Cynulliad Cenedlaethol Cymru

Good morning, and can I welcome everyone to this morning's meeting of the Llywydd's Committee? Members will be aware that our role is to scrutinise the funding allocations to the Electoral Commission, and we welcome members of the Electoral Commission to this morning's meeting. Before we go into the item, just a bit of housekeeping. Can I remind Members, if you require simultaneous translation from Welsh to English, that is available on the headphones via channel 1. If you require amplification, that is also available on the headphones via channel 2, or that might be channel 0, depending on which box you've got.

There are no scheduled fire alarms for this morning, so can you please follow the directions of an usher if one does take place? If you have any mobile phones or other equipment, can you please either put them on silent or turn them off, so it doesn't interfere with both the meeting and the broadcasting equipment? We have received apologies from Joyce Watson, and can I welcome Hannah Blythyn, who's substituting for her today? We've also received apologies from Llyr Gruffydd and welcome Heledd Fychan, who's substituting for Llyr today. And we've received apologies from Janet Finch-Saunders, who is unable to attend. Does any Member wish to declare an interest at this point in time? I see everyone's saying 'no'.

In that case, we move on to our scrutiny session of the Electoral Commission financial estimates for 2026-27. And can I ask the therefore members of the Electoral Commission—I'll go from left to right—to introduce themselves and their posts within the commission, please?

Good morning. I'm Niki Nixon. I'm the director of Communications at the Electoral Commission.

Vijay Rangarajan, chief executive. 

Bore da. Good morning. Rhydian Thomas, head of the Electoral Commission in Wales. 

I'm Chris Pleass. I'm the director of corporate services.

Thank you very much for that, and you will be aware that, obviously, at the end of the meeting you will be able to access at some point a transcript for the meeting as well. We'll talk about that at the end, anyway. Okay, before we start our questions and our scrutiny session, perhaps you'd like to give us a few opening remarks.

I'll turn to English just now. We're very aware and very grateful to you for the support that you've given to the activity of the commission. Obviously, that brings with it a degree of scrutiny, which you are doing this morning, but we're also grateful for the kind of patronage that you've shown when we've done our media round-tables and our work with young people. I hope that that will continue in future.

So, next year, of course, is the big one; it's the election year. As you will see from our budget, our budget will reflect that event. As you know only too well, we are facing an election that has new constituencies and a partially new way of voting. I think, as Peredur points out to us often, it's not a completely new way of voting, because it sort of reflects the list as it stands. But I don't think—if I can just make a few general remarks—I don't think there has been a time when the scrutiny on the process of elections, and on votes, and how donations happen, how parties are financed, and eventually how people are persuaded either way to vote for particular parties, has come under such scrutiny. I think that the rise of more populist parties throughout Europe, throughout the United States and in the UK has been instrumental in, perhaps, shifting the balance that we've seen in the past between two main parties. So, this is a very complex situation that we find ourselves in.

There's also—and I'm sure that our chief executive will be able to talk to this much more in the coming few minutes—there's also been an emphasis on deepfakes, on the use of social media in a non-beneficial way, and especially a rise in harassment and vilification of candidates and especially, if I may say so, women and candidates of ethnic minority. As such, our core budget has also had to reflect a strengthening of our IT structures, of our cyber security and our ability to, perhaps, detect and call out deepfakes, if necessary, as we go on.

It's not enough to have this injection of money for a six-week period leading up to an election. It has to happen for months leading up to any election and, indeed, I would argue, all through the year, every year. So, you will see that the Speaker's Committee has actually strengthened our general finances substantially and I think we were the, if not the only, public body that received such a substantial strengthening, financially, from Westminster.

So, as you will see, our budget today also reflects a need for strengthening with regard to the elections in Wales. So, we stand before you willing, as always, to share our thoughts with you as to how we can best progress to make this election safe, trustworthy and where the general public are not either misled or feel that somehow the election has been won or lost on unfair grounds. So, that's our role as a commission, and I hope that you will be able to hold us to account as to how we're going to do it. Diolch yn fawr iawn.

Thank you for that, Elan, and I can assure you we will try and keep you to account for that.

Yes, I'm sure you will.

It's also pleasing to hear that the Speaker's Committee has actually decided that it's allocating more funding; perhaps more can be shared our way as a consequence of that as well. Okay, thank you. There are some points you made there that we have looked at in the estimate, which are the ICT, for example, and a few other things, and protections. You've highlighted, perhaps, some of the challenges that are facing all elections and candidates in the elections, and all members who are elected as representatives in the future. But we need to make sure that (a) we get value for money, and (b) that the accounts and the estimates reflect the need as well.

So, in line with that, we've got some questions we wanted to ask you. I'll start with some nice, easy ones. Last year we talked about communication with the committee, and I just, I suppose, want to ask the question as to how you feel the commission has ensured that its communications with this committee have improved as a consequence of our recommendations last year. So, Rhydian, perhaps if I come to you.

Probably, yes. I think the process has certainly improved. We've got an engagement plan, a communications engagement plan, which we've drafted alongside yourselves. We have planned meetings two or three times a year. You were very welcome in coming to our office earlier this year, and there are written briefings that we provide to you. And we will try and include you, as well, in the work that we will be doing around the election of next year, to look at the booklet, which I'm sure we'll come on to talk about later on.

I think, maybe, on a slightly wider point, there is still some work to do there in terms of Members of the Senedd more generally. We've been trying to attend the marketplace events that the Senedd holds regularly, and provide briefings to Senedd Members. But I think there is potentially more that we can do there in terms of ensuring the messages that we have are communicated effectively. We have talked previously about setting up an all-party group relating to democracy and elections, and that's something that we still intend on doing, potentially—maybe one shadow meeting ahead of the elections themselves next year, and one following the elections. But that will form part of—. We will, after this meeting, once again provide you with an engagement plan for the next calendar year, and we would welcome any feedback that you have, or any ideas or any improvements that we can make to that.

I think, David, also we have tried to answer all the points that you raised with us last year, if you'd want to look at those. I think all of them—. Possibly, one of them might not have quite achieved it, but I think most of them have been achieved, of the points that you raised, which you were concerned about last year.

Okay. And your plan for engagement, it would be very helpful for the committee to see that. And in a sense, I also link into, now, the—. We talked about the disparity between the timings of budgets of the UK Government compared to Welsh Government, and that hasn't changed. So, how are you managing to address that change, that difference, between the two sets of budgets? Because the UK Government's budget will be shortly delivered, and the Welsh Government's won't be done until the new year. So, how are you managing to make sure you're actually benefiting from all that? It's Chris, I assume, on this one, is it?

Yes. I'm happy to talk about that, David. So, I think the first thing to say is, it is, of course, extremely challenging and probably unique in the public sector. So, we're effectively having to budget six months ahead of the normal budgeting round. And we have very little discretion to change our budget assumptions once we've done that. So, once we go through the Senedd's and the Scottish Government's estimate process, effectively, because it's proportionate or apportioned, we're locked into the Westminster budget. So, it is very challenging. We approach it very seriously and rigorously. So, we run a number of iterations involving the board. And one of the things that I've looked at carefully since I started with the commission about a year ago now is our forecasting performance, and it is definitely improving markedly.

So, I think it remains challenging. We would very much like to understand opportunities to align our budgeting cycles, and it's something that I'd like to talk to the Speaker's Committee about as well. We've got a private session next week. So, I'm very keen to do that. It would absolutely improve the quality of our forecasting, because of course it's carried out more proximally to the financial year, and we would be able to come up with a more robust estimate. As you know, we do have an error correction mechanism during the year. So, we only draw down from Wales what we are expecting to spend. So, every quarter, we reforecast our expenditure for Wales, and we would only draw down what we need to cover that spend. So, there is an in-year correction mechanism, should we need to change our forecasts.

Vijay, do you want to add?

I was just going to expand on Chris's answer. We've got a lot better at the short-term forecasting now, and actually cost controls and actually knowing precisely what is in the budget and what we're going to spend—where we're not spending it, obviously, we either don't draw it down or it comes back. So, you'll see a list of contingencies that are just not being used—fine; they will not come.

The other thing, obviously, we produced last year was a forward five-year budget, trying to give a real sense of forward look, because a lot of our bigger projects and plans don't deliver on an annual basis but really are spread across two, three and sometimes four years. So, I think we've shared all of that. Taking up the recommendations that the committee made last year, we've split those down, so you can see very clearly the core, the project—we separate that out—and then the event costs, because they have very different profiles.

I think that longer term profiling is incredibly important, not least because it's part of the glue that also binds together the other complexity, which is multiple different corporate plans on different timescales—so, trying to actually bind that so you can see what projects address the needs in Wales, the needs in Scotland, the needs across the UK. And they are similar, but they will deliver in different places for different elections at different times, and, again, there's a bit of work to do on making sure that is coherent.

Could I just add? I think one of the lessons we knew over the last two or three years has been the need to avoid supplementaries, if possible, and therefore good forecasting is of the essence to that. However, I think there is one area in which we have not submitted a budget to you, and that's the automatic registration, mainly because we have no idea how it will work and when it will come into being, and therefore, if that did come into being within a reasonable time frame, we would have to come back to you on that one. 

We will be asking questions on that specifically, on that point. So, do you want to ask that question now, Peredur, since you were—?

It's more—. Can I just—?

Yes, do the two.

It was on the timing. But I'll come on to that as a second part, because you've raised it now. With some of the work that the Finance Committee has been doing with directly funded bodies, and you've engaged with that, I'd be interested, for our legacy report as a Finance Committee going into the election and then for a Finance Committee to take forward in the new Senedd, in that timing issue, and especially around supplementary budgets. We have engaged with the Government. The Government at the moment haven't decided to do any changes, but any reflections as to what you'd want, or anything further you want to add on the challenges that come, especially around supplementary budgets and not being able to lay one without the Welsh Government laying a budget, might be of interest for us. But that was more of a comment rather than anything.

But, coming on to voter registration, can you give us a quick taste of how that pilot has gone in Wales, Rhydian, and when are we likely to see any results from that? Because there are some constituencies that have tried to do it. Are there any early indications of how it's gone?

I think that the work generally has been very smooth. We've worked effectively with the Welsh Government and the four pilot areas that have been trialling AVR over the course of the year. The statutory deadline for the publication of the report is the end of December. We'll be publishing well ahead of that. Then it will be for the Welsh Government to decide what to do in the future in terms of AVR—whether this Welsh Government or a future Welsh Government—and whether they decide to roll it out in its entirety, whether they decide to use one of the models included within the pilots, thinking about specific audiences and how they're included and not included. We will obviously want to play a full part in that discussion with the Welsh Government and with other partners as well, especially the electoral management board and local authority leads in Wales. So, we'll be publishing the report in mid December, and we can ensure that the committee has a copy of that report when it's available.

Are you getting any sense that the Westminster Government, because they're talking about it as well, are looking over at those pilots and using them as a basis to see what could work?

Yes, very much. We've been talking to the Westminster Government, and the Ministry of Housing, Communities and Local Government in particular—and you've probably seen that in the Government's elections strategy—about how their proposed AVR system is going to interface with this one. I think at the moment it has left us with a lot of questions, and those will need to be worked out. In particular, does the UK-wide AVR system, which is going to be able to draw on some of the bigger data sets, like passports, driving and vehicle licensing, benefits, some of the CIS database, for example—how will those interface with the local data sets that the pilots are using? So, how much council data do you mix together? That's a big question; I don't think there's yet a clear answer to that. 

And second, once that data is mixed and a returning officer has sufficient assurance that somebody should be added to the register, to which register? Is that going to affect, for example, the Welsh local government register as well, and how will these multiple systems work together? Our initial view on this is that it doesn't make sense to have several different parallel overlapping automatic registration systems, but if you've got electoral registration officers in whatever area of the UK, they should be able to draw on as much data as is needed to assure themselves that somebody should be added or should be removed from the register. That is why I think that there is still an inclarity as to what's going to be budgeted when, what IT system is going to be used and how to actually make this really efficient for electoral registration officers, because it should simplify their lives and, most importantly, simplify the lives of voters who will find themselves on the register.

There are two other issues, which are interlinked. One is the big digital ID roll-out, and again the timescales of that still are not set, over the next couple of years. But if there is a UK-wide digital ID system, that is clearly going to have a massive interface with automatic voter registration in particular if there's a one-login system that allows a form of trustworthy identity. So, I think that a lot of this is still to be worked out in the course of the next couple of years. It does have an enormous benefit at the end of it for voters, I think, and for administrators and for political parties, but we can't yet see exactly what system is going to be implemented when and hence estimate what the costs are going to be.

There are some very interesting constitutional issues with regard to who administers and who does what. 

I think that, on the Bill, there have been discussions already with the Senedd. A legislative consent motion will be required, I think, for this Bill. In particular, if it starts to get into, for example, electoral registers in Wales and the local government register, it will definitely be needed. 

Can I, very quickly, add some reassurance to the committee? We're talking about AVR for the 2027 elections, but we have included within the budget your general voter registration campaign. So, it's not as if we haven't thought about what will happen ahead of the 2027 elections; there is some funding there for a registration campaign. What we are saying is that, if there is going to be significant change in terms of the introduction of AVR, we would need to change that campaign, and we may even need to include a booklet or some form of voter information to it. So there is funding there ahead of the 2027 elections and the election campaign, it's just that we haven't included AVR specifically at this point. I don’t know if Niki would want to add to that.

I think that's exactly right. One of the things we'll need to know is, if AVR does go ahead, who is captured by it, and that would impact the way we approach our campaigns. As Rhydian says, for the moment, we're planning to go ahead with the registration campaign as is, and that's in the budget for 2027.

When is the go/no-go on that? When do you get too far down the line to—? Because there's planning that happens way upfront on this sort of stuff.

Yes. If the registration campaign needs to change significantly, it would be helpful to have 12 months' notice. For something like a booklet to provide additional information, six months' notice is sufficient, but really just as much notice as possible, because if it's a significant change for the campaign, we'll need time to create new materials.

But in the meantime, if something significant changes, then you'd need to come back to us and say, 'This is the cost of it'.

Yes.

It's interesting to hear what you're saying about the combinations and different options available. As you say, it's quite complex in terms of how one has a knock-on effect on another. Therefore, it affects the budget allocation for you as to what needs to be done. So, it's something we'll keep a close watch on, I think. It's an interesting point.

Since I've been talking about the difference in the budgets, I will ask a simple question. We all know as well, I'm sure you're aware, that we're in the final year of this Senedd, and the situation following the recent by-election causes difficulties for the current Government as to ensuring that it gets a budget passed. It's a hypothetical question, but clearly, if that fails to happen, then the Government and everything else will operate on a 75 per cent allocation. Have you taken into consideration what that means for the commission? Because it affects you as well.

We have. We've looked at what a 75 per cent allocation would look like. Clearly, just to break this down, in terms of the core costs, we would try to find a way to just keep running, doing what we needed to do, trim some of the projects, reprofile. I think the really difficult part would be the event costs. Maybe I can turn to Niki on what would we do, because about three quarters of all the money is event-related for the 2026 and 2027 elections. For the 2027 election, we've budgeted a broadly similar amount, because the same number of voters will need to be reached across Wales. The scale of that would have real and quite obvious impacts, and we've thought about how that would work. Niki.

All we can really cut within the campaign are opportunities for voters to see it. A reduction in cost can be done, but it does mean that the campaign would be placed in fewer places. It would have an impact particularly on under-registered voters, I think, because our campaigns are weighted towards under-registered voters. The less we're spending, the fewer we can reach, essentially. We plan our budgets in tiers, so we tend to opt for the highest tier, because it helps us reach the highest number of people with the message that they do need to be registered. So, that's unfortunately what would have to be reduced.

I ask this question because the estimate we've been given is for 1 April 2026 to 31 March 2027. Come 1 April 2026, there'll be very little campaigning left to do on registration, because there's a cut-off point for that. I appreciate there's also the analysis of the election, which takes place following, and that's a big issue, but I would have thought the campaign materials and everything else should be covered under this year's budget.

There is a large amount that is covered under this year's budget, but actually, the majority of the advertising actually happens in those final three weeks before the registration deadline. We know that people are very motivated by a deadline, and that's when we tend to see the spike in registration.

The majority of the spend goes on actually placing those adverts in the three weeks. I think it's £750,000 that we have got allocated to those three weeks for placing the adverts. You're right, all of the creative is done, all of the materials already exist, but we would be paying in those three weeks for them to be placed across the channels that reach the most people. So, it would have an impact on the Senedd campaign.

So, the payment of that comes out of next year's budget, not this year's budget. 

Yes.

Okay. Thank you for that.

I think I'm right, Niki, that when we look at cost per voter, actually, the most difficult ones to reach are also, obviously, the most costly.

Yes.

And that's part of the problem when you're trying to reduce your budget.

Okay. The Senedd Cymru (Member Accountability and Elections) Bill—I'm not sure if you're aware of it—I don't see it in your estimate. I understand possibly why, because it's only just been introduced. Looking at it—because I chair that committee—I don't anticipate a heavy cost to you in that. Have you looked at it? Do you anticipate any additional costs in relation to providing information on the Member accountability Bill, particularly as there's an element that talks about elections? Part 3 is all about the election, not about Member accountability post the election, basically.

I don't think we would incur any—. We'll double-check this again, and I know we're giving evidence to your committee in a couple of weeks' time and we're also providing written evidence, so maybe that will be the place where we can dig into it in some more detail. But I don't foresee there being any significant add-on costs to what we will be doing. The focus from our perspective will be on guidance and making sure that administrators know what they need to do, and also that candidates and political parties will know what they need to do and can and can't do as part of any recall process. So, we wouldn't necessarily be looking at any public information or public awareness around recall.

Keep an eye on Part 3, because it goes wider than just candidates.

Yes.

If I can just add, I think these are the sorts of costs that we need to absorb, really. It's a big piece of legislation, it's got some very innovative things on recall and Member accountability. We're spending some time analysing it and obviously we've already been advising teams here about how the recall system could work and some of the provisions of that. Were it to happen, we would then be into advising on how to actually implement such a system. So, I think it's one of the issues we will absorb, unless it actually needs a large public campaign, but that's unlikely, as Rhydian says.

That's something we'll be discussing in committee when you come to us. But it's a question as to whether you had figured this into the estimate, in this meeting. Peredur.

I'm just wondering, on the back of the Caerphilly by-election, there was lots in the media and obviously some of my constituents were coming to me saying about some of the misinformation that was being shared. They were going to the police and then they were going to the returning officer and then they were going to you, and they seemed to be doing that merry-go-round and not getting very far because everybody thinks it's somebody else's issue. Have you given any thought as to how you clarify who deals with some of that, especially in light of the Bill and looking at some of that deception element that's in the last part of that Bill, that communication and understanding of who's responsible? Because you have the rules, but the police do the investigation, and it gets very confusing for people who are aggrieved, that they're being misrepresented, especially as groups of people. So, I just wondered if you had any comments on that element.

This is probably a wider point in terms of how we react to misinformation, disinformation, abuse, intimidation, all of these things. We have had discussions with police forces. There are single points of contact within each police force that focus specifically on elections and election work, and we've started to talk to them about what we could potentially do together ahead of the Senedd elections next year, and it really boils down to two elements. The first element is a general one, ensuring that the police forces are able to support the management and administration of the elections—polling stations, counts and so on—and to be aware that it is a significant event that may need additional resource this time around, and we'll be having discussions with senior police officers at the start of next year.

But the second element—and more to your point, perhaps—is that we will want to produce, alongside the police forces in Wales, the College of Policing and ourselves, and potentially the management board, some additional guidance and resources—for candidates, to explain to them what the situation and the process is, and also slightly wider, to the public at large. I think we all want healthy debate in our democracy. That's really, really important, but it's about not crossing that line from healthy debate to intimidation, abuse, harassment, whatever it may be. So, there will be some messaging there for the public, and I think that could include who you could go to. We already say that in our guidance now to candidates, but I appreciate that the general public probably don't look at the Electoral Commission's guidance on a daily basis. So, we will be pushing that message out there. I don't know whether you want to mention anything from a UK-wide basis.

Well, I think we're all facing, and we're internationally facing, the same issues on misinformation. There are a number of things we can do. I'm not sure it's the complete answer though, certainly for the commission, because, fundamentally, nobody regulates campaigns or political free speech, and we certainly don't have a remit to do that.

I think the first part is trying to produce good information that people can trust, and get it out. The Welsh election information platform is going to be really important, so that people know—and we can point them always to—where the trustworthy information is, and they can check it. The second is work with the social media platforms. There's a lot of work under way with Ofcom, as they implement the Online Safety Act 2023. We're talking a lot to their advisory committee on mis and disinformation, with Lord Hallam chairing it. How can some of the social media companies act better in terms of taking down what is clear mis and disinformation, or abuse, intimidation and just threatening behaviour? Some do it well; some don't do it well, and there's a real issue there.

The third, and the one that is also worrying us in particular for this May's elections, is the incredibly rapid rise of deepfakes of politicians. Frankly, the barrier to entry to produce them has fallen so dramatically, anyone can do so. We've seen one of a Conservative politician pretending that he's defecting to another party recently. It was very convincing. So, we're looking at whether we can actually start to try to detect these things in real time during the May elections. It's not easy. Again, this is going to be, probably, a pilot we will try to run. But, if not, the voter effect is that voters don't trust anything they see, because someone may say it's a deepfake, somebody won't. There are technical ways to detect deepfakes now, and not just the obvious six fingers or crossed legs or whatever, but you can look at some of the images, the videos, and detect these. And even having something that makes this less of a space of complete impunity I think will help. But that is a worry, because we've seen this in other elections around the world, with deepfakes being used at short notice to try to influence voter behaviour and, actually, destroy voter trust. So, I think there are a number of things.

The final one is just overall education and digital literacy, which obviously Wales has done, but there is a lot more that we will need to do to educate not just young people but all members of the public on where to go to get good information, how to spot less good information and be appropriately sceptical of it. So, that seems to be an ongoing set of issues.

And just to remind you that, if you didn't hear the statement from the Counsel General on Tuesday, when she introduced the Bill, it's unlikely to be in law by the time of this election, but, of course, the principles are something you just highlighted that you want to ensure are enacted.

Okay, one final question from me then, and then I'll pass on to my colleagues. The estimate is, actually, I think, a 32 per cent increase from last year. Even the Senedd's own Commission budget is only a 21 per cent increase. So, percentage wise, that's quite a large number. You highlighted the fact that it seems that the campaign funding is in the first three weeks of the next financial year, but that's still a heck of an increase, when a lot of preparation would have been done for the Senedd elections this year, and the local authority elections are not new next year. Even the 16, 17-year-old voters, that's all been done previously. So, why is there such a dramatic increase in the budget from an estimate we saw, which was a lot less than even from last year's, and then we're going up again to 32 per cent this year? Why the increase?

Just to try and put this in context, the commission's corporate plan published last year indicated the profile of expenditure to support the Senedd over a five-year period. So, 2026-27 is very much a peak year, with significant increases, as you've highlighted, before reverting to around this year's cost level the year after, 2027-28, and then from 2028-29 to more of a steady state figure of about just under £2 million per annum plus inflation. When we look at the increase for 2026-27, as, I think, Vijay mentioned, about three quarters of that is very much in the event space, and Niki might want to talk a little bit more about the increase there. The balance of 25 per cent relates to our core costs, so just over a £200,000 increase. The main drivers of that increase are growth, general growth, in the organisation, so, scaling up as an organisation to meet the demands, the ambitions of the corporate plan, and increased investment in IT, and we can probably talk a little bit about the benefits of that. 

We'll come on to that, don't worry.

And thirdly, inflation, which is a relevant factor. Inflation is relatively high at the moment, but the large driver of the increase is absolutely in the event space. So, we've touched on that, but Niki, do you want to just pick up a bit more on that?

Yes. So, while there are no changes for the local elections in 2027, the electorate is still the same size, and we treat all elections with parity, so we want to work just as hard to ensure people understand how to participate in those elections. The difference in cost—. We won't be paying for a booklet, which we're doing this year to explain some of the changes that are happening for the Senedd elections, but actually we will still be doing all of the same campaign activity. We will still be having adverts on television, billboards and lots of campaign activity that reaches young people as well. It's just as important for the local elections, obviously, as it is for the Senedd elections. So, the campaign costs, if you remove the booklet costs, are comparable.

Okay, if you remove the booklet costs then they're comparable—I'd expect not necessarily a 32 per cent increase, a slightly smaller increase, as a consequence of that. And you've used 3.6 per cent inflation, I think, or is it 3.8 per cent, whereas the Office for Budget Responsibility and the Welsh Government have used 2.2 per cent in this budget. So, you're actually using the current inflation figures, rather than projected inflation figures.

The source of the figure is ONS, and the 3.6 per cent figure is already looking quite low compared to the most recent forecast. So, it's a number that is moving quite a lot at the moment. It's tending to trend upwards as well, but it's the best assumption we have, drawn from the ONS data at the point at which we produced the estimate. And it's consistent with the approach we use for inflation for Westminster and the Scottish Government as well.

Heledd, do you want to ask a question?

I can come in later. It was just on the no booklet for—

Well, it's been raised, Heledd. Do you want to ask it now?

Yes, I just wanted to follow up on that, because obviously 16 and 17-year-olds—. Is there a risk, if you're not preparing for any booklets and so on, that there may be an over-reliance on social media? There may be more trusted sources with the booklets. I was just wondering: who is your target audience with that booklet this year, and why do you deem that it's an unnecessary expenditure next year? 

So, the audience for the booklet is everyone. I realise that might not sound too targeted, but it goes to all households. And the reason we're doing it is because there are a lot of complicated changes to explain—things like the new voting system and how those votes are counted, which we're including in that booklet.

Those won't be necessary for the local elections. But we will still, as I say, be doing a wide campaign. Certainly, we will be using channels that we know young people use, so, things like TikTok and Snapchat and the various social media channels. But we will also be doing the usual mass advertising—so, as I've said, television, billboards and a wide range of advertising channels to reach everybody. And we are certainly, in 2026, trialling some new approaches. We are going to be advertising via Tesco and Co-op in-store. We're also looking at advertising through Gumtree, trying to reach home-movers, who we know tend to be under-registered.

So, we certainly won't be just relying on social media. We will have an eye on the entire advertising landscape to reach a mass audience, but primarily to reach those under-registered audiences. So, the budget is weighted 70 per cent towards under-registered voters, who do tend to be young people and recent home-movers.

Thank you, Chair. Perhaps we can return to this later when we get on to—

Your question.

Yes. Thank you.

Peredur.

Thank you. Just going on from your question, Chair, with the 32 per cent increase, you talked at the beginning about the Speaker's Committee investing in the Electoral Commission, and that element. How much of that is reflected in this budget, as in just passing the same metrics across, and how much of this is a specific Welsh element, if you like?

Just before, on that, when you talk about investing in the commission, I think it was a realisation of the centrality of the commission and of the landscape that we are living in and how important, really, public trust is. So, I think it was actually a reflection of the Westminster Government's desire to make us a force for good. Sorry, that was my point at the beginning.

It's good to have that context. But then is that passported across then to us, effectively a decision made in Westminster? Whilst I might well agree with it, it's just for us to understand the ins and outs of the budget.

In summary, our core and corporate plan costs are allocated on a population basis. Our event costs are determined on a direct basis, so, driven by those activities where the beneficiary is Wales.

And with your core costs, I think a 9 per cent increase in core costs compared to 2025-26, has been driven by the number of staff working in UK-wide issues, as you laid out in the corporate plan. So, are those costs, effectively, 5 per cent as a percentage split? Is that the methodology that you use?

Broadly, that's correct, yes. 

And then the Welsh spending element has increased by £9,000 or 22 per cent compared to 2025-26. You've noted that the investment in the Welsh office will be able to recruit two new members of staff to develop the work of the corporate plan. So, what additional investment is the commission seeking for the new Wales offices? How does that give us value for money for—? What do we get for our money, I suppose, for want of a better way of saying it? Can you explain what those two roles are? Are they two full-time equivalents based in Wales, or have they got UK-wide elements as well? Because I know sometimes certain roles are part-funded or fully funded for Welsh elements.

There will always be an aspect of a UK-wide element in my work and everyone else's work, but, specifically, they will be brought into—. We have began on the road towards a Welsh corporate plan, looking at the corporate themes that are in the UK-wide corporate plan. I think what those two members of staff will do, they'll focus very much on regulation, new regulatory measures and the administration of elections, linking in with political education as well, and begin that work that we will see included within the Welsh corporate plan this time next year.

We'll also be bringing a couple of temporary members of staff in to deal with bilingual public information queries ahead of both the Senedd election partly, but mainly for the local government elections too. But the two main new resources will be full-time members of staff based in the Wales team, working on regulatory and education matters.

But then, looking at support and improvement, that has gone down by 13 per cent. So, can you talk through what that means? Are those people—? Are they people or are they—? What's moving, in that sense, and will that have a negative impact on the support and improvement that you make? Or are you—? Or is it just people moving around? Could you unpack that for me a little bit?

I think it's more—. There will certainly not be a reduction in the support that we provide to local authorities, both post Senedd elections and running into the local government elections. If anything, that support will increase. So, I think it's just happened as a result of different changes and different movements within the commission. But maybe we can look into that in a bit more detail and write to the committee after today, just to give you a bit more reassurance on that.

Because what you're saying and what's in the numbers doesn't quite work. So, a clarification of that could help.

Of course. 

And then we were talking about—. You were developing a set of measures to evaluate whether increased investment in regulation offers value for money last year. Can you give us an update as to where we've got to with that?

In the corporate plan, we set out a number of value-for-money indicators in both KPIs. Those were: are we actually producing what we say we will, particularly in the regulation space, and are we giving the guidance in good time to people? We are tracking that. It's actually proved quite difficult to work out what is the actual cost and benefit of some of the investment in regulation. Where I think we've got a much clearer picture now is the investment in regulatory support. So, what we have scaled up significantly—and this is partly at the request of almost all political parties—has been help to train them, and volunteers and agents in particular, in how to navigate the labyrinthine complexity of electoral law, which is getting even more complex. So, we have scaled that up. We've produced quite a lot of material and—obviously, this is being done in Wales and in Scotland, as two the biggest elections coming—our training packages are going into how you obey the law and allow people not to trip over some of the very easy [Correction: 'tricky'] process issues, and how to actually evaluate, for example, notional or candidate or national expenditure. So, all of that we've seen be both effective and welcomed by the parties, and reduce, actually, the number of queries that we've got. So, I think we can actually draw quite a clear experiment of that that has worked.

On the overall guidance—and I might just come back to the point, Peredur, you raised on support and improvement—that's just a minor decrease, but it's a UK-wide decrease where we went through a bulge before the general election, of increased support and improvement, largely on training. It's just come down, and Wales is benefiting from an overall reduction, then. Guidance as a whole is staying static, and I think the Welsh share is about £40,000, because we're actually producing quite a lot of guidance in Wales—but probably more in Wales now because of the new system. But, again, that's just a pro-rata cut of the guidance function, which works across the UK for—

So, it shouldn't have a negative impact on what you're doing, because it's just a natural rebalancing of where it was before. 

It's a core function, that part, yes.

If I could just add, I think, and just building on what Rhydian was saying earlier, that the organisation, the commission, is very much trying to recruit the best people, irrespective of their location. And we've very recently based a number of staff in Wales, because they're the best people that we've been able to recruit. So, for example, a senior lawyer is joining the Cardiff office, who will focus on Welsh legislation. We've also recruited a couple of people joining my HR team in recruitment, who understand the local recruitment market there particularly well. So, we are seeing people based in the Cardiff office on a national basis, but, I think, with specific benefits for Wales.

With an understanding of what's going on in Wales, so it helps with that element.

Yes.

And you've talked about the corporate plan and, obviously, developing the corporate plan for Wales. We talked a lot last year about the timing of that and whether it should be aligned or not. Any further thoughts on that? Is it helping that the UK corporate plan now is in place, and the follow-on for a Welsh corporate plan, because that's what you implied last year, that that would be a logical way of doing it? The problem is it straddles elections. That's the complexity of the timing. 

Yes, and I think the work that we've undertaken on the UK corporate plan and the themes and the key activities included there has probably helped, and will probably help, in terms of the development of the Welsh corporate plan, which we'll submit to the committee this time next year. So, yes, I think we are in a pretty good place in terms of—. In an ideal world, they would all be streamlined and come together. Now the Senedd is moving to four-year terms, that might make life maybe even more complicated, depending on the timing of the UK parliamentary elections, for corporate planning. It's been a significant amount of work that the commission has undertaken over the course of the last 18 months, in the publication and preparation of the UK corporate plan, and I think that will bear fruit when we come to developing something for Wales. But, of course, the Welsh corporate plan will include what it needs to include from a Welsh perspective, as well, and what you, Members of the Senedd, and what any new Welsh Government will want to focus on in terms of activities, moving forward.

And finally for me, Chair, it looks as though the biggest increase is in campaigns and corporate identity, and that's nearly £600,000, 54 per cent in cash terms. That's a huge jump. Can you unpack that for me, for me to understand what it means, because I think it's down to corporate identity and it's down to the events, and that element? Maybe, Niki, do you want to give us a breakdown of what that all is included, and what it's paying for, effectively?

Yes. That increase is driven by the fact that the year is bookended by two sets of elections, which didn't happen last year, so we start the year with those costs that I mentioned in the first few weeks of April on the Senedd elections, and then we end the year with the run-in to our local government election campaigns. So, we have those two sets, and that's driving up the costs. We've got, as I said, the booklet for the Senedd elections, although those costs sit in this financial year rather than the next financial year. We're working also on various materials that we can share with voters. So, for instance, we produce things like animated videos to help explain the voting system slightly more simply, particularly for younger people, so hopefully that explains what's driving those increased costs.

And from a numbers point of view then, Chris, the 54 per cent there, or the £600,000, how much of that 32 per cent increase—? How significant is that, going back to the Chair's first question, really, on the difference between the 22 per cent and the 32 per cent?

It is a big factor in that difference. The total increase for the year is about £850,000, so it's a huge proportion.

Okay. Thank you.

A question on that—the last question. It is a large increase, and I understand your argument, as you highlighted earlier in the meeting, that there's a large element in the first three weeks of the next financial year and there'll be an element running up to the local authority elections towards the end of the financial year. Part of it is focused on increasing voter registration, voter involvement. How are you going to assess that, to ensure that it actually is delivering value for money, or that the money you're spending is actually delivering results?

Yes. So, the evaluation and the value for money is absolutely key when we're designing the campaign. We split our evaluation into three sets of metrics, I suppose. One is campaign metrics, so we look at things like impressions: how many times was an advert was displayed? How many people saw it or had the opportunity to see it? If it was an online advert, how many clicks did it receive? And all of that is tracked. Then we also measure awareness, so we run research with members of the public and we ask them if they are aware that they need to register to vote—did they see the campaign? So, we check campaign recognition. And then the third bucket of metrics is around voter behaviour: did the campaign actually change the way people behave? Did it lead to an increase in applications, and not just an increase in applications, but were those additions to the register rather than duplicates? So, those are the sorts of things that we're looking at for each campaign.

We look at a previous campaign before we start the next one, so we can learn all the lessons. Particularly with technology changing so quickly and new channels coming online, we're constantly looking at innovation and where we can try new channels to reach different people. So, as I've said, this year we're doing in-game advertising. We're doing Snapchat and TikTok. But we're also looking at things like the adverts within supermarkets that we think could help us reach a different set of people. So, we're constantly looking at how the campaign can evolve and how we can take advantage of changes in the advertising market.

Just one thing. As Niki says, we're trying to reach under-represented groups. Some of these are not reachable through mass media and some not even through some of the social media channels. So, we rely on our partners quite a lot to reach some of the really difficult to reach groups. Rhydian might want to say something about, particularly, some of the partnerships that we've developed here in Wales. But those are crucial for, say, the homeless, or some disability charities are incredibly good partners of us.

The social media world is moving very rapidly, and if it's helpful, once we've done the evaluation, we can say the number of impressions. But just to give a sense, from May 2025—we've finished evaluation of that—Snapchat was at 9.4 million impressions. So, that suddenly really worked for us. Amazon Prime did quite well, with 2 million. And we were using Zoopla to reach the new home movers, who quite often then fall off. And that was 6.3 million impressions that came just through the Zoopla channel. So, that's 17 million, and we managed to reach them.

One of the big debates that the board has regularly and needs to continue to have is what is the right balance between the very traditional things, like a paper booklet landing with a thud on everyone's doorstep, and these new channels, and actually where does the balance of value for money lie, to be honest. We try to keep these costs down, and I think that it's going to shift significantly, in particular if AVR comes in, because we won't need to run registration campaigns, but what then does actually reach people in the end? So, I think, over the next year or two, there will be a significant rethink of how we actually do our campaigning, still with the same aim of reaching everybody to tell them, 'This is how you participate in democracy'.

Heledd. 

Thank you. [Interruption.] Sorry.

Go on, Niki, you first and then I'll bring Heledd in. 

Just one final point on evaluation is that there is a lot of scrutiny internally, as you would imagine, because of the sums of money, but we are also using external best practice as well. So, we're very much looking at the Government communication service's approach to the evaluation cycle, and we've had the campaigns looked at by the National Audit Office to make sure that we really are achieving value for money. 

Heledd. 

Thank you very much. If I may pick up on the point in terms of impressions, though, because obviously there's a huge difference between impressions and the impact they have. So, how do you balance that evaluation in terms of understanding how effective it is, because yes you can say that there are impressions, but does that lead to someone registering to vote or voting? I just wondered how you evaluate that as opposed to a booklet as well, which is able to contain more information. Also, I note that, for 2026-27, it's a £13,000 increase in terms of digital communication, but also within that budget line it seems to be that materials for schools is included. There's talk about digitising the training for teachers. So, how do you balance all of those to ensure that it's the most effective way of not just knowing that someone may have seen something but actually that they've engaged with it?

That's a very good question, Heledd. One thing that I'd just like to say, before Niki comes in on the detail, is that, as Vijay has just said, this is something that the board keeps in mind all the time. So, exactly what is the relationship between an impression and an actual registration? And our stats on the actual registration and the spike that happens after a campaign show a very, very decent correlation. But, as well as that, the question for all of us, and for the board in particular is: at what point do you reach a cusp where you think that a booklet has gone past its sell-by and that the heritage media, as I might call it, is no longer what's required and whether we are now entering a digital-only phase, where everybody over—? I think that there will come a time when people who are older than me will be digitised and that the booklet may no longer be necessary, but we haven't reached there yet. It is something that we keep in mind.

I'm asking specifically about under-represented groups, and specifically young people. We go into schools a lot and, actually, young people are desperate for more information. They don't feel that they have enough information. They understand that they have the vote, and some may have registered to vote, but they tell us that they want more information and, often, they ask us for booklets, because they trust that source because they're bombarded online. So, I just wanted to explore if that's something that you're working on with schools, with young people, and also with elected representatives to understand what voters are telling us that they'd like to know as well, to be able to vote.

Absolutely. We have been providing and publishing education resource and education material since before the 2021 elections, both for educators and for young people themselves, and we've very recently created a new suite of education resources, working alongside the Senedd Commission and the children's commissioner. I think that a copy of those resources has been passed on to this committee. That's a pilot at the moment that we're working on, and I think that there are 26 schools involved in that across Wales, and we'll action any feedback that we receive from them by the end of the calendar year. We also have a youth voice partner in Wales, Children in Wales, and we work very closely with them on activities, and really to use the young people that are involved with Children in Wales as a test, if you like, for our material and our messaging, and then hoping that they can go out and talk to their peers.

I think after the 2026 elections, what we are intending to do—. Much of this comes down to political education and citizenship education and the future of that area of work in Wales. What we will do after the 2026 elections is convene leaders in the education sphere in Wales to talk to them about the future of citizenship education and what we can do ahead of the 2027 elections. What we are hearing anecdotally from schools—. We set up schools groups both at a primary level and a secondary level, because there is a lot of resource available both from the commission and other organisations.

The issue is the resource available within schools themselves, i.e. time for teachers to be able to talk about these things, and secondly, maybe critically, the confidence amongst educators to talk about these things. There is a lot of nervousness and anxiety out there about talking about politics these days. What we want to focus on is our democracy—not politics with a capital 'p', but politics with a small 'p'. It's talking about our democracy, how you can take part, what the Senedd does, what the UK Parliament does, and leaving party politics to people like you and candidates. So, there is a job of work there in inspiring further confidence within the teaching community, and in finding time as well, and we have been having discussions with the Welsh Government on that.

To the question about how we measure the impact of some of this, it is very difficult to draw a direct line to whether or not somebody turned up and voted. As I said, we do measure whether or not people register to vote. In the four weeks to the general election in 2024, we saw over 117,000 people in Wales register to vote, and that was while our campaign was running.

In terms of the booklet, what we will be doing is carrying out public awareness research, so we will be talking to people who received that booklet and asking them did they read it, did it help them understand, did it give them the information that they needed in order to participate in those elections. That is definitely something that we will be evaluating closely.

Heledd, do you have any further questions on that particular point?

The only thing I had is that, obviously, you'll be evaluating, but you've made a decision that you're not budgeting for a booklet prior to the local authority elections in 2027. I just wanted to understand how that evaluation might help shape your budget, or if there will be a need to revisit or reprioritise the budget following that evaluation.

We do have additional information in the budget for next year for additional activity. It's true that it's not for a booklet; it is for things like materials that can be given out to some of the partners that we work with. But if we find that some of the information that was in the booklet was particularly useful to people, we would look to replicate that sort of information potentially online, which would be a much more cost-effective way of doing it, because it wouldn't involve the printing and the postage, which of course is a really major element of the cost of the booklet.

One of the things that we've had to home in on this year, of course, has been the actual ballot paper, which is going to look very different. And so you've got to give a sample model of that ballot paper to everybody within a booklet, and within our website.

A copy of the ballot paper will be included in the booklet, alongside information on registration and what the Senedd does, what Members of the Senedd do and how you can cast your vote. We can share draft content of that booklet with Members ahead of time if that would be easier.

But it's important to note that it's not just the booklet. There will be accompanying resources that we will be using. There'll be an 'easy guide to voting' video, there'll be other online resources that we'll be using and providing to partners across Wales, and some of them will be specifically tailored for their audiences, so they can use the messaging that's included within the booklet in different forums and in different ways with their key audiences.

Can I ask one additional question, though? Hopefully, this campaign is going to be hugely successful, everyone is going to understand the new constituencies, the new way of voting and so on, but then you have the challenge for the 2027 elections that it is, again, a different voting system. So, in terms of then explaining the changes—. I just want the reassurance, really, that you have the budget in mind for communicating all elections, because of the differences there. There's obviously a focus for the Senedd election, but in terms of that budget scrutiny, I just wanted to know that that's something that you're considering as well.

Can I just come in, Heledd? Thank you. Yes, we do, and that is actually why I think, as Peredur pointed out, next year's budget is a spike; it's because we've got both these elections running, and we're going to spend roughly the same amount on the local elections in 2027, partly in order to make sure that people know there's an election coming, partly to explain the different systems that will be run, but also partly to just explain things like you won't need to bring voter ID, and we have to keep on being clear on some of the differences, the diversity of different electoral experiences across the UK. We have to keep reminding people so that they know what they actually need to do. I think that diversity is a real strength, and experimentation like AVR is great, but we have to keep on making sure that people understand the election that's coming and what they actually need to do. And that takes last-minute—. You know, frankly, people's attention turns to it in the weeks before the registration deadline and on election day. That's when their attention is focused.

Okay. Hannah.

Thanks, Chair. Apologies, as is always the case, time is being eaten away—

Keep going. Don't worry.

—which is not unusual. But if I can, I'm just going to turn to some indirect costs in your 2026-27 estimate, with the resource allocation increased by 9 per cent due to the additional IT roles, investment in IT and cyber security. I know you've already touched on that earlier, around the increasing threat of deepfakes and misinformation, and also around your work to address gaps in capacity and capability in things like finance, procurement, human resources. So, notwithstanding what you already said about that kind of investment in IT, I wonder if you could expand on how that investment will ensure that you're in a strong position, not just to tackle potential hacking incidents but how that would work in practice in terms of the things you've referenced around the increasing concern around those deepfakes and the spread of misinformation, and how that could influence forthcoming elections.

I'd be delighted to, and thank you. First of all, thank you to the committee for funding us in past years to actually make an investment in IT for particularly recovery from the cyber attacks in 2021-22, which are still in everyone's minds.

We have changed significantly and vastly improved our cyber security and essentially shut down all our legacy servers, moved everything into the cloud, put up multiple different walls of security. We've got external accreditation on the cyber security now. So, we were aiming for Cyber Essentials, which is the kind of core certification. We passed that, we've now passed Cyber Essentials Plus fully, 100 per cent, which means that we've got the top level of accreditation. It doesn't mean it's perfect, and I'm afraid it has to keep on going. We are still coming under enormous numbers of cyber attacks. I think on election day last year, we had over 60,000 cyber attacks just on that day. Two really major denial-of-service attacks came in just before that. And that just does not stop.

I think—and I'm afraid it's part of the new normal—all our political parties and candidates and the Electoral Commission, anyone involved in this, should expect to be cyber attacked quite frequently. So, we're expanding our work with political parties to help on their cyber defence as well, because they have access to the electoral register and are crucial. The way that our chair, John Pullinger, has phrased this is that the electoral system, the democratic system, is critical national infrastructure and needs to be protected as such, so that's how we're approaching it.

It also enables us to do some really important things, and I think I'll just link back to the digital ID. This is potentially going to be quite game-changing for registration and how that will work. But we, digitally, are aiming to put more information online for voters. At the moment, Democracy Club wonderfully run a polling station finder, but we're aiming to build on that and our new IT enables us to do that.

We're also starting, cautiously, but we are actually getting under way with this, some AI-enabled projects that should enable us to make some savings. Just to give you one example, some political parties, the smaller ones, give us receipts on paper for their spend. Someone has to sit down, enter that, classify it—that is a perfect task for an AI, and we're testing out whether that can be done. That will just be a straight good efficiency saving if we can have our own in-house. It's something that other parts of the public sector are already using, like the Office for National Statistics. So, there are a number of really specific use cases. And just to be clear, we are not going to be using AI, certainly not in the short term, to produce guidance or answer questions, because we don't trust that enough to do so. You would need quite a lot of testing. But some of the incoming information we can, and our IT will help us do that now.

You just foresaw my follow-up question about the application of AI, because, obviously, it is a very broad thing and needs lots of things that can have different concepts. It's reassuring to say that you're not quite where you'd want to—. It's not been tested enough for that. Can you just come back to the second part of my question around, I think, previously, addressing gaps in capacity, particularly in departments such as finance, procurement and human resources? I'm looking at you now, Chris. 

I'm really happy to talk about that. Since I joined the commission about a year ago, it's been a huge focus for me. We have invested in finance and procurement, for example, and HR, and I think we're really seeing a great impact. As part of the run-up to the Speaker's committee in Westminster last year, the National Audit Office carried out an independent review as to how we were doing in terms of building capability, and they were extremely positive about that, so we've got some independent assurance.

I mentioned earlier that our forecasting performance is improving, and that is both marked and also measurable. But I think it's been a really good investment in terms of the quantum of the work that we've had to carry out. Over the last 12 months or so, the procurement function has had to manage over 150 live procurements. At the moment, we've got about £50 million-worth of contract value under procurement, so we need very robust and professional functions to manage such a significant sum for the commission.

If I can just add to that, having a good and really well-functioning procurement function has also meant some significant savings have been achieved. We've had some good, robust commercial negotiations that have resulted in several hundred thousand pounds of sheer savings, just when you go with initial bids, comparable market bids, and then actually work out what we can do in the end.

I don't think our contract management over life cycle was very good in the past, and it's a lot better now, particularly when we've got some of the really big contracts, the ones that Niki's referred to, in communications, advertising, spend, our IT contracts as well, and some of our leases. They're are all now much more tightly managed, and that is actually giving proper bottom-line straight financial dividend as well. 

Before we move off IT, can I ask one question? I appreciate we've had the cyber security issues in the past, and the increased cost to protect those systems. You just mentioned you're looking at more ways in which you can communicate, with social media and everything else. Technically, therefore, you're opening more doors to your systems. So, are you confident that the money you're spending is also going to be able to protect, as you go on in the future, and open those doors—that the systems will be protected and, therefore, the funding, actually, you're putting in place will be effective? 

Thanks, David. That's an extremely good question. That is very high on our—. As we start each of these projects, we obviously set a project initiation document and say what needs to be done. One of the crucial parts—. For example, if we expand, and we probably are going to expand, our work on TikTok, because that's where voters are, and we need to explain this to wherever voters are, we'll need to have very strict rules as to what systems it can run on, and make sure that, for example, we don't put any personal data onto those systems, and make sure that all our data is tagged.

This has been another enormous piece of work, which is almost done, but it's been going on over the last year. It's a full-scale data tagging exercise. We have a lot of metadata, so we know when something is, say, personal data, we know what it's doing, we know how it can be done, which is the essential underpinning for security—only some people with some roles or some systems can access the tag data. It's also incredibly important for any further use of AI, because, again, you can really restrict it. But, yes, we will put some strict security rules around the internal data. The personal data clearly gets treated more sensitively.

Just to say, we're looking hard at what's going to be a complicated but even more protected set of data, which will be children's data, because if we are dealing with 16-year-olds on the register in England and in Northern Ireland—we already have that here—how do you make sure that the 14 and 15-year-old attainer data, particularly in the context of AVR, is properly protected? So, we're talking to the Information Commissioner's Office a lot about exactly the data issues there. 

Sorry, Hannah.

Thanks, Chair. The accounts depreciation has increased by 22 per cent. Can I ask what you mean by that being driven by adverse effects of attribution based on population, and also what assets have you acquired to account for that increase as well?

I'm happy to talk about that. The main two drivers of the increase in depreciation are buildings and IT. We spoke a little bit about IT, but, as you take out new leases or extend your leases, that drives capital spend and therefore depreciation. So, we're looking at London, but we're also in the process of moving to new accommodation in Edinburgh and also expanding our Belfast office, so all of that will increase your depreciation.

We've also recently invested in a stock of new laptops, so, modern hardware. One of the key features of our new laptop stock actually is very robust security features, so, just following up on the previous conversation, you need good hardware to maintain security. Actually, we've just had our headline results from our annual staff survey, and our staff satisfaction in relation to their IT kit has increased by over 30 per cent, from a rather low 47 per cent to a probably public-sector-leading 78 per cent, so we're fantastically pleased about that.

You've just highlighted depreciation, Edinburgh, London, Northern Ireland—what proportions of that, therefore, are allocated to the Welsh budget? 

Again, it's on a proportionate basis, so our depreciation charges are allocated on a population basis.

So we're covering depreciation for buildings in other parts of the UK—an element of it.

That is the implication of the way that those costs are allocated, yes.

Thanks, Chair. 

If I could, does that then, the IFRS stuff, is it just on a proportional basis rather than—? 

It is. It is.

So, you don't differentiate your office in Cardiff compared to Edinburgh, or—?

No. It's probably worth just adding that, as we expand on more of a UK-wide basis, and we spoke earlier about expansion in more staff in the Cardiff office, but also expanding in Edinburgh and Belfast, those are much cheaper locations to base staff than London. So, overall, our lease costs should reduce as a result of that on a per capita basis. So, all of that will flow through on a UK-wide basis, ultimately.

And of course Cardiff has been the beneficiary of that system, two or three years ago—

With the new offices, yes.

Thanks, Chair. You've led me quite nicely to my next question, because you mentioned staff and IT, and I've got a question around staffing. You've included in your accounts a budget of £60,000 for the 2026-27 pay award, which is a 35 per cent increase compared to 2025-26, and it says it's based on 3.6 per cent inflation and a provision of 1.75 per cent associated with the pay review implications. Now, the Welsh Government has budgeted for a 2.2 per cent uplift for the public sector pay element, so the committee would like to know how you've estimated that 3.6 per cent inflation when calculating the pay award.

We spoke a little bit earlier about the basis for the 3.6 per cent inflation. That is our standard approach and source for information—the ONS forecast at the most recent point in time. Forecast inflation is actually increasing, so that 3.6 per cent looks potentially quite low.

The 1.75 per cent is an assumption as we move into the first year of operating pay progression and performance-related pay. I think we've spoken to this committee about that. We've had long-standing issues with our pay regime in the organisation and carried out a complete overhaul to a much simpler system that allows for progression through a smaller number of gradings. The way that the system works—and Vijay may want to talk about this as well—is, over a number of years—and this is a model that has been already in place in a number of departments—it becomes self-funding, and that is because, as people leave or are promoted, they leave at a higher cost, and then that post is filled by somebody coming in at the bottom of the grade. So, in terms of the overall pay bill, over a number of years it effectively becomes self-funding. I don't know if you want to add to that, Vijay.

Yes, I'm very happy to. Overall, we follow, first of all, the inflation figures, as Chris has said, but we follow the overall civil service pay remit as well, and we've stuck to that this year as well. So, our system is very, very broadly akin now and much more similar to the overall UK civil service pay system. There'll be a base increase for people who are performing okay, and then some people will get a higher pay increment, but it should move to be completely self-funding within that, like most of the UK Government departments, and no doubt Welsh Government departments are as well over time. But it's now, at last, a very similar system, with a proper performance management system built into that.

And just to come back on your initial question on depreciation, the other point just to put on the record is that these estimates do not contain any capital, because the capital is entirely funded by Westminster, but that's why the depreciation falls here—in a sense, it's two parts of the same coin. The capital comes from there, but then there's a depreciation charge on the capital asset, which is what we then record in these accounts.

You said that there's a higher increase for some, so is that based on performance-related pay?

It's based on an—. So, it's not based on an inflationary rise. That's the same as the civil service as a whole. It's based on moving the system from one that has no performance-related pay in it to one that has performance-related pay, and every year that amount is going to go down, to the point where it becomes completely self-funding as we get into that model.

So, how is the pay award distributed across the different pay grades, then?

The same system applies across all the different pay grades. Everybody in the commission is now on the same system, and everyone will have the same performance management system applied to them.

Before we go on, we're getting close to 11 o'clock and we have a few more questions. Are you free to stay for about 10 minutes or so beyond that?

Perfectly, yes. 

Okay, let's keep going then. On the pay issue, you said you thought 3.6 per cent was a little bit low. We've had this question before on pay. Do you have contingency within your budget, an estimate, to actually go higher if that figure is incorrect? Because last time you came back with a supplementary to us.

So, in that circumstance—. Actually, that was rather similar to what happened this year when we were setting the pay award, where inflation was somewhat higher than our budgetary assumption. We would, as a first call, attempt to absorb that in-house. So, we would look at the overall budget, particularly the overall pay budget, to see if we could absorb it. I would say that the potential for us coming back to seek additional funding if inflation moves somewhat would be rather low. If it moved hugely, then we'll need to look at our ability to absorb it, but that is what we would do as a matter of course.

Peredur. 

In the Finance Committee, we heard from the Cabinet Secretary that the Welsh Government believe that inflation next year will trend towards 2 per cent—OBR forecasts, and that aspect. You're saying inflation will go up next year.

That was the most recent forecast.

This is a constant conversation in the Finance Committee. It's apples and oranges when it comes to what inflation are you looking at. So, is that based on CPI or is that based on—?

CPI H we use, yes. 

Right, okay, so it's a different measure. Because it gives a very different outcome when you look at that. Is that an appropriate measure to use?

It's the one that we've discussed with the Speaker's Committee and agreed it's appropriate. We also speak to our peers as well to ensure that we're using the same assumption, and we do. So, we consider it to be appropriate, but, clearly, it's something that can be debated, for sure.

Yes, okay. 

Hannah. 

Thanks, Chair. I'll try and hurry through this, because I know there are other questions after mine as well, and we are, as you said, pressed for time. I just want to turn to parts of the UK corporate plan, but also, in correspondence with the Cabinet Secretary, she noted the disparity between the estimate presented to the committee of £3.6 million and the projection of the Electoral Commission's five-year plan for Wales of £1.79 million, and the projected figure—I'm making sure I've got all my figures right—and the projected figure of £3.9 million set out in the UK corporate plan for 2025-30. So, how will you assess delivery against your current corporate plan for Wales and how it will feed into the next plan? And can you explain why there is this disparity there?

I'm happy to start with that. I think the simple answer is that the Wales corporate plan was produced about five years ago, so that number is a very old number now. The scope and direction of the commission's work has changed significantly, and we did and might want to talk about that in relation to Wales. And our recently published corporate plan, I think, is genuinely an ambitious and transformational plan as well. So, the simple answer is that the world has moved on in five years significantly, and our corporate plan figure is a figure that is a reasonable estimate for the work that we want to do and need to do.

Only to say, you know, I've mentioned already that the current UK corporate plan is an ambitious one, as we've discussed with you previously, and we would want the Welsh corporate plan to be similarly ambitious, and also based on the conversations that we would need to have with Welsh Government, and looking at policies that they would want to implement in terms of constitutional reform, moving forward. So, things have developed exponentially since five years ago, when we put together that initial five-year plan.

So, how do you see the UK corporate plan benefiting your activities in Wales?

Maybe I can come in on that. To quite an extent—. So, some parts are separate. Let me just give you one example: the big scale-up of the education work is being entirely funded by Westminster, because it's on the introduction of votes at 16 in England and in Northern Ireland. However, all that material will, clearly, be used across the UK, and we can use it for all the things that Rhydian's already been setting out in terms of education in schools across Wales. We translate it—everything here in Wales, we will translate all of that. All the videos and so on I think we can use extensively. But because it's—. It's pinned on votes at 16 in England and Northern Ireland.

More broadly, our capabilities and the work we're going to do on some of the threats to democracy, we're certainly using the work we've done on that, and will continue, for elections in Wales. So, for example, thinking about deepfakes, again, that's just part of, I think, core business for us now. We've touched on mis and disinformation, Chair, already. And then, partly looking at the significant amount of work that we have under way on the abuse and intimidation of politicians and candidate safety—. I think, after the general election across the whole UK, including in Wales, we saw a real uptick in some horrendous treatment of candidates, and we got that from the candidate survey, we got that from the police, we've had that very directly from political parties, who shared with us really openly all the examples of what happened on the campaign trail, at hustings, at counts. There's a lot of work to do on that. We've touched on some of the work with the police, but I think there is a significant amount of work, some of it set out very clearly by the Speakers' conference in Westminster, which reported very recently, which now needs to be done to try to tackle this. So, all of that is very much core to our corporate plan, because we see that—people not standing to be candidates or not wanting to be in political parties is a pretty major problem. So, we will be working, but with a lot of other people. It's not purely in our remit; there are a lot of others who will need to work with us, and I think we're aiming to use a lot of this for 2026, certainly 2027.

Just one final one from me, Chair, because I think a lot of it has been covered, and it's just to ask how you will ensure that this committee is engaged with the development of the next corporate plan for Wales. 

Rhydian, you may want to talk about the exact timetable. I think, essentially, we will come and consult all of you informally soon after the Senedd election has happened, and then you'll obviously see iterations of the plan. We'll probably, actually, want to write, I think, to set out some of the underlying themes and just test that that is indeed what your successor committee wants to see in it, and then there'll be a formal process. But do you want to—?

I think that's right, Vijay. We would probably have discussions with you prior to the Senedd election as well. We are starting to think about what could potentially make it into the corporate plan this side of the election, so we'd want to have some exploratory discussions with yourselves relating to that. But the real work then will kick in, clearly, after May, when we have a new Welsh Government and a new Senedd and all of the various potential policy changes that will come with that.

Okay. I'm going to leave the last couple of questions to Heledd.

I think the partnership work that Rhydian has outlined is particularly effective in reaching under-registered groups. In terms of the advertising campaign and the other elements of our communications activity, I don't see us needing, necessarily, a bigger budget to reach under-registered groups because, as you mentioned, we already weight 70 per cent of it towards those groups. It's more about paying very close attention to the channels that people are using, the channels that are most effective, and reviewing that on an annual basis—so, swapping out channels that aren't particularly effective for those that work better. And that's why I've mentioned some of the innovation that we've been looking at. One of the things that we trialled recently was using things like digital speakers and the Alexa technology for people who, perhaps, have accessibility needs that are better met by an advert that they can interact with vocally. So, I don't think that it's a matter of layering further elements of the campaign on top of what we already have; it's about swapping out less effective channels for more effective channels. 

If I could just add one very brief thing, just to link this back to the discussion we've already had on automatic voter registration, we are focusing on under-registered groups, and with automatic voter registration, depending on the data source and data quality, who is then under-represented is going to change significantly. We have good data on children, under-16s—they've been given national insurance numbers, and so on. They are currently very under-represented, hence that emphasis that we have at the moment, but that may change. So, I think the really hard evaluation—and, Heledd, you're absolutely right to refer to the evaluation of this—the research that we will need to do is to work out, over time, who then is under-represented, who do we need to reach, if anyone. I doubt it's going to be a perfect system, but it should be an improvement. The commission continues to say that there are between 5 million and 8 million people in the UK not on the electoral register who should be. This may be a system that really helps that, but then we'll need to look at who it is who doesn't—. There are other barriers to participation in democracy, and I think that that's very much the focus of the corporate plan.

Thank you. We've now gone beyond that 10 minutes, so I think it's time now to say 'thank you' and call this session to an end. You will receive a copy of the transcript. Please, if you identify any inaccuracies, let the clerking team know so that we can correct the record. So, once again, thank you for your time today, and thank you for the extra 10 minutes as well. It is important, as you appreciate, for us to understand the value-for-money agenda and to ensure that the money that we either approve or disapprove of will be spent in the best way possible for the people of Wales, and therefore it's good to have this discussion as to how we see it going forward. 

Could I say that this may be the last time that we meet formally, perhaps, before the Senedd election, unless there are terrible budgetary issues coming up?

Unless you have a supplementary budget, possibly, yes. [Laughter.]

And I'd like to thank you on behalf of the commission for the support and the interest that you have taken, and thank the committee generally. You've been fair and tough, but you have also been very supportive. Thank you very much.

You haven't seen our letter yet. [Laughter.] Thank you, Elan. We wish you the best. Obviously, there's a very busy time ahead of us, and you, to make sure that the election goes smoothly, but also in how we assess the election afterwards and what worked, what didn't work, how did things pan out and was our money spent wisely. And that's something to be done in the next Senedd. So, thank you very much for your time. 

For Members, therefore, I'd like to propose, in accordance with Standing Order 17.42(vi), that we resolve to exclude the public for the remainder of the meeting. Are Members content to do so? I see that everyone is nodding and saying 'yes'. Therefore, we will now move into private session.