Public Administration and Constitutional Affairs Committee 

Oral evidence: Review of the 2024 general election, HC 487

Tuesday 7 January 2025

Ordered by the House of Commons to be published on 7 January 2025.

Watch the meeting 

Members present: Simon Hoare (Chair); Richard Baker; Markus Campbell-Savours; Charlotte Cane; Sam Carling; Lauren Edwards; Peter Lamb; John Lamont; Mr Richard Quigley; Luke Taylor; Michelle Welsh.

Housing, Communities and Local Government Committee members present: Chris Curtis and Mr Lee Dillon.

Questions 1 - 80

Witnesses

I: John Pullinger CB, Chair, Electoral Commission; Vijay Rangarajan, Chief Executive, Electoral Commission; and Jackie Killeen, Director of Electoral Administration and Regulation, Electoral Commission.

Examination of witnesses

Witnesses: John Pullinger CB, Vijay Rangarajan and Jackie Killeen.

Q1                Chair: Good morning, colleagues, at this first meeting of 2025. I hope that you all had a good Christmas break, and a happy new year to you, one and all. This morning, we have before us the Electoral Commission. You are very welcome, and I will ask you to introduce yourselves and your roles in just a moment.

I want to welcome four new faces around the Committee. I will start with our visitors, Chris Curtis and Lee Dillon. They serve on the Select Committee for the Ministry of Communities, Housing and Local Government. As we know, that Department authors elections policy. Gentlemen, you are very welcome. Thank you for finding the time to take up our invitation. We are grateful.

We have two new permanent members of the Committee: John Lamont and Sam Carling. Again, you are both very welcome. Thank you for agreeing to serve. Colleagues and I look forward to working with both of you.

Let us now ask our formal, official guests to introduce themselves for the record. We will then turn straight to questioning.

John Pullinger: Good morning. I am John Pullinger, the chair of the Electoral Commission.

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

Jackie Killeen: I am Jackie Killeen, director of electoral administration and regulation for the Electoral Commission.

Chair: Thank you very much indeed for that. John will ask our first question.

Q2                John Lamont: Thank you, Chair, and good morning to the witnesses. Your report suggests that the 2024 election was well run. Could you tell me the criteria by which you judge that?

John Pullinger: Certainly. The first thing to say is that the 2024 general election was particularly challenging to run. I would cite three reasons why it was challenging. The first is the scale of the boundary changes that took place; 584 constituencies were on different boundaries. The second was the implementation of the 2022 Elections Act, which brought in a raft of very significant changes. I am sure that we will come to some of those in questioning later. The third, which you will know in particular, is the timing of the election, particularly the impact in Scotland where there were specific challenges because of the holiday period.

The running of an election requires the mobilisation of not just thousands of candidates and campaigners in parties, and outside parties, but tens of thousands of electoral administrators, staff of the Electoral Commission and others. The fact that it was well run should recognise the amazing hard work that all those people put in. I would like to take this opportunity to pay tribute to them for that work.

There are two criteria that we use to say that the election is well run. The first is that we ask the public. We had a large-scale survey of the public after the election; 83% of people said that it was well run, which compares well historically. The main reasons given by people who did not say it was well run or expressed concerns were about the first-past-the-post voting system; the feeling that there was not enough information about the candidates or parties; or the timing question, particularly in Scotland.

That is the first criterion, on which we had a positive response from the public. It varies a little bit by different demographic groups but, across the board, it is pretty high. The second criterion is based on the performance standards that we set for returning officers. We set and monitor performance standards across the piece. There were only two areas where there was a failure to meet those standards. In those two areas the concerns were about errors in the count. In neither case did the error affect the outcome of the election. We are following up with those returning officers to work through improvements that they could make, but that is a pretty good score.

On the basis of responses from the public, and our assessment of performance standards of electoral administrators, we concluded that the 2024 general election was well run.

Q3                John Lamont: Obviously there are lots of participants in an election, whether it is candidates, parties, the Royal Mail or the printers. As a candidate, I certainly feel as though we get lots of oversight from the Electoral Commission. I suspect that political parties feel the same. Are you confident that the other participants—namely, the Royal Mail, printers and returning officers—get the same level of scrutiny as political candidates and political parties?

John Pullinger: Levels of scrutiny? We have significant concerns in some areas. Some of the problems this time related to supply-chain issues, with contractors and others not being able to deliver quickly enough, particularly postal votes. In our report we highlight the need to really look hard and to scrutinise whether those systems are resilient enough.

For a long time, we have expressed concerns that the whole process is dependent on very small numbers of contractors, where there are challenges in them delivering against very short deadlines in a no-fail situation. Any one element of it that falls over is problematic. We heard, and you observed, that there were some problems in some areas and some near misses in several areas. We see our job as providing some of that scrutiny and highlighting it in our post-poll report, as we have done, but particularly in relation to printers we think there are significant issues that need to be addressed.

Q4                John Lamont: Do you think that could have been avoided if the commission had engaged at an earlier stage with the printers and Royal Mail? For example, in my constituency a Royal Mail postman was on holiday for a week, and the postal ballots were not delivered because there was no replacement. The printers were obviously not getting the ballots into the system fast enough. If the commission had engaged with them at an earlier stage, could those problems have been avoided?

John Pullinger: Maybe Vijay will say something about Royal Mail because he spent particular efforts with them, and maybe Jackie will say something about the printers and our discussions with the Government Department that has the responsibility in that area. Certainly, we have been expressing these concerns for some years. We have, ourselves, sought to engage. I think your question is well judged, if I may say so. We need to get beyond engaging, to actually making sure that the system has stronger resilience and is less prone to the kinds of failures that we are now really concerned about. Would you like Vijay to say something about Royal Mail at this point?

John Lamont: I suspect that may come up in further questions.

Chair: We will deal with the Royal Mail in a separate section.

Q5                John Lamont: I have one more point. You mentioned, under your performance criteria and the standards that you set yourself, that there was an issue around a count. I am assuming that you were flagging the Putney count in Wandsworth.

John Pullinger: Yes. Wandsworth was one, and Richmond was the other.

Q6                John Lamont: Did the commission have staff at the Wandsworth count?

John Pullinger: No, I don’t think we did.

Q7                John Lamont: At what point did the returning officer notify you of the problems about the declaration of the result?

Jackie Killeen: It was afterwards. We then provided them with advice.

Q8                John Lamont: What sanction has been imposed on Wandsworth’s returning officer?

Jackie Killeen: We will be working with them to make sure that lessons are learned. They didn’t meet elements of the standards. That was the conclusion of the performance standards assessment in that instance.

John Lamont: My question was in terms of sanctions.

Chair: I don’t want us to drill into that much detail. Let me turn to Richard Quigley.

Q9                Mr Quigley: I have a short question, but it will probably be a longer answer. What do you think the reasons were for non-voters having less trust in the election? Were they the same reasons that voters had?

John Pullinger: I think it is unsurprising. It is something we have seen as a consistent pattern. You see it in other areas as well. People who have experience of the system and have been through it have seen for themselves that it is effective. They have gone away and then said that. For people who have not, one of the reasons why they may not have voted in the first place is that they have concerns that it will not be well run, and that it is generally not for them. We are seeing increasing incidents of that. We might come to turnout later. We saw a dip in turnout on this occasion. The two things are bound together; people’s concerns about whether the system is going to be well run and their propensity to vote are tied up with the same set of issues in their minds.

Q10            Mr Quigley: Is there anything in particular? You mentioned first past the post. Feel free to go with what you think non-voters really thought were the key issues. You can pick up lots through the media, but was there anything that came out as a key message?

John Pullinger: There are the three things that I mentioned. We asked people who said they weren’t satisfied, “Why weren’t you satisfied?” The top three reasons were the system of voting, their perception that they did not have the information they felt they needed on candidates and parties, and, in Scotland in particular, concerns about the timing of the election.

Vijay Rangarajan: We do research more broadly about why people do not engage in the system and why they do not vote. In addition to the points that John mentioned, there is, “We don’t trust the system.” That comes through fairly loud and clear, not for a large number of voters but it is there. “We don’t like politicians. We don’t see why we should engage.” The biggest one of all is, “I’m too busy with running my own life.” The reason why most people say they don’t go to the polls on a particular day is that they have something else to do.

It is very understandable, but there is a part about engaging voters, particularly young voters. We see very sharp differences in the groups who give those answers. About half of young people are not on the register and do not think it is important to be on the register. They do not think that engaging in politics is important. We may come back to some of the education work we think is important later on.

Chair: At this juncture it may be helpful if, not setting the scene, I make this observation. What this Committee is very keen to lead the charge on, as it were, is this. We had a general election last year. The result was so clearcut that one or two errors, or mishaps, made no difference overall to the result.

We had a huge amount of change to electoral law in the last Parliament. What we are trying to do our best to establish at an early stage in this Parliament is the resilience of the administration of election law and its delivery to give robust, reliable results, irrespective of whether the overall result is a landslide for one party or another, or more importantly when it is incredibly tight and people need to know that the result is legitimate.

Quite a lot of what we want to look at this morning—I urge you to couch your answers with this at the forefront of your mind—is future-proofing a system where there are clear operational changes going on, but also usage changes as far as how the electorate engage in the election process.

Luke, do you want to come in on this question?

Q11            Luke Taylor: It is just to add an extra point that has only just occurred to me from the discussion of the question from John. You mentioned Wandsworth. I recall that in 2015 I was a candidate in Battersea. It was only because I was viewing the figures while they were being checked but before they were declared that I spotted that my result had been recorded incorrectly. The practice in 2015 was that numbers were written on a piece of paper and tallied up by hand. I was particularly interested because it meant that I was going to come in fourth rather than third, behind a Green. I spotted it and pointed it out to the returning officer, and it was corrected.

I am not saying it is Wandsworth specifically, but there seems to be a lack of consistent procedures in tallying or adding up. While the result being declared incorrectly is probably quite rare—or the winning result being incorrectly declared—there is a risk that, more broadly, errors are introduced into the system. There definitely needs to be a better system. Is that something you have looked into? Should there be more consistent procedures within the count to tally results and to add up various ward figures, for example?

Jackie Killeen: We have been working on improving the guidance that we give returning officers on managing the count and all aspects of the count. We review it after every election to make sure that any issues that have come up and have been identified in one election are taken forward into the guidance for the next. We have tried to improve and tighten that guidance year on year.

Where there is a known issue in one area, it will be factored into the degree of engagement and support we have with that administration team in the run-up to, and during, the next election. We have teams working with every team of administrators across the country. We do a risk assessment based on their previous performance, what we know about what is happening in that area and any issues that have arisen in a previous election. We work closely on those to make sure that, as far as we are able to ensure, those errors do not recur the next time round.

Q12            Chair: Jackie, you just used the word guidance.

Jackie Killeen: Yes.

Q13            Chair: I can provide guidance to all sorts of people. Whether they take the guidance is entirely up to them. I think this is germane to a general election and less so for local, because there are differentials there. For a general election, should there be a uniform way—a more robust way—than merely guidance? It is a national election covering the whole of the UK. Should people who sit in this place know that our counts are all conducted in exactly the same way; that the criteria for recounts is exactly the same; and that how things are tallied and recorded is exactly the same? Underlying that question is picking up what Mr Taylor and others have been talking about. Is exhortative guidance sufficient?

Jackie Killeen: The two main tools that we have are the guidance and the resources that we provide for returning officers in relation to counts, and then also the performance standards, where we can judge whether the standards have been met, including in relation to counts. The guidance that we give is consistent.

Q14            Chair: I’m not querying whether the guidance is consistent. Moses issued the ten commandments, not the ten guidances. I am not talking about the post-event review, but the conduct of the actual event itself. Should it be stronger than guidance? Should there be a rubric that all returning officers follow at a general election, and if so, who authors that? Is it you, as a commission, or is it the Government Department through legislation in order to have guaranteed parity of quality?

John Pullinger: The guaranteed parity of quality is certainly something we need to have. The performance standards framework is designed to give that. It creates an objective test that we set, monitor, assess and report on.

In terms of your question on whether we should do something about that, our post-poll report highlighted the things we need to do something about. There is quite a list of those. This is something where we think the current system has delivered successfully for us.

You are right to say that now is the time to look at risks that might materialise next time, when there might be a much closer result. We had a lot of close results this time as well. We have been tested in individual constituencies in this kind of situation. Our conclusion from this time is that there is not one single result that was contested. There were only those two examples of a failure to meet our performance standards, so we do not think this is an issue that we should see as a priority for making changes. The comments that you and colleagues have made reinforce the significance of our performance standards and the importance that we should be placing, and do place, on assessing and assuring ourselves that they are being met uniformly across the country.

Jackie Killeen: Perhaps I could add, John, that over the course of our next corporate plan period we will be doing more work on the performance standards, to make sure that we use them as effectively as we possibly can and draw out all the learning that comes out of our assessment of how they have been met at a more granular level, so that we can make sure that we are tackling any issues that might be at a lower level at this stage to prevent them from becoming more significant issues down the line.

Q15            Chair: Were there to be “Thou shall, thou shan’t” commandments—the rubric for delivering a count methodology—who would author it? Is it you as the commission, or is it the Government?

John Pullinger: Under the set-up we have at the moment there is quite widespread acceptance across the system that we are a suitable author for the guidance. Certainly, in our post-poll report, we tested whether people feel that the guidance we issue is helpful and relevant.

Q16            Chair: I am talking not about guidance but about an actual tick-box rubric which everybody is following to give that confidence. Many of us would say—I would probably agree with you, Mr Pullinger—that our electoral system has held up to pressures and challenges over the years since 1832. Casting forward, it is about resilience and robustness at a time when society becomes more questioning about the efficacy of officialdom to deliver on their behalf.

Vijay Rangarajan: The reason why we are not giving you a completely clear answer is that it would depend exactly on what we were mandating. Certainly, there are some things we can indeed mandate and put in the performance standards or the polling station handbook and say, “You need to stick to that and there is no room for discretion.” We allow returning officers a large amount of discretion in general because they are the people responsible for delivering the election in the legislation, and because every constituency is very different. There is a need for quite a lot of discretion. The question we are touching on is: how far should there be a difference?

Q17            Chair: But it is not, is it? Constituencies are not different. Constituencies are made up of people who qualify to vote, and who vote on exactly the same day or during exactly the same period. It makes no difference if you are urban or rural, in the north-east of Scotland or the deepest south-west of England. It is a general election. I would take the point for a local election, but for a general election I remain to be convinced.

Vijay Rangarajan: We would say that the rules are similar, and we would try to enforce them overall. For example, the activities of a returning officer running a very large and very rural constituency will be quite different from how they give their count or how they assemble their ballot papers in a much smaller and denser urban constituency. There are some areas—I think we are touching on the detail of electoral law—where we would like to see significant modernisation and simplification of electoral law. That is why, in some cases, it would be Government and Parliament legislating who would need to make some of these changes. For example, it is very difficult for returning officers to use of some of the electronic systems in polling stations because of the way that electoral law is written and has not changed in those areas for 25 years.

On Mr Taylor’s point, we have seen significant variation in the way counts are done. Some are done on paper. Some people use Excel spreadsheets, and some use electoral management software. Several of those fill me with worry, not least given the cyber threat and the possibility of error. Our systems are baked in, so there are many eyes on every element. That is good and gives a degree of resilience and testing, but we can do better in modernising some of those systems, not least for information for candidates and for voters, just to make sure that our systems are more robust. As Jackie said, that will be a theme of our coming corporate plan—bringing some of those systems into at least the 20th century.

Chair: Gosh, that really is a “hold on to your hat” sort of thing.

Q18            Mr Dillon: To bring us back to voter turnout, in particular the turnout in the last general election, which got close to the record low turnout of 2001, could you expand on what you think the reasons are for that low turnout in this general election specifically, rather than generically?

Vijay Rangarajan: In general, turnout fluctuates, as you know. It goes very up and down. We are less concerned by the individual turnout in a particular election. But this was low, down at the 60% mark. It was interesting that turnout seemed to be lower in constituencies where people were more confident of the outcome. There is a degree of contestation driving turnout.

The thing that worries us overall, and much more, is the long-term secular decline in turnout. Going back over 20 or 30 years, we have seen turnout fall very steadily on average—again fluctuating by election—up till now. That is coupled with a degree of alienation from the process of voting, from the political system and a feeling of “Why vote?” and seems to come through quite strongly in the public polling. It is that rather than the specific turnout. It is a marker; it shows it has gone down yet again, but it is the long-term secular decline over 20 or 30 years that is a real concern for us.

There are things we can do. We have been running some very large campaigns. We have been trying to assure people that, for example, your vote is yours alone, in the campaign, “Your Vote is Secure.” We have been running work based on voter security. Clearly, we want to step up the work we do on education in order to try to help on this. All of that will help. We also need to work very closely with the political parties, who have a massive role to play in voter turnout. It is a shared challenge, and it is one that we will need to try to reverse.

Q19            Mr Dillon: The Electoral Reform Society has some research on voting systems, linked to turnout. Countries that have a PR voting system have a 77% average turnout, compared with 69% for non-PR. Do you have any views on that at all in relation to the system we have here and what that does to our turnout?

Vijay Rangarajan: Looking at the different electoral systems within the UK—where we are a bit more comparable—we have not seen a strong link between the voting system and turnout. We have seen a link between the importance of the issues being raised and contestation of particular seats and particular elections, where people are energised to turn out because they think their vote will count more. I am not sure we can draw a direct link to the voting system itself from that.

Q20            Chair: If you look at Scotland and Wales, where obviously there is a mixed element of election for Senedd and Parliament, the turnout figures do not indicate any wild differential between them and Westminster.

I would suggest that the declining number of people who are engaging in political processes in mature Western democracies is not a unique phenomenon. Is there a country or sister organisation that you could look to and say, “That really worked to help drive turnout and reignite interest and engagement”?

Vijay Rangarajan: We have been looking to and working closely with our colleagues in the Australian, Canadian and New Zealand electoral commissions. They are parallel and face very similar challenges across the entire range of issues.

Q21            Chair: Australia, of course, has mandatory elections.

Vijay Rangarajan: I was going to say that. Rather than going all the way to mandating voting and registration, there are ways in which the Australians, in particular, have managed to reduce some of the barriers to voting. Those have been partly in automatic registration. They have a very advanced and sophisticated automatic registration system, which, if my memory serves me right, is delivering them high 90s levels of both completeness and accuracy, which is remarkable.

We have seen, for example, on polling day a spike in people wanting to register to vote. They have woken up and thought, “Oh, there is an election, and I do want to vote but I’m not registered.” Being registered in advance might help those people. It is not a gigantic number, but it would help. There is a number of different interventions one could do to help the registration system, which means people could vote.

The Australians also have, as do the US elections, quite an advanced set of ways in which people can vote through early voting in different areas to make it more useful. Again, we would be interested in exploring that. Some might work. Some might not work in the UK. What tends to happen in the UK is that people use postal voting as a way of more flexible voting. In this election that has driven an absolutely record number of postal votes. We had a tremendous number, which put the system under some strain. It was the sheer number going through Royal Mail and that returning officers had to process.

Returning a bit to Mr Lamont’s question, elements of the timetable and the calendar were very compressed. We are looking, and have made recommendations, at the calendar for postal voting in order to give a little bit more time. We think that people will continue to use postal voting at a very significant level, which is 20% of all voters at the moment.

Q22            Mr Dillon: You referenced the postal voting in America. There, the postmark, as long as it is on election day, gets counted, but they have an elongated time after the election before an Administration takes over. They also have mail-only ballot boxes on the streets. What percentage of postal voters would we need in this country before we would have to look at the whole system of how we deliver postal voting, rather than relying on the traditional Royal Mail delivery network?

Vijay Rangarajan: I would say that we should be looking at that now, not least because of the issues we had with the timetabling and the printer timetables interface at this election. More broadly, it is to maintain trust in the system. We asked voters, “Are you confident there isn’t fraud, and did you experience any pressure in your vote at the ballot box or in postal voting?” We are not seeing an increase in fraud levels or in concern about postal voting, which is good.

There is a broader issue which, I think, is shared internationally about the way that postal systems are evolving. They are not getting stronger. The challenges are growing. In the end—this is something we have been discussing with our colleague electoral commissions—how do we manage to improve postal voting and maintain it? Early voting may be part of the answer. That was the part I was referring to. We are in quite close contact with Royal Mail about how they can help. To be honest, there has been a very good discussion with them about the changes that are happening, with their new chief executive and their excellent elections team, who did a very good and augmented job for the general election.

Lauren Edwards: I was born in Australia, so I declare having experience—

Chair: I am not sure that is a declarable interest.

Lauren Edwards: —of that system, which I think is a very good one. I know I am a lone voice in being supportive of mandatory attendance at a polling station or mandatory voting, where you can just go along and mark. One of the other reasons is that it is quite a cultural event in Australia. I hazard to suggest that barbecues at polling stations are incredibly popular and certainly work to bring people out.

Chair: That’s called treating the electorate, isn’t it?

Lauren Edwards: No, because it is run independently by charities.

Chair: Independently treating the electorate.

Q23            Lauren Edwards: It is a community event. There are definitely no party political votes for sausages or anything like that. I was going to ask about one of the things they do in Australia that you did not mention, in terms of looking at the way that elections are administered and how changes can be made to increase turnout, and that is the day of elections. I have always been slightly perplexed about why there is an insistence to have one in this country on a Thursday, when it seems to disincentivise people who want to go to work. We obviously have lots of issues around schools being open, which limits the number of polling stations that we can have. Do you have any particular view, or any objections, as to why it has never been proposed for moving to a weekend, or a Saturday, when countries like Australia operate on that basis?

John Pullinger: I don’t think we have any objections to it. You talk about traditions in Australia. This has been a tradition in the UK for as long as I can remember. I don’t recall the commission doing any significant review of it over its lifetime. It is not something that has been raised. In terms of Mr Hoare’s description of the position of this inquiry, thinking about things that might be done that would improve the voter experience and make it more likely that the elections will be well run, the day on which the election is held is a relevant thing to think about.

While I have the floor, and linking your question to Mr Dillon’s earlier question, we have increasingly good experience of different approaches, from Wales in particular, but shortly also in Scotland, of looking at alternative ways of doing the election. We evaluated a series of pilots from experiments done in Wales that looked at early voting, having voting booths in further education colleges, and a variety of things like that. We have some experience of testing those things. There is opportunity in our system to run pilots if there are areas willing to try things in a different way that will then help us get the experience to judge whether a wider change might be possible.

Q24            Lauren Edwards: Have you had any discussions with the Government around doing a pilot approach? Would you need to engage central Government, local authorities or both?

John Pullinger: Both, yes. We have not, no. For the reasons that I have said, this has not been something that has been raised particularly. The questions that you raise come up quite often, but we are trying to highlight things where we know there is a specific challenge that we experienced in the election. Those are the things that we put in our post-poll report. In the 2024 election, the timing of the election in July was a problem. The fact that it was a Thursday was a much lower one.

Q25            Sam Carling: Earlier, you started to talk a little bit around disengagement in young people and them not feeling that politics was important in general. Your report says you have a view that “high quality education about democracy and elections” is important to encourage particularly young people, but people in general, to vote. Could you go into a bit more detail about that?

Vijay Rangarajan: Thank you. First of all, the data shows quite clearly the disengagement. I won’t go through all the numbers now, but essentially for 16 to 25-year-olds only about half think it is important to be engaged in politics. All that matters is when it is affecting their lives.

We have very interesting differential data, obviously, in Wales and Scotland, which both have had votes at 16 for some time and have been doing voter education in a different way. We recently convened 40 civil society organisations, a very broad spectrum around England, and have submitted some evidence to the Department for Education in their curriculum review on PSHE and how that should be done.

To briefly summarise the views that we put in to them, first of all, there is reticence by some teachers to teach about politics because they think they are getting into party political territory and parents may be unhappy about that. We definitely think that we and others, like the Association for Citizenship Teaching, can help with that by providing material that is neutral, certified, tested and shown to the political parties to help to teach it in schools. There is something important about showing, explaining and working with young people to show why their vote really matters, and that it is not a theoretical process where you go and do something strange but it matters to your life, and to bring that to life rather more.

There are two linked, very large, issues where we think an increased focus on education, right the way from about 11 through to sixth-form college, is needed. The first is mis and disinformation and digital literacy. That is something we could bring out more as part of all of this. “Don’t trust everything you read on social media. You’re going to see some stuff that isn’t true. Judge it in this way.”

The second—where you and many of your colleagues will play a central role—is in making politicians human to young people. Again, we will publish this data. We have one very worrying graph. When we ask members of the public, “Is it okay to abuse someone in the street?” and as a second question, “Is it okay to abuse a politician?”, if you are over 50 or 60, 97% of them say, “No, neither is acceptable.” I am afraid that when we get down to the 20s, it is still pretty unacceptable to abuse a member of the public, but only 60% of them think it is unacceptable to abuse a politician. There is a secular change all the way down the age gradient.

We will no doubt touch on the abuse and intimidation issues coming, but this is an issue which could get significantly worse. Part of that is about information. Part of it is simply about contact with MPs, with councillors and with candidates coming into schools and, again, helping and enabling that as far as we can. Part of it is about helping young people to actually vote—to go and get used to the process.

One thing we are fairly sure of, and the data is pretty supportive, is that voting is habit forming. If someone has voted by the time they are 16 or 17, they will continue voting. I have heard from many of the political parties that if they knock on a door and someone is over 35 and has never voted, it’s not unusual for them to think, “I’ll go and spend my time elsewhere because they are unlikely to go and vote.” There is something about education and bringing candidates and politicians into schools, changing the citizenship curriculum and spending a bit more time on this which could have real democratic and long-term benefits. I am sorry that is a long answer, but we are going to make it quite a theme of our next corporate plan.

Q26            Sam Carling: That is really helpful, thank you. The statistic about the proportion of younger people who think it is okay to abuse a politician is really shocking. Often, in the public discourse around that, you don’t picture a young person as a typical person who would abuse a politician. It is quite interesting in terms of where that is coming from. Would you say that lack of education awareness about democracy is a problem in other age groups as well? Is there anything we can do about it?

Vijay Rangarajan: Yes. I shouldn’t have given the impression that we would only be focusing on young people. There is education and things we need to do with a series of groups who face particular barriers. To give you an example, we have been doing quite a lot of work, but would like to do more, on disabled voters and their access. There is a perception that their access is difficult. Many of them, as we have seen, know that they can go to a polling station, but they do not realise the accessibility help that they could get in a polling station. Again, there is more communication and information. We are doing that primarily not through our comms channels or media work but in conjunction with various civil society organisations with good links with disabled groups.

Private sector rented accommodation is another very under-represented group when it comes to voting. Again, there are things we can do. Some of the work on registration would particularly help that, because they tend to move quite often and not keep their registration updated. Some ethnic minority communities, the Gypsy Roma Traveller community in particular, and people who are unemployed, fundamentally find it harder and are under-represented. There is no one size fits all on any of this, but each takes a different form of communication, education and then a campaign approach as we approach an election.

Q27            Sam Carling: I have one more question, if I may, Chair. The misinformation and disinformation point that you mentioned is something we have had a lot of discussion about recently. Can you speak a bit more about the action the Electoral Commission took in advance of the election in that space?

John Pullinger: One thing that I did a year or so ago was to convene all the other regulators that have some experience in this space—Ofcom, the UK Statistics Authority, the Advertising Standards Authority, the Charity Commission and a few others—to try, first of all, to work out who was doing what. We also created a hub that people could come to, which the Electoral Commission ran with the support of the others. It provided support for voters on how to handle mis and disinformation when they came across it.

We now have an ongoing relationship with those groups. We meet very regularly and will continue to meet from hereon in. As the landscape evolves—we have not mentioned generative AI yet in this discussion—over the next few months and certainly over the next few years, it has the scope to significantly transform the way in which we receive, interpret and trust, or do not trust, the information that comes our way. We need to keep across it, but it is not something that the Electoral Commission can do alone. In fact, we do not have any statutory powers around the content of campaign material. Others have the opportunity to speak, and we certainly have the opportunity to convene around it.

The core element, as Vijay described, is helping people to be equipped to challenge information when it comes their way, so that they are sceptical about claims that do not look like they can be real. They may well not be. We have been asking people whether they felt that they came across examples of mis and disinformation during the last general election campaign. A lot of people did, but the responses were, “Well, we ignored it,” or “We didn’t believe it.” They were generally quite encouraging responses, but we know that the landscape of AI mis and disinformation feeds into concerns about trust, particularly in relation to social media.

One of the stats that really impacted me when we looked at that was when we asked people about their trust in the system and the source from which they get their understanding of the news and the world. There was a significant association of people distrusting politics whose primary source of news was social media. Again, there are issues that we need to think very carefully about. As implementation of the Online Safety Act goes forward, there is going to be a lot of experience in Ofcom about that and we are working very closely with them to monitor what is happening. We also work with institutions like the Alan Turing Institute, which did a very nice review of the impact of mis and disinformation on the election. It was generally reassuring this time, but, looking forward, we should be much more concerned.

Going back to the Chair’s statement at the beginning, we need to be thinking about what we need to address between now and the next general election. This is certainly one of those things because it has the scope to undermine people’s confidence in the system. That is the key criterion by which we should judge ourselves.

Q28            Chair: Mr Pullinger, what speed are you putting on this? If stuff requires legislation, you know how slowly the legislative machine grinds in this place. The general election, let’s presume, will be in May 2029. It is not that far away. Is this a piece of work that, with recommendations, gets completed within, for example, the next six months?

John Pullinger: The first thing to say is that recommendations in this space are really difficult. The discussions that took place in the last Parliament around the Online Safety Act demonstrate how difficult they are. The key thing that we need for a democracy to be and remain healthy is freedom of speech; the ability for people to have opinions, and to be respectful of opinions that are not their own. However vehemently put, we must have respect for that. Navigating the boundary between trying to prohibit something deemed unacceptable while at the same time respecting freedom of speech is really difficult, and will require significant parliamentary debate.

On your question, it needs to be done at pace, but thoughtfully, if we are not to have unintended consequences. Parliament is debating it, because it is not just about elections. It is about all kinds of aspects of our public life, and people’s trust in all kinds of issues. Vaccinations would be the most obvious analogy that you might draw. How do you successfully navigate the maintenance of freedom of speech against the corrosion of a system that we all need to uphold? It has to be very careful.

Countries that have tried to step into the space in the elections arena have had some difficulties. Australia is an example that has attempted it. Ireland would be another one where they have looked into it. We are looking at the experiences of other legislators and trying to bring them into our own evidence. We are certainly looking at it at pace, because it is changing at pace, and if we need to do something before the next election it will need to be done sooner rather than later.

Chair: I am conscious of time, and I have a long call list of people who want to take part in our proceedings. John wants to come in. Then I will turn to Markus, and then we will have a canter round the issue of postal voting.

Q29            John Lamont: Thanks, Chair. My question is about where people get information about the activities of MPs—less at election time, although it comes up at election time. There is a website called TheyWorkForYou. I have spoken to you about it before, Vijay. It purports to tell voters how MPs have voted on different issues. I am sure that if you go to my profile it says I voted to pump sewage into the water, take free school meals away and various other things. I had a very abusive constituent on the doorstep during the election campaign who was adamant that I had voted against same-sex marriage, despite the fact that as an MSP, along with Richard, I voted and spoke in the Scottish Parliament in favour of same-sex marriage when it was being debated. Because of some obscure amendment to some Northern Irish legislation they had said I was against same-sex marriage, which caused my constituent, rightly, a lot of anger.

What control can you exercise over websites like that? It is more of an issue for Government MPs than for Opposition MPs, because in government you have to defend Government business and, often, to take difficult decisions, as I have discovered over the last few years, but there needs to be some control over the types of information that voters are given in allowing them to inform themselves more accurately than they do now, particularly for that website. I have raised this directly with the website, and they refuse to budge on how they report my voting record, despite the fact that I was able to provide clear evidence of how I had voted on that issue in the Scottish Parliament. There are other examples and I am sure that other MPs have had different, or similar, experiences with voters as a consequence of information on that website.

John Pullinger: It would be difficult for the Electoral Commission to have influence in that particular example, because our remit is tightly around elections. I spent 10 years being responsible for the parliamentary website, back in the day. TheyWorkForYou was setting itself up then. There is certainly a role for the parliamentary website to make sure that there is clarity on the statement of record, and that people can interrogate it effectively. The parliamentary website has come on in leaps and bounds in my time. In its original conception, TheyWorkForYou was filling a void. A group of enthusiasts was creating something, but they were meeting a demand from people for accurate information. You clearly have a concern, which I know is reflected among Members, that they are not fulfilling a public service by giving accurate information about what MPs say—

Q30            John Lamont:  It is the opposite.

John Pullinger: The key is to make sure that there is a source where people can go.

Q31            Chair:  It is context—contextualisation.

John Pullinger: Yes.

Q32            Markus Campbell-Savours: Vijay touched on this before: the Elections Act introduced new accessibility requirements and I wondered if you could tell us more about how they have worked in practice. Correct me if I am wrong, but I got the impression that they caused challenges for some of our councils in identifying suitably accessible polling stations, which may have led to some reduction in the number of polling stations, or their consolidation, or to polling stations being located where they might not have been as accessible for some people. You mentioned the perception of what is available to people. In some cases, is just having a polling station close to them not the most important thing for people with accessibility issues?

Vijay Rangarajan: Shall I give a little bit of the picture? We have been pushing on accessibility for quite a while. That sometimes means changes to where polling stations are, or making sure that they are accessible for anyone with, say, a mobility need. Of disabled voters, 96% said it was easy to get inside a polling station and vote, so we are already quite pleased. That is quite a good statistic. However, only 49% of disabled voters agreed that information, equipment and support was available to them, so we think more needs to be done about support for disabled voters.

Across the whole of the general election there were only 117 requests for specific equipment. That is a remarkably small number. You can ask in a polling station for someone to help you with whatever your disability is, but we interpret 117 requests as meaning that people did not realise they had the right to ask. Administrators mostly found our guidance on that useful, but we continually talk to them about what more could be done. The assistive technologies are changing all the time, which helps a lot. Overall, it is not a bad picture at the moment, but there is more that we can do to help administrators and, I think, to help voters with any form of disability to vote safely and securely.

One of the complaints that we heard at this election was about people being helped to vote by, say, a presiding officer, and not feeling that their vote was secret. There are ways, particularly for blind voters, to use assistive technology that will help with that, and we would like them to be rolled out more broadly.

Finally, there is tension between having a polling station very close to you and that polling station being accessible. That is something that returning officers have to navigate carefully. You can put it—I think there is one—in the front room of a pub, but it may not be easy to get into, and some people might be put off by its being there. There is a balance about where they are.

Q33            Markus Campbell-Savours: You mentioned some stats on how many people felt that access to the polling station was achievable. What was it before? Prior to the changes, was there a big issue that they were designed to deal with?

Vijay Rangarajan: Yes, there was an issue with disabled voters’ perceptions in the last decades that, first, it was difficult to get in to vote and, secondly, there was very little help for them. Thirdly, there was a lingering perception, interestingly, among voters with mental disabilities, that the old rules on peers and madmen not being able to vote somehow left a social tinge and that they were not welcome voters. We have been working with Mencap in particular, as well as Mind and several other charities, to get the message across that they are indeed utterly legitimate voters who should be there voting, and that returning officers have an obligation to help them vote. There are a number of different groups to work with. Jackie may want to talk more about the experience of administrators.

John Pullinger: I was just going to highlight, if I may, Jackie, that a key recommendation in our report is that more needs to be done for disabled voters. They continue to face challenges, but the Elections Act 2022 made significant improvements. On some of the issues with polling stations, two things are happening together. Returning officers are finding it much more difficult to find polling stations, not just accessible ones. It is reported back to us that that is becoming more and more challenging.

The improvements for disabled voters this time were twofold and quite important, and appreciated. The first was making it much clearer who can be a companion to a person who needs someone to assist them when they go to a polling station to vote. That has been taken up and well supported by many.

The second thing is making it much clearer what sorts of support people can expect in the polling station. Blind and partially-sighted voters are a group who face challenges, because of the nature of completing the ballot paper. The responses we had from the RNIB, the Thomas Pocklington Trust and other organisations that work with blind and partially-sighted people have been encouraging. “Progress, but more to do” is, I think, where we would put this, but we would particularly focus on blind and partially-sighted voters, and people with learning difficulties. Generally, our guidance for polling stations is about making sure that the basics are there—ramps, doors, the ability to get into the building, and helpfulness of staff. Those are the basics that people need to be able to cast a vote in comfort and in secret.

Q34            Chair: Mr Pullinger, you said en passant that returning officers are finding it harder to find polling stations. Why?

John Pullinger: Jackie, do you want to come in on this?

Jackie Killeen: They have to balance a set of considerations—availability, familiarity and accessibility. It is the ability of the potential polling station venues to meet all those needs. The timing of an election is an additional factor to consider, which adds to the set of issues that administrators have to weigh up when planning how to deliver an election.

To add, if I may, to what John and Vijay have said, based on our experience from the May and July elections, when we saw the new accessibility measures being implemented for the first time, we think there are probably more things that can be done. We would probably want to emphasise that more information, both offline and online, should be available to disabled voters and voters with accessibility needs. One helpful thing could be to include it on polling cards, for example, so that people who might not be online could request, in advance, information about accessibility measures. That might reduce one of the barriers that prevents people from going in the first place. The other thing we will do is to look at whether our voter information tool can be enhanced to provide better information on accessibility, as well as the location of polling stations. We will continue to look at how our own guidance to returning officers and administrators can be improved.

Chair: Richard Baker, very quickly.

Q35            Richard Baker: Very quickly, I declare an interest in that I used to work for the learning disability charity, Enable, which worked with the commission to provide information to people with disabilities, particularly on requirements for voter ID. Vijay, you mentioned the work of the RNIB in this area, and their report, “Turned out”, which found that 73% of respondents did not know that they could request reasonable adjustments. Building on your point, Jackie, isn’t it vital to get more information out to those people, in the appropriate accessible formats—not just for blind and partially-sighted people but for all those who require alternative communication formats?

John Pullinger: Our primary recommendation is increased awareness, and working through other organisations, as well as returning officers, and, through our own work, to double down on getting the message out there.

Q36            Chair: Let’s turn to postal voting, with a first question to establish a principle, I suppose: is the postal voting genie now out of the bottle? It used to be a quite rarified commodity with very tightly drawn reasons for requiring one. People can now have a standing postal vote. Given the drive for accessibility, and potential changes in the electorate and the franchise, give us your thoughts on the operation and resilience of the postal voting system. It is coming under increasing demand and has greater regulatory complexity. A principal deliverer, Royal Mail, which is coming under new ownership, is generally dealing with a lower volume of daily envelopes and so forth, so that its workforce patterns are changing. What of Royal Mail’s resilience with regard to delivering, in a meaningful way? Allied to that—going back to the point about uniformity and a general election—is the fact that we now have general election days, and vagaries in when and how postal votes are issued, that can affect outcomes, depending on what the dominant issue, scandal, policy announcement, or whatever may have been. Tell us something about the timing and the future resilience of a service that seems to have increasing demand and a more challenging environment in which to deliver.

Jackie Killeen: We absolutely recognise your description. The data would indicate that, yes, the genie is out of the bottle. Over the last couple of elections we have seen a growing trend towards postal voting; 20% of the electorate registered for a postal vote at this election, up from 18% in 2019. This was also the first UK general election at which people could apply online for a postal vote, and 1.3 million people did that after the election was called, which in turn put a huge additional strain on administrators. If there is ease in a facility for voters, they then expect the whole experience to be equally smooth; but the plumbing behind that online front door is not quite the same as with that first experience. While 89% of postal voters said that they were satisfied with the process, we are all aware that there were definitely some issues this time around. A number of people did not receive their postal votes in time to complete and return them before polling day. On this election, I would probably add—

Q37            Chair: Just pausing there, is your hunch that, as an issue of importance, that may have been affected disproportionately because it was holiday time, and a particularly popular one for people without kids, who might go away before the great onslaught of buckets and spades at beaches? Is it therefore a distorted figure, or is it part of a trend?

Jackie Killeen: You could say that it is part of a trend, but we think that that trend was probably assisted in this election, particularly in Scotland, and probably also in Northern Ireland, by the timing of the election. The schools had already broken up in many local authority areas by the time the election took place, so the increase in postal vote applications in those areas would have been higher. It is impossible to say that with complete certainty, but it would indicate, and we became aware of, a higher number of applications from Scotland, and from Northern Ireland, although they have a different way of applying there, so it was not quite the same.

Q38            Chair: Jackie, you mentioned Scotland: I am trying to get a handle on whether the data is reliable or whether there are freak things that distort it, but my understanding is that as far as the print contract for Scotland was concerned, the person who let the contract to the printers did not tick the priority print box, so that the Scottish ballot papers were just the next job off the taxi rank, rather than being triaged because there was a set of prescribed dates, and that that added to the problem in Scotland, with the school holidays and everything else.

Jackie Killeen: There are a couple of things to say on that. One is that people probably do not understand the mechanics of what happens when you apply for a postal vote. They are not printed on demand. The printers print in batches and, in this instance, they printed in two batches. People who applied early after the election was called and whose postal votes went out in the first batch would have been far more likely to receive their votes in time. People who applied later and whose postal ballot was included in the second batch were more likely not to have received their vote in time.

In Scotland, there were absolutely issues with the fulfilment of some of the print contracts, particularly a production issue quite late in the day. We then saw administrators and others pull out all the stops to try to make sure that ballots were printed, fulfilled, returned and delivered in time. Arrangements were made by a number of local authorities to allow people to return their postal votes in particular areas, but it highlighted the fragility that we mentioned earlier in the supply chain and supply market for postal votes in particular—that behind-the-front-door experience.

We reflected on that, and we think there could be a couple of improvements. One is improving the information given to voters both before and after they apply for a postal vote—what to do next when they apply online—so that they can understand whether this method of voting is going to work for them or whether, perhaps, an alternative method such as proxy voting might be more suitable. We would also recommend, and have recommended, looking at the rules for reissuing postal votes. At the moment, they can only be done four days before polling day. That does not give administrators much flexibility to step in and help when something like that has gone wrong. We recommended looking at the funding available to make the system more robust overall, so that administrators can deliver the system that voters expect to experience when they apply for a postal vote.

Q39            Chair: Given that the uncertainty of timing of a general election has returned since the abolition of the Fixed-term Parliaments Act, have you issued guidance to returning officers who are letting contracts to print the packs that they should ensure that they have a priority claim on the printers’ work dockets?

Jackie Killeen: The relationship with the printers sits more with Government, but we have been reviewing how we can focus administrators’ attention on the issues and challenges, and how they make sure that they address those in a timely manner. We have reviewed a lot of our guidance, as you would imagine, post poll, and at the moment we are working on the polling station handbook for the next election, and on other pieces of guidance. Yes, we are absolutely looking at how that can be strengthened, in so far as it is within our locus to do that. Is there anything you want to add, Vijay?

Vijay Rangarajan: One minor thing: one thing that confused voters was not being able to hand in a postal vote to a council office or directly to a returning officer, and having either to put it into a Royal Mail post box or take it to a polling station. We may want to look at that. Obviously we need to think carefully about the numbers and potential for fraud, and so on, but it could be easier to be able to hand in postal votes. That was a change that was made and that we may want to review.

Jackie Killeen: Could I add one last thing, Chair?

Chair: Yes, please.

Jackie Killeen: On your point about future-proofing, we would say that the trend is likely to continue. If we also see an extension of the franchise so that younger people are included, we need to invest in this system and address the issues in it overall, so that it is fit for purpose in the future. That is also in the context of what Vijay referred to as declining postal systems globally, which probably leads into some of the overseas voting issues that you may come on to separately.

Q40            Chair: That review: when, by whom and how?

Jackie Killeen: It is probably a number of us together. We absolutely want to contribute to that, but it is probably for Government as well as the suppliers and administrators.

John Pullinger: We are already discussing that with Government—how we get it going with some urgency so that improvements are in place by the next general election.

Q41            Richard Baker: Given the situation in Scotland, which affected John’s constituents, as he said earlier, and people in my constituency who, on polling day, did not have their postal ballot, I think that the Chair’s question is very important. You said that there was responsibility in government in Scotland to ensure that the contract was prioritised and that the supply was there, in terms of printers and suppliers, so that the votes could be sent out. What discussion has taken place to ensure that that decision is not repeated in the future, that the appropriate priority is given, and that people can rightfully expect that within those parameters they will receive their postal votes in time and, yes, hand them back in, if councils make that facility available? Not all councils did. Obviously, preferably they would send them in by the normal process.

Jackie Killeen: We discussed this with the Electoral Management Board for Scotland, which has a locus in working with administration teams in Scotland. Review of the contractual arrangements, the resilience of those contracts and making sure that they are fit for purpose for the next Scottish elections as well as the next general election is top of everybody’s agenda on this, to reassure you.

Q42            Chair: Thank you. The Royal Mail and resilience. Linked to that is that we have a tendency, don’t we, to take a sedimentary approach to these things? “We have voted this way since whenever”; then something else changes. I think I know Vijay’s answer to the question, which is that we slightly retrofit and keep our fingers crossed that the system will somehow rise to a new challenge and deliver. Everyone will be knackered at the end, but they will have delivered. There is an argument, isn’t there, that effectively says, let’s have a fundamental scrapping of everything that we have at the moment, and start from scratch, to find a robust election system fit for the next 60 or 70 years? What happens when Royal Mail says “We can’t deliver all these pieces of paper”?

John Pullinger: I don’t think we would say we need to scrap everything, but we do say that the whole canon of law and guidance, and everything that sits underneath it, has become so unwieldy that it is almost impossible to administer or follow, whether you are an administrator, a candidate, a campaigner or a party. It inevitably leads to confusion for voters too. That feels, to us, unsustainable. The last Government very helpfully commissioned the Law Commissions of England and Wales and of Scotland to have a thorough review of how electoral law could be consolidated. They produced a report that has been supported, in general, by all the political parties. We believe that that is a critical piece of work that needs to be taken forward with some urgency.

At the moment we have been stretching and stretching the system and adding more and more complexity to it, and there will come a point when it fails if that is not addressed. We see those risks growing. We look at risks to the system and see the acute risks that are upon us on election day, but there are chronic risks that build up. The risk of the canon of law itself being problematic is a chronic risk that has been growing very substantially. We have said some positive things about the Elections Act in this meeting, but the extension in terms of the quantity of law has been a challenge for all of us to get our heads around. I paid tribute at the beginning to the staff involved; it was a huge piece of work just to understand the law we were now expected to implement.

We believe that there is a starting point in the work that the Law Commissions have done. There is a process in Parliament for taking through legislation that consolidates the law, and we are certainly working very hard in the commission. Indeed, I convened a panel of some of the leading lawyers—particularly those who represent parties and are involved in adjudicating cases that come to court about election matters—and there was unanimous support for this as something that should and can be done. We are already taking that to the incoming Government in Westminster and saying, “Look, we need to get on with this.” It will not be a straightforward thing to do, but we believe it is an area where there is likely to be cross-party support.

All parties have already given it their support, so the question is where it fits in the timetabling of legislation in government, and how confident we can be that it will be taken through without turning itself into a kind of enormous activity that takes up more time than people are prepared to spend on it. I certainly agree 100% with you that this is something that needs to be dealt with, and the sooner it is dealt with the sooner we mitigate the growing risks that we already see arising.

Q43            Chair: What is your assessment of fraud and postal voting?

John Pullinger: When we asked the public about fraud in relation to postal voting and in-person voting, the answers were not so very different. There is a small proportion of people who indicate there are concerns about being coerced into voting in a particular way, which is a particular concern in relation to polling stations, and comes up in reports from observers, and in relation to postal voting where you do not know what is going on inside a household. We see it, and certainly we hear significant reports from some areas of the country that there is that concern.

Our response at the moment and for several years now has been to run a campaign with Crimestoppers, titled “Your vote is yours alone”, and to get across, particularly to communities where we think it is most likely to be a concern, the message that people have the right to vote, to make their own mind up about what they are voting and to be able to vote in secret, whether postal or online. We think that the systems put in place for both postal voting and for in-person voting have improved the secrecy of the ballot. Vijay has already quoted statistics from this general election that there is a high level of confidence that the system is protected from fraud, but we continue to hear the concerns, and we see there are concerns, and we think we should keep putting out the message that voting is something that is for the individual and for them to have the opportunity to do it in secret. We are not proposing new or enhanced measures, but we are continually vigilant to those that we see. We are currently—

Q44            Chair: What about the existence of election centres that pop up in some communities where the community is encouraged to all come together with their ballot papers and postal ballot papers and have it as, effectively, a community event? That is not necessarily fraud, but it can lead to coercion, and it can also lead to women in particular, young women specifically, feeling very constrained with regard to the freedom by which they can cast their ballot.

John Pullinger: I used the word “coercion” in my response to you earlier, and we would certainly bracket that within the issue of fraud. That is the question we are concerned about. Our response is, as I said, to double down on getting the message across to those communities that your vote is yours and these are the places where you can get help and assistance, and that coercion and fraud is a crime. We work very closely with the police to get that across, but it continues to be the case that there are very small numbers of cases that come to court and are prosecuted. The primary response that we can reasonably make is the public awareness response. Vijay or Jackie may want to come in.

Q45            Chair: It is where public awareness and cultural norms collide and where public information comes off as the loser.

Vijay Rangarajan: Chair, we are in process of collating all the UK fraud data from the general election. We aim to publish that in March or April. We are working with the police and we have been keeping a close eye on the cases the police are investigating, as well as looking at the perceptions of voters as to whether there is a difference in their perception of secrecy and coercion in a polling booth and in postal voting. We do not see at the moment—this is at quite a high level—an overall difference. About 80% of people in both postal voting and in the polling stations think their vote is secret. Given a sliding scale of 1 to 5, that is a 4 or a 5. We are also not seeing differences when people say there are elements of coercion. It is very small, but we are not seeing a lot. That could mask the issue that you are saying, which is that it is very localised in some areas. In some ways, we would applaud groups who get together and say, “Come on, let’s all fill out a postal vote,” as long as they were not pressuring people to vote a particular way. That is the issue that we were looking at.

Q46            Chair: That is contrary to the 1872 Ballot Act.

Vijay Rangarajan: It is very important that people are able to enter a secret ballot, particularly with a postal vote—we agree—but if people are saying, “Let’s all fill out our ballot paper,” and get together, that is fine, and it could be a good way of encouraging people to fill out their ballot papers. We have seen a very small number of actual cases with the police. This is incomplete data. At the moment, there are about 160 cases, but most of those seem to be on imprints and a couple of bits of candidate financing and spending, and a couple of areas of false statements were coming through. Those are being investigated at the moment. Quite a lot are being resolved. The number of cases of things like personation are zero to very small digits. It is an extremely small number. We will publish all this in March or April.

Chair: Mr Dillon.

Mr Dillon: My questions were on Crimestoppers numbers versus what we think the actual number of issues are out there, so I will wait for the reports. Thank you, Chair.

Q47            Chair: I think there was a more robust approach to policing and a greater drive towards uniformity of the police’s role in helping to deliver the election this time. Is that a trend that you see continuing and hope to improve?

Vijay Rangarajan: Chair, you are absolutely right. Before the election, we produced some joint guidance with the National Police Chiefs’ Council particularly to help police officers, for whom obviously this is not necessarily their day-to-day issue, on how to interpret electoral law, which is complex in this area. We think we are going to have to do a significant amount more on that, particularly—we will discuss this in the Speaker’s Conference—on abuse and intimidation because that is where some of the issues will come up. The police put a significant amount more resource into it, as did Operation Bridger, and many other parts. There was great gratitude to them and to Home Office colleagues in particular for work on it. Given the threats that we have seen, it is definitely an area we are going to have to do more on.

Chair: Thank you. Charlotte and then Richard Baker.

Charlotte Cane: Is this on overseas voting now?

Chair: No, I have you down for postal as well. Let’s deal with postal at the moment, and then we will come on to overseas.

Q48            Charlotte Cane: Mr Pullinger, you touched on the timing of the election in several answers and in your opening remarks, and that clearly had a significant impact, in your view, on postal voting. Have you considered what the implications would be of going to fixed-term elections and how that might help, or having restricted periods such as holiday periods when elections would not happen?

John Pullinger: Parliament recently debated the Fixed-term Parliaments Act. The key thing that we have made recommendations on is to give enough time for the elections to be run in an orderly way. In relation to this particular occasion—because July elections are very unusual—we were very clear in our report that the fact that it was scheduled for July was problematic and that particular account should be taken of people who are likely to be away on holiday, and to recognise that the situation is different in different parts of the UK, in this case particularly Scotland and Northern Ireland. I do not see a need to make recommendations beyond that. The timing of the election under the procedures we have at the moment is for the Prime Minister to take a view on, and we are just spelling out, which is the limit of what we can do, the consequences of doing it at a different time, and for the Prime Minister to be accountable for the decision that is made as a result. We are clear that July did not work well, and we think it will be problematic to do that again, but beyond that I don’t think we would make a recommendation—I am looking at you, Jackie—would we?

Jackie Killeen: That is not where our focus has been. On a smaller scale, one of the things that we recommended is looking at the election timetable, the period between when an election is called and polling day, to see whether that can be amended to enable administrators perhaps to have an earlier deadline for postal voting applications so that they are able to process them and issue them in a timelier fashion. The election timetable is very compacted. There is a lot of work for administrators to get through, and we can see the cumulative impact of that. One of the things we have suggested to Government when talking to them about that is looking at the actual election timetable rather than the bigger-scale issue that you are talking about.

Q49            Charlotte Cane: There is a balance, isn’t there, between the electorate becoming fully aware of exactly when the election will be and the impact of that on their other plans, and therefore having time to apply for postal and proxy votes, and making it easier for the administrators, unless you are talking about making the whole election timetable longer?

Jackie Killeen: We did not see a lot of evidence of support for that, so we were focusing more on what seemed more likely, which was whether there is any potential flexibility within the existing timetable as it stands—when the deadlines fall within the election timetable. As you say, that would sit alongside one of our main recommendations, which is about the information provided to voters as well, so that they can understand whether the different absent voting options are going to work for them in their circumstances. We think that has to be improved and increased for the next time. That is a key piece of learning from this election.

John Pullinger: The debate on the Fixed-term Parliaments Act shows the tensions in both directions. The sweet spot is a very small place to be. As Jackie says, we are trying to navigate within that sweet spot rather than have a fundamental reform, which would be very challenging to give effect to.

Q50            Chair: The Fixed-term Parliaments Act, let’s not forget, was only introduced to provide stability to a unique peacetime coalition, not to ease the delivery of electoral administration.

John Pullinger: Sure, but the debate was really instructive in terms of the tensions as to whether you should have more time or less time. That is all I am saying.

Q51            Chair: The old working presumption was that the administrators, in all its general sense of the term, needed five weeks from the Dissolution of Parliament to election day as a minimum. Is that still the same, do we think, or is it now up to six?

John Pullinger: From an administration point of view, the longer you have the better, particularly when it is unexpected and from a standing start. From a campaign point of view, there are pressures the other way. That is why I describe it in terms of a sweet spot.

Q52            Richard Baker: Jackie was talking about reforming the timetable of elections. Your report on this election makes a number of proposals around making rules for reissuing postal votes more flexible, and other forms of early voting have been discussed today. How optimistic are you that these reforms can be brought in ahead of the next UK general election, presuming that is in four or five years? How are these proposals being received by those who have the authority to bring them forward and to effect change?

John Pullinger: We are having positive discussions with the Government. It depends on the Government to do some of these things. We have identified the issues and the risks of not dealing with them, and so far, so good in terms of having positive discussions.

Q53            Richard Baker: Do you think there is good evidence that these changes will make the system more robust and effective in the future?

John Pullinger: For some of the changes we are talking about, we have already highlighted implementation of changes in Wales and our evaluation of pilots in Wales, and there have been encouraging responses to some of the reforms that have taken place there. Our evaluations show they can be successful. We will be evaluating some pilots in Scotland over the year or so ahead, and that is a chance for us to get increasing confidence that this can be done. We need to get on with it, as the Chair suggested. It feels like a long time until the next general election, but given the sequencing of processes and the need to get them in place well before so that people can administer them successfully, the time is actually quite short.

Vijay Rangarajan: We have major elections in 2026, so we are trying to take some of the learning of this into those elections and not just wait. That will be part of the timetabling issue, the way postal votes are done, and we have been working extensively with the Welsh and Scottish Governments on the conduct orders for those and trying to improve them to make the changes in advance. There will be another set of learning out of the ’26 elections that we will also need to use in ’28/’29, whenever the general is.

Q54            Chair: Will that continue to reflect on the evolutionary potential of digital and IT?

John Pullinger: Absolutely.

Vijay Rangarajan: You raised, Chair—and John responded—the set of legal issues that would help to avoid the sedimentary problem you alluded to. There are some fundamental data and IT issues that we would also like to look at quite closely. First of all, the register in many ways is quite an old-fashioned thing. Changing that and making it more robust and more resilient to cyber-attack and many other things would enable many of the changes that we are all discussing here. It is a limiting factor in many cases for what we can then do.

As Jackie has already mentioned, we can increase the amount of digital and offline voter information. It is curious that we do not have a list of elections. Neither the Government nor the Electoral Commission maintain the list of polling stations and accessibility, so I would like to do that and do it properly so that voters have access to what polling stations, what elections and what candidates there are and so on. There are some quite fundamental things. We can modernise the system without changing major elements, but that will enable some of the other changes that we then want—the register, the underlying data and some of the electoral management system issues. We have a patchwork of different electoral management systems. It comes back to the point on uniformity that you mentioned, Chair. They are not very uniform, and the way that things are run is not uniform between them, and we would like to see them raised to a higher bar.

Chair: Thank you. Let’s turn to overseas voting and let’s turn to Charlotte Cane.

Q55            Charlotte Cane: Thank you, Chair. The first thing before we discuss the detail of overseas voting is how robust the data that you have is. My understanding is that most local authorities have not published details on the percentage of overseas voters and the split of the countries that they are voting from. Is that correct?

Vijay Rangarajan: Unless Jackie wants to correct me, I think it is correct. Yes, we do not have the complete data as to the split of overseas voters yet published by constituency. We have a smallish survey of a group of overseas voters as to how their experience was, and the numbers from the various portals about what the application rates and the registration rates were. That is the data that we have.

Q56            Charlotte Cane: What are you doing to try to get a fuller picture?

Vijay Rangarajan: Partly, talking to overseas voters themselves. We think, from what we have seen so far, that many of them found it incredibly difficult to vote at this general election. They are a group of people relatively newly enfranchised by Parliament, many of whom found it incredibly difficult to get a vote back.

There were reasons for that. Partly, the international mailing times did not help. It was not just that. There are some very interesting distinctions between different countries. The numbers that we have at the moment are that there were about 170,000 applications to register. Of the new citizens abroad, only about 26% were aware of the changes. There is a large number who simply are not aware that they are eligible to vote. The return rates varied enormously by country. We think 75% of voters in France were able to return their ballot in time, but only 32% of voters in Spain were able, which seems an incredible disparity. Clearly, distance played a part. Only 6% of UK voters in Australia were able to return their ballot on time. That means that there were a lot of enfranchised people who were not able to vote. It was partly a function of the printing times and people registering, and partly a function of the sheer time it took to send and then return a ballot paper.

It is an area that we need to look at extensively and possibly quite fundamentally in terms of the changes. How should people vote? Should they be able to use a ballot paper sent out electronically and return it on paper? There are some issues with that. Would it be possible to tell because they had printed it out that it was a different ballot paper? Should we be able to use our network of consulates and embassies to enable that? We have begun a discussion with the FCDO about what is possible and how that could be done. Should we be using things like telephone voting? This is another good Australian example. They use it. I saw it at work in Australian Capital Territory and more broadly. It is quite secure. There are ways in which we could enable that. It is an area where we would like to propose some pilots to experiment and to discuss what is possible. Fundamentally, we do not think the experience of overseas voters really was a good one at this election.

Q57            Charlotte Cane: It is ironic given what you said earlier about high turnout in Australia for their elections that there was such a small proportion of our voters living in Australia who managed to get their postal vote back in time. You sound reasonably hopeful that we might be able to set up a structure that would allow overseas voters to use our consulates and embassies to ensure that their vote gets through in time.

Vijay Rangarajan: We are in the early stages of discussion at the moment. There are a number of issues—for example, the security of the ballot. How do you handle people who are outside, say, where there is an embassy or consulate? Are they sending it to them? There is a lot of real operational detail on safety and security and privacy of the ballot that we need to work through. It is an area where we will need to work incredibly closely with Government, MHCLG, FCDO and others to see what is doable.

Q58            Charlotte Cane: Would there need to be legislative change to facilitate that and enable it to happen?

Vijay Rangarajan: It almost certainly will require some legislative change to allow that to happen and a significant series of guidance changes. Who is the returning officer? Where is the ballot sent back? We would be in the situation where people would be able to be registered in different constituencies in the UK. How would we manage to get the correct ballot paper to them abroad and for them to be able to get it back to the right returning officer in time? One thing that is very different in a UK general election from either the Australian or the US approaches is that they have a significant period after the election when ballots can return—up to 30 days in many cases. We want a definitive result, and Parliament wants a definitive result, to know which MPs are elected within a couple of days of the general election, so our timeframe is much tighter than for other countries, and that is one of the things where we will need to test what is even logistically possible.

Q59            Charlotte Cane: One of the other recommendations you had was to make sure that people are more aware of the ability to use proxy voters. Some issues that arise from that are not just for overseas voters but for UK ones as well. The proxy voter obviously has to be someone you trust and someone with some connection to you, so they may not be physically in the constituency in which you are voting, and they in turn can then get a postal vote. It is how we make that clear to people, because I find with UK voters as well there is lack of clarity. Would there be any scope if you are a shareholder in a business for something where you can give your proxy vote to a trusted officer of the business? It doesn’t have to be someone you know. Have you considered anything like that?

Vijay Rangarajan: I don’t think there is a limitation we put on who you can choose to be a proxy voter. The issue we heard from a number of overseas voters was that they did not have somebody they trusted sufficiently in the UK to make a proxy voter. It is by definition not a secret ballot; someone else knows the way you voted, so there was reticence to do so as well. That is one area where we will try to give as similar an experience as to UK voters. It may not be exactly the same, but at the moment they have a very different experience.

Q60            Charlotte Cane: Do you have plans in place for communicating with overseas voters to make them aware of all the current options, at least?

Vijay Rangarajan: Yes, we are thinking of it in terms of the campaigns we are going to run before the next general election, which is the crucial one, and how we will communicate. That was why the figure that only about a quarter of them were aware that they could register was very interesting indeed. We have a major job to communicate internationally and get the message out.

Q61            Charlotte Cane: How do you find those people in order to communicate with them?

Vijay Rangarajan: If I can delve back into my previous life a bit, I was posted in Brazil as ambassador for four years. We knew a lot of the Brits in Brazil, and many of them had dealings with us of various rates. We knew a lot. We also had communications channels that we used to broadcast things. We used social media a lot. In every country, the local embassy, consulate and ambassador will have pretty good communications channels. You would want to use some local media. I would go on the radio or the TV and say, “There’s an election coming up. Please register to vote. These are the opportunities.” We did not push that so much this time, but we will probably want to do it more as part of the discussion with FCDO.

Charlotte Cane: Thank you. Thank you, Chair.

Q62            Mr Dillon: You mentioned the ability to run pilots. For them to benefit us in our next general election, they would need to be run prior to that. Are there any intentions to run pilots in the Scottish elections in relation to the schemes that you have outlined?

Vijay Rangarajan: On overseas voting?

Mr Dillon: Yes.

Vijay Rangarajan: No, I don’t think we have any pilots scheduled for 2026. We are deep in finalising the conduct orders for the ’26 elections, so it is already too late. The timescales, as the Chair already said, are quite slow. It is already too late to make any significant changes to the way the elections in May ’26 will work.

Q63            Mr Dillon: Do you think the pilots would be for the next general election if it was May when we anticipated one?

Vijay Rangarajan: Yes. If we can get the pilots done and then whatever legislative changes are needed in time, we would love to have something in place that gives overseas voters a better experience of the ’28/’29 general than they had at this one.

Q64            Luke Taylor: Very briefly, because I am conscious of time, you mentioned Australia, Canada and New Zealand as electoral systems, processes and organisations that work well. In terms of overseas electors, do you have examples where systems are in place that you look at and say, “That seems to be quite a sensible way to do it”?

Vijay Rangarajan: The Australian one is very interesting. I might be wrong, but the largest Australian polling station in the world is their high commission here in London. They run a very impressive operation, and my Australian counterpart and I went to see how they did it. Clearly, other embassies and consulates from other countries do the same. There is quite a lot of experience as to how people can run this abroad.

Q65            Luke Taylor: Importantly, your point was that most countries use their consulates and their embassies, which we do not.

Vijay Rangarajan: Yes, quite a lot of them. The ones that run our kind of system use them, but some of their underlying systems, like the electoral register, are significantly more advanced. You can go into an Australian polling station from a different area and they will print out the ballot paper for exactly the area that you are registered in. You fill it in and it will then all be processed. That is because they have a register that spans all Australian voters, and they know exactly which ballot paper you are entitled to. We could not do that at the moment with the technology and the state of the register that we have. We cannot inform political parties that they have already lobbied someone or that a voter has died, and they or their family should not be hassled. That is something we hear from political parties. There are a number of fundamental underlying system changes that would enable all those things to be done.

Luke Taylor: That is helpful because it informs the first question that needs to be fixed before we can do other things better. Thank you.

Chair: Okay. That’s overseas voting. Let’s turn to voter ID and fraud. My speaker list tells me I have Luke, Michelle, Peter and Chris Curtis.

Q66            Luke Taylor: Thank you. I am quite keen on evidence-based policy. The implementation of voter ID was suggested to be introduced to tackle electoral fraud, and in particular personation. From the figures that we have seen in various reports, it was suggested that in 2019, after which the previous Government brought in the voter ID rules, there were 34 allegations of personation, which is 0.000057% of the 58 million votes cast in that election, and then two people were cautioned or convicted out of 58 million votes, which was 0.000035% of votes. We then saw turnout drop, and from the report as well, the suggestion that up to 1.9 million people said that they, from surveys, did not vote because of voter ID. Do you think that the impact of voter ID was a proportionate response to the problem it was trying to solve?

Vijay Rangarajan: It was intended to solve a number of issues. We have had voter ID in Northern Ireland now for about 20 years. It had some effect. We published a bespoke paper on it in early September, looking at the general election and trying to learn some of the lessons from it. To set out what we found, a small group of people did not bring voter ID to the polling station. Most of those went away, got ID, came back and were able to vote, but 0.08% did not vote. That is 16,000 people, therefore, at this general election who did not vote or were unable to vote and they should have been able to. It has had some effect.

Q67            Luke Taylor: Sorry, they arrived at a polling station and then didn’t return. That was 0.8%, you said.

Vijay Rangarajan: It was 0.08%—16,000 people.

Q68            Chris Curtis: Is that literally turned up and spoke to the polling clerk? Obviously, there may have been people who got discouraged before that who you are not counting in those numbers.

Vijay Rangarajan: Absolutely. That is people who spoke to them, and were then turned away and came back. We will come back to people who were discouraged in just a second.

The system of voter ID that was surprisingly underused was the voter authority certificate. I think 210,000 people applied for a VAC between January ’23 and the deadline for the general election; 26,000 VACs were used out of 40-odd million voters, so it seemed remarkably low. Part of the solution was trying to give people voter ID who did not have other forms. That has informed therefore, and the evidence has therefore informed, our recommendations that we should expand the list of acceptable IDs. We have made those recommendations publicly and to Government. The veterans card is being debated at the moment. There are a number of others that could be added to target some of the groups who find it harder to engage. There are various travel discount cards that would work and, crucially, digital ID as well. At the moment, it is producing a piece of paper that someone has to bring. If you could have a digital ID, and we found some with the right security standards, it would help people tremendously to be able to show their phone at a polling station.

Finally, for those who do not have it, we recommend that the Government—in this case, it would be something that would have to come through Parliament—look at vouching or attestation. If someone knows you very well and they are a registered elector and they say, “Yes, I know this person and I attest formally that they are this person,” which many councils can also do, they ought to be able to vote. All of that is with a view to broadening accessibility. Some people are put off by it. In our data we saw that. This was from public opinion research asking people whether they voted. It is not just people who voted. Of those who said they did not vote, about 4% gave voter ID as one of the reasons. It is not an absolute link. They had many reasons for not voting; voter ID clearly was a factor. It has put some people off.

Before the election we managed to run some significant communications campaigns to get knowledge of the voter ID requirement right up. To be honest, we were concerned before the election that there would be a real differential between different regions. With England having done it and Scotland having not done it, you can imagine some of the political reaction had there been a real difference in the number of voters turned away. That did not happen. We were broadly at the high 80s to 90% awareness across the whole of the UK and it was pretty uniform, so that was good. We will have to continue that. We now see that having made the change, and Parliament having legislated for the change, people are getting used to it. Most brought some form of voter ID.

Our data is not that great, but it might have had some increase in trust in the election that you had to show an ID. Concerns about personation exist. You are absolutely right; the numbers are incredibly low. The concerns about that and any forms of mistrust might have been helped by it. Hopefully, it will become very much like Northern Ireland where it is taken for granted that you bring a form of voter ID for an election.

Q69            Luke Taylor: To recap back to the beginning, though, was voter ID and personation a problem that the Electoral Commission saw prior to 2019 when it was put into the manifesto of a political party?

John Pullinger: We were very unusual in not having any check on the person turning up to vote, and this was something regularly reported by international observers. The Electoral Commission for about 20 years was saying, “Look, we need to do something about this.”

Our position has been that we need to do something about it because it will increase confidence in the electoral process, but one person turned away is one too many. People should be able to exercise their vote, and that is why we are now in the position of recommending changes that will deal with those people who were either turned away or put off. As you indicated, the number turned away at the polling station is the tip of an iceberg of people who felt that voter ID was a deterrent to them. We must increase our public awareness. We must also look at how we can make changes to the voter authority certificate and to the list of forms of ID that would be acceptable, particularly for those who were put off. The jobseeker’s allowance card would be another thing that would directly go to groups of unemployed people who we know were put off more than others. The 18+ Oyster card would be another one where we know young people—

Q70            Chair: I don’t want to drill too much into the detail of what could or could not be permitted. I just want to ask this very quickly before I bring in Michelle Welsh. On the data of people who were turned away, our presumption is that they were people who were qualified to vote and on the register but did not have ID, not just people who turned up but who were not on the register and were turned away. Is that correct?

John Pullinger: Yes.

Chair: Fine, thank you.

Q71            Michelle Welsh: You have just talked about the voter authority certificate and expanding the list of accepted IDs and introducing vouching as a way of improving the voter ID system. You touched on in practice what impacts you think those measures would have, and I would like to hear some more with regard to that. In addition, I would be really interested to know what work has been done or is intended to be done with regard to where the people who did not have voter ID were. Was it a particular socioeconomic group? Where was the voter ID actually a hindrance and a barrier to people voting? I say this as someone who is a Member of Parliament in North Nottinghamshire where it was very clear where that barrier was and who the people were who were not voting. It is probably very different in different areas.

As a barrier to voting, who was that a barrier to? At the moment, we are talking about this on the whole big scale of, “If we have this ID and we have that ID,” but what work has been done to identify the people who need that ID? Where is the missing gap in our system? People approached me when I was out and said, “I’m not voting. It’s the first time I’ve never voted because I don’t have any ID,” and then we discussed it. I know what my personal perspective might be on it, but it would be interesting to know what work you have done with regard to that.

John Pullinger: When the policy was first introduced in 2023 in the local elections, the group that seemed to be most cut off from voting were disabled voters. Vijay quoted a figure of 4% for the population as a whole. It was about 9% for disabled voters who felt that voter ID was a barrier to them. It was not necessarily because they did not have ID; they just had not understood what needed to be done and hadn’t got the support necessarily to ensure that they were producing ID. I observed that in polling stations. It was better by the general election. That was an example where once we can identify the specific barriers and people who are concerned, we can work with civil society organisations and directly with voters and returning officers to try to address the question.

For the general election, the groups that stood out were unemployed people particularly and people in more manual occupations. I mentioned the Jobcentre Plus card. For them, some of it was because they did not have a form of ID. They did not have a passport or a driving licence for all kinds of reasons—or some of the others on the list. Expanding the list should make it much more acceptable for those kinds of people.

In our report, we have a series of lists. Those were the two groups. For 2023, it was disabled voters, the unemployed and people in more manual occupations who were particularly affected, and we can see potential solutions for them. There are other smaller groups. Homeless people are often less likely to have ID, and it is about how we work with their communities. Vijay has already mentioned Gypsy and Traveller communities. Trans communities are ones where visual identity may be difficult if people look very different now from how they did on any form of ID that they have. Some of these are quite precise.

Our survey work and our links with civil society organisations have given us a good insight at quite a granular level as to whom we need to work with. Our recommendations are designed to increase the list: to go directly to people who have some forms of ID but not the ones currently on the list; to create in the voter authority certificate an alternative to those; and particularly whether a digital solution can be found. Finally, there is the vouching option, which we have seen work in Canada, where someone who is already on the register can say, “Right, this person who is with me is who they say they are and therefore is eligible to vote.” Has that got to the question, or do you want to drill down a bit deeper?

Q72            Michelle Welsh: Yes, that confirms what I thought. Are there parts of the country or constituencies where it was more prevalent? Is there more work to do in a certain council area than others? Have we got down to that level with regard to it? Does it affect the north of England more than it affects the south of England? Does it affect mining communities more than it does inner city? Where does it affect people the most at the moment? Have we got to that level?

Vijay Rangarajan: We do go down to that level of detail. We will publish the data on this. We do not see a strong geographical or urban rural distinction. By far the strongest distinction was on social grade: C2, D and E gave us very different answers from other social grades, particularly about voter ID putting them off voting. They gave a stronger response. The others, as John said, were unemployed, disabled people and those renting from a social landlord. Those were where there was a statistically different reaction to voter ID. Our proposals to expand the list are trying to pick up the forms of ID that those people are likely to have. Most people have some form of ID that is sufficient to prove their identity. It depends a little bit on the security tolerance that Parliament chooses. None of the other differences was statistically significant.

Q73            Michelle Welsh: Did it have more of an impact on women or men?

Vijay Rangarajan: We could not see a strong gender difference either. We will publish the breakdown tables. I am very happy to go into that in more detail. The four groups we put in our report were those where there was a clear statistical difference, and the strongest was social class.

John Pullinger: There were some differences by ethnicity that are also a potential factor, and it is about working with different communities so that there is real understanding and appreciation of how it will work in practice.

Chair: Forgive me. Peter wanted to come in on this.

Peter Lamb: I don’t believe my point is relevant any more.

Chair: Okay, thank you, Peter. That is useful for time management.

Q74            Chris Curtis: A lot of the discussion we have had today has come back to you as an organisation being able to provide data and recommend changes that the Government are therefore responsible for enacting, and that is the process that we have gone through with voter ID. What makes many of us quite uncomfortable with that process is the way we saw it play out with voter ID, where for some of the reasons that were brought up so far I am not convinced at all that your strong recommendation in 2014 to change to voter ID in any way balanced the risk management. How can you reassure me, given the data we have heard so far, that you correctly balanced the risks against the very small number of people who were enacting personation in order to come to that conclusion?

John Pullinger: We try to give as much of an accurate perspective in this report as we can on where we are. The position we are in is better than many feared it would be, but not as good as it needs to be if we are to achieve the balance that you have described. At the outset, we were very clear publicly and in our conversations with the Government that, if the voter ID legislation was to be successful and what we needed to maintain was a healthy democracy, it needed to meet the security requirements, which were the dominant thing in the mind of the Government when they proposed this in the first place. They need to meet those concerns, but not at the cost of accessibility.

Our evidence base was from Northern Ireland. We had seen that within quite a quick period people became used to doing it and it had been attuned to the way in which people were able to respond to the policy. We have not seen for many years now any issue at all with voter ID in Northern Ireland. We expect with the changes that we have made that the same will be true in the rest of the UK as well.

The final criterion was workability for local authorities, which we have not talked about. We had real concern that it would be extremely difficult to administer and likely to result in long queues at polling stations because it would be very slow getting through. The responses we have had from local authorities and our own observation is that that did not happen. Local authorities did a magnificent job of training their staff and of organising systems, and it was certainly very slick everywhere I went. The remaining block is to get us up to the kind of situation we have in Northern Ireland in terms of accessibility.

Q75            Chris Curtis: With respect, that was not quite the question. I appreciate that there were some bad things that some people said were going to happen or we feared might happen that did not happen. What worries many people, or certainly worries me, is that there was a recommendation made by the Electoral Commission to bring this in in 2014.

There seemed to be two arguments: first, to stop personation, for which, as we have heard, data did not exist, at least compared to the many other forms of electoral fraud that were significantly higher. I cannot see why that justified the damage it has done to people’s ability to vote in this country. The other one was perception of risk, which also does not seem to stand up based on the data. If you look at the polling data at the time, there was a very high level of trust and satisfaction in our democracy. When you ask people what they thought were the biggest problems with our democracy, the idea of personation was incredibly far down people’s priority lists.

There does not seem to me to be a justification to make that recommendation based on actual risk. There does not seem to be justification based on perceived risk, from the polling data that we had at the time. It just seems very strange to me that you as an organisation recommended that change given those two facts.

John Pullinger: The recommendation came from our observation of what is happening in other countries. Our own observation that people are expected to provide some form of verification of who they are in so many other settings—

Q76            Luke Taylor: Can I cut in there?

Chair: Mr Taylor, let Mr Pullinger finish.

John Pullinger: Many people are surprised that we did not have it. Certainly, in the conversations I had in 2023 when we first introduced it, many polling station staff were surprised. They said, “This is a risk that may not have materialised much yet, but it could. Let’s close the door before the horse bolts.” We were always clear that we should not introduce it unless we met the other tests, too. I feel some regret that the way it was introduced in a sense created it as more of a divisive policy than it might have needed to be. There was an urgency in the way that the Government wanted it in by 2024. Mr Hoare was in another role at the time, and we had many discussions about it.

Chair: We did indeed, Mr Pullinger.

John Pullinger: The need to do it in a way that generates public acceptability in advance was something where we could have done better with hindsight. I remain content with our recommendation, but if we had been starting again now, with a more careful approach to making sure we really had taken into consideration the challenges Ms Welsh’s interventions indicate for certain communities and the risks we knew were likely to materialise, we could have dealt up front with some of the things that have now been dealt with in Northern Ireland but are still to be dealt with in the rest of the UK.

Q77            Chris Curtis: I am still not completely convinced on those arguments either. You don’t say, “We are going to make a change because other people are doing it,” in order to solve a problem that does not exist. That does not feel like a particularly compelling argument either; neither does, “Well, it worked in Northern Ireland eventually. In four or five election cycles, people will get used to it.” You are still disenfranchising many people in the election cycles in the meantime. I am not completely convinced by this argument at all that the recommendation was justified.

I have two very quick separate questions. You are a fantastic statistician, John. We acknowledge that a lot of data is missing and the evidence base is missing for this decision as well. We do not have good data on how many people do not have acceptable forms of ID and all questions like that. What kind of data and evidence do you think that we should be starting to collect in order to do this? Related to that, a lot of the data and evidence that has been collected has been through online polling. I say this with a background in polling, particularly online. Do you think that is an appropriate way to collect the data, given that most people who are not registered to vote and most people who are unlikely to have ID are also unlikely to be signed up to online panels?

Before this scheme was introduced widespread in local elections in 2023, there were pilots in two very specific local areas that tested it out beforehand, which did not gather evidence, as Michelle said, about how it might affect different communities accurately and about how it might implement differently across different places. Do you think that implementation process should have been, in hindsight, far more widely piloted than it was before it was introduced?

John Pullinger: To your last question, yes, with hindsight it should have been, because you need to get down to the granular detail. On the broader point, you know my background; we need far better data across the whole system to be able to have a really clear understanding of how well it is working. It is something Vijay and I have a lot of conversations about. He has already mentioned some of it in relation to some of our earlier responses.

The key to getting there is probably through the registration system and getting much better information about who is registered so that we can look in a much more granular way at the gaps—who is not registered and what is happening to those who are. You are right; online polling is becoming a much less reliable way of getting evidence even than evidence you used to get in the past, and that evidence was generally about just getting broad averages about what is going on rather than helping you to look at the patterns of behaviour under the bonnet.

We need to be much more sophisticated in the way we think about information and gathering data and reporting it, and making it available to you, particularly at constituency level, which obviously matters to you, and looking at different local authorities, how they are performing, what is happening in different wards and what is happening within different communities—we mentioned private rented communities and social housing areas—to really see what is happening in places like that. “What is happening for people like me so that we can be helped and supported in our ability to exercise our vote?” It is a very significant piece of work, which is part of our corporate plan that Vijay has already mentioned. We will be trying to raise our game working with others, including the Government, on getting that data so that we and you can hold the whole system much more effectively to account.

Q78            Luke Taylor: You talked about broader international systems and comparisons with voter ID systems and there being many countries where voter ID is a thing and it is accepted. Was a comparison made with countries where a nationally available—compulsory or optional—general ID card is available? You look at New Zealand, which does not; it is in all of our examples of good electoral systems and it does not have a national ID card or voter ID. We are a very different country culturally from a lot of countries where, particularly in Europe, you have ID cards. Was that taken into account when we looked at other countries and said, “Oh, well, we’ll get used to it”?

John Pullinger: Yes, but it is undoubtedly a lot easier when you have a national ID card that everybody is required to hold.

Q79            Luke Taylor: That seems self-evident. It does not seem to have been taken into that much account when implementing this incredibly dramatic policy.

John Pullinger: There was a significant amount of work done, primarily within government, to look at who holds which sorts of ID. We think now that about 3% of people do not hold one of the IDs currently on that list. Certainly, that work was done. The Government led the work with us and others to make an assessment of where the gaps are, and that is what created the idea of the voter authority certificate. There were substantial numbers of people who did not have one of the accepted forms of ID, so something needed to be done to enable them to get something specifically for that purpose. The intention was right. The implementation has not been very good, as Vijay described. The voter authority certificate was only used in 26,000, which is a tiny number of cases given the number of people we think do not hold one of those forms of ID. Let’s reform that and let’s increase the list of accepted ID, and that will narrow the gap. From your comments and Mr Curtis’s comments, you are looking, as we are, for a very substantial improvement, because one person turned away is one too many, and we would all hold to that.

Q80            Luke Taylor: I finish my questions by asking whether the fundamental principle of voter ID is being evaluated. If it is a system that cannot work without other changes, such as a national ID card, which obviously is not on the agenda, is the principle of it being evaluated, as well as how to make, in my opinion, a bad system work?

John Pullinger: The approach that we have taken is in the report. We think the right approach now, having implemented it, is to make it work better.

Chair: I am conscious of time, and there are two statements that colleagues may want to take part in after Health questions. We wanted to cover electoral administration, which we have slightly dipped in and out of en passant during earlier questions and answers. If colleagues are happy, I propose that we effectively table in written form those questions to the commission and receive a written submission on them. We are going to hear from the Association of Electoral Administrators in due course anyway, so they will certainly be able to put flesh on the bones. Is everybody happy with that approach? Bottoms have got numb enough as it is. Time to move on.

Thank you, Mr Pullinger, and your colleagues, for attending this morning. It is a lively issue, as we all know. I am tempted to say that, because we were all elected, the system worked terribly well. That I appreciate would just be too superficial an assessment of where we are. We are really grateful to you for your time and also—I say this in closing—for the work that the commission does in general in helping to play a very important role in maintaining the integrity of our electoral system, which, as we know through all sorts of pressures, comes under an awful lot of intense scrutiny from time to time. Thank you very much indeed.