aware, and I can double-check on that, there has been no—I've not received any complaints about that at all, put it that way. 

Thank you. Nick. 

Diolch. Good morning—yes, it's morning, still. [Laughter.] Given your information technology systems are cloud-based, how do you ensure the security of taxpayer information and what are the advantages of this approach?

We had people in yesterday doing some testing of our systems, as it happens. The first thing to say is—and I've been guilty of this—thinking about cloud systems as being sort of potentially less secure than something you can see in the corner, where you can hear the thing whirring—'That's my server.'

The mainframe.

Yes. I can secure that—I can put something around it. I'm not sure, actually, that that's really the case. I think actually having a cloud-based provision of your IT systems is probably even more secure, to be perfectly honest. What we do is use the expertise of others—we leverage the fact that our systems undergo lots of testing. They were accredited before we went live and we used the National Cyber Security Centre's accredited people to do that checking for us to make sure that the systems are in place and we do that regularly. So, there is an ongoing process of checking the systems. From my perspective, I suppose, and obviously I'm not a cyber security expert—that's not where I come from—but, as a chief executive, it's what do I notice about—'Hang on, are these systems really secure?' I suppose if you permit me one very, very brief story: we've now moved on to having electronic pay slips, as well as digitalising other stuff, so I wanted to send my pay slip to my personal e-mail address so that I could keep a record of it, for a variety of reasons. So, I sent the e-mail from my work account to my personal account and it set off a security warning because the system had detected that somebody had sent out a national insurance number from a WRA system to outside of the WRA system. Obviously, there's nothing wrong with the chief executive sending his own personal pay slip to himself. I actually got some sense of confidence from that. Okay, so I've actually used the system and, in a way that I haven't expected, it has automatically provoked something that tells you that the systems work. 

Only yesterday, we had a National Cyber Security Council-accredited person in doing some penetration testing for our new software, which will go live on the twenty-eighth of this month. I don't want to quote this individual precisely, but essentially he was saying that he was very, very impressed with the quality of the security that we had and that it was industry best practice. So, we rely on the support of experts who know about that. I think the cloud services—we use Welsh Government's wider infrastructure to do that, and obviously we're in Microsoft Azure, so you're protected by the whole of Microsoft's security package too. So, I'm pretty confident with how it's going at the moment.

Just playing devil's advocate for a moment, because I hear what you're saying about the security and about your ambition to be a digitally advanced organisation, and that's all commendable, but there is a large cost to that, isn't there, relative to sticking with—well, I imagine there is, maybe there isn't—previous systems? So, I'm just wondering how you balance, for a smaller organisation, the costs of maybe trying to be as digital as possible with the benefits you're actually getting. 

I'll start, you carry on. I think part of the answer is that the cost profile looks different. The costs fall in a different place. So, if we're at the stage where 97 per cent of the transactions we deal with just go straight through, then you don't have the investment of both time and money in members of staff having to deal with it, so the cost profile is different. We benefit from the enormous investment in security controls and assessments of our provider, so we don't have to make that investment ourselves. But having said that, there are some elements of cost that we pay.

I think we're probably at the forefront of the exploration of what digital—public service, I should probably say—but digital service providers have to contend with in terms of what it's like to do cloud versus not. And one of the things that we're starting to get a better understanding of is that you don't have big spikes of capital investment when you have a refresh—you don't have that because you're renting. But when you come to perhaps develop new software, then you're creating an environment for a period of time in order to test the new software before you then put it into your system, so you're using that capability for a period of time and then you're turning it off. So, you're only paying for what you're using. So, that means that there's probably an ongoing cost, and in terms of Government accounting terms it's not capital, it'll be departmental expenditure limit costs, so that will be different too. So, does it cost more? I think, from the experience we've had thus far: no, it probably doesn't cost more than the alternative, which is to have a heavy capital cost and then depreciate over time. That's not what we'll need to do. And for a small organisation, actually it allows us to be much more nimble and agile and allows us to leverage skills from others that we just have no hope of being able to create, all of those, for ourselves. 

Are Revenue Scotland going down a similar path or have they stuck more with the old-fashioned way of doing things, so to speak? I'm not even going to ask about HMRC because I imagine that they are still dealing with systems that they've had for a while.  

I'm not an expert on either of those organisations. I think what's really important to say is that the systems that the WRA runs are different to what Revenue Scotland runs. So, we run our own finance system; that's not what Revenue Scotland does. So, we're not sharing Welsh Government's service, we're providing our own in order to bring data together from what people are paying and from filing the returns. So, we just run a very—it's quite difficult to make a comparison. Revenue Scotland are in the process now of refreshing a large-scale capital investment in a new product. I don't know the extent to which that will be cloud-based or not. I'm sorry, I don't know. 

It was just out of interest, really—you didn't have to have knowledge about all the other organisations in Britain. Anyway, that's fine.