Good morning, everybody. Welcome to the Housing, Communities and Local Government Select Committee. I am Florence Eshalomi, the Chair of the Committee. Can I ask Committee members to introduce themselves, please?

I am Maya Ellis. I am the MP for Ribble Valley.

I am Andrew Cooper, Member of Parliament for Mid Cheshire.

I am Sean Woodcock, MP for Banbury.

Good morning. I am Andrew Lewin, MP for Welwyn Hatfield.

I am Sarah Smith, MP for Hyndburn.

I am Will Forster, MP for Woking.

Good morning. I am Lee Dillon, MP for Newbury.

I am Lewis Cocking, MP for Broxbourne.

I am Gagan Mohindra, MP for South West Hertfordshire.

Could I ask our guests to introduce themselves, please?

I am Alan Renwick. I am a professor of democratic politics and deputy director of the constitution unit at University College London.

I am Maria Sobolewska. I am a professor of political science at the University of Manchester.

I am Toby James, professor of politics and public policy at the University of East Anglia and co-director of the Electoral Integrity Project.

Thank you for joining us this morning. This is the first session of our inquiry into modernising elections. Elections are always in the news, and, as politicians, we cannot get enough of elections.

I will start off with a general question. How healthy would you say democracy is in the UK at the moment? It is a nice easy one.

As academics, we often like to spend a lot of time defining democracy. Democracy involves lots of aspects. Democracy is more than just elections. Democracy requires constitutional rights and protections. A real democracy also involves people being able to have good living standards and access to healthcare and education. Focusing just on elections, because the inquiry is looking at modernising elections, the Electoral Integrity Project evaluates elections around the world. We produce rankings and league tables in terms of election quality. The UK is ranked 24th out of 39 countries in Europe, so in the bottom half of countries in Europe, and 53rd out of 170 countries worldwide, so that is fairly good. Clearly, there is scope for being a lot better. The areas of strength in elections are the professionalism and independence of electoral officials, who you will hear from today. There is really strong integrity in terms of processes, results, the counting processes and so on. The weaknesses, in brief, according to the indexes, cover many areas, the electoral system being one of those. It is a disproportional system in an era of multi-party politics. There are problems with voter registration, with nearly 7 million or 8 million people either missing from the electoral register or not completely registered. There are low levels of participation at elections, with difficult voter ID laws. The role of money is also an ongoing challenge in elections.

My specialism is more looking at voters and what voters think. From the point of view of British voters, we have a relatively well-working democracy, but there are increasingly issues and voters are increasingly dissatisfied. I would mirror what Toby just said, which is that the feeling of not getting the choice that you voted for in elections is increasingly present. We are definitely looking at the breaches of the system and the system not working in terms of translating how the voters voted into who represents them in Parliament. That is a huge issue, but I know that this is outside the scope of your Committee. The one that is inside the scope—again, Toby already mentioned registration—is that we have one of the hardest systems of registering voters in the world, and this really needs to stop. It is such an easy thing to remove. It is an administrative obstacle. It is not really needed. There are many countries that show us how it can be done much more easily. If we were to change one thing, that is the one I would have gone for.

That is a crazy statement to make.

Is it a crazy statement?

“It is very difficult to register to vote in the UK”. It is free; it is easy.

I said that it is much more difficult than in other countries, yes.

I agree with everything that has been said.

Which one? Do you agree with Lewis or Maria?

I agree with what has been said by my colleagues. The UK has a very long-standing tradition of democracy and has maintained healthy, free and fair elections, but there are significant difficulties in, particularly, the electoral aspects of democracy today. As Toby and Maria have both said, participation in elections has fallen. Satisfaction with the democratic process, particularly, as Maria said, with being able to make a choice that people are happy with, has declined significantly. That is one point around public concerns with the state of democracy. Secondly, we see significant problems in the functioning of aspects of the system, notably, as Toby said, the role of money in politics, which is a matter of significant public concern, but also misinformation and disinformation in politics and widespread problems with harassment and intimidation of candidates and others involved in the electoral process. Thirdly, looking around the world, we see significant examples of democracy being undermined, or at least attempts being made to undermine democracy, by international and domestic actors. The resilience of the UK electoral system to those kinds of challenges is something that needs to be thought about very seriously. Fourthly, there are significant questions around the core of the electoral system. I am not someone who has ever taken a strong view on what the electoral system should be. There are advantages and disadvantages with all sorts of different electoral systems. First past the post, proportional systems and other systems all have good arguments in their favour as well as against. We risk, given the changes in the party system, having an election next time round that is simply indefensible in terms of the degree to which individual candidates in constituencies clearly have the support of their constituents and also the degree to which the Government that are elected have wide support around the country. At the very least, we can no longer presume that the case for first past the post is clear. That would merit further investigation.

You have all touched on the electoral system. Briefly, what do you think the Government should do in response to the concerns that you have all flagged up?

It would be a major reform. When you have major electoral reforms, there are often disagreements and it can lead to discontentedness. Some of the research that the Electoral Integrity Project has done with Westminster Foundation for Democracy is to think about how you do electoral reform. The key mechanism for doing that is to do that on a cross-party basis and to have a specified group with a remit to look at options, explore the evidence and listen to different perspectives. A specialised committee would be the way forward.

The Government have suggested a commission. Would it be something like that, or do you have something else in mind?

I would add to what Toby has said. The cross-party element is very important in process, but also engaging members of the public is absolutely crucial. There has been a proposal put forward by the APPG on Fair Elections for a national commission on the electoral system. As I said, I do not take a view on what the electoral system should be, but this is a debate that we increasingly need to have and it needs to be a conversation that engages the public, not just those who are directly involved in elections.

The Government’s election strategy is titled “Restoring trust in our democracy”. Notwithstanding what you have all just said, does the Representation of the People Bill meet that ambition?

It offers many steps that go in the right direction, in terms of improving access to virtual registration, clamping down on some problematic aspects of the role of money in politics and taking some measures, at least, to address harassment, disinformation and misinformation. In all these areas it could go further and I would very much support many of the calls that have been made by many different organisations to go further. To take one example from those various different areas, in the area of the role of money in politics, there are some measures in the Bill. Clearly there are more measures coming as a result of the Rycroft review, which are very welcome. There are some areas that still have not been addressed that the Rycroft review hints at but does not make recommendations on. It is indefensible that it is possible to donate any amount of money to political causes. Equality is a fundamental principle of democracy. I cannot think of any conceivable defence of the position that someone can donate £10 million, £15 million or £20 million, or whatever it may be, whether they are living in the UK or not.

How would you expect elections and political parties to be funded? There is no public appetite for public money to be spent on political parties.

A number of people have suggested quite a high limit in terms of donations—say around £1 million—at least as a starting point, in order to establish the principle that it should not be free for people to donate any amount.

Would that apply to entities such as unions as well, or is it just individuals?

These sorts of details need to be worked through. That is a stumbling block that has led to indecision on this matter for a very long time. There are differences between donations by corporate entities that are controlled, essentially, by one person and entities that bring together many people. If it is the case that donations through trade unions are done on an opt-in basis, there is a good case for saying that those are treated differently. If they are done on an opt-out basis, there is not a case for saying that they should be treated differently, in my view.

In the spirit of clarity and transparency I will declare that I was funded by a trade union for my last election, and many of my Labour colleagues on the Committee are probably supported by labour unions as well.

Picking up on this thought of where the generally well-meaning legislation does not go far enough, I will pick, as you have probably realised, my favourite topic of registration. This is to do with the extension of the franchise to the 16 to 17-year-olds. This is a great example of a missed opportunity. These young voters are one of the least well-registered groups and this is an area where we have actually made things worse by moving to individual registration. Under the old system of household registration, often these were the kinds of voters who were registered by their parents. They have had this sudden change in how they register, but, because they are also less likely to know about or be interested in politics, they do not quite have the will to go and register themselves or inform themselves to vote. Marrying extension of franchise with some form of automatic registration of this group would have been a great opportunity to improve participation in elections. This is not an unreasonable demand, mostly because these voters, unlike the other voters in this country, are within the reach of the state in a way that almost no other group of voters is. These are predominantly people who are still in formal education and generally in touch with various other formal systems, because they are not released until they are 18 from all these kinds of state-based systems. We could marry registration of these voters with giving them national insurance numbers, for example, or opt for more of a University of Sheffield-style system of prompting schools to get them registered as they go through the relevant point in their lives and relevant point in the school year. This is the kind of thing that would have been relatively cheap and easy and that legislation really should have included, but an opportunity was missed on that.

Would you not be targeting certain aspects of the electorate differently to others? I am a Conservative, but there is a historic understanding that maybe younger voters may vote for non-Conservative parties. Therefore, in effect, you are gerrymandering.

How could you say such a thing?

I agree. I think all people should vote, at whatever age, for the Conservatives.

That is when they are enlightened.

More seriously, is there not a concern that, if you are practically targeting or auto-enrolling certain aspects of the electorate, that may benefit some political parties?

It is very important to delineate the automatic, automated or Government-assisted registration of voters from the compulsion to vote. These voters will be registered and enabled to make their political choice, inform themselves and become interested in the campaign, but they still can make a choice not to vote. We are not advocating for Australia-style compulsory voting systems. It is just removing a considerable administrative obstacle. You have to remember that to register not only do you have to go on the website yourself, but you have to have your national insurance number to hand. You often then have to have photo ID to vote as well. I would advocate that the national insurance number should become a form of photo ID as well, although this Act also is trying to extend the number of IDs that we can use. Generally speaking, those kinds of voters are being assisted more than other voters because they are more disadvantaged. They are less likely to have a photo ID of the sort that is accepted. They are less likely to have a bank card if they are 17 or 16, if the legislation goes through in the current form, and much less likely to be able to access that kind of online service and register themselves. We are only temporarily assisting them. In the long run, I advocate that all voters should be somehow assisted or at least prompted by the state to register.

I remain nervous about targeting certain groups over others. In my view, it should be all or nothing, but I am sure there will be further discussion on that.

I would share that nervousness, to be honest. We need to be very careful of moving to an arrangement where some people have automated registration and other people do not.

We are going to come to that.

We did a study of 159 countries, looking at the voter registration system in place. It shows, similar to research that Maria and colleagues have done, that it improves both the completeness and the accuracy. There is some discussion in the parliamentary debates around whether this would compromise accuracy. It would not. It actually stands to improve completeness and accuracy, so it is very much a win-win. This assisted and automatic voter registration is therefore a vital move forward, and the Bill enables that. What is missing is a timetable. There is no detail about when this would be implemented or which data sources would be implemented. In practice, what is happening is that a lot of technical decisions will come forward in secondary legislation and then in guidance that will be developed, presumably with the Electoral Commission and Association of Electoral Administrators. That technical detail is really important. Which groups? Who is going to be assisted? That is an opportunity going forward to include other political parties to ask questions, civil society groups and academics, so that there is a consensus about how to develop that. One route that other countries have undertaken, for example, is to require a Minister each year to report to Parliament to say what progress has been made in terms of implementing automatic or assisted voter registration and to give some statistics about what has been done. That makes sure that there is continuous parliamentary scrutiny, beyond this Parliament, going forward, because assisted and automatic voter registration is not a one-off. You are not going to flick a switch and suddenly this is going to happen. You are going to need to tinker and change the data sources. That continuous process and the scrutiny, like we are having today, is really important.

Do you think this is a missed opportunity to consolidate electoral law? There is obviously a panoply of legislation involved in elections.

Yes. You have witnesses coming forward later on who will definitely make that point very strongly. There is the recommendation—I do not know, Toby, whether you remember how old this is—to merge all the disparate pieces of legislation that we have about this. It definitely is a missed opportunity here, yes.

I am going to move on to some of the bits inside the piece of legislation. The Government say they will pilot innovative approaches to voter registration, but, as has been alluded to, there is not yet any detail on these. What approaches would you recommend that the Government test, and by when?

Toby is the expert on this. I would just say something on the “by when”, and then Toby can come in. It is very important that we are not in a piloting phase at a time when there might be an impression of bias being created by having pilots going on in some places rather than in others. There is one concern there. I know that you have someone from one of the boundary commissions coming. The next boundary review is due to report in 2031, which I think means that they will be using the 2029 registers. If you have pilots having taken place in some constituencies and not others at that time, that will skew the boundary review, and that would be unacceptable. On timing, it is important to finish pilots and roll something out in advance of that.

We talk about pilots. It is worth thinking separately about assisted voter registration and an automatic voter registration. Assisted voter registration could work very simply. For example, when people apply for their passport or renew a passport using the gov.uk portal, we can build a tick box into that process so that people’s details are automatically sent on. The research we have done has focused on passports being one mechanism there, DVLA applications being another and universal credit applications being another. The key thing is that these pilots have to happen very quickly and we do not want those pilots to take place close to a general election. Otherwise, it is going to put a lot of pressure on to the electoral administrators. We have seen from Wales that those pilots can take place and can deliver results. We have seen in Australia that you can move through a system within 18 months to delivery. The clock is ticking and the detail is not public yet.

I want to talk about the Rycroft review. We got the recommendations at the end of March. I will start with Toby and Maria first on your view on those recommendations. Alan, when I come to you, you described astonishing weaknesses in the Bill as originally drafted. Do you think that the Government’s most recent announcements in response to the Rycroft review tighten that Bill as well?

I will be relatively brief. I welcome the review. It does an important job in terms of focusing our attention on the prospect of foreign interference. That is a major threat that is facing democracies around the world. One thing I would like to draw attention to in this is the electoral register, because this is missed. We currently have the open register, which is sold to anybody, anywhere around the world, for any purpose, and that is not monitored. The Representation of the People Bill proposes to amend that, so it is going to be opt-in. However, it would be a bizarre situation for us to sell the details of the public to anyone for any purpose, because that would allow them to be micro-targeted for the data, which could be mixed with other bits of data that are available on the dark web. That is something that is missing from this conversation. I would strongly urge the Committee to move to abolish the electoral open register.

I mostly just agree. From the perspective of voters, one really important job for this particular Bill is to ensure that the processes are fair and equal for everyone. That security concern and the case of funding that Alan raised really chime with voters, so these are very important areas.

The Rycroft recommendations should be implemented in full, in my view, and particularly recommendations on corporate donations. That cap should apply to profits, not revenues, and should apply to the total donation from a company over a year, rather than to any individual entity. These are absolutely vital.

Are there any other changes? Alan, you have just said some there. Are there any other changes to the political finance framework that anyone would recommend on top of Rycroft’s 17 recommendations?

I made the point earlier about donation caps. Also, controls over spending that do not just cover the campaign periods should be looked at. We know that a huge amount of activity now takes place throughout the electoral cycle, so looking again at those spending controls seems important.

The regulated period forces a change in behaviour, because you will see a ramp-up until you hit the regulated period and then a drop back down to be within regulated spend.

I want to take us back to votes at 16 and automated voter registration and how the two need to work effectively together. I am a supporter of votes at 16. It is already in place in Scotland and Wales for devolved elections. I do not see any reason why voters in England and Northern Ireland should not have that same right, but I am really interested in your expert view on how this interacts with automated voter registration, especially with the provision in the Bill for young people to register at age 14. I would like your views on how we can make that work as effectively as possible in tandem, please.

If we talk about who is not registered, it is young people, so voter registration should be integrated into the schools process. That is the most systematic way of doing that. The difficulty there, and the one tension point, is nationality and nationality data. At the moment, there are specific nationalities that are eligible to vote in some elections but not for others. The situation is simpler in Scotland and Wales, where it is residence-based voting. I would add that there are very strong arguments for residence-based voting for parliamentary and local elections as well, but the nationality is a technical issue. Do schools have data on nationality? Should they ask pupils for data on their nationality? There are some arguments for why they should. This would facilitate voter registration. There are other aspects that are important and relevant. There have also been some concerns there that schools might pick and choose pupils on the basis of nationality. That is the question and the tension point. I think the only way to make votes at 16 work is to enable registration in schools.

It is not necessary for the school to have nationality data if they have an assisted process, in the sense that they prompt and assist the pupils to access the existing online registration system. Then the children in question might confirm or not confirm. Of course, all those online registrations are then checked by the electoral registration officers, which means there is this additional check of which registers the kids will be eligible to be on ultimately. I agree that this is such an important part of the reform. To assuage the Conservative fears about treating other voters differently—

I do not think that what I said was party political.

No, but assisted registration would therefore meet that criterion of not necessarily automatically registering, but assisting that population, which is much easier to access.

I have one quick point there. We actually have automatic re-registration. That is not really acknowledged. That was a reform in the last Parliament. We have automatic re-registration. It is called canvass reform, but basically data sources are used to automatically re-register people if they have not moved. The main beneficiary of that is older populations, who tend to be less mobile. Introducing this for young people is almost like for like in many senses.

It strikes me that year 9 in the school year is the logical time to do this. That is before they start the GCSE syllabus, because most of those courses are two years, and that would then link with age 14. Please tell me if you have a different view, but that seems like an opportunity. I am interested in your views on other barriers facing young people. One thing I have encountered on the doorstep, and I suspect is true for colleagues, is that I often meet young people who say, “I don’t feel like I know enough and therefore I won’t vote”. I find that less common among older voters. There is another barrier there and a feeling that they should not, when of course we need to make sure that they are very clear they have just as much of a right to vote. I am interested in any quick reflections on what other work we could do in that space.

The always tricky point of civic education in schools is one that would have to be discussed by the Government, because this is an area that is always controversial. People always worry about how much lead the teachers can give the children. There is a lot of research showing that earlier franchise actually enables those young people to take those kinds of steps into their own hands and inform themselves. Knowing that they are eligible might spur a lot of them on. Especially if they are already registered as voters, that might be a good step for them to take steps to inform themselves and educate themselves too.

That is an absolutely vital point. I am a bit concerned that the Government, in response to questions about this, have replied by saying that they are introducing changes through the national curriculum review, which will not be implemented in time for the next election. A clear focus on thinking through what needs to change in citizenship education, in terms of the curriculum but also in terms of supporting and enabling teachers to navigate this minefield of ensuring that they are impartial, and seen to be impartial, by all of their pupils and parents, is incredibly important in order to deal with this.

Thank you very much. You have given us some thoughts and some areas that we will definitely continue to pursue in the course of this inquiry. Witnesses: Tom Jenkins and Laura Lock.

Welcome to the second panel of our inquiry on modernising elections. Can I ask our two witnesses to introduce themselves, please?

I am Laura Lock. I am deputy chief executive of the Association of Electoral Administrators, a membership body for people running elections and electoral registration in the UK.

Tom is joining us virtually.

I am Tom Jenkins. I am acting secretary to the Boundary Commission for England.

Good morning. Two of the major reforms in the Bill are automated voter registration and votes at 16. What assessments have you made of the impact that these changes will have on your work?

We think that it will have significant impact on the work of electoral registration officers who work in local authorities. We are still waiting for an awful lot of detail on what is going to be the system for it. We think that it is really important that we are looking across the whole of the electorate as to how it is implemented. Votes at 16 obviously has particular challenges. By way of some background on a point that was made by an earlier speaker, Maria, when we moved to the individual electoral registration system, just over a decade ago, we saw a significant drop-off of what we call attainers, so 16 and 17-year-olds who will be eligible to vote at the next election. In 2011, we had just over 400,000 attainers on the December register. In 2023—the last figures that we have—it was around 120,000, so a 70% reduction in young people being registered before they attain the age of 18. We are concerned that, unless something is done to address that, that is going to be even more of a problem for 14 and 15-year-olds. We are also concerned about the impact of any reform that is looking at passport data, for example, or DVLA data, whichever it might be, and how that is not going to be uniformly available for all elements of society. We know that the number of people holding a passport ranges significantly across the country. We are also going through a local government reorganisation process at the moment. Access to data in local authorities is a huge issue, with many of our members reporting that they cannot access young people through education records. We also have particular issues relating to registering older people in care homes and things like that. Access to data is going to be vital and we welcome the Bill looking to enable better data sharing. Exactly how that appears once all the secondary legislation is laid will be very interesting to see.

For us, it is difficult to determine what the impact will be without knowing the full scale of what the auto registration process is going to look like. In Wales, they have introduced some pilots, which have increased the number of electors on registers by between 2% and 8% in the three or four different local authorities that have introduced these pilots. Until we see more detail around what pilots will actually look like, it is hard to determine how that will have an impact.

Laura, what would you need to feel confident about the implementation of these reforms? Your organisation has indicated that more detail is needed, particularly on the automated voter registration. What are the key features you are looking for that would make this work?

For us to feel confident, we need to feel that there is the resource available to do this, and that includes time. When votes at 16 was implemented in Wales and Scotland, we had a very clear window where we could get the young people registered to vote before they were enfranchised to vote. At the minute, we believe the plans would be for both of those things to happen on day one. We do not just have elections every four or five years when all of you are due for election. We have elections every single week. We think that it is really important that EROs have the time to share the message, “Get registered now. You will be able to vote from whichever date is picked”, and that we have a really clear window to register those people. With automated registration, a lot of the issues have been raised by colleagues. It is much more difficult in England to register people than in Scotland and Wales because of the nationality issues. Even if you are an EU citizen, some EU citizens are eligible to register for local elections and some are not. There is different registration for parliamentary elections. It is really difficult. We also have anonymous registration for those who have concerns about having their information in the public domain. It is not simply the case that you find the name of someone who lives at a property and get them registered. Canvass reform was mentioned earlier. We had a change a few years ago where, instead of having to fill in the form for your house every year saying who lives there, now we can send an email or a letter that says, “These are the people we have. If that is okay and still correct, let us know”. We have already done some data matching, so we have checked DWP records to confirm. If we do not think it is right, we ask them to confirm the details. For the approximately 70% of people who are automatically registered, all we are doing is data matching at the moment. There is no data mining process. If DWP records said, “We think these two people still live there, but we think another two people also live in the property”, we still would not be requesting their information. We think that there are many facets to this that all need to be joined together. We need civil society organisations, schools and everybody to be working together to try to make this actually work in practice.

Going back to the Elections Act 2022—I think I was the only one elected when that was introduced—lots of key areas changed, including the introduction of voter ID, which was a big change. I think at the last general election the Electoral Commission highlighted that 4% of voters said they did not vote because of voter ID. In my view—and I think many colleagues would share that—that is 4% too many. If only one person is not able to vote, that is really worrying. In terms of some of the changes, you have expressed concerns about the use of bank cards as voter IDs. Why would you say that that is an area for concern?

We represent electoral administrators who employ people who work in the polling stations. A lot of these people work one day a year or maybe one day every two or three years and the process is so complicated. They need to understand so many facets. We feel that now adding bank cards to the list of ID is another challenge for them to navigate. If you are going to water down the photographic ID policy that far, you would be better off just getting rid of it in the first place. There are other things that we should consider, particularly with votes at 16 coming. At the same time as voter ID, you could implement voter authority certificates, so you can apply for free photographic voter ID. You could look to make that an online digital ID that potentially could be applied for on the day. It makes checks against Government records and national insurance numbers to allow that as an on-the-day digital form. That would be something that would help that 4% of people who did not have something available.

On the proposals that you have outlined, Laura, the Government are looking at using digital ID, because I think most people have moved to an automated version. They have moved to having digital cards in their wallets, so that will be acceptable. Do you think that those small changes will make a difference, especially to your colleagues who are administering the elections on, as you say, a weekly basis? There are elections going on every week.

Yes, I think a lot of those will help them. Moving things away from a very paper-based process, which we have in a lot of areas, will help. When we are talking about modernising elections, there are still some huge things that we could look to do to bring the process forward in leaps and bounds. We advocate having an online nomination portal, so you no longer have a reliance on delivering paper nominations. When we come to parish electors, we are talking of thousands of pieces of paper being delivered. The Bill talks about candidate ID, and we do not support candidate ID for all polls in the format that it currently is because of the burden and our concerns about how that could put people off standing, particularly at parish council level. Colleagues in Cornwall accepted over 1,700 parish council nomination papers last year. For each of those people to have to find a copy of a passport or driving licence would be something of a challenge, I suspect.

You raise some valid concerns. I can think of two different bank cards I have, one with my initial and one with my full name, so that would cause issues on the day for polling.

Laura, on the subject of ID and particularly the fairness of it, what assessment have electoral administrators made on the fairness of older persons’ bus passes being acceptable forms of ID when young persons’ bus passes are not? I know that we had a question from Gagan earlier about the unfairness of young people being enrolled, but is that not the other side of the coin: that we have a very fragmented system that sometimes favours one group of people and sometimes favours another?

Absolutely, yes. With votes at 16 coming, we are even more concerned that those people may be less likely to have a passport. They would not have a driving licence and things like that, so it is going to be really important that we consider that. I know that some studies have been done as to how many of those people would have a bank card and how much having bank cards will help those people. My 15-year-old stepdaughter does everything on her phone. It is all Apple Pay and things like that. I suspect she could not even find her bank card if she tried to. We need to understand how much that will help young people to be able to vote. Also, with postal voting, you have to have your ID verified now to apply for a postal vote. With registration, you have documentary evidence. We can help people to register using documentary evidence. We are not allowed to do local data matching in the same way, so we would have to go to documentary evidence, which would be harder for young people. For local data matching, council tax bills are not in their name. They are unlikely to be applying for benefits. There are an awful lot of things that need to be thought through to enable those people to actively participate in democracy.

To ensure that young people can vote when it is 16 and 17-year-olds, do we, as a country, need to look at what ID young people can vote with, so whether that is a young person’s bus pass or railcard or other things? Do you think that the current system would not work?

The current system would be a great challenge. We need digital, on-the-day voter authority certificates, so that young people could do things very quickly and instantly then. As well, we need to be considering, given that we have implemented voter ID, what the evidence is showing as to whether using bank cards is diluting the system too far and actually we would be better off going back to the previous system that we had.

As part of the Children’s Wellbeing and Schools Bill, we had unique identifier numbers for children to be able to track them through social services. Could that be a lever that you could link in with?

Absolutely, yes. With any data that the electoral registration officer or returning officer can access, they can hold that record securely and use that to enable people to register, and to vote, potentially.

I have some questions for you, Tom, about automatic voter registration. We heard today that about 7 million to 8 million people are not currently registered to vote, and some of that may be through choice. I have asked Ministers a number of times around whether they are all going happen at the same time for the same general election. It has kind of been alluded to that the answer is no. That is blatant gerrymandering if they are not all done at the same time for the same general election. Could you give us your insight from the Boundary Commission’s viewpoint? The next general election is planned for 2029. There should be another boundary review in 2031. If, say, 4 million of those 8 million are done, how are you proposing to deal with that through the Boundary Commission for England and the new boundaries for the general election after the one that takes place in 2029?

Ultimately, for the next review we will have to use the register as it is on 1 December 2028, unless legislation is changed. Currently, in the Parliamentary Constituencies Act, it states that we have to use the 1 December 2028 register. In order for us to change that, there would have to be some change to the primary legislation, which would also have to go through committees and be approved by Parliament. In terms of how that could be changed, you could postpone the next parliamentary review or not add the auto-registered electorate until after the 2028 registers are published. Until we know when those pilots are coming in, how many of those pilots and at what scale, it is hard to determine what could actually happen to the next review.

If all the voters who are currently not registered are not automatically registered by 2028, will the next Boundary Commission proposals and review be fair for the next general election?

We are unbiased politically, so we do not have a view on that. We will have to use the register as it is on 1 December 2028, unless we are directed to do otherwise.

To take my constituency of Broxbourne, I have two councils, so two authorities that deal with electoral roll. One of those has automatically registered people and the other one has not. Would that be a fair boundary at the next general election for me to stand in the Broxbourne constituency?

I do not want to sound like a broken record. If that is the state we are in in 2028 and those people are not registered, that is the register we will be using, regardless of whether people agree it is fair or not. That is not down to us to determine.

You have to determine whether each of the constituencies are within the 5% threshold across the country. With some voters automatically registered and others not, there will be loads of constituencies out of that variance.

Our final determinations would still be within the 5%, though. We would not propose anything outside of the legislation or what constraints the legislation put on us.

I do not understand how that can be true if half the voters are automatically registered and the other half are not.

We work with the electorate on the register. We would draw a line on 1 December 2028. Regardless of where people are registered, that will be the register we use to determine the next round of boundaries, if you see what I mean.

Going back to the initial question, if the implementation does not start by 2027, which is what the Electoral Commission has expressed as well, you do not think this would happen in time for the next general election.

I cannot see how it would come into force in time for us to make a distinct change to how we would usually go through the process of the review.

Tom, Lewis and I are from different political parties, but this is a really important question to get right. I appreciate that your role is strictly non-political. Would it then be fair to say that doing any pilots by parliamentary constituency would, in your view, be a fair and equitable way to go around this?

A lot of parliamentary constituencies cross local authorities, so you would have to consider which ones or how that would actually work in practice to do it by constituency. You would not necessarily have whole local authorities within a constituency, which would make it tricky for electoral administrators as well.

As it stands, there are no dates for any pilots in any constituencies, so it would be hard to predict what those pilots would then bring up. This is a question that we would definitely need to put back to Ministers in terms of the timescale. In give or take a few more weeks, it will be May 2026. Time is ticking on this. Thank you very much for that, Tom, and thank you very much, Laura. We are going to move on to our third panel. Witnesses: Vijay Rangarajan, Sam Hartley and Alison Williams.

Welcome to our third panel in our first session of our modernising elections inquiry. Can I ask our guests to introduce themselves, please?

Good morning. I am Alison Williams. I am general counsel and director for devolution and law at the Election Commission.

I am Vijay Rangarajan, chief executive of the Electoral Commission.

Good morning. I am Sam Hartley, director of policy, research and greater engagement at the Electoral Commission.

I was after the views of the panel on the current Bill, but also the ability for the Executive to be able to postpone or cancel local elections. Do you think that the Bill should be strengthened so that that cannot be allowed, or that a higher threshold needs to be met before a cancellation? What is your overriding take on how that resonates with the public, in terms of the state of democracy in this country? As we heard in panel 1, our democracy probably does not rate quite as highly as Members around this table would like it to.

We do a series of tracker surveys. We try to survey a large range of the population after each election. We do other surveys of candidates, as you know. Overall, on the state of democracy, I would like to start there. Satisfaction with the system of registering to vote or the process of voting is high. The percentages are in the 70s to 80s among the British population. That is good. Satisfaction with how democracy works is really quite low. It was 38% in 2024, up to 47%. It fluctuates quite a lot. Elections are well run. It is up at the 79% kind of level. The figures that are really dire are on political finance, where “Is the spending and funding of political parties, candidates, and campaigning organisations open and transparent?” is down at about 15% to 18%, which is about as low as you can get in a public poll. The perception that political misinformation and disinformation and the information environment is a problem is right now up at almost 70%. We are seeing that the overall threat picture is changing. Voters are fundamentally worried about where money is coming from and need reassurance on that. They are worried about the information environment. That is really going up in all the public polling we do. That is a combination of information that they get, misinformation and disinformation, allegations of foreign interference and deep fakes. Those are all rising rapidly. The third real threat in our system is the growing abuse and intimidation of political party candidates and elected representatives. All those three act together, I am afraid, where you see evidence of online co-ordinated campaigning, which then gives rise to aggression towards MPs, say, and vice versa. We have seen all those together. I will not go into too much detail, because we put a lot of evidence to the Speaker’s Conference on this and are acting on those recommendations. Overall, those are the threats that we see rising and that our polling is showing, and we think the Bill goes a very long way to address them, actually. There is very strong support from us overall on the Bill, particularly the political finance measures. It is strengthened in the way Philip Rycroft has recommended. Mr Dillon, to take your point on local government elections, as you know, our position is very much that scheduled elections should go ahead. The uncertainty in whether elections should go ahead is not welcomed by voters. It creates problems for candidates, who may find themselves standing or not standing. We are seeing this at this time. The certainty that there is a fixed period and people have their say is really important. We do not think that there is a need for a legislative change in the Bill. There are the powers that the Secretary of State has to postpone elections. We have said what should be done. They are postponed where you have a very short period where someone would be elected and then replaced by, say, a shadow authority. There is lots of precedent over decades about when it makes sense to postpone. Where you have situations such as covid, clearly there was a need to postpone some elections, but they should be rare and really justified, and not be justified on the basis of resources in councils. That was where we had difficulties with the last set of proposed postponements and why we came out strongly against that. Electoral administrators, who do a fantastic job, were telling us that they could run them, they had the resource to run them and their local councillors wanted them, mostly, to run. The chief executives were clear that they could go ahead and run those. It was very important and so we are delighted that the elections on 7 May are indeed all going ahead.

Since this Bill was first talked about until now, there has been a shift from the talk of automatic to automated. What do you think happened in that time to shift from a more insinuated opt-out system to what we are now talking about, which is piloting and potentially making things more efficient? Do you think the digital ID fallout contributed to that at all?

Automated and automatic registration can both exist at the same time. There are other countries that we have studied in real depth. Particularly, I would hold up the Australian example. There are some really good examples there that we can use. They use a combination with automated, which means that, when you are applying for different things such as a passport or your driving licence, or other business processes like that, you can tick a box that says, “Add me to the register”. The system then has enough data on you to add you to the register. Those are all really welcome. They can help reach some particular groups that are harder to reach. We know who is not on the register. A very large number of completely eligible voters are not on the electoral register. Our position is that they should be on the register as easily and accurately as possible. Both have to be true and these automated systems can help with both. To give you an example, we know that those in private sector rented accommodation are often underrepresented in the register. They move very often. When they go and re-register for a driving licence change or changing their tax address or whatever, those processes could just shift them on to the new register. Those are really good examples of automated registration. There are people in care homes. There are plenty of groups that find it hard to register. Full automatic registration, in the way that I think your previous witnesses were talking about, is also a very good thing. We would like to see that implemented as soon as possible, but it is more complicated than just inserting this into the business process. It will involve data matching by electoral registration officers and they will need to use a wide range of sources. There is no single source of the truth in the UK data environment. It would involve looking at, say, passport office data, DVLA data and elements of schools data, which would be very important for attainers, and building that into some of the processes we already have. When you are 15 and three quarters, you get a letter with your national insurance number. Rather than just a letter saying, “We are going to tax you for the rest of your life”, it would be nice to have something that said, “Here are your rights. You can now vote. You are now on the register. Get involved in how your money is going to be spent”. There are elements there and we can build how that will work. To come back to the discussion you had with your previous witnesses, we are really concerned by the differential impact of automatic registration around the UK. We understand that, following their successful pilots that we evaluated, the Welsh Government are intending to go ahead with automatic registration soon, in the next year or so. That will mean that the Welsh local and Senedd registers will have a lot of people on them, but they will not necessarily transfer that into the parliamentary register, which is a recipe for extreme confusion. What we would like to see is automatic registration and these processes being uniform across the UK, including in Scotland and Northern Ireland, where it will help some of the canvass issues, and done before the general election, for exactly the reasons on the boundary review that Mr Cocking was alluding to. It is going to be difficult otherwise, if we have different registers in different places using different systems.

The crux of my question, though, was around why there is a sense that that was the ambition, but, within the last year, that has been backtracked on and there has been a lowering of ambition. We have heard today and in other conversations about the technical complexities of putting that in. Is that your understanding as to why that has been watered down, or is there still an ambition to get to that automatic opt-out system, either before the next general election or as soon as possible? What timeline is the ambition now?

The ambition is there for a full automatic registration system piloting the various data matching powers, but it is not there to do this before the next general election. That will bring two issues. First of all, some parts of the UK may have automatic registration and some parts will not. Secondly, we will hit the boundary review issues that have already been alluded to. Sam, I do not know if there is anything more you want to add.

From our point of view, we would urge the Governments to talk to each other on this. Vijay has highlighted the issues of divergence and the risks to not just the voter but also the electoral administrator community of having to maintain in Scotland and Wales two different types of register, and voters not knowing whether they are on the right register at all, because of the confusion. The added challenge across Government of running a fairly large-scale digital project to get us to a point where we can have automated registration involves all the Departments talking to each other and using the big databases that we have referred to in HMRC, DWP and so on. It looks incredibly unlikely that it is going to be in effect by the general election, and that is a real shame.

Just to follow up on that digital ID point I made, from your understanding, would a stronger digital ID system have made that quicker? There is a lot of frustration and fear around digital ID, but there is a lot of public support for automatic voter registration. Are we doing enough to align what digital ID enables us to do when it comes to things such as democracy?

We should be looking at every type of ID possible to increase accessibility. From our point of view, there is not yet enough detail about the digital ID programme for us to know whether it is going to assist or potentially slow down the automated voter registration system. Anything that can break down the barriers for people to get on the register and access the right to vote would be a good thing for us, but it seems quite early when it comes to digital ID.

Just lastly on this topic, are there any specific methods of registration that you would most like to see the Government pilot? They have committed to pilots, so what would you most like to see?

In terms of automatic registration, we would first like to see the automated steps, so getting it involved in the other business processes, which are already highly digitised and secure. That is the work of gov.uk and the Government Digital Service, and they are thinking about how to do all of that. Second would be to pilot the correct data. What we found in our evaluation in Wales, which did pilots in four local authorities, was that local council data could get you a very long way in terms of accuracy. They added 16,000 people to the register in those pilots, of whom, a while later, 83 were taken off the register, mostly because they had moved, and that is exactly right. You get it as a snapshot in time. It then degrades. You need local data as well to really maintain the accuracy of the register. It produces a far higher-quality register and one that reduces one of the barriers to people participating. We would like to see them rapidly pilot the main datasets. As Sam said, that involves a lot of discussion in Government on sharing that data. Quite a lot of the technical infrastructure already exists in the IER data system, which was set up for individual electoral registration, but it will need some more to be built. Then we would like to see something uniform come in around 2027, if that is possible. That ambition is really important so that we do not hit a problem of either lack of uniformity or just continuing to have 8 million people or so not on the register in 2028‑29.

That is really helpful. Just moving on to votes at 16, we heard earlier that a Government response about how we are going to support 16-year-olds to be better informed will not come in, in terms of the curriculum review, until after the next general election. I know you have done research around how young people feel about voting at 16. What are the highlights of that? In the next five years, what do those 16-year-olds now most need from Government to support them as they are able to vote?

Before I pass to Sam in a second, I will just say one thing. Those 16-year-olds will need a good form of voter ID. That is really clear from the start. That is where the digital ID may come in. As your previous witnesses have said, there is a huge appetite among young people for more information. This is where I pass rapidly to Sam.

Before I answer the question about education, this links again to the voter registration piece, which is that, if Parliament is going to legislate for votes at 16 across the whole of the UK, we want to make the process of them getting registered as easy as possible too. That is a crucial part of this. You are right. It is nice to be able to say that we do not need to wait for the curriculum review, because the Electoral Commission produces a lot of material and educational resources. We visit a lot of schools as well. We know that young people want more education in schools. We know that over two-thirds of them want to learn more about democracy at school. We also know that they trust what they hear in school much more than they trust social media, which is heartening, in many ways. It also is a great opportunity for us to roll out some civic and democratic education. Only 30% of them have experienced what they would call high-quality education about politics at school. On the curriculum assessment review, one of the things that we welcomed and advocated for was the idea of citizenship education at primary school level as well, to really start the education in terms of how young people participate. To turn back to the work that we do, you can take a look at our website and see the resources that we have rolled out. We have had some really good feedback from teachers and practitioners when it comes to the first set of resources that we have rolled out. Over 90% of them feel much more confident to teach high-quality, impartial democratic education at schools. We are conscious and aware that some teachers are nervous about teaching this sort of education in schools. One of the strengths that the commission’s resources bring is the impartiality and independence side of things. Annually, we run Welcome to Your Vote Week. Many Members will have been involved in that in the last few years as well. Last year, we reached over 600,000 children and young people through our partners directly, and over 1.5 million indirectly through our digital channels as well. This is very much the start of a large programme of education and engagement work that we are rolling out with the support of the Speaker’s Committee and Parliament in order to address the very challenges that you have raised.

Just finally, the AEA has expressed concern around fast timelines for implementing votes at 16. Do you share this concern?

As usual, the Gould principle and the legislation, including the secondary legislation, will need to be in place six months before. The fast timelines will be an issue, particularly on how we get those attainers to be registered. As Sam said, we are going ahead immediately—and have done so for the last year or so—with the education work, because a lot of them need to know what they are going to be doing if they are going to be registered to vote for the next election. The final thing is that they will need to have some form of digital ID. Having a digital voter authority certificate, or whatever digital ID, in place by then is really going to be necessary. All three things—knowing what they are voting for, being able to make their own minds up, and being able to vote, being on the register and having the ID to be able to register—will need to work together.

Vijay, the Rycroft review was published at the end of March. What was the commission’s reaction to it? Are all the recommendations deliverable by the next general election? If not, which ones should be prioritised?

I will just give a very brief summary and then pass to Alison. We gave a very strong welcome to the Rycroft review. We are seeing an increasing threat from all the things that Philip described, which were various attempts to influence our politics from outside the UK. His recommendations, particularly on company finance—so profit, not revenue—and on tightening up some of those rules further are very welcome and we hope will be taken forward in the current Representation of the People Bill. Quite a few of his other recommendations are also doable. The biggest one is recommendation 4, on an overall regulatory system for candidates and non-party campaigners over the whole year. That may take some time to work on.

Just to build on what Vijay has just said, we very much welcome the review and the fact that Philip Rycroft has concluded it in such a timely way to enable many of the recommendations to be accommodated in this Bill. That is the first thing. There is an urgency, which means that we welcome the timing of the review and Government pressing ahead with the recommendations, but there is no doubt that it is a challenge for all of those recommendations to be taken on board in this timescale. It is very much a challenge for MHCLG. As Vijay said, in terms of priorities, the Government have, in a sense, already set the first two priorities in relation to the announcement that they have already made on the overseas donations cap and the crypto assets moratorium. We welcome those, but they raise a legal certainty challenge in the sense that they have been announced, but we have yet to see the legislation. There is an additional legal certainty challenge in the sense that the intention is that they apply to elections, which are ongoing now, but that will require, in certain respects, legislative consent motions from the Scottish and Welsh Parliaments. There is an absolute challenge there in relation to detail, but, given the importance of the recommendations, our commitment is to, first, work very closely with MHCLG officials to ensure that those recommendations can work and, on the other side, work with the regulated community to support those. We have already published provisional advice, pending the legislation, to support the regulated community to prepare for those. In respect of the other recommendations, as Vijay said, we very much support the recommendations on corporate donations. We have said, as the Committee will be aware, that we do not think that the revenue-based test in the Bill at the moment is sufficient. Our view is that revenue should not be a gateway to give to as many recipients as possible. Profit should operate as a UK-based cap. That is what Philip Rycroft recommends, and we hope that the Government will take that recommendation on board. The other recommendations that we consider can be prioritised are the non-legislative recommendations that might be less glamorous but are no less important for effective enforcement, so how the system works and joins up.

Thank you, Alison. That is really helpful. Vijay, you said earlier that almost the weakest part of our democracy, based on opinion polls, is the concern about finance and dodgy donations. The largest single donation in the history of this country was done recently, with £9 million given to Reform by someone who has made their money from cryptocurrency. We genuinely do not know where that money has come from. The review recommended a ban on crypto donations until we have proper regulation and a proper framework in place. What challenges do you envisage in putting in that regulatory framework in such a fast-moving, difficult system? Effectively, should we not just ban crypto donations?

We have not been in favour of a complete ban. That is partly for the reason that it just pushes back the boundary of what crypto is. What happens is something called an off-ramp, where someone makes a donation in crypto. It is then converted into sterling—into fiat—and then someone makes a donation to a party. We would like to be able to see—and enable parties to see—that whole chain and to do the sorts of know-your-donor checks that they do. What we would like to see in the Bill as the end of the moratorium—the Government have been clear that Parliament needs to be happy with this—would be that we would produce statutory guidance laid in front of Parliament. If Parliament agreed, that would enable some forms of crypto potentially, if they met the rules, to be used to donate to political parties. That is partly because the particular issue we have seen in crypto donations—we met the crypto industry association and many of their providers last week in probably the first of many meetings—is that we and the parties cannot always trace who the actual donor is. That is the thing that is weaker than a normal bank account, where, when you get a transfer, you get information on the name of the donor. Parties need to be able to do that because that is what electoral law stipulates they need to be able to do. As you rightly say, this is changing so quickly that it is perfectly possible that, in the future, there will be crypto systems that will, indeed, give sufficient assurance to a party that the donor is who the donor really is, and sufficient transparency, or maybe even more transparency. We are not there yet. As that regulatory system moves and as the FCA and others put their systems in place, it is possible that this will come. We do not want to close the door completely, but we do want to ensure that any donation made via cryptocurrency meets the same level of certainty that is coming from a permissible donor as for any other currency donation coming in.

The other thing that we should add is that it was good to see the Government announce that they are withdrawing their strategy and policy statement on the Electoral Commission, because it is important, given all the issues we have highlighted today, that the Electoral Commission is free to be independent and able to do its job.

Can we have some idea of the assessment the Commission has made on online disinformation and misinformation, and its prevalence and impact in recent UK elections?

In recent UK elections, it is the dog that has not barked. We have not seen any evidence that it has affected the outcome significantly—there is an enormous “but” coming—but it is rising enormously as a threat. What we have seen, for example, in the recent Hungarian elections was a scary level of massive information manipulation. What we saw in the recent Irish presidential election was another issue, which was one of a really credible deepfake. There was a very beautifully done deepfake of Catherine Connolly, the Irish presidential candidate, withdrawing from the election. The deepfake was released three days before the election. It got plenty of views and got taken down after a while. That sort of attempt to influence one of our elections—clearly, it could happen for the May elections—is the reason why we have set up a deepfake scanner, and we have made very public what it is doing. It is trying to look at what is going on in the election information environment, and then we will work with candidates and political parties to try to remove this. For example, we take down or contradict it if it is just wrong. If they say the election is cancelled or a candidate has defected—remember the example of one of your colleagues, George Freeman, who was, indeed, wrongly deepfaked as defecting—all of these things are factually wrong, so we would then correct them or contradict them, particularly if it is a matter of electoral process, which is where we are circulating. The other thing we are seeing, though, is an increasing use of deepfakes, because the barrier to entry for these has fallen so dramatically. Just this morning, we are seeing one party use a slightly AI-generated image. In another one, there appears to be an AI-generated candidate in Scotland. We have a number of these that are beginning to be used in political debate. They have the effect of fundamentally misleading voters. Our core concern here is voters beginning to not believe what they are seeing or what is being told to them in a campaign, and that is, clearly, deeply corrosive of political trust. We are very grateful to political parties for sharing their codes of conduct. Basically, we boiled them down to a set of principles for campaigning. That includes not using material such as this, which is designed to mislead voters. The reason for the principles was so we could also send those to all independent candidates to maintain a level playing field of campaigning. We are thoroughly worried, and this is going to be a growth area where we are going to have to keep on working on this.

Have the Government missed an opportunity in the Representation of the People Bill to tackle some of this?

To be really honest, I do not think that we know. We could not come up with the legislative provision that the Government should put in the Bill. It is quite hard to do. We may come up with more once we have run the deepfake pilot and then seen what the effect is and what is going on. We monitor the playbook of people trying to interfere in operations in other countries quite extensively. In some cases, there is legislation in place that may be useful, but it has not necessarily grappled with it. One of the problems is that the legislation tends to ban things and put criminal offences in place, but they can take a tremendously long time to then act on. Working out the provenance of a deepfake, who was behind it and then prosecuting them takes months to years, whereas the damage is all done and the election has happened. It has happened in seconds. We are taking the view that we are provenance-agnostic. We do not really mind who has done it or who is behind it, and we will never find out in time, but, if it is wrong, we will simply go out and stop it, try to get it taken down, contradict it or correct it where we can. There is no other way of trying to protect some of the integrity of the election.

Is there more that can be done beyond legislation in order to protect our democracy?

The short answer is yes. We are going to do a report on the deepfake pilot and the other information environment. You will see this rise more and more as one of the themes of our post-poll reporting. It used to be about electoral administration—we will still cover all that and make sure that is right—or about regulation of the various processes involved, but it is increasingly the information environment where some of those threats can be.

Just to come back to the education piece that we were talking about earlier, we have talked about the extent to which the threat is real and coming, but this is a public education and public confidence issue as well. Our opinion polling shows that nearly four-fifths of people are concerned about the untransparent use of misinformation and disinformation and how that affects elections. Our polling shows that people—and not just children or young people—want to have better media literacy skills to be able to address, identify, challenge and judge the digital information that they see as well. Part of our education and engagement work is not just about accessibility and involvement, but also about giving people the skills to challenge what they see online.

I just want to cover an example that links this to the Philip Rycroft review. As of this morning, Storm-1516, which is one of the pro-Russian bot networks, is circulating quite extensively an AI-generated image of Nigel Farage with Jeffrey Epstein. They are trying to do the rounds. We are seeing this stuff. We can now detect it, which is useful. We may not be getting massive traction, but we are seeing direct attempts at this sort of intervention into British politics.

There are lots of hands up from colleagues on deepfake. You mentioned the deepfake review pilot. When will that conclude?

It started last week. It will end just after the elections, in the middle of June. We will then do a full evaluation and report, which we would be very happy to come back and discuss.

There is a growing amount of evidence now about the impact of algorithms, and particularly around children and young people. If you look at the gendered impact of algorithms and what direction that points young people in politically, it is quite worrying and quite stark. I was just interested in whether you, as an organisation, have been considering that particular element of how algorithms are radicalising our young people in different directions and shaping their political views. Taking a step away from the education that you might provide in the classroom, if that is where they are spending a lot of their time and getting a lot of their content, that has a potentially very dangerous consequence as we move to try to register them to vote at an earlier age. I am just interested in whether you had looked into that.

We have begun to look into that. I would say three things on that. First, there is definitely a trend that quite a lot of the AI that we are seeing at the moment is just clickbait. Because people are monetising it gently—although it is not much, they do want money for this—it does then tend to go for the most extreme because they are the things that people will then click on. The second issue—and I am going to come back to Sam on this—is how we do the digital literacy work in schools, where everyone is concerned. The third is what we are calling algorithmic bias. There have been credible allegations that, in other elections, some social media platforms have twisted or tuned the algorithm to preferentially promote tweets of one political persuasion and preferentially suppress tweets of a different one. Sky News has done a good investigation of this. There is some good academic evidence that it happened in the US election. That would be incredibly concerning, because voters will not realise that that is happening to them. They will think their feed is their feed and they are seeing stuff as normal, but, during the election period, maybe they are seeing a really skewed example of what they think the usual people who they follow are doing. That is a real concern, so we are looking at that at the moment. Is there more you want to say on digital literacy?

Only that we work with expert partners in this field. I may not be the expert in this field, but we have partners who we work with who give children and young people the confidence to challenge what they see and not take everything that they see on social media at face value.

It puts a lot on a young person to be able to differentiate and understand what they are subject to, and then to act differently to that. Would you support stronger guardrails to stop these algorithms being able to influence our young people?

We generally would, bearing in mind that we are looking at democratic engagement and the influence on elections and electoral integrity. As Vijay said, we talk to Ofcom and to the wider group of organisations involved in safeguarding, but our focus is very much on democratic engagement and electoral integrity.

The other thing is that we are trying to get into the spaces that young people are in to provide a trusted voice and to point to other trusted voices on what the correct information is, which is also equally important. If they are not sure about something, the message is, “Go and check it and then make your own mind up”.

Going back to deepfakes, can I ask briefly about the resource that you have at your disposal to deal with this? What I am alarmed by, going back to the words that you used a moment ago, Vijay, is that the damage is sometimes already done by the time you have the ability to act. I remember seeing a report about the last general election that said that about 15% of people made their mind up on how they were going to vote within the last week. If there is a scenario where, in either a local or a general election, we are flooded with deepfakes and disinformation just at the point when people are making their judgment, I fear that, if you do not have the resource, as you said, it is too late. I am interested in how you are looking to scale up resource and if you have an ask for support from the Government to help you to do that.

At the moment, we have the resource, and we have found that within our budget, because we have managed to work with the Home Office. There is a part of it called ACE, which is its innovation capability, which runs the deepfake challenge. It evaluates the different deepfake tools that you can use. It has been incredibly helpful and we have worked with it to contract with several private sector providers that run the tech. The system is now up and running. As I say, it is fascinating to be able to see what is bubbling away at the moment. There is nothing major so far, though these examples are all percolating through. We are funded to run it over May, and then we will want to come back and think about it. If it did not work, we will not continue it. If it did work, should it be scaled up and made more permanent? Then there will be a resource issue about how to run this. There is an element of this being an arms race as well, because the AI companies continually manage to improve their models, which can generate better deepfakes, and different systems are needed to catch them. This is going to be an ongoing challenge. The reason why we have the system up and running all the time—24/7—at the moment is exactly because we will have to react extremely quickly, and increasingly so as we get closer to 7 May, if we see some major deepfakes.

I will move on from deepfakes, because you and colleagues have covered it very well. Something else that can be disinformation that unreasonably influences elections is political betting. Has the Electoral Commission made an assessment of the risk of that, where either an outside influence or a political party disproportionately bets to change the odds to influence behaviour?

We have not made an assessment of it, but we talk to the British Polling Council, significantly, which has a set of codes about how to avoid being manipulated into polling that is there to influence rather than to inform. We also talk to the Gambling Commission, not least given past history of people betting on the dates of general elections, just to make sure that our systems join up and we ensure people are not using gambling markets, the betting market or some of the prediction markets as a way to influence rather than just bet, but, clearly, as a group of people who should not be betting on those markets at all.

I find that relatively reassuring, but should you not make an assessment of the risk of that unreasonably changing people’s views?

I will take that one away. It is a very good idea. So far, we have not done so, but I would like to talk to my counterparts in those other organisations and see if we should.

I would just like to come back briefly to the point you were making about algorithmic bias. In broadcast media, Ofcom will enforce due impartiality during the election period, whereby political parties and independent candidates must be given due weight across broadcasters’ TV and radio coverage, which clearly made sense when most people were getting their news from broadcast media. I am just wondering whether, during the regulated period, there should be due impartiality in terms of how a social media company’s recommended algorithms work during the regulated period in particular, whether that is YouTube recommended content or the political bias of content that is shown on Twitter or on Facebook.

I would defer to Ofcom in terms of whether that was impartiality, but we would take seriously any attempt by a social media company to deliberately influence what voters were seeing or to bias their information sources. We would want to identify it and report back. I do not think that there is anything in UK law at the moment that would apply to that situation, and so that is one that we would want to think about and come back with a proposal as to what could apply, and not just at election time. Between elections is when an awful lot of people’s thoughts and influences happen in terms of social media, but significant political bias in the information environment is an increasingly long-term worry.

I would just add that, as Vijay said, we want to build the evidence case to address this. After every election, we evaluate it from start to finish, essentially, and publish the report of that later. We will be looking at all influences on the elections in May this year and then taking decisions after that.

That would be helpful, because the data shows that, during the 2024 general election, social media was the second most common news source for a number of people. That is only going to continue, so Andrew’s question on that is interesting.

Among young people, TikTok makes up 69% of their basic data source.

I am struck that we have been talking about deepfakes only during elections. You only have to spend a bit of time on YouTube to look at the extent to which deepfakes are operating 24/7 with regard to what the Prime Minister is saying. It feels like we are missing what is happening right now. My line of questioning is to do with reforms to the conduct of elections. As our Chair said earlier to the previous panel, the Elections Act, in 2022, brought in a requirement for photographic ID, notionally to provide some confidence that the person who was casting their ballot was the person they said they were. Prior to that, in the last 10 years, say, how many people were convicted of an offence of personation?

It is a very small number—four or five. The number of people convicted of personation has been tiny. We have seen growing support for voter ID among the population. When it was introduced, it was controversial. I have just looked at the latest figures. Some 65% were in favour of voter ID in 2024. It is now up to 73% in favour. That is partly because people get a sense that it is indeed an element of security in the process and they want to trust the process. Through some very large communications campaigns, we have managed to get awareness of the need to bring voter ID high. It is around 90%, but it is falling off, so we are re-running those campaigns in the areas that need voter ID for the 7 May election.

Such a small number of people have been convicted of it. We heard earlier that 4% of people did not vote because of the requirement to have photographic ID. That is probably around 750,000 people, if that was replicated at the last general election. Is it proportionate to have introduced a measure such as this, based on the conviction of a single-digit number of people that, on the other side of the ledger, is causing perhaps 750,000 people not to vote? Is that a proportionate measure, in your view?

We have said that we would support keeping voter ID, because chopping and changing brings its own difficulties, but also expanding the list of IDs. Using the example of Northern Ireland in particular, where voter ID was introduced 20 or more years ago, it has now become completely accepted, and we see a very small number of people who cite voter ID as a reason for not voting. As time goes on, people get used to it, but there is a differential in the kinds of voter ID that are available, particularly for young people and for older people, and so we do support broadening that set of voter IDs, in particular into the digital IDs, as Ms Ellis said earlier.

Yes, indeed. You have expressed concern about the use of bank cards, as proposed by this Government, as a form of ID. What should the Government prioritise instead, if not bank cards?

We have proposed some of the photo IDs that meet the right security standards for young people. There are a couple of DWP cards and some travel cards that are possible. Digital voter authority certificates and digital voter IDs are really where we are going to be going at the moment. Our concern on bank cards is partly the electoral administration issues for polling station staff, which Laura Lock alluded to earlier, and partly because they are intrinsically less secure. We cannot always see the first name of someone. Chair, you said that you have one with an initial and one with a first name. It is much harder to use a bank card, and there is no uniform format at all. Having said that, they are very broadly held by members of the population, so it does solve accessibility, but there are better ways to do that.

The Government are piloting flexible options for voting in four areas at the local elections. When will you make your assessment of the success or otherwise of those pilots?

In common with the post-poll reporting we do normally, we will aim to get those out before the summer recess. I cannot guarantee that, but that is our intention at the moment.

The Government are piloting early voting and central voting schemes. Are there other forms of flexible voting that you would like to see tested?

We are very interested in them. Three areas in Tunbridge Wells are going to be doing early voting. That will be very interesting to see, because that is the largest one of those tests. Telephone voting would be very interesting to try. We have seen that work well, particularly for disabled and blind voters in Australia and New Zealand, who use that well. It is very secure. We can explain how it works, but it is possible to get a code in advance and phone up. There are two people listening to it, and they cannot identify the voter. That will be interesting, partly for disabled voters. There need to be some pilots on overseas voters, who get a very poor deal at the moment. If you are a perfectly legitimate overseas voter, it is very difficult for your ballot paper to get there and back in time. We would like to see further work—and we have asked the Government for this—to see how we can properly enfranchise legitimately registered overseas voters to be able to vote properly.

Finally, because we ran out of time five minutes ago, you have repeatedly called for the Government to consolidate, simplify and modernise electoral law. What are the risks of not doing this?

The risks are manifest. We are conscious that we have been calling to safeguard the system, so we are going to get legislative patchwork through this process. We are calling on the Government, after this process, to commit to begin the process of consolidation with a view to legislating in the next Parliament, understanding that it is a lengthy process. To give an example for voters and the way that it impacts them, if you want to understand how to register to vote in Wales, you need to refer to 10 pieces of interlocking UK and Senedd legislation to work out whether you should be on the register and how to get on it. That is just one example of how the legal framework is complex, and it has a real impact on accessibility and voter confidence as a result. Electoral administrators have a difficult time, but they are very expert in picking their way through that. They have been doing it for decades, but the fact is that this is part of the critical national infrastructure. When one starts to see the electoral law framework as part of the critical national infrastructure, as Philip Rycroft sets out, then one starts to see the threats and the vulnerabilities, and the ability to exploit it if it is as vague, complex and fragmented as it is.

Indeed, you have differing requirements for nomination papers and returns of expenses for town and parish council elections versus unitary authorities. It is a mess and all over the place, so it is that sort of thing that you think we should be consolidating and tidying up.

It is indeed, as well as looking at modernisation in the way that the Committee has explored with us today.

We think that this is the 27th Representation of the People Act.

Hopefully this legislation will see some consolidation. Thank you for coming before us this morning and giving us some thoughts that we will definitely come back to as we proceed with this inquiry.