Skip to main content

Representation of the People (Annual Canvass) (Amendment) Regulations 2019

Volume 800: debated on Thursday 31 October 2019

Motion to Approve

Moved by

That the draft Regulations laid before the House on 14 October be approved.

Relevant document: 3rd Report from the Secondary Legislation Scrutiny Committee (special attention drawn to the instrument)

My Lords, the annual canvass is an information-gathering exercise which electoral registration officers, appointed by local authorities across Great Britain, are obliged to conduct each year to ensure their local electoral registers are as complete and accurate as possible. This currently involves sending a form to each residential address with a prepaid, pre-addressed envelope, which households must legally respond to, whether or not the details the ERO holds for that address are accurate. This is followed up with a further two written reminders and a household visit if the household does not respond.

This one-size-fits-all approach to the annual canvass, which incorporates numerous prescribed steps, takes little account of differences within and between registration areas. The current regime is heavily paper-based, complex to administer and stifles innovation. It is expensive and inefficient, for both government and EROs, and is financially unsustainable in its current form. Furthermore, it is clear that the current process leads to confusion for the citizen. If you have lived at the same property for 30 years, it seems nonsensical to complete and return a form every year stating the same thing.

As part of our commitment to making the process of electoral registration as smooth and simple as possible, in 2016 and 2017 we worked with 24 local authorities to design and deliver pilots across Great Britain to test whether or not there were potential alternatives to the current annual canvass process which could be more efficient and at least as effective. The evaluation of the pilots provided a strong body of evidence which has informed the development of a new, less prescriptive and burdensome canvass model that will still act as an effective audit of the electoral registers.

These regulations implement this new model. The most significant change is that the new model moves away from a one-size-fits-all approach and instead provides for a more tailored canvass. Households that have not changed since the previous year can follow a more streamlined, cost-effective process, enabling the ERO to target their resources to where responses and updates to the electoral register are likely to be required. These households will be identified through a new data-matching step at the outset which will inform the ERO which households are likely to remain unchanged. The pilots showed that between 57% and 83% of households across the pilot sites remained unchanged since the previous year. By identifying these properties, the ERO will be able to focus their attention on the properties likely to require additions to the register, and specifically to direct more resources towards registering to vote those hard-to-reach groups.

The data-matching step will involve EROs matching their data on registered electors against data held by the DWP and, where relevant, locally held data sources. Where the data the ERO holds on registered electors matches data in a national or other locally held dataset, the ERO can have a level of confidence that the details they hold on their electoral register remain accurate. The ERO will then follow one of three routes for each property. Route 1 is the matched properties route. This will be used for properties for which the data indicates that the names the ERO holds on the electoral register for that property are likely to be complete and accurate. By introducing route 1, we will align the audit of the electoral registers with citizens’ expectations, as they will now no longer have to take any action regarding the canvass unless updates are required to their household’s entry. Route 1 will also save EROs conducting a resource- intensive canvass exercise for a property where there is a high likelihood of the correct people being registered on the electoral register.

Route 2 is the unmatched properties route. This is the default route and will be used for properties for which the data-matching exercise has highlighted that there may be a change in the people who are currently registered, or not registered, for the property. This route is similar to the current canvass process but allows the ERO to use e-communications and telephone calls to communicate with electors instead of hard-copy correspondence. Route 3 is the defined properties route. This will be available for properties for which the ERO believes they can more effectively and efficiently obtain the current list of residents using an alternative approach. The ERO will be able to identify a “responsible person” to provide the most up-to-date list of people who should be invited to register in respect of the property. Examples of such properties are care homes and student halls of residence, where a care home manager or a university accommodation manager are in a unique position to provide the ERO with the information of everyone who lives in the accommodation.

In respect of all three routes, these regulations allow for more efficient and modern communication methods, such as emails, text messages, telephone calls or a short letter encouraging electors to respond using online channels rather than via post. With the introduction of these communication methods we have ensured that safeguards are in place, such as by guaranteeing that each property will be contacted and given the chance to update their details.

Noble Lords will wish to note that the provisions outlined in these draft regulations relate only to the parliamentary registers across Great Britain and the local government register in England. Under the respective devolution settlements, responsibility for the local government registers in Scotland and Wales is devolved. The final canvass reform policy was agreed between the Minister for the Constitution and counterparts in the Scottish and Welsh Governments. It will be for the Welsh and Scottish Governments to introduce complementary legislation to cover the local government registers in Wales and Scotland. We have worked closely with officials in the Welsh and Scottish Governments to enable them to create complementary statutory instruments, which they are due to lay in their respective legislatures in the coming weeks. This should enable these reforms to be in place across Great Britain registers by the beginning of 2020. This shows that, even in these most politically divisive times, Administrations of different colours can work collaboratively to ensure that the fundamental pillar of our democracy—maintaining a complete and accurate register enabling participation in elections—is given support to flourish and improve.

The EROs fully support these reforms to the annual canvass. Given their frontline experience administering the process year on year, they are best placed to understand how important modernising this process is. These regulations are the culmination of three years of collaboration with key stakeholders, such as the Association of Electoral Administrators, and the Scottish Assessors Association, which represent EROs and electoral administrators. A public consultation was also run on these new proposals, garnering responses from electoral administrators throughout the country. In addition to this, government officials visited every region of Britain presenting the proposed reforms to groups of electoral administrators ahead of the publication of the final statement of canvass reform policy in September. It is worth noting that feedback from the electoral community on the proposed reforms has been positive. In his response to the 2018 consultation, Peter Stanyon, chief executive of the AEA, said:

“We believe that using data to deliver a better experience for citizens is the right approach to take. Electors cannot understand the necessity to confirm their details each year and allowing them to be contacted without needing to respond is a step forward. It will also deliver much needed cost savings to local authorities”.

The Government have also worked closely with the Electoral Commission on the design of these reforms. Section 8 of the Electoral Registration and Administration Act 2013 provides that the commission must be consulted on certain changes to the annual canvass process. The Electoral Commission’s report on these regulations, laid alongside them, was also overwhelmingly positive. In particular, it noted:

“The canvass reform proposals should result in greater efficiency, allowing Electoral Registration Officers … to focus their increasingly limited resources on areas of greatest need thereby better meeting the objective of the canvass”.

The Government were also required to consult the Information Commissioner’s Office on the regulations. The ICO noted:

“We are satisfied that the draft SI accurately reflects the aims of the project and is correctly limited in scope to deliver the Canvass Reform … [T]he Cabinet Office has, so far, weighed the risks and benefits of the new scheme, considered its necessity and proportionality, and sought to mitigate the risks identified.”

Noble Lords may note that the Secondary Legislation Scrutiny Committee has raised some concerns around the Explanatory Memorandum, which is being rewritten to address those concerns and will be published next week. However, it has raised no concerns with the policy or statutory instrument itself. Also, it is important to note, in this momentous week with the Early Parliamentary General Election Bill 2019-20 coming through Parliament, that this draft statutory instrument has no impact on the registration for this election, and in the future will continue to maintain the high standards that electoral registration officers achieve in ensuring that all eligible electors are able to register to vote.

In conclusion, the current annual canvass process is not fit for purpose. Put simply, these regulations will give EROs greater flexibility to decide how to canvass their local areas, providing them with opportunities to identify where greater efficiencies could be made locally. They will also make the citizen experience more streamlined and user-friendly. These regulations are uncontentious, largely technical, and have the support of all major electoral stakeholders, as well as the Welsh and Scottish Governments. They also meet our 2017 manifesto commitment to,

“continue to modernise and improve our electoral registration process, making it as accessible as possible so that every voice counts”.

By bringing the annual canvass process into the 21st century, we are doing that. I beg to move.

My Lords, these regulations are important, but they appear to be being put forward in some haste. I understand that they will be taken today without debate in the House of Commons, which, ironically, leaves this place again having to scrutinise issues fundamental to our democracy.

Our Secondary Legislation Scrutiny Committee said that the Government’s Explanatory Memorandum,

“failed to provide a coherent overview of the intended policy changes”.

It described the Explanatory Memorandum as “impenetrable”, which does not inspire confidence. Therefore, there is a need for some caution—perhaps even a little suspicion—about these regulations, given the history of the issue of the annual canvass.

Perhaps we should remind ourselves that all parties agreed with the principle of individual electoral registration, but, when the legislation was being prepared, the Conservative side of the coalition was keen to drop the annual canvass. Those of us most concerned with levels of voter registration were strongly opposed to this, and Parliament agreed that it would be retained.

The proposals now, for a more targeted annual canvass, are sensible in principle. The present system is inflexible and, as the Electoral Commission warns, some necessary steps must be taken to help increase levels of registration if we are to proceed in this way with the annual canvass. I seek some assurances from the Government in that respect.

First, we should look at an issue of principle. For many things, such as a subscription to an information service, the data principle must always be one of opt-in. But you do not need to opt in to the right to have NHS treatment or the rights that are automatically afforded to every citizen under the law. So you should not have to opt in to the right to vote. If we make changes to the annual canvass, progress must be made towards automatic voter registration, in which any engagement with a state body showing that you are eligible to vote should result in your inclusion on the electoral register. If HMRC, the DVLA, the DWP, the passport authority, your local council or your university know where you live and that you are legally entitled to be on the voting register, you should be included automatically, as a right. If electoral registration officers have legitimate access to other data—for example, from credit reference agencies—they should be able to use it.

The new proposals for the annual canvass provide for an element of automatic re-registration when the details of people in a household have not changed. However, a survey conducted by the Electoral Commission some time ago showed that most people think that the electoral registration process is automatic for everyone and that they do not need to do anything to get registered. That is partly why so many registration forms are discarded, so many people are omitted from the register, and so many are unable to attend a polling station or get a postal vote for an election.

In 2015, the Electoral Commission estimated that 8 million people were missing from the electoral register. If they had been included then the outcome of the general election that year—and of the referendum the following year—might well have been different, as the youngest people were the most likely to be omitted from the voting registers. We are now preparing for a general election campaign knowing that the most recent estimate from the Electoral Commission is that around 9 million people who are entitled to vote are either not registered or not registered correctly. One does not need to know much about elections to know how important those numbers are in helping to ensure a fair and democratic outcome. Given that polling day is six weeks away today, what steps are the Government now taking to try to ensure that as many people as possible are registered in time to vote in the election on 12 December?

We have recently seen the Government spend £100 million on propaganda about leaving the EU today, which we are not doing. We have also seen them waste a lot of money producing 50p coins to commemorate leaving today, which we are not doing. But what we are doing is having a general election on 12 December, so how much money will the Government spend on advertising the need to ensure that people are registered to vote?

On the issue of principle, will the Minister confirm that the right to vote is not something that should require an opt-in principle but is a right that should be afforded automatically? The change in methodology for the annual canvass is said to result in a saving of £20 million per year. Can the Minister confirm that this sum will in future be reinvested in helping to ensure that as many as possible of the 9 million people who are unregistered, or incorrectly registered, will be included and enabled to vote in future?

My Lords, I declare an interest as someone who was involved with the Electoral Registration and Administration Act 2013. I remember the discussions with the noble Baroness, Lady Finn, and others within the Government about the change from household to individual electoral registration; and the frustrating discussions we had then about data matching and a wider modelling of individual data held in the Government’s hands, which would have made this all a great deal easier. When the Minister says—six years later—that we are bringing electoral registration into the 21st century, that is a bit generous. We are bringing a little more modernisation to electoral registration but it is not that modern yet, compared to a number of other Governments who are now beginning to bring data together.

I entirely agree with paragraph 7.2 of this dense document, which says:

“In its current form, the annual canvass … is heavily paper based, inefficient and outdated, leaving little scope for digital innovation”.

This modest change does not take us very far down the road towards digital innovation.

I strongly agree with my noble friend Lord Rennard that we have to move towards an automatic system for voter registration for all citizens. I recognise, and the noble Baroness, Lady Finn, would probably agree, that in changing the system the Conservatives are more concerned with ensuring accuracy and the Liberal Democrats are—as I suspect the Labour Party also is—more concerned with ensuring that we also get completeness. The gap between those we know are eligible to vote and those on the register is a real problem and a scandal. We all know of the problems in the United States between a right-wing Republican Party which does its best to limit the number of people on the register and the Democrats, who are much more concerned with completeness. We do not want to go further down that road in this country.

I am impressed by the way that in a number of other countries—for example, Estonia—digitisation and the use of government digital data has taken them a good deal further towards providing a unique identifying number for individual citizens. It is a means of access to the data which government hold on you; we in this country have so far failed on that. The verification proposals have not been very successful, after bringing in Experian and a number of other private companies. This is something that we should look at in the next Parliament. Indeed, some of us ought to propose an ad hoc Lords committee, precisely on this. Perhaps the noble Baroness, Lady Finn, and I might then return to the subject on which we battled with a number of people some years ago.

We have seen with the Windrush scandal, and may see again on the question of settled status for EU citizens, that people for whom various government agencies must have held data were unable to demonstrate that they had certain rights and had lived here for a number of years. There are various obstacles as to why the Government do not put that data together: they are legal and administrative, and there is sometimes rivalry between government departments. The move towards a unique identifying number, which is far from the old debate that we used to have on ID cards, would then feed back into making sure that every citizen was automatically on the electoral register. That would take us, at last, into the 21st century—25 years too late by then, probably.

This also links into the question of voting age. I have been converted to supporting a voting age of 16, partly because it would bring young people into the system when they are still at school. The noble Baroness, Lady Finn, will remember the remark that people would make as we debated this. We were told that young people in their 20s did not register for anything: they forgot to register for their doctor and did not register for a whole range of things. That left them, particularly young men, outside the system in which government knew who and where they were. That is unfortunate. If we move to a voting age of 16, young people will go on the register, and will have to be taught about citizenship while they are still at school—another scandal which we need to resolve. The idea that all British citizens have certain automatic rights and obligations will be strengthened.

I therefore give a lukewarm welcome to this mild move. It takes us a little further forward, but it does not take us forward far enough. When some of us suggest in the next Parliament that we should look at the broader question of the collection and maintenance of government data, and individual access to it for each citizen, I hope that it will have a warm welcome in this House.

My Lords, I should first declare an interest as a vice-president of the Local Government Association, as it is local authorities which do lots of the work here. I support a review of the current model of the annual canvass of electors and the aim of making it easier and cheaper to administer. That is important at a time when local authority budgets are squeezed but, like the noble Lord, Lord Rennard, I am a little concerned that this provision has come here very quickly at the end of a Parliament. I would have liked a much longer debate and the time to have looked at this more carefully. That said, I would maybe give it a very lukewarm welcome; that is the best you will get out of me today.

My concern is that if we make changes, we should not damage the primary purpose of the canvass, which is of course to get more people on the register and ensure that it is accurate and complete. That is very important. Whatever changes we are making here today, we know that there are maybe 7 million, 8 million or 9 million people who are eligible to vote in our country but who are not registered. We are doing very little about that, but it is an issue. Yes, we want to be streamlined and cost-effective, but we also want to ensure that we are doing the job properly and getting those people registered to vote. My concern is that we risk not getting that right.

The Government’s new model is a hybrid of the models that they have been trying over a number of years. That in itself is a little strange. Can the Minister confirm that the model proposed has been properly tested and that it is not just a case of, “We’ve tested bits of other models and put them all together”? I fear that that is what the Government have done, without actually testing the model.

I am concerned that the pilots have not been enough. We have talked about the 24 local authorities, but are we confident that the number of authorities involved, and their breadth and scale, has been right? What will the impact be on underrepresented groups? The Electoral Commission did a study of the accuracy and completeness of the register. We all know which groups are more likely not to be registered: 71% of people aged 18 to 34 are registered compared with 93% of people over the age of 55; 58% of private renters are registered to vote compared with 91% of homeowners, and 75% of people from black and ethnic-minority backgrounds are registered, whereas the figure is 84% for white ethnic backgrounds. Various key groups are not registered. Will the Minister set out how the new model will ensure that those groups are registered? If it does not do that, it is a complete failure. We know the people who are not registered. What are we going to do to get them on the register? That will be the test of whether the new model works.

I am all in favour of using technology and other methods and moving on from paper-based systems, but the new model has to work. That is what worries me. I agree with the comments of the noble Lord, Lord Rennard, in this respect.

The regulations get at best a very lukewarm welcome from me. I am concerned generally about where we are. When we get back in the new Parliament—whoever is in government—we need to look at what we do about the millions of people who have the right to exercise their vote but who are not on the register.

I thank the three noble Lords who have taken part in this debate today. I am grateful for their generous, warm—albeit slightly lukewarm—welcome and for not attacking me too much; it is much appreciated.

The noble Lord, Lord Rennard, asked whether the regulations had been debated in the House of Commons. They were debated at 11.30 am today in the Delegated Legislation Committee; the Minister for the Constitution answered questions on them, and they were approved.

The noble Lords, Lord Rennard and Lord Wallace of Saltaire, asked about automatic registration, which has come up quite often in debates in this House. They probably know what my answer is going to be. We feel that automatic registration is not consistent with the principle of individual responsibility and ownership of registering to vote. There are technical obstacles; for example, no one dataset is currently identified which would allow a registration officer to establish all aspects of eligibility to register to vote, in particular nationality. Moreover, the accuracy of other datasets would also be vital. If the data underpinning any system of automatic registration were not robust, it could lead to multiple or duplicate register entries for the same individual, increasing the risk of electoral fraud.

My Lords, to reinforce that point, government actually has all this data, which demonstrates who people are and where they live, but it has not yet been pulled together. The question that we have to discuss as part of the broader issue of government data management is precisely how, with guarantees on privacy et cetera, that can and should be done. We all know that Google now knows much more about us than do the Government—it probably knows how often I breathe per minute. It is important that, on a cross-party basis in the next Parliament, we discuss how we might take government towards a system which would allow conveniently for automatic registration.

I hear what the noble Lord says. That is a good idea and we should perhaps debate it. It is obviously above my pay grade to say exactly when and how, but I have no doubt that it will be debated in this House in the next Parliament. I will take back to the department what the noble Lord has said.

The noble Lord, Lord Wallace, also brought up votes at 16, which, again, I think will be debated. As he knows, the Government have no plans to lower the voting age at the moment, having been elected on a manifesto commitment to retain at 18 the current franchise for parliamentary elections. Again, I think this is a matter that will not go away, and I am sure that we will discuss it at a later date.

The noble Lord, Lord Rennard, asked about people registered to vote for this general election and what we were doing. As the noble Lord will know, EROs have a statutory obligation to maintain a complete and accurate register on a year-round basis. They are currently finalising this year’s annual canvass and will have recently written to those people identified through the canvass who are currently not registered to vote. These people will be sent invitations to register. As always, people will be able to register online at the GOV.UK register-to-vote website. This process is quick and easy for everyone to use. The registration deadline for the 12 December election is midnight on 26 November. The EC will be running a publicity campaign to raise awareness of this.

The noble Lord also asked how much money we were spending. I am afraid that I do not have that to hand, but I will write to the noble Lord before we leave this House on Tuesday to make sure that he gets a reply to that. He also mentioned people being omitted from the register. This is one thing that we feel will be good about this new plan, as it will give EROs much more time to concentrate on the hard-to-reach groups, which we know only too well are not registering, instead of having to worry quite so much about people whose circumstances do not change.

The noble Lord, Lord Wallace of Saltaire, asked about digitalisation and what we are going to do on that. The new system, when it goes ahead—the national data-matching service—will be provided by and through the Cabinet Office digital service. This is a highly successful service, which already provides for the verification of electors’ identity at the point of application. The data-matching service will be done seamlessly through the electoral registration officers’ software systems, with no extra burdens on electoral administrations or citizens. The electoral administrators use the Cabinet Office digital service daily to download and upload registration applications. They hold the service in high regard and already have the required IT infrastructure in place to support the connection to the service. The intention of the proposed reforms is to identify properties where the household composition has changed, allowing EROs to target their resources towards these properties, as I mentioned earlier.

The noble Lord, Lord Kennedy, talked about the pilot evaluation. Qualitative evidence suggests that staff in all piloting authorities experienced considerably less pressure under the alternative canvass models and emphasised significant cost savings in printing, postage and staff time. It was apparent that there was no interest in returning to the legislated canvass, viewed by EROs as a repetitive process and a backwards step from modernisation attempts, such as encouraging electors to use emails instead of paper forms. The feedback from the EROs, through engagement with the Association of Electoral Administrators, has been that electoral administrative staff are strongly opposed to further delays, because they feel that this is a way forward to really engage better with the citizen.

Noble Lords also asked why we are doing this so quickly. The reason is basically that this was originally set out as an SI on 14 October and was then going to be debated in mid-November. Bearing in mind what has happened with the general election and that we are not going to be here, if we did not do this, it would mean that they could not start doing the annual canvass on 1 July 2020. That is the reason that we have rushed it through today, I am afraid. I completely understand that this has perhaps not given noble Lords the time to look at this and for us to have a full debate, but that is the reason.

I think that I have answered all questions. Would a noble Lord like to jump up and say if I have not? No—good. In closing, I thank everybody who has spoken. The regulations presented here today will make the canvass process simpler and clearer for citizens and give EROs greater discretion to run a tailored canvass that better suits their local area.

Motion agreed.

Sitting suspended.