Certainly. I am chief executive of the Association of Electoral Administrators, or AEA, and we are the professional body that represents those who deliver the electoral process across the United Kingdom. It includes some returning officers and some registration officers, but primarily it includes those who many of you will have come across, who actually deliver the nuts and bolts of the electoral process in the field. We are a body that represents their interests, such as liaison, training and the like, across the board.
Q182 Peter, thank you so much for joining us this afternoon. It is excellent for the Committee to have the benefit of your expertise. I wonder if I might start with two questions. The first is very general. Could you talk us through what the work of a boundary review, and after a boundary review, looks like from your perspective? To take an example, the next boundary review will finish by July 2023. Could you talk us through what will then have to happen to implement those boundaries?
Certainly, Minister, and thank you. The key point is that these are the building blocks of the democratic system. The hard work is not necessarily directly to do with the elections process, but is more to do with the production of the electoral register. In terms of how the process works for administrators, the actual involvement in whether the proposals are right, wrong or whatever is not quite at the same level as that for local government boundary reviews. It is more about providing support to elected representatives and others regarding statistics and the like, to make sure that all the relevant needs are met so that the boundary commissions can come forward with their proposals, and councils and the like can make representations through the various processes available to them.
When presented with the final outcomes, the task starts. The key point is to revise the electoral register, so a lot of work goes on to ensure that the building blocks are correct. That does not just mean the parliamentary constituency boundaries—how they interrelate with local government ward boundaries, council divisions, parishes and the like—but, following on immediately from the constituency boundary changes, there is a need to look at all the polling districts, polling places and polling stations for the elections themselves. A lot of technical work goes on behind the scenes to make sure that on polling day, the elector arrives at their polling station in the correct area, with accessible venues and things like that.
One of the huge challenges—this goes back to the outcome of the previous review, which obviously is being effectively terminated—is the fact that each individual registration officer works in the individual building block of their local authority, but parliamentary constituencies do not follow those boundaries. One of the dangers of the previous review was that an awful lot of cross-boundary work needed to take place, which means liaising with neighbouring local authorities. That sounds reasonably straightforward, and in most instances it is, but it often means that different software systems are used for the electoral register and there are different working practices.
Although we all work according to the same legislative background, there are different ways of interpreting that locally. That means trying to ensure consistency across the piece, with the electors and candidates at elections receiving the right level of service and being able to be involved. Where there is more cross-boundary work, more elements of risk come in. Effectively, when it is under their self-control, it is a lot easier for local authorities to deal with those sorts of things. It is really a communication beast between individual registration and returning officers once the actual boundaries are agreed.
Q Thank you very much, Peter. To introduce a term that we will come on to in Committee, we often talk about the Gould principle, meaning six months of preparation time for administrators and others at the working level before an election takes place. Will you explain the value of that for administrators, and why six months is a helpful amount of time for you?
Absolutely. That came from Sir Ron Gould, who did an investigation into—I think, from memory—the Scottish independence referendum, where there had been some very late changes to legislation. Anything can be planned for. With elections, as you all know, the period ahead of the polls becomes very pressurised. A longer lead-in to any significant change—a constituency boundary change would be significant—is welcome, and six months is certainly the minimum that an election administrator would want.
In the case of these boundaries, the fundamental point to bear in mind is that the electoral registers will need to be reshaped and put into their new building blocks. Whatever the case, we have
If this boundary review were to throw up some significant boundary changes—which would not be unexpected, given that, certainly in England, the data from the last Q review was from 2000—and given the principle of a bare minimum of six months between any major change and elections, what period would be the most appropriate or comfortable for electoral administrators to go from completion of a boundary review to an election based on that set of boundaries?
If I were to ask for tomorrow, that would be helpful, but I am not sure that is going to happen. In terms of the lead-in periods, we welcome the proposed spring timescale for boundary commissions to submit their reports to the Speaker. An ideal timescale would be elections taking place in May 2023, with preparations for an electoral registration cavass kicking on immediately after those May elections finish. We would then certainly look to have something by early summer at the very latest, so that, over that autumn period, as the canvass takes place, the amendments can be introduced to registers in the time for the revisions to be published on or by
Q On registers and their accuracy and completeness, we know that no electoral register is either 100% accurate or 100% complete. Obviously, there is a discrepancy between the numbers in the December 2019 register and those in the March 2020 register. Can you say something about that? We have heard different figures, but the difference between the number of people on the December 2019 register, at the time of the general election, and the number on the March 2020 register may be in the hundreds of thousands. People will have fallen off the register between December and March, so could you explain why that might be? [Interruption.] Did you hear the end of my question, Peter? I was just finishing when the bell started.
Yes, I did. Ironically, the most accurate register of electors is arguably the register that is published with the additions the month after a major poll. In the case of the December 2019 general election, applications were flooding in, but what happens over the elections process is that people are deleted from the register as a result of returned poll cards information coming through to registration officers. Ironically, it is usually the month after an election, when the updates are made, that we have the most accurate version of the register. You may well see drop-offs from the register because your processing-through information has been returned to registration officers as part of poll cards going out, postal votes for deceased electors being returned, and other such issues.
One of the huge things with regards to the
The register on
Peter, I wonder whether you can describe polling districts and polling stations in more detail. You took me slightly by surprise. You said that when you have constituency boundary changes, you then have to do a review of polling stations and polling districts. I am slightly unclear about what that means and why that is. Is it because you might have a split polling district, or is it just par for the course? Can you give us more detail on that part of your statement? How does it play into constituency changes?Q
Yes, certainly. The legislative background is that a local authority must subdivide every constituency in its area into polling districts, and then designate a polling place for polling stations. If there are changes to boundaries within a local authority area, they might not replicate the situation that is currently in place, so there would need to be a review of the provision to ensure that the newly defined constituencies and the building blocks within them are still applicable to the electorate at that stage.
We have just come to the conclusion of the statutory period for polling district review. The next one is due during the period between
Q To clarify, are you speaking about the review that takes place if a polling district is split in a constituency? Some polling districts might be dropped out of a constituency—some polling districts are coming in and some are being dropped—so you are splitting wards. Is it about redoing polling districts if a polling district is split? I am slightly unclear about the meaning of the exercise if the polling districts have not changed, even if they may have changed constituencies.
There are instances where a review would be needed—whether that is a full review or a light-touch review—to ensure that the scheme is appropriate for the electorate at that stage. There are examples—this is from my personal experience—of where a boundary change has a polling station in one constituency but it moves to another constituency in a shared district because of the nature of the buildings available. That will add a degree of complexity, with two constituencies going in where previously there had been one, so there would be a need to make sure that each of the layers there still related to the constituency.
May I first follow on from the question asked by the right hon. Member for Elmet and Rothwell about polling districts? When a local authority makes polling districts, is it simply an administrative process done on numbers and geography? Is there political or democratic input into that? How does it work?Q
It is a local authority decision, generally in full council. It depends on how individual local authorities approach this, but there is a need within the statutory process to seek views from those affected in the area and those with special skills with regards to accessibility and disability, for example. Ultimately it is, in effect, a geographical and numbers exercise, but it also takes into account what is best for the needs of the electorate in that area, which is where the political aspect comes in, with the council making that decision for the subdivisions.
Q My own constituency of City of Chester has split wards, with some shared with Ellesmere Port and Neston and one shared with Eddisbury. What administrative difficulties or issues do you have to deal with in terms of split wards? Let me ask a further question: imagine you are an administrator and the Boundary Commission has given you a couple of constituencies in your area that share wards. Do you roll your eyes and think, “Oh God, that’s a bit more work for us,” or is it quite easy to get on with split wards between different constituencies?
That much depends on the relationship between the local authorities. On the split wards situation, the returning officer responsible for running the parliamentary election in that area must comment on the review potentially undertaken by the other local authority. It very much depends again on what local practices are. The ideal situation for an administrator would be to have full control of all the areas—the subdivisions, polling stations, districts, staffing and so on —as that makes life easier for administrative arrangements. It is not insurmountable; it is purely about the local practice.
It gets slightly more complicated when we talk about combined polls. If you have a local government election and a parliamentary election taking place side by side, that adds to the degree of complexity. If it is a stand-alone parliamentary election, it is not quite as difficult to administer.
Peter, the Bill allows you to consider ward changes that have not necessarily come into effect yet. For example, in Salford, where I used to be a councillor, there has been a boundary review that should have come into force in May, but obviously the election has been delayed. Considering that, is there a preference about which set of boundaries you use? Do you find the newer, updated boundaries more useful for keeping electorates within quota and drawing more coherent seats?Q
We welcome the fact that the Bill provides for an understanding of the situation closer to when the decisions are recommended by the boundary commissions. One of the big issues is that where ward boundary changes have taken place and the new constituencies follow the old ward boundaries, there is an awful lot of complication in trying to explain that to electors and trying to change systems to reflect a system no longer in place. When you look at a map and see a boundary going straight through the centre of a ward, you are sometimes puzzled about why that is the case. You go back to how it was, based on the previous situation. It is far preferable for the parliamentary constituency situation to be closer to that of the local authority, purely for the administrative reasons of ensuring that you de-risk the possibility of sending electors, postal votes or ballot papers to the wrong area. We would always welcome the latest situation, which is as close as possible to the review, being the one that is enacted and rolled out in the electoral registers themselves.
Q If there were a situation where you could draw more coherency from the old set of boundaries, would you ever use a mix and match approach? Using the example of Salford again, most of the changes are in the east of the city, where the population has gone up quite a bit. The west is relatively unchanged, so you could leave the seat of Worsley and Eccles South pretty much intact, but you would need to heavily redraw Salford and Eccles.
In many respects, it is the certainty of what the boundaries are. One of the difficulties of the 2018 boundary review was that the boundaries had changed so significantly in some areas that it was trying to replicate them back to the areas themselves. Where registration officers are aware that a previous system—for want of a better phrase—will be the preferred system, as long as that is known well in advance, it is easier to administer than if there is a sudden change to something later on.
Q Is it fair to say that an element of the disruptive change that will be an inevitable part of this review will be down to the fact that local electoral geography has changed substantially over the last 20 years?
Absolutely. It comes back to the electoral figures that are being dealt with. Certainly, the proposed reduction of seats from 650 to 600 exacerbated it. It is 20 years since the review was undertaken, so there will be significant changes in some areas. Over time, hopefully they will be negated as we go forward, but yes, it is difficult to cope with at the moment because it has been a long time since the last boundary review.
In local authority A, the electoral registration officer will cover the area for that local authority, maybe giving that register away. That is reasonably straightforward in terms of polling stations and the like, but slightly more complicated with absent votes and postal votes. There need to be agreements about who will be leading on each individual process. In some areas, the give-away authority will administer parts of the process for the authority that has taken it in, because of software incompatibility or different approaches being taken.
Most of the challenge is about: how do you mirror local authority A’s working practice on to local authority B? Despite the fact that the law that everybody is working to is exactly the same, there are local practices that are slightly different. That comes down to the real nitty-gritty of things like how many staff are appointed to polling stations, the processes used for the opening of postal votes and things like that. It is more an administrative approach that is difficult, which means that the respective returning officers need to communicate very closely with each other, to make sure that there is no element of doubt as to the way in which processes are administered.
It would really depend on the nature of the split in the area, but, generally speaking, it is far easier to manage a constituency within a local authority area in which you are normally running elections. Equally, splitting down to polling districts, and going lower than the ward building block, may be preferable in some areas, but it could add different issues, depending on the nature of those splits. We would probably be able to cope with the odd one here and there, but if it were across the board of a local authority on a consistent basis, I could foresee that being as complicated as it would be across boundaries.
Q Are there any sources of information that electoral registration officers would like to have access to and that they currently cannot access, which would assist them in maintaining an accurate electoral register?
Much of what is going on as we speak in terms of the changes to the canvass process is about data. As you are aware, the new IER process involves inviting people to register. More access to data that allows registration officers to target those who could potentially be on the register would be welcome, be that local, national or regional. It depends on the type of data source; equally, it needs to be the right sort of data so that register updates can be done in an accurate and convenient manner.
I think the Department for Work and Pensions database is, at the moment, pretty robust in terms of checking. The Electoral Commission has done a lot of work on other sources that we have been a party to, including HM Passport Office and the Driver and Vehicle Licensing Agency. Each comes with its positives and negatives; there are lots of pros and cons. One of the things we want to avoid is the provision of data for the sake of the provision of data, because sometimes the data that we already have is more accurate than the data coming in, throwing EROs off course in terms of registration.
Thank you for your evidence, Peter. If you mentioned this earlier I did not catch it, but we have been talking a lot Q about polling districts. Could you confirm how often, on average, polling districts actually change? I have been an MP for 15 years and I could probably count two or three times we have had changes in polling districts, which should always be as a result of boundary changes for wards. Is that typical, or are they usually more regular than that?
It will vary across the UK. A statutory review must be undertaken every five years. One has just finished, and the next one is due to report between October 2023 and January 2025. In some local authorities, polling district reviews are undertaken after each major poll, just to make sure that the scheme is suitable. It depends on the fluidity of local authority areas.
Q I raise this because the boundary commission talked about how difficult it would be to look at polling districts as a unit of currency, as it were, because they change so often. How could those changes be better monitored? Iain McLean mentioned the need for more investment in geographic information systems. Is that a problem, or are the two issues separate?
They are separate, as some local authorities will have access to far better mapping tools than others. The simple answer to the question is that basically the polling districts are left to each individual local authority. How they are reported to a national sub-dataset may be inconsistent across the UK, unlike ward boundaries and constituency boundaries, which are on the public record. Because it concerns local authorities, they do report these things but there is no up-to-date central database of every single polling district sub-division, as far as I am aware.
I am not sure that I am qualified to say that GIS would be the answer to that sort of situation. Better and more complete reporting of where changes have occurred would be beneficial to all those involved in the delineation of boundaries, whether that involves GIS or something else.
Thank you for your contribution so far, Peter. I will also ask you about polling districts, and will declare an interest at this stage: in addition to being a Member of Parliament I am also a borough councillor at Charnwood Borough Council, Quorn and Mountsorrel Castle. I will talk about Quorn. In Quorn, there are two polling districts. The reason there are two—and the reason they are where they are—is that we have a football club at one end of the village and a village hall at the other end, and they are the polling stations. Is that the kind of thing that happens across the country? What is your advice on that? Do the locations of the polling stations denote polling districts as opposed to something else?Q
That is a fair summation. The legislation is currently worded to say that you start at the top and work down; the reality is that most polling district reviews are based on working upwards, based on the availability of premises. The key point for any review of polling districts is that the locations—the polling places—must be accessible to the majority of electors. In the case you have described, the decision, which was presumably made by the local authority, is that there are two good venues with good accessibility, so it would make sense to use both venues in that situation. In other cases, there will be a surfeit of venues, making it far more difficult. It really depends in many respects on what premises or locations are available. In some locations you see temporary buildings, such as portakabins and caravans, because there is physically nothing else for returning officers to use.
Colleagues, I know there are more questions to ask, but I must end the session now. On behalf of the Committee, Peter, I thank you for your time and the evidence you have provided. We are very grateful.