Draft Electoral Registration Data Schemes Order 2012
The Committee consisted of the following Members:
† Ashworth, Jonathan (Leicester South) (Lab)
† Banks, Gordon (Ochil and South Perthshire) (Lab)
† Bradley, Karen (Staffordshire Moorlands) (Con)
Campbell, Mr Gregory (East Londonderry) (DUP)
† Crouch, Tracey (Chatham and Aylesford) (Con)
† David, Mr Wayne (Caerphilly) (Lab)
† Glen, John (Salisbury) (Con)
† Harper, Mr Mark (Parliamentary Secretary, Cabinet Office)
Hillier, Meg (Hackney South and Shoreditch) (Lab/Co-op)
† Lewis, Brandon (Great Yarmouth) (Con)
† Newmark, Mr Brooks (Lord Commissioner of Her Majesty's Treasury)
† Raynsford, Mr Nick (Greenwich and Woolwich) (Lab)
† Redwood, Mr John (Wokingham) (Con)
† Rogerson, Dan (North Cornwall) (LD)
† Roy, Lindsay (Glenrothes) (Lab)
† Wright, Simon (Norwich South) (LD)
Kate Emms, Committee Clerk
† attended the Committee
Second Delegated Legislation Committee
Tuesday 19 June 2012
[Mr Joe Benton in the Chair]
Draft Electoral Registration Data Schemes Order 2012
10.30 am
The Parliamentary Secretary, Cabinet Office (Mr Mark Harper): I beg to move,
That the Committee has considered the draft Electoral Registration Data Schemes Order 2012.
It is a pleasure to serve under your chairmanship, Mr Benton.
The powers to make the order are in sections 35 and 36 of the Political Parties and Elections Act 2009. By enabling the sharing of specified data between the Department for Work and Pensions and local electoral registration officers, the instrument will provide the legal basis for the second electoral registration data matching trial. As the Committee will know, an initial round of data matching schemes took place in 2011. In the Government response to pre-legislative scrutiny and public consultation on individual electoral registration and amendment to electoral registration law earlier this year, we announced that we were minded to use data matching to simplify the transition to individual registration for the majority of electors, subject to further testing. That is what the order enables us to do.
Before I set out what the order does, let me provide the Committee with some background. I think that both sides of the House agree that we need to improve our electoral registration system: we need to make the register both more accurate and more complete. We also need to ensure that it is not vulnerable to fraud, but that people find it as easy as possible to register. In the initial trials in 2011, we compared the electoral register against other public databases primarily to identify people who were missing from the register, giving electoral registration officers the chance to contact them and find out whether they wish to be added to the electoral register. The 2011 schemes were also aimed at identifying potentially inaccurate and/or fraudulent entries on the register. In such cases, the electoral registration officer would then be able to take the necessary steps to remove them. The evaluation of those schemes told us that further piloting work was required, and I will say more about that in a few minutes.
An unexpected benefit of the 2011 pilot schemes was the discovery that it should be possible for a significant majority of existing electors to be confirmed as accurate to an acceptable standard by matching the electoral register against data held by the Department for Work and Pensions. We hope that that will deal with one of the criticisms of some of our individual registration proposals, that there may be a risk of a reduction in the number of registered eligible voters. In the transition to individual registration, those electors whose details were confirmed through data matching would be automatically placed on to the new individual electoral registration register without having to make a new application.
We want to test that proposition in the set of pilots run under the order, but with a larger sample so that we have some good, robust evidence and Members on both sides of the House can be confident that it will be successful. The order enables us to test the capability of the data matching process to confirm existing electors on the register. It will involve a range of areas in England, Wales and Scotland. The results of the trial will be evaluated by the Cabinet Office and the Electoral Commission and will enable us to confirm that this matching process will assist the transition in 2014 and enable us to refine the process for carrying it out.
The order enables DWP to provide electoral registration officers with the data necessary for these schemes. The 14 local authorities planning to take part in the trial are listed in the schedule to the order. I want to express to them and the Department for Work and Pensions my thanks and appreciation for the constructive work to date. The Committee will notice that 17 local authority areas are listed in the schedule. Guildford and South Ribble are included not to take part in the confirmation testing, but to pilot online electoral registration. We will not pilot it end to end, including verification, as that would trespass on the House’s deliberations on the Electoral Registration and Administration Bill, but we will look at front-end data collection.
One of the useful aspects of our new system is that we will be able to get some idea of how many electors are likely to take up an online method of electoral registration; we will also be able to see whether there are particular segments of the population for whom that method of registration would be most attractive. That will do two things: it will enable local authorities to have a better idea of the different channels that people will use to register, and it will help us to deal with some of those segments of the population who are currently not as well registered as they ought to be. I suspect—it is a hunch at this point—that younger voters in particular would be the most likely to use an online registration method.
Since the draft order was laid, Colchester council has, unfortunately, had to withdraw from the pilot scheme. That does not affect the validity of the order, because inclusion in the schedule does not compel authorities to take part, or of the results, because we have enough local authorities taking part in the pilots for the results to be robust from a research perspective.
The draft order stipulates that, before any data can be transferred, a written agreement setting out all the data protection requirements must be in place between the electoral registration officer and the DWP. It also sets June 2013 as the date by which each of the schemes must have been evaluated by the Electoral Commission. After the pilots have ended and the evaluation has been done, all the personal information and data created and held for the purposes of the pilot schemes will be securely destroyed. The Information Commissioner’s Office has been consulted on the draft order and commented that if the schemes carried out under it confirm the results of the previous pilots, the transition to individual registration will be simplified and many individuals will not be required to provide additional personal information.
As mentioned earlier, the evaluations of the 2011 schemes also concluded that further trials were needed to ascertain the potential of data matching to identify potential electors who are missing from the register.
The draft order will enable such schemes to be carried out in some areas but, if it is decided to extend the schemes to include further areas or data sets, a separate order will be laid at a later date.The Committee should also note that the Electoral Commission briefing confirms its support for the draft order and recommends approval by the Committee. I hope that the Committee sees the merits of the second trial and the benefits it will bring to work to strengthen our electoral registration system, and approves the draft order.
10.37 am
Mr Wayne David (Caerphilly) (Lab): Our intention is not to oppose the draft order, but I have a number of questions to ask and comments to make.
The second tranche of pilot schemes is important because, to be frank, the pilots in the first set were not that successful. The Electoral Commission assessed them objectively and, while accepting that they had been useful, nevertheless noted a number of deficiencies and said that the data provided were far from satisfactory. I therefore welcome the second lot of pilots.
On the number of pilots, the Minister explained that there have been three withdrawals from the list in the schedule, so we are talking about 14 rather than 17 pilot schemes. Is he confident that the remaining 14 represent an accurate cross-section of the country? Interestingly, the first pilot schemes included none from Wales, whereas on this occasion, although Wales is represented, industrial south Wales is not. Is he comfortable that the schemes in their totality are likely to produce an accurate understanding of what is possible in different parts of the country? Also, how has the list been drawn up? What were the criteria for selection? Were local authorities more widely approached and is the list those who responded? What method was used by the Government to come to their conclusion?
Secondly, I want to ask about the methodology to be used in the pilots. As the Minister has indicated, these pilots adopt a different approach that is very much based on achieving confirmation—data matching the register from the 2014 elections with national insurance records—ahead of the IER commencement date in mid-2014. Is he satisfied that the consultation has been robust on the methodology to be used? Is the Electoral Commission, for example, entirely comfortable? Will he give a commitment to make no further change to what has apparently been agreed? What will happen if the data matching exercises show a significant drop, or any drop at all, in the numbers of those on the electoral register? What will happen after that? Will there be more pilots, or will the Government simply sit back and say, “That’s a shame. Unfortunately, it was an overestimate on our part, but there you are, these things happen”?
Are there contingency plans? If these pilots are as unsuccessful as the first ones were, will the Government conduct more pilot schemes to examine carefully whether its procedures are the right ones to ensure the greatest possible completeness of the new electoral register? The Minister is on the record as saying that the Government believe that approximately 66% of entries on the register could be confirmed on the basis of the first pilots, but I looked very hard for evidence to substantiate that statement and found none whatsoever. I am therefore somewhat concerned about what the Government hope will be achieved by the second round of pilots.
As was evidenced in last night’s debate on the Floor of the House on the Electoral Registration and Administration Bill, there is real concern about the completeness of the electoral register both during the transition and when the new register is established fully on the basis of individual electoral registration. We are concerned that, as the order makes clear, the pilots will be evaluated in June 2013, which is a long time after the Bill will presumably reach the statute book. This is very much a case of putting the cart before the horse. As we argued last night, we would prefer the common-sense approach of having the legislation first and the pilot schemes afterwards.
That concern is about the legislation, rather than the pilot schemes dealt with by the order. With that major reservation about the Government’s entire approach, we will not press the order to a vote, but I would be grateful for a response from the Minister to my specific questions.
10.43 am
Gordon Banks (Ochil and South Perthshire) (Lab): It is a pleasure to serve under your chairmanship, Mr Benton.
May I pick up a point that my hon. Friend made about Wales and apply it to Scotland? There are seven areas of Scotland in the order, but all are in the central belt. Where is the logic in that? Why have seven central belt areas?
My main point relates to the fact that by 2014, new registrations will be required to be individually registered, but all postal votes and proxy votes will also require such registration. When in 2014? How will today’s order, in those seven areas, impact on the electoral register in Scotland at the time of a potential Scottish independence referendum in autumn 2014?
10.44 am
Mr Nick Raynsford (Greenwich and Woolwich) (Lab): I shall not detain you or the Committee long, Mr Benton. I simply wish to make a couple of observations, given that my area was involved in the previous pilots and will also be involved in the latest ones.
I discussed the previous pilots in some detail with the electoral registration officer in Greenwich, who made clear to me his doubts about the effectiveness of much of the data matching in those earlier pilots. He stressed very forcefully that the annual canvass is by far the most effective way of ensuring representation of the largest possible number of people.
We recognise that we are dealing with a tension between the implications of individual registration, which we understand the Electoral Commission believes is appropriate, and the risk of losing coverage of people who in the past have been registered by their relatives—I am thinking particularly of younger people—or others who, for a variety of reasons, may not take the trouble to get themselves on the register, as against the benefits of traditional family registration, where one member of the household has registered other members. We understand the arguments and reasons for individual registration, but there are risks. My hon. Friend the Member for Caerphilly has highlighted the risk of a fall in the numbers on the register. It would be helpful to know what the Government intend to do if that were to materialise.
The main point I want to stress is the importance of ensuring that there is proper evaluation of the implications of the various approaches. I take the Minister’s points. Some groups—particularly younger people, one suspects intuitively—may be more comfortable than others with online registration, and that option should be made more readily available. Clearly, however, there will be areas where that is not an effective and appropriate response and other measures may be required if we are to avoid gaps in the electoral register. It would be helpful if the Minister said more about how he proposes to discuss with representatives of electoral registration officers throughout the country the appropriate mix of measures, including continuing the annual canvass—in my view, that is terrifically important—online registration and other measures, to ensure that we maintain the most comprehensive coverage on the electoral register.
10.47 am
Mr Harper: Let me take the points made in the order they were raised.
I am grateful for the support of the hon. Member for Caerphilly for the order. To answer his question about the areas taking part, one dropped out—Colchester—and the other two I mentioned were never going to do confirmation. Based on our previous experience, I am pretty confident, because once we had started the process, worked with the authorities and engaged them in it, they all stayed in. Any that dropped out from the previous pilots did so at the beginning of the process—they looked at the scope of what they were engaged in and decided it was not for them. However, all the authorities in the schedule other than Colchester are already involved in discussions and are clear about what they need to do; they are appropriately resourced and very enthusiastic. They want to take part because they want to make sure that this process works.
On how the local authorities were selected, all those in the scheme are volunteers, which is helpful and better than our having to press them into doing it. They cover a variety and range of local authorities in Great Britain. As the Electoral Commission said, they are not a representative sample, but they represent a good cross-section of the different types of authorities. There are some urban ones and some rural. Whichever authorities we pick, someone is going to say that somewhere is not covered. The hon. gentleman said there was not one in Wales last time, but now that we have some in Wales, they are not in the right parts. In a similar vein, the hon. Member for Ochil and South Perthshire said that we have not picked the right part of Scotland. We want to make sure we work with electoral registration officers in Scotland to make sure that there are no particular issues there. So that we have a broad cross-section across the country, in choosing the areas to be involved we worked closely with the Electoral Commission and the Association of Electoral Administrators. The Electoral Commission is satisfied with the broad cross-section we have.
The hon. Member for Caerphilly made a good point about methodology. One of the reasons why the new pilots specifically focus on confirmation is to make sure we have a good methodology, that it is evaluated, and that people are confident in the results. Our evaluation
of the schemes last year indicated—the Electoral Commission also said this—that we should adopt a more prescriptive approach to the methodology used in the pilots. That way, there will be clear guidance for each of the pilot areas and we can draw robust conclusions from the evidence. We have developed that methodological framework with the Electoral Commission for this year’s pilots.The Electoral Commission wrote to me stating some of the concerns they laid out in the briefing note they gave the Committee. I wrote back confirming that we will address all those concerns, and I have done so. We have agreed on the important points and are now working on the details, but when the pilots begin we will have an agreed methodological approach that we have agreed with the Electoral Commission.
Mr David: Could a copy of the Minister’s letter to the Electoral Commission be placed in the Library?
Mr Harper: Yes; I am very happy to do that. The letter addresses each of the points and sets out what we going to do.
All the pilots have been asked to agree to the methodology before the work begins. We are also ensuring that we learn the technical lessons from last year’s pilots about the exchange of data between the EROs and the DWP. We can make a real step forward on the sophistication of the algorithm that matches the data and decides whether or not we have a match, so that EROs do not find the process as resource intensive. I am sure that the right hon. Member for Greenwich and Woolwich will know from talking to the EROs that a lot of resource was involved and that a lot of the data generated were not very useful. By making the algorithm more sophisticated, we can get the confirmation required, without generating as much work for EROs.
Mr John Redwood (Wokingham) (Con): Does the Minister have in mind a figure for the amount of fraudulent registration under the current system that he hopes to tease out by better practice?
Mr Harper: One of the problems is that we do not have a very good handle on fraudulent registration—either deliberate or inadvertent. My right hon. Friend puts his finger on a good point, on which we have been clear throughout: the process is about accuracy and completeness. I gently suggest that the hon. Member for Caerphilly seems to be too focused on completeness, and I did not hear him mention an important focus on accuracy. The Government have been very careful about that. We understand people’s concerns about those who are eligible to vote who are not on the register. That is very important, but it is equally important that those who are not eligible to vote should not be on the register. Right from the beginning, when I set out our proposals, through to how we have conducted the pre-legislative scrutiny and introduced the Electoral Registration and Administration Bill and the order, we have tried to consider both aspects, so that we end up with a system that is more robust and in which the public have confidence. Members on both sides of the Committee think that that is important.
Mr David: I do not want to repeat the arguments that we had on the Floor of the House yesterday; but for the record, it is important to say that of course we want to see accuracy, but we honestly believe that the Government have not placed enough emphasis on completeness. It is vital that everyone who is entitled to vote is included on the register.
Mr Harper: I have taken great pains at all stages to make clear the Government’s position. All the work that we have been doing with officials, the Electoral Commission and IEA, all the messages we have given out and the fact that we have significantly changed our approach in number of key areas in respect of removing the opt-out, introducing civil penalties and the language that we used, as well as the confirmation strategy that the pilots are intended to support, will, if successful, take a lot of the risk out of the system. That will ensure that we have an accurate register, because we will have confirmed that those electors are real people who live at the addresses that they have given, without having to write to every elector in that short period. That is a sensible step forward.
The hon. Gentleman said that we need more pilots. Obviously, we will look at the evaluation from the current pilots, but I hope that we will not need more of them, although that depends on the circumstances. If, based on robust methodology and clear evaluation, the results of the pilots reveal that we need to test some specific things, we will look at that again. The option to pilot again remains, but we are pretty confident that how we have set out the pilots and the robust framework that we have agreed with the Electoral Commission will not just give us the confirmation that we require, but, once the commission has published its results, will convince Members as well.
The hon. Member for Ochil and South Perthshire asked about locations, and I addressed that in my response. He had a specific question about the Scottish referendum. Clearly, a date has not yet been set for the that, and we will have to look at it when it happens. He will notice in the legislation that there is some flexibility about when we switch some of this on. I am not going to start speculating—we do not know when the date will be—but I am confident that any unexpected electoral event will neither be disrupted nor disrupt the transition to individual electoral registration.
Gordon Banks: The Minister will, I hope, agree that large aspects of rural Scotland will be lost from the pilots, because of the centrality of the pilots’ location. From a population point of view, that is perfectly acceptable, but from a rurality point of view, that is unacceptable.
I want to push the Minister on the autumn 2014 issue. What impact will the order have on the electoral registration of people in the seven areas in Scotland?
Mr Harper: On the first point, rurality is an important issue. I think the hon. Gentleman will recognise that, if we were to try to cover rural and urban areas in Scotland, Wales and England, we would end up with a large number of pilots. Rural areas are represented in the pilots, although not in Scotland, but we have covered all parts of the UK and different types of local authority.
On the specific issue, the pilots will have no impact on 2014 at all. They will enable us to be comfortable that our confirmation process will work and ensure that we can move a significant number of voters over to the new register without having to write to them to ask them to provide identifiers. Clearly, that is helpful in making the strategy much less risky.
When the date is set for the referendum, we will look at what impact it will have. I will not start speculating, because we do not know when it will be. The Government’s position remains that the decision should be put before the people of Scotland at the earliest available opportunity, so that a decision can be taken. The Government are confident that the people of Scotland will make the right decision and will want to stay part of the United Kingdom. Obviously, the negotiations between my right hon. Friend the Secretary of State for Scotland and the First Minister are ongoing and are yet to reach a conclusion, but the hon. Gentleman can rest assured that, when the date is set, we will be happy to set out for the House what, if any, implications there are for individual registration. We can then ensure that colleagues are comfortable that there will be no issues. I am happy to come back to have that debate at the appropriate time.
Mr Redwood: It might help the Minister to know that Berkshire is different from the other home counties in having a complete set of unitary authorities. It is not represented in the list that the Minister has given us today, but we do not feel bad about it at all. We think that the issues that affect us will be captured by other authorities that are not directly comparable, and I hope that the same will be true for Scotland.
Mr Harper: I am grateful to my right hon. Friend for that helpful point.
The last few issues were raised by the right hon. Member for Greenwich and Woolwich. He made some sensible points, as I would expect from a former Local Government Minister. He talked about data matching and canvassing for identifying unregistered voters. That is a good point. One of the interesting things from the data matching that we carried out last year in respect of identifying missing voters is that he is quite right that the annual canvass is still effective, which is why we are keeping it. One of the debates that we had on the Electoral Registration and Administration Bill yesterday was about the power that we have retained to get rid of the canvass at some point in the future. Northern Ireland has demonstrated that it is possible to have a register that is more accurate than and as complete as the one that we have in Great Britain without a canvass, but we found that a canvass in Great Britain was cost-effective. We do not want to shut off the possibility of moving to a better system in the future. However, the Bill contains controls that would ensure that, if we wanted to get rid of the canvass, the Electoral Commission must report on whether the alternative proposals would be at least as effective. An order would have to be laid before Parliament, thus enabling both Houses to vote on it. That is a robust check. The right hon. Gentleman is quite right that the canvass is robust.
The right hon. Gentleman also made a point about locally appropriate methods. I have been keen to ensure that EROs still have the power to use the processes that they think are most effective locally. I know from my
own experience that local authorities that have a high population turnover often adopt a very different approach. In some cases, they do not mail out the canvass at all; they do a 100% doorstep drop and hold a conversation with electors, because they find that much more effective. We therefore want to provide local authorities with a range of tools.One of the things that we have set out, which I hope will reassure the right hon. Gentleman, was that confirmation in our previous pilots was 66% on average. The figure varied to some extent; I think it was plus or minus approximately 10% in different areas. We will consult local authorities on the funding mechanism this summer. We have made it clear that baseline funding will be available to them, but we recognise that some will not be able to confirm as many voters as others and will have to do more work. The funding mechanism will reflect the difference in effort required, to ensure that local authorities can properly fund the work that they need to do.
I hope that my concluding point will make the right hon. Gentleman more relaxed. Moving from household to individual registration is a big change, but it carries opportunities as well as risks. In Northern Ireland,
for example, I saw the benefit of the system with younger voters. Now that the electoral officer engages with them directly through schools, Northern Ireland does a better job of getting 16 and 17-year-olds on the register ahead of their 18th birthday than we do in Great Britain. The electoral registration officer and his team do a better job than relying on mum and dad to do it.I have also had feedback from organisations that represent disability organisations, which think that for a lot of disabled people an online model of registration is much more accessible and much more likely to get them registered than a purely paper-based system. They think that there is a real opportunity here. Yes, there are some risks, which we need to manage and mitigate to ensure that we avoid undesirable outcomes, but there are some opportunities in getting some of the under-registered groups to register. I hope that I have been able to answer the Committee’s questions and that it will approve the order.
That the Committee has considered the draft Electoral Registration Data Schemes Order 2012.