final logo red (RGB)


Select Committee on the Electoral Registration and Administration Act 2013

Corrected oral evidence: Electoral Registration and Administration Act 2013

Tuesday 16 July 2019

3.30 pm


Listen to the meeting

Members present: Lord Shutt of Greetland (The Chairman); Lord Campbell-Savours; Lord Dykes; Lord Hayward; Lord Lexden; Baroness Mallalieu; Baroness Pidding; Baroness Suttie; Lord Wills.

Evidence Session No. 1              Heard in Public              Questions 1 16



I: Peter Lee, Director, Constitution Group, Cabinet Office; Simon James, Deputy Director, Elections Division, Cabinet Office.





Examination of witnesses

Peter Lee and Simon James.

Q1                The Chairman: Good afternoon. Welcome to this evidence session of the Select Committee on the Electoral Registration and Administration Act 2013. You have in front of you a list of interests that have been declared by members of the Committee. The meeting is being broadcast live on the parliamentary website. A transcript of the meeting will be taken and published on the Committee website. You will have the opportunity to make corrections to that transcript where necessary. This is our first public session, and a number of members may have relevant interests. They have to put them on record the first time they speak.

We will make a start with Peter Lee and Simon James of the Cabinet Office. I have a note here that, under the Government’s own guidance, the Cabinet Office, as the Department responsible for this Act, should have produced a postlegislative scrutiny memorandum within five years of Royal Assent, in January 2018. We do not think we have seen this. Have you brought it with you?

Peter Lee: We will produce the memorandum after the Committee has concluded its discussions. Our logic for this concerns the point at which we thought we would be able to understand the success or otherwise of the policies the Act underpinned, in particular the extent to which the introduction of individual electoral registration has done what we set out to do, which was to maintain or even improve the completeness of the electoral register while improving its accuracy.

We had one report on that in 2016 from the Electoral Commission, and the second one was just after the reforms, at which point the system had not really settled down. The next one is due in or around September this year, at which point we will have a better idea and more data about how well the Act has done what we set out to do with it. Our conclusion was that we would provide that postlegislative report at a time when we could demonstrate effectively whether the legislation had done what we set out to do.

The Chairman: You think that will be in September.

Peter Lee: The Electoral Commission produces a regular report on the accuracy and completeness of the electoral register, and the Committee, as I understand it, will conclude its deliberations towards the end of the year, with a report, I presume, at the start of next year. It would make sense for us to have both pieces of information before we provided our conclusions on the effectiveness of the legislation. At the moment, our plan is to aim for that kind of timescale.

The Chairman: You are saying that this report will be at the back end of this year or early next year, are you?

Peter Lee: That is our plan at the moment.

The Chairman: What can you say at the moment about the Government’s assessment of the Act and its implementation? Are you able to take a view as to whether it has fulfilled its objectives, and what key lessons have been learned?

Peter Lee: As for the introduction through the Act of individual electoral registration and what followed from that, and the changes that have been made to the administration of elections, we can say how successful we think they have been so far. The missing piece we are waiting for is more hard data about the impact they have had on the extent of registration.

Through IER, we set out to close down a very significant fraud risk. Previously, when people registered to vote, we did not know that they were who they said they were. We are now much more certain because of the way the system works. It has brought us into line with other modern democracies. The idea of registering on a household basis, where the head of the household completes the form on behalf of other people in the household, seemed at the time and continues to seem anachronistic. In that case, we have modernised the system and closed down one of the most significant fraud risks in our system of electoral registration.

The initial indication from the 2016 report by the Electoral Commission is that it has held completeness at a similar level to the preIER period but improved accuracy by about 4%, which is statistically significant, so we have a more accurate report than we had before. There were a lot of people on the register before who had moved, effectively ghost entries on the register, and they have been removed. As I say, we will know more about that later in the year.

It has also enabled the implementation of the Register to Vote website, which has been a modernising step change in the way people register to vote in this country. About 75% of people now register to vote online. That was enabled through the legislation. Previously, everything was paper-based. The vast majority of citizens are now used to having an online option for completing these kinds of transactions, so that was a big and important reform.

That is not to say that it does not present us with challenges. I suspect we will talk later on about the further modernisation to the canvass system which the Government are planning, and how we need to continue driving increased registration, particularly through hardtoreach groups. There is still a lot to do, but the fundamental objectives on registration through the Act have been completed.

The Act made a series of pragmatic reforms to the way in which we administer elections. Probably the most significant was the move from 17 to 25 working days between dissolution and polling day. That has proven particularly important and necessary when we have had shortnotice elections outside the fiveyear cycles set out in the Fixedterm Parliaments Act, the 2017 general election being a case in point. Instead of having 17 working days to prepare for that, the electoral administrators had five weeks to conduct an election. When you consider how complex our elections are now and the number of people taking part, including people overseas, it was a very important reform.

Simon James: To make sure that these points are not lost, although I am sure the Committee is fully aware, there were also a number of quite significant electoral administration reforms in the Act. This is not an exhaustive list, but these are the ones I would highlight.

Peter has mentioned extending the electoral timetable. We made changes to the timing of polling place reviews in Great Britain to bring them into line with the parliamentary timetable in the Fixedterm Parliaments Act. We allowed voters queuing at polling stations by the close of poll to be issued with ballot papers. That was a direct response to issues seen at previous general elections. We allowed police community support officers to enter polling stations as police constables can do currently. We enabled regulations to be made to place electoral registration officers under a duty to give notifications about rejected postal votes. We provided that the Secretary of State may withhold or reduce a returning officer’s fee for reasons of poor performance.

There are a number of quite discrete changes there, but each of them, in its own way, has contributed to a modernisation of our electoral administration.

The Chairman: Thank you for that, but I am just a bit concerned about this business of the Government’s response. We have somebody lined up to come next week, but do not forget that, starting in the autumn, we are meeting every week with two evidence sessions, an hour apiece. Then we have to prepare a report. We will start that after our final one in December. If we do not get this government response, it will be pretty late in the day.

It seems to me that we really ought to have this earlier, because the people who come to give evidence could well say, “Yes, we have studied this Act and this is our preliminary view. Now we have studied what the Government have had to say about it, and we do not think they are right”, or whatever. Is it not possible to get this much earlier, so that at least the people giving evidence to us can have sight of it before they give the evidence?

Peter Lee: Those are all extremely valid points. Perhaps I can take that away and work out what to do. Without trying to unpack that now, I think we would be keen to try to do that in the light of the Electoral Commission’s next report on completeness. I absolutely hear what you are saying. It is important that the Committee gets full evidence, a lot of which I hope to be able to provide you with today. We will then come back to the Committee clerks with what we can do.

The Chairman: That is a helpful response.

Q2                Baroness Mallalieu: What are your main priorities now, and what are you actually working on now in relation to still improving electoral registration and administration?

Peter Lee: On registration, I will touch on three points and then hand over to Simon to talk about where we are on electoral administration. One point I have already touched on is canvass reform. We have ended up, post IER, with a costly, bureaucratic and somewhat outdated system of conducting the annual canvass. A major priority for Ministers is to complete the reform of that, which we will be doing between now and the canvass in July 2020, so over the next 12 months.

Basically, that consists of two broad elements. First, we will use what we call a national data step, which we used for individual electoral registration. We take the registers from each local authority as they are and compare them to national data, in this case the DWP data, which is the most complete database that we have. That will show that a very high percentage of people, probably 70% to 80%, have not changed their address, changed their name, got married or done anything that would require them to change their entry.

At that point, we will also enable electoral registration officers to take a more flexible and lighttouch approach to those people. At the moment, they have a very heavily standardised approach. They have to send out the household form, they have to chase it so many times, and then there is a door knock. It is all paperbased. They will now be able to do that on a much more lighttouch basis. They will be able to use relatively modern technology such as phone, email, text messages and so on. They will also be able to take a more flexible approach to chasing up those who have apparently changed or where it is unclear.

At the moment, we have this system whereby people will get one form that says, “You need to report whether anything has changed”. If it has changed, they have to go off somewhere separately and complete their registration to vote. You do not register to vote through the canvass system. It is a reform aimed at providing a much more proportionate and flexible set of tools for EROs.

That has huge benefits for the EROs themselves and cuts out vast swathes of bureaucracy from the system. It benefits the citizen, who will no longer get endless letters through the post. We can approach citizens who have not noticed they have had something through the post through other mechanisms that they might be more likely to respond to. It also a quite significant saving to the taxpayer.

We hope that, through this process, the current government contribution, which stands at about £15 million a year, will go to zero. We also hope that some local authorities will be able to make savings themselves simply by cutting out huge swathes of bureaucracy. That is the most significant thing we will be doing over the next 12 months, and we will be doing it in conjunction with the Scottish and Welsh Governments. Those are devolved areas, but we have agreed to deliver that reform with them so that the system is the same across the whole of GB. Northern Ireland has had a different system for some time.

The second element is democratic engagement. As well as having a streamlined and efficient system of registering to vote, Ministers are very keen to encourage people to take part in the democratic process. I always hesitate to talk about the views of politicians, but it is generally accepted across the political spectrum that getting the maximum number of people to register to vote is a good thing for society. Some of this is about hardtoreach groups. There is a particular issue with people who are homeless or who are frequent movers. They have just as much right to register to vote and take part in the democratic process as others, but find it more difficult to do so. This is also about getting hold of people at a young age and making it a habit to register to vote.

Baroness Mallalieu: What are you doing about that?

Peter Lee: On young people, attainers and universities, we have appointed 1,000 young democracy ambassadors. People respond better to a message that comes from peers rather than from outside. Through recent legislation, we have put a set of requirements on universities to provide certain information but also to encourage registration among students, which is a group we are keen to get to.

On people who are homeless and frequent movers, we are running a programme through an internal government body called Policy Lab, to work with all parts of the sector, including people who are either currently or previously homeless, on what practical changes we could make as government, and what already existing best practice across the sector we can bring together, to remove barriers and to encourage people to take part in the democratic process.

There is a huge amount of stuff, and I do not want to delay the Committee, but if you want more on that we can provide it. There is a load of really great stuff in there.

Another example of what we have done, and I am incredibly proud of the team for this, is that we have enabled survivors of domestic abuse to register anonymously. Previously, those who were resident in a hostel or sheltered accommodation of some kind were particularly reluctant to register to vote if it meant that their name and address might be available to others. We have made it much easier for them to register to vote anonymously. Why should they be prevented from taking part as well? It is often not big-bang stuff; it is practical. Sometimes they are the smallest of changes; sometimes it is just about nudge, communications and best practice.

Lastly, our Register to Vote website is a great asset. We want to make sure it is as robust as possible. It is a slightly funny website that has to cope with very lowlevel demand for 99% of the time and then a massive spike immediately before a major electoral event, which we saw, as everyone will recall, in 2016 during the referendum. We are looking to move to a cloud technology platform, which means that we have almost infinite capacity to surge, as is required before a big election. We had 600,000 trying to register to vote in a single day.

Baroness Mallalieu: Is that up and running?

Peter Lee: We have already made a series of changes since 2016 to hugely increase the capacity. We now have capacity that would easily have got us through the 2016 surge, but we want to move it to the next stage and do it more cheaply and effectively through a common technology that is being used across the world now. That will be up and running by 2020. We are just in the process of letting the contract.

Q3                Lord Wills: I should declare my interest as president of the Wiltshire Association of Local Councils and a member of Transparency International UK’s advisory council. I suppose I should also declare an interest in that I was the Minister responsible for the Act that first introduced the prospect of individual electoral registration.

I have two brief questions on what you have just told us about the motivation. You say it decreased a significant fraud risk, and I assume you chose your words very carefully.

Peter Lee: I did.

Lord Wills: First, what evidence do you have that actual fraud has decreased? Secondly, you say that you have held completeness. I wonder whether you are disappointed by that, given the Government’s huge efforts in the runup to the Act and subsequently, which you have just described. They are all excellent and really commendable measures. Despite all that, completeness has held pretty steady.

Peter Lee: I think you know the answer to the first question, but I will answer it anyway. As you well know, the actual evidence of fraud in people registering to vote is not high. It was previously difficult to tell when people were committing fraud. If someone was effectively, on behalf of their household, completing the form in a fraudulent way, it was very difficult to pick up. It is pretty obvious where the risk is; we just did not necessarily have huge amounts of data to demonstrate that it was going on. It was more a case at the time of closing down a risk channel than tackling a problem for which you could say, “We have 1,000 cases and we believe it will be zero after this”.

We know for definite that one of the most important gateways into our democracy is now much more secure. We know that the people who are on the register have had to go through some kind of checking process against nationally held data, so we are much more confident at that end. We should bear in mind that the other end of the electoral process, the act of taking part in voting, which we might get to later on, has relatively few safeguards. It is really important for us to know, at least when the register is being produced, that there are not loads of people on it who should not be, by and large.

When it comes to disappointment, at the time there was always bound to be a degree of churn. We are fairly confident that around 600,000 entries on the register were probably nonexistent—in that people had moved on and were on the register several times in different placessimply because when we went through the whole process, that was the rump we ended up with. We had used the data match and gone through a very intensive process of working out who all the unmatched people were and where they were. At the extreme end, we tried to contact people up to nine times. Once you have done that, if you have 600,000 people left, they are probably not real.

With that degree of churn, holding the completeness at about level by 2016 was a reasonable achievement. I do not want to use pejorative words such as “disappointment, which are possibly more for Ministers than civil servants, but there will be a degree of it by September when we find out where we have got to if we are not driving more people to register to vote. We know that in the 2017 election more people were registered to vote than ever before, but that is partly because the population had increased and partly because it was a hugely significant electoral event. We will have to see how it is.

It definitely does not mean that we would in any way slacken our attempts to persuade as many people as possible to get registered, including those who struggle at the moment.

Q4                Baroness Pidding: You talked about engaging with underregistered groups, but you did not mention what we are doing in BAME communities. You also mentioned young ambassadors. Can you tell us a little more about how they are appointed and what they are doing?

Peter Lee: On the appointment process, I might need to provide that subsequently, because I do not have to hand how we went through the process of appointing them.

Having done a few events particularly around National Democracy Week last year, I have discovered that it is quite hard for people like me to engage young people to want to take part in the democratic process. That is the philosophy behind this. If you have a set of young people spread around the country who are incredibly passionate about getting their peers to take part in the democratic process and the importance of it, that is a good tool in the armoury, alongside the other stuff we have done, such as providing schools with school packs and activities during National Democracy Week.

One thing I took part in involved a bunch of the democracy ambassadors coming to the Cabinet Office and having a three or fourhour session about what they thought, what they had found worked well and what did not work. Young people are much more likely to respond to that. If we can get them to do that as they are on the cusp or in the early period of being able to vote, they are more likely to carry it on through life. If they vote the first time they can vote, the evidence shows that they are much more likely to vote for the rest of their lives.

On the BAME community, again through National Democracy Week we did a number of activities with people from all the groups that we know are less likely to vote. We did quite a bit of stuff with people who have various disabilities or learning difficulties, who are entitled to vote but who statistically do not vote as often.

The Electoral Commission has provided material on how to vote in ways that are more accessible. Through things like National Democracy Week we have been trying to get the message out. We and local authorities, which did a huge number of activities, focused in particular on the groups that are less likely to take part.

On the specifics, I could definitely provide the Committee with more information on what we have been doing for black and minority ethnic people, if that would be helpful.

Q5                Lord Campbell-Savours: My question is very brief, and it requires only a yes or a no.

Peter Lee: They are the worst answers.

Lord Campbell-Savours: Do the Government make a contribution towards the annual canvass?

Peter Lee: The annual canvass is funded by local authorities out of the Revenue Support Grant.

Lord Campbell-Savours: Is it completely funded that way?

Peter Lee: Yes.

Lord Campbell-Savours: Secondly, what was the £18.2 million referred to in your document on page 20: “initial allocations and additional funding, as a result of the justification-led bids”? Very briefly, is that linked in any way to the annual canvass?

Peter Lee: Individual electoral registration adds additional costs on to the canvass, because we run the canvass and then local authorities have to register people to vote. Because that cost more than the previous system, whereby they basically sent out a single canvass form to each local authority, under new burdens we provided additional funding for local authorities to cover the cost.

Lord Campbell-Savours: In the event that the annual canvass is finished, will there be a saving to the Government?

Peter Lee: The annual canvass will not finish. It carries on, but it is done in a different way, in effect.

Lord Campbell-Savours: Yes, but will there be a saving, not just to the local authority but to the Government?

Peter Lee: Yes.

Lord Campbell-Savours: That could be the driver behind it as well.

Peter Lee: It certainly makes it more attractive for the Treasury.

Lord Campbell-Savours: Thank you. That is all I need.

Q6                Lord Lexden: I am a trustee of the Hansard Society. That is the interest that I declare.

Could you please tell us how the Cabinet Office works with the Electoral Commission and the devolved Administrations on registration matters? Perhaps Northern Ireland could be included in the answer. How do you coordinate activities among the various offices and ensure maximum cooperation?

Peter Lee: I may well ask Simon to chip in on the elections side. We work incredibly closely with the Electoral Commission. I chair the lections and registration board, which meets every couple of months. That provides overall strategic steer for our work on elections and registration but also connects the Electoral Commission, the local authority sector and the electoral administrator sector at that level.

We talk to the Electoral Commission daily on all matters. When we are in the runup to electoral events, it will be even more often than that. They are on all our main governance boards, such as the electoral integrity board, which looks at the introduction of voter ID, and the registration reform board, which is driving the canvassreform implementation working group. Ministers engage very regularly with the chair of the Electoral Commission.

It is a very close working relationship, while respecting the very clear boundaries between us and the Electoral Commission. We have very clear and separate roles.

The Chairman: On that very point, though, is there absolute clarity about the rights, duties and responsibilities between you, the Electoral Commission and, indeed, the electoral registration officers? Do you send postcards to the electoral registration officers, or do they only deal with the Electoral Commission? It would just be nice to get that clear.

Simon James: I am happy to answer that. In all our work, we are acutely aware of where legal responsibility lies. The responsibility for maintaining a register is legally with the electoral registration officer. That is a fact in blackandwhite law. Exactly the same goes for the running of a poll: that is the responsibility of a returning officer. You add to that the Electoral Commission, which has a responsibility to provide guidance and various other responsibilities.

We are helped by the fact that those things are written down. We are also helped by the fact that each of those organisations, for different reasons, is fiercely proud of their independence. The Electoral Commission would not take kindly to being told what to do by the Government. I am fairly sure they would say that to you themselves, and ditto local authorities and so on.

There are times when responsibilities are at risk of being blurred. At those times, we have to be very careful to write things down and to make sure that we understand the basis on which we are doing things. That individual legal separation of responsibility is very helpful for us, but it does not stop us talking, being very open and working on a daytoday basis with each other.

Peter Lee: To touch on the devolved Administrations, helpfully within the constitution group we have not only the teams responsible for elections and registration but a team responsible for devolution. The main source of expertise on the devolution settlements is within the constitution group.

I am now incredibly keen that we work really well with the devolved Administrations, and as far as possible we coordinate and keep divergence between the ways in which electoral systems work to a minimum, but we also respect the devolution settlements.

Both those things are important. We must respect the devolution settlements, but all the devolved Governments and the UK Government are clear that we need to make sure that the systems do not get confused unnecessarily. We have made it entirely possible to implement votes at 16 in Scotland through all our systems while not introducing that within the UK.

Q7                Lord Hayward: Can I first declare an interest, to the extent that I receive hospitality from ComRes for an annual lobby briefing on elections? In fact, the lobby briefing is dependent on the regularity of elections, so in some years there is more than one. Can I also indicate for the record that Mr James and I have worked together previously on boundary reviews? I do not think it causes any problem of conflict, but I just wanted it on the record.

In relation to boundaries/registration, when I first started working on this, local authorities had four different computer systems. It was complete chaos trying to get information about electoral rolls even from one authority, let alone all of them. We now have a position whereby we have a database, but at each election huge numbers of people try to establish whether they are on the electoral roll, in some cases several times.

Are there any plans to have a central online system so that people, while they cannot alter the system, can check whether they are on the electoral roll?

Peter Lee: No. That is the oneword answer. There are no plans to do that. Let me explain where we have got to, because this is not something that we have not thought about. Local authorities and EROs regularly raise it as an issue, particularly around election time. It is an issue that people will register to vote unnecessarily, particularly now it is so easy. If anyone has done it online, it takes about three minutes. That has driven a greater number of duplicates than may have been the case before.

We took a good look at this. Our conclusion was that the cost and complexity of doing so outweighed the benefits. We developed some numbers on that. We estimate that the cost to local authorities of the current duplicate situation is about £1 million in additional bureaucracy. In effect, they get an application in, and they have to work out that it is a duplicate and then reject it. The cost of a cloudbased IT system that would enable that checking was kind of in the ballpark of £7 million, but that would require a centralised database, which the Government do not have any plans for. This is a longrunning issue for the Government. To have one that anonymously sucks in all the local authority information is considerably more expensive than that. There is a basic point about where the costs to the taxpayer outweigh the benefits of doing so.

The other thing, more positively, is about what we have found to be effective. Since we started trying to tackle this problem through other methods, we have found that a big-bang IT solution might not be the most effective way of tackling it. Using behavioural economics, behavioural science, nudging and communications can deliver the same results for virtually nothing.

Some things that we have done, such as changing the messaging on Register to Vote, have had a significant effect. Our way of measuring it is the number of people who, once they start going through the registration process, drop out. That almost doubled during the last registration window, because we changed the messaging to make it clear that if you have previously registered to vote, you have not moved house and so on, you do not need to register to vote.

If we can solve the problem through that and then do similar things locally about the way they communicate around election time to reduce the duplicates to a more minimal level, you can get most of the benefits for a fraction of the cost.

Lord Hayward: I have a very quick observation. The words “government IT system” and “success” are not often used in the same sentence, but you are suggesting that it would cost only £1 million to ring up councils and ask to check information on hundreds of thousands of people. Can I have a written analysis of how you come to those two figures?

Peter Lee: Yes, absolutely.

Lord Hayward: To be honest, I do not think they are credible.

Peter Lee: I made the same point when I saw them, but I reached the conclusion that, even if they are wrong by two or three times, it will still be a significantly lower cost than the alternatives. I share your point. We want to use big bang IT solutions only when they are necessary. We are not convinced that, in this case, a solution that requires the creation of a national database, which has a whole set of separate issues, is one Ministers find attractive.

Q8                Lord Wills: Can you tell us what your priorities are for how the Government work with electoral registration officers and, indeed, the Electoral Commission, to make sure that registers are well maintained and elections are effective administered? Can I ask you particularly to look at funding in relation to that? How far are you monitoring what local authorities actually spend, particularly on registration? They have been particularly hard pressed over the last few years. Do you keep tabs on whether they are cutting back on the money they spend on this?

Peter Lee: Yes, but where we are providing direct grant. Where they are funding activity through RSG, it is virtually impossible now to tell. The transparency and the reporting requirements in local authorities have changed dramatically over the last few years. Tracking and making sense of whether they are spending more or less on registration activity is not something we or the Electoral Commission do.

We have a measure of performance management. It is not so much about what they are spending as what the outcomes are. Are they achieving the outcomes we would expect? This is the sort of thing that particularly the Electoral Commission does. It is responsible for providing a bit of oversight of the activities of EROs. It is quasiperformance management. The responsibility lies with the EROs, though. They are legally responsible for ensuring that their registers are as complete and accurate as possible.

We do know how much we are spending in order to deliver IER, because that is direct grant. That has been coming down since we first introduced it. First, it is always hard to tell up front how much things are going to cost, but also they have discovered efficiencies and better ways of doing things. They have been able to drive down that cost to about £15 million a year. Beyond that, no. We and the Electoral Commission tend to look at outcomes. There is a mechanism by which, if it thinks that local authorities are not doing what is necessary, it can recommend to the Cabinet Office that we issue a direction to a local authority saying, “You need to up your game”, but thus far we have never had to do that. That suggests that the activity we need them to carry out is largely being done across the country.

Simon James: On the running of polls themselves, which was part of your question, we work very closely with local authority officials and the Electoral Commission. Peter has listed a number of our groups, but we have one called the electoral coordination and advisory board, which is a regular meeting with the Electoral Commission, the senior regional returning officers across the country, me and other representatives from the Government. That meets regularly, but also meets frequently in the runup to major electoral events, because major coordination is required between each of the different organisations there. We meet and speak frequently.

On the funding side for the administration of polls, it is worth reminding ourselves that it is a separate pot from the funding of registration, which remains a local government function. The administration of centrally run polls, very importantly, comes from the Consolidated Fund. We regulate that flow very carefully. I do not think I am giving too much away by saying that local authorities would say that we scrutinise that too much. They would like less scrutiny, but there is a balance there between giving local authorities the freedom to make the decisions they understand, about how to run a pollwhich local school or hall to book for the count—and value for money for the taxpayer. We have a very close eye on that process.

Q9                Baroness Pidding: Mr Lee, you touched earlier on one of the purposes of the Act: tackling fraud. How effective has it been and what evidence do you have of that? I know you said you were waiting for the reports to come through, but what evidence makes you think it has already been effective in tackling fraud? What can be done to make it even more effective?

Peter Lee: As we discussed at the beginning with Lord Wills, the actual data was not strong on the number of cases of fraud. It was hard to track when the Act was introduced, and that was acknowledged when the Act went through Parliament. We know for a fact that a fairly obvious route for fraud has been closed down. We are confident that people on the register are who they say they are, because we are able to check them against a national database. The ease with which one could commit fraud in registering to vote has dramatically reduced.

The main focus for Ministers is now at the other end of the process, on voter ID. That is the other big gap in the defences: it is relatively easy to commit fraud at a polling station. If you have the necessary information, you can do so. Ministers are focusing on that now. Simon is probably best placed to talk about that.

Simon James: I am happy to mention that briefly if it would help. Rather like the statistics Peter was giving, the Electoral Commission produces regular statistics on personation; that is, pretending to be somebody else in a polling station. Critics will say that those figures are relatively low in terms of justification for the Government’s response, which is voter ID.

I would say exactly the same as Peter. Just because those reported figures are low, it does not mean the incidence of fraud is low. There is probably a fairly big correspondence between the number of people who have their car stolen and the number of people who report that their car has been stolen, because it is pretty obvious. If you were not going to vote anyway, and somebody went into a polling station and used your vote by pretending to be you, you might not know about it unless it was detected. There is a missing link here between detection and reporting, and the actual levels. We find that at both ends of the system.

Q10            Lord Campbell-Savours: Can I take you back to when the Act was originally introduced? We were told that the driver behind this Act was fraud; that is how it was sold to Parliament. I remember the debates in this place, in which I was involved. If it had not been for fraud, do you really believe we would have introduced the legislation? What do you think the driver was?

Peter Lee: There were probably two drivers for it. One remains the case regardless of whether you believe that fraud was taking place. It is a strong argument that there was a very easy way to commit fraud previously. It was genuinely an anachronism that the head of the household took part in the registration process on behalf of other people of voting age within that household. When this system was put in place, the head of household was presumably defined as the man of the household. That is an antiquated system that other parts of the world had got rid of already.

The philosophy behind it was that each person should be individually responsible for taking part in the democratic process. This is a system whereby that could happen. We moved away from a system where the patrician in the household decided.

Lord Campbell-Savours: But you are accepting that one of the main drivers was fraud.

Peter Lee: Yes, one of the main drivers was fraud. You do not need to know that a crime has taken place to spot where you have a potential weakness. If you are not checking that people are who they say they are when they register to vote, it is a fairly big weakness.

Lord Campbell-Savours: Have you dealt with that problem?

Peter Lee: It is impossible to say that no one could ever commit fraud in voter registration, but it is now quite difficult. Because of our ability to check people’s details against the national insurance database, it is extremely difficult to commit fraud. It is much more difficult, but not impossible. To a certain extent, if you are that desperate to commit fraud within the democratic system, at the moment it is easier to do so through the voting process rather than the registration process, where it is now pretty difficult.

Lord Campbell-Savours: We have identified 17 target areas where there is a major problem or there has been a major problem. They are referred to in a number of documents. Some of us argued at the time that there was no need for the legislation at all, and we needed to spend the money on rooting out the problem where it existed. How do you respond to that? Let us take my county, Cumbria. We had no problem. We have a whole system set up in Cumbria doing all this. Many of us believe it to be completely unnecessary. We feel that we have had it imposed upon us to resolve a problem in a few areas of the country.

Peter Lee: In general, we have discovered, through the more worrying parts of our electoral system, that you have to use all the tools in the armoury. You can stand an argument up that says, “It is a nationwide solution to tackle a problem that you might perceive as being localised or relatively small”, but we know it works. We know the system will make it very difficult to commit registration fraud. Targeting it where it was happening would have been very difficult, since we did not know at the time, as we acknowledged, exactly where it was happening.

Lord Campbell-Savours: We did, because I produced in Parliament a whole list of the areas that were at risk. They were put on the record, and many of us believe that was the case. We asked for targeting, but the Government insisted on national coverage to cover these individual areas. Even to this day I am trying to find out why. Why were we ignored?

Peter Lee: I was not there at the time, so I am defending my predecessors of five or six years ago. It is always an alternative to try to enforce the existing rules. Enforcement is expensive and invasive, and it is invasive to people who are not doing anything wrong. It is a little like when you crack down on speeding: you get the people who are speeding and they do not like it, but others do not like it either. It is the same when you crack down on parking. You can always do enforcement of the existing rules, but if you can change the rules in arguably a relatively simple way—

Lord Campbell-Savours: I will ask you a final one, very briefly. Was there an element of political correctness in this?

Peter Lee: Crikey.

Lord Campbell-Savours: I am sure you know what I am driving at.

Simon James: I accept the points you are making. This is not a case of either/or. I would say that the Government can change the law and introduce a nationwide system, but that does not mean they should not continue to do other initiatives. For example, in 2018 the Government, alongside their voter ID pilot, piloted in a number of areas—I mention Tower Hamlets, for example—more stringent measures and checks on postal voting. That involved an injection of cash, and it was targeted. It was done with the voluntary participation of Tower Hamlets. I need to be very clear about that: we asked for volunteers.

The two work in tandem. Just because one has changed the law to introduce a nationwide system, it does not mean that cannot be done in tandem with more targeted responses.

Q11            Lord Wills: It is a fundamental point. After all, fraud was a big justification for the Act in the first place. Of course, I accept the difficulty with the data. I absolutely take your point about the difficulty of proving when it is hard to prove, as it were. I am interested that you rely on personation as the fraud this was designed to tackle, rather than postal votes. You have just mentioned postal votes. My understanding is that postal votes are by far the biggest risk and that it is a much easier fraud to commit than personation, for example. I wonder whether you could say a few words about how the Act tackled the problem of postal vote fraud.

Simon James: I am very happy to outline that once I find the correct page. I am happy to focus on what is in the Act, but there is a missing piece here that talks about what the Government are doing further to the Act, which takes us beyond the measures outlined in the Act.

Lord Wills: Specifically on postal voting.

Simon James: In terms of what the Act did, I would give the example of notifications about rejected postal votes; that was there to enable electoral registration officers to write to people who had voted by post where an error had been found in that postal vote. That was one way of perhaps signifying that there had been an attempt to use fraud in that postal vote.

Lord Wills: How would you know it was an attempt to use fraud rather than a simple error?

Simon James: We are back to the issue of data. If I may, the Government have announced more recently things that are far more pertinent to tackling postal vote fraud than were in the Act.

Lord Wills: Given that fraud was promoted so heavily as a reason for bringing in the Act, from what I have seen so far, although I am happy to be corrected, there does not seem to be much attempt to tackle what is likely to be the biggest cause of fraud in postal votes.

Simon James: At the time, the Government took a view that the main need for reform at that stage was in registration. As part of that, there was a sprinkling of measures on postal vote fraud. The Government, having worked on that and having implemented a new system, are now seized of a number of areas where they need to do further work on postal vote fraud. It is probably a question of picking priorities. The Government recognise that there are a number of areas where postal voting is open to fraud. That is why they have made announcements more recently than the Act.

This is all part of seeing the way the Act has worked, and how the measures have changed any of the reporting or the understanding of postal voting, which is why the Government are keen to announce more and do more on postal vote fraud.

Lord Wills: I am sure we will return to this in due course.

The Chairman: On that very point, perhaps it would be useful if you could drop us a note on what the Government have in mind to do now on this postal vote issue.

Q12            Lord Dykes: Moving on to an intriguing and rather elusive subject for some, how do the Government support the EROs in helping to identify and register underregistered groups? What examples of best practice are there? How are the Government sharing these in their work and notifying people?

Peter Lee: I touched on this briefly earlier. We have a programme of work on democratic engagement. Alongside generally seeking to drive people’s engagement in the democratic process, it is particularly focused on identifying and registering underregistered groups.

We talked about some of those before, such as students and young people. We have put in place legislation that requires universities to work much more proactively. For example, Sheffield University has combined its student registration system with its electoral registration system. When you go to register each year as a student, you are encouraged to register to vote. These sorts of, in effect, costfree measures are the most likely to be successful in nudging behaviour, rather than larger solutions.

We will be running a second National Democracy Week in October; we ran one last year. It is about spending a week, with local authorities, EROs and civil society, focused on getting the message out there, particularly to underregistered groups and through them to others, about the need to take part. Through that, we identify the barriers. We do not know what the barriers are for some underregistered groups. Often, only they can tell us. A lot of work is already going on at the local level through EROs. It is not just the Government driving this. Increasingly, our role is to coordinate and convene EROs, civil society, the Electoral Commission and others involved.

We are doing the things that only government can do. If something requires a legislative change, such as anonymous registration for survivors of domestic abuse, that is something only we can do. If something requires funding, that might be something only we can do. It is about bringing all our expertise and roles together, which we do through the democratic engagement programme, to target our resources on where they can make the biggest difference. I have talked about some of the examples already and I have committed to provide specifics to the Committee so you can see more of those. We largely drive this through the democratic engagement programme.

Lord Campbell-Savours: Can I ask a supplementary to the question asked by Lord Wills? Going back to the £18.2 million I referred to before, which is on page 20 in your brief, last year we allocated £18.2 million as initial allocations and additional funding. How can you possibly allocate additional funding when you do not know what the funding provision is in the local authorities? Surely, before you hand out more money, you have to find out what they are already spending. I find it hard to understand why you have not come today with the figures telling us what is being spent on this provision nationally by all the local authorities collectively.

Peter Lee: We have a twostage process, which we think over the years has got closer and closer to the amount local authorities need to spend to deliver IER. At the beginning, we expected the burden to be higher just because this was a new burden for local authorities. We have a two-stage process. We use an economic modelling approach to come up with what we think local authorities should be spending to deliver IER in their area, and then we have a justificationled bidding process, where they come back and say, “That was not quite right for these reasons. This activity will not be funded if we only get this allocation, so we are going to need more money.

Lord Campbell-Savours: So the local authorities have asked you for additional money.

Peter Lee: Effectively, the second stage is a bidding process, yes.

Lord Campbell-Savours: They ask you for it, so why do you not ask them how much they have already spent?

Peter Lee: Because of the way the funding is allocated, we cannot track local authority budgets—

Lord Campbell-Savours: I am not asking you to track them; I am asking why you do not ask them how much they are spending on this area.

Peter Lee: Part of the process by which we work out how much they are going to need is working out what they were previously required to spend.

Lord Campbell-Savours: You said “required to spend”; I am asking how much they have spent.

Peter Lee: I do not think we go back and ask specifically, “Have you spent all this money? Have you spent it all on electoral registration?” I can check.

Lord Campbell-Savours: Can you try to get us some figures in the course of this inquiry, which is going on for several months, on what is actually being spent by local authorities in this area?

Peter Lee: We will get you what we can.

Lord Campbell-Savours: The taxpayer is entitled to know how much money is being spent on electoral registration nationally by local authorities, just like on housing, social services or anything else.

Peter Lee: The only cautionary note I would offer is that for some time the Government have not tracked exactly what local authorities are spending—

Lord Campbell-Savours: You could put in a freedom of information request to all local authorities, if you wanted to, and then get the figures for us.

Peter Lee: I will provide the Committee with the best answer we have on that question.

Lord Campbell-Savours: Yes, not a dodge answer. Thank you very much.

Lord Wills: Could you do it over time? It could be over five years.

Peter Lee: I think five years is how many we have, yes.

The Chairman: We had better crack on.

Q13            Baroness Suttie: I declare an interest as a trustee of IPPR and an associate of Global Partners Governance.

You have already touched on how the Government are working with the devolved Administrations. I wondered whether you wanted to add any more on how we ensure that individual electoral registration works across the UK. In Northern Ireland, IER has been in place longer. What lessons can we learn from Northern Ireland?

Peter Lee: The main thing worth touching on with regard to electoral registration is the approach we are taking to canvass reform. This is something we have agreed and will implement at the same time as the other devolved Administrations. We implemented IER across GB. It was already in place in Northern Ireland. The same issues exist across GB with the annual canvass. We are reforming that even though it is a devolved responsibility.

We have coordinated and corralled ourselves together, politically and administratively, to do this, so it is as common an approach as is possible while respecting the devolution settlement. From its perspective as the centre of expertise on devolution, the Cabinet Office is very keen to make sure that we respect the devolution settlements but also manage those things in as efficient a way as possible.

The main lesson that was learned and then carried forward through the implementation of IER was about the need to make sure the transitional arrangements were robust and sufficient so that the register was as complete as possible by the time we finished. The big concern was that, if you did not do that, you ended up losing an awful lot of people off the register, which was also a concern as the legislation passed through the House. We learned an awful lot from the way that IER was implemented, and we provided an exhaustive set of steps, once the datamatching process had taken place, for those who were not matched, to ensure that as many people as possible were suitably registered.

One thing with devolution is the ability to look at how the different nations are doing different things, and to compare and contrast. Northern Ireland has taken a different approach to the annual canvass. Our conclusion on canvass reform is that we need to carry on conducting an annual canvass. In Northern Ireland, they abolished the annual canvass. From the Northern Irish experience, we were concerned that you would start to lose the completeness and robustness of the electoral register if you did not conduct an annual canvass. That is important for us not only on the electoral side, but for things such as jury service, in how people are selected. We took their way of implementing, but we have also learned and drawn lessons from the different way they have gone about the annual canvass. 

Baroness Suttie: Have you learned anything from the Scottish experience on improving registration of attainers?

Peter Lee: I might turn to Simon. This is basically about moving to votes at 16, having attainers at 14 and 15.

Simon James: I do not have any information to share. We will possibly write to you on that if we have any relevant information.

Peter Lee: One reason that voting at 16 remains unattractive to our Ministers is that it is more difficult, with younger attainers, to make sure that the pipeline into voting at 16 works. I can fairly confidently say there are no plans from our Ministers to implement voting for 16plus. It has been interesting to see the problems that have occurred on younger attainers, but we will provide you with more details.

The Chairman: We will give the question on ID pilots a miss, but perhaps you could write to us about the ID pilots, the potential of rolling it out further and so forth. If you could spell that out, it would be helpful. Because of time, we will give that one a miss for the moment, but it is important and perhaps you could come back to us.

Q14            Baroness Pidding: What are the key problems in relation to the annual canvass at present? Some proposed reforms were made by the Government in October 2018. What difference will these make? Can you tell us how the Government have assessed the pilots?

Peter Lee: Yes, and I will do that very briefly. The key problem with the annual canvass is that it is too prescriptive, clunky and costly. At the moment, there is no opportunity for local authorities to tailor their approach to the annual canvass in a way that suits local circumstances. If your local authority area has a very widespread and low population density, you are following the same process as if you were an innercity local authority area.

The annual canvass also focuses too much on people who are already registered to vote and whose information we already have. Through the datamatching process we are going to be put in place, we will take a very different and much more lighttouch approach. EROs will have the option to take a much more lighttouch approach where we know people have not moved or changed any of their details. They can focus on those who have much more effectively. The ability to focus resources is not just about saving money, it is about being more effective. That is what lies behind it, and that is the problem we are seeking to address.

The main thing we drew from the pilots was that a combination of approaches worked best. We tried different pilots doing different things, and we drew from that the need to use some sort of datamatching process both nationally and locally; so we do a national data step and then the local authorities can then use council tax information to check some of the 30% who are left once we have done the DWP check. The different communication approaches and how effective they are mean that a mixed approach will probably have the best set of results. Rather than going for a single approach in the pilot, we concluded that, if you take the best of all of them, you can provide flexibility and probably the most effective way of having the most complete electoral register.

Q15            Lord Lexden: Can we turn to the sections of the Act that provide for a range of further administrative changes to elections? What are the Government doing about them?

Simon James: As we mentioned previously, the second part of the Act contained a range of administrative updates and areas for modernisation. The Government keep these under regular review. I would mention here that after every major poll the Electoral Commission has a statutory requirement to publish quite a substantive report into the running of those elections. Obviously, the Electoral Commission is independent, and the Government are always interested to see what the Electoral Commission has found through its many conversations and surveys with electoral administrators and, ultimately, the public.

We take note of whatever the Electoral Commission says in those reports. I would highlight just a couple of areas. For example, the Commission’s report on the 2015 general election, which was the first one after these changes were implemented, highlighted two things of particular relevance here. One was the ability of people to vote if they were at the polling station in a queue but had not yet voted at 10 pm. That did not go well in 2010 in a very public way. The Commission noted that worked well. In its 2015 report, the Electoral Commission also set out the benefits of the extended timetable that we introduced. It is not surprising to know that I very much agree with those assessments by the Electoral Commission.

Those are the formal ways in which we keep track. As Peter said, we talk on a daily or weekly basis to administrators, who are not backward in coming forwards in reminding us of areas where reforms are working and where further reform is required.

Lord Lexden: Are reports on elections to the devolved legislatures also carried out?

Simon James: That is absolutely right. The Electoral Commission has the same duty. Now we are in a situation where the running of certain types of polls in Scotland and Wales is the responsibility of the devolved Governments, but those reports exist as well.

The Chairman: Finally from me, what other reforms to electoral administration are the Government considering to ensure the good running of elections and public confidence in the electoral system? Mr James already referred to further work on the postal vote so you might want to spell that out a bit further, but tell us about anything else that you have in mind.

Simon James: Thank you very much. I will be brief. The Government are committed to ensuring that our electoral system works effectively and is both secure and accessible. Those are the two themes that go through the Government’s work in this area.

As for the security of the electoral system, we have not so far mentioned the then Sir Eric Pickles’s report on electoral fraud. When it came out, it contained 50 recommendations. The Government have picked up 48 of those and committed to moving forward on them. Those are all about strengthening the electoral system. We have touched on one or two of them; for example, voter ID. We will happily write to you, because there is a lot to say on voter ID. I flag up that we are imminently producing an evaluation of the most recent round of our voter ID pilots, which took place at some local elections in May 2019.

On postal voting fraud, I would highlight the Government’s manifesto commitment to reform postal voting to ensure that elections are secure. The Government have said that they are looking for opportunities to introduce measures that will prohibit party campaigners from handling postal votes and apply a limit to the number of postal ballots that any one individual can hand in at a polling station. This is sometimes called postal vote harvesting, where people are able to turn up with a Tesco carrier bag full of postal votes. The Government are interested in that and looking at opportunities to make that no longer possible. Our intention is for that to be backed up by a new offence. That will require legislation, and the Government are looking for opportunities to bring forward relevant measures.

I would also highlight ongoing activity across government to ensure that our democracy is protected. There is a wider programme of work here that goes beyond the Cabinet Office to partners across government and other relevant agencies, on strengthening resilience and looking at our regulatory framework for campaigning.

Finally, the accessibility of elections is a key priority. In September 2017, we launched a call for evidence to find out how people with disabilities experience registering to vote and voting itself. Our work continues with our accessibility of elections working group, where we have very good representation from different groups representing those who, for whatever reason, may find it less easy to vote. We are working in partnership with them on ensuring accessibility.

Q16            Lord Campbell-Savours: You have spoken to us now for an hour and a quarter. Have you given similar evidence to the House of Commons committee looking at these matters?

Simon James: I have not.

Peter Lee: Ministers have appeared before the relevant Commons committees.

Lord Campbell-Savours: Was that to deal with these matters?

Peter Lee: I would have to go back. We can probably provide you with a list of appearances of Ministers at the—

Lord Campbell-Savours: It is the matters that interest me.

Peter Lee: I am pretty sure we have covered voter ID and the like. If it is possible, we will provide you with a list of our Commons appearances to discuss these matters. As far as we are aware, they have not specifically covered the implementation of this particular Act. I certainly do not remember that happening.

Simon James: I would stress Peter’s point. Those were ministerial evidence sessions rather than official evidence sessions.

The Chairman: Thanks very much indeed for coming.