The Commission has carried out statistical research into patterns of non-registration

2.1 The Commission carried out research[15] by checking a sample of electoral records for England and Wales to the 2001 Census and the Labour Force Survey. The best estimate for non-registration at 15 October 2000 lies between eight and nine per cent. This rate means that around 3.5 million eligible people in England and Wales were missing from the register at that date.

2.2 Mobility was identified as a key factor in non-registration, (35% non-registration amongst those resident in a location for 2 to 3 months, but 2% for those resident 10 or more years). Non-registration was high among private renters and those in metropolitan areas, especially inner London. People from some minority ethnic groups were not as likely to be registered as their white counterparts, as were young people and those living away from home. Finally there were those who did not register as they were disenchanted with politics and voting. Few non-registrants saw benefits to being registered.

2.3 The Electoral Commission aims to encourage people to register to vote, but they cannot measure changes in registration levels except over the long term because there is no centrally held register. This situation also makes it difficult to determine the impact of campaigns. It is not difficult to count the names on an area's register, but it is hard to establish its level of completeness. There is no reliable record of the total eligibility in a local authority against which the register can be checked.

Our survey of Electoral Registration Officers supported the findings of the Commission's work on registration

2.4 Our survey of Electoral Registration Officers supports the Commission's findings on the completeness of the register. More than one third of respondents, however, thought non-registration was higher than 20 per cent. Electoral Registration Officers have the local knowledge of their areas and are best placed to estimate the levels of non-registration based on that knowledge, but their estimates have not been verified due to the point previously made in paragraph 2.3 above, that there is no source against which they can be checked.

Figure 3: Around 94 per cent of Electoral Registration Officers we surveyed believed the register was at least 80 per cent complete

The Electoral Administration Act 2006 establishes a duty for Electoral Registration Officers to ensure the registers are as complete and accurate as possible

2.5 Following the enactment of section 9 of the Electoral Administration Act 2006[16] each Electoral Registration Officer must take all steps necessary for the purpose of complying with his duty to maintain the registers. [17] The Electoral Registration Officer has to send a form to every property in the area and follow up non-returns through postal reminders and door-to-door visits (the annual canvass). Following up the initial mailing is compulsory under the new section 9A of the Representation of the People Act 1983. Door to door enquiries where no response has been received are now mandatory. Where the return for the household listing individuals and their details is unchanged, some Electoral Registration Officers have set up a system for a telephone or internet return to be made confirming that there are no amendments.

2.6 Following the initial mailing of the form, some electors will register immediately by returning their forms or using telephone or internet registration service where this is available, and where they are confirming that there have been no changes to the household. Our survey and interviews with Electoral Registration Officers found that they considered personal canvassing the most effective means of increasing registration, following the mailing of the registration form (Figure 4). Electoral Registration Officers believed, for example, that some people waited until the canvassers came round to fill in the form. The effectiveness of personal canvassing comes from the local knowledge and experience of the canvassers.

Source: NAO Survey of Electoral Registration Officers

The annual canvass is a resource intensive and demanding process which depends on local knowledge and experience to reduce under- registration and inaccuracies. One local authority we visited employed 110 canvassers for up to 5 weeks. As Figure 5 shows, recruiting and retaining canvassers was identified as an issue by 44 per cent of Electoral Registration Officers in our survey.2.7

Figure 5: Many Electoral Registration Officers find it difficult to recruit staff for canvassing

Q 4.8 In your opinion how easy do you find it to recruit sufficient canvas staff?
	Respondents	Percentage
Very easy	0	0.0
Easy	15	12.6
Average	43	36.1
Difficult	32	26.9
Very difficult	21	17.6
N/A	8	6.7

Source: NAO Survey of Electoral Registration Officers

2.8 During our fieldwork we spoke to some canvassers who expressed concerns about personal safety and also about the difficulty of gaining access to places such as flats or gated communities. One authority had produced detailed health and safety guidelines covering threats of violence and other hazards. Other difficulties cited included the time of year of annual canvass (September-November) which means it is dark, cold, and wet; lack of trust and wariness of people answering unexpected calls; lack of understanding of other uses of the register, such as for debt collection. There is also a lack of awareness of electoral registration in general, including eligibility requirements, nationality requirements and the fact that voting is not compulsory.

Some groups of non-registrants are particularly hard to reach

Our survey of Electoral Registration Officers and interviews with other stakeholders found that some groups of non-registrants are hard to reach. This finding is consistent with the Commission's research into under registration, which identified as hard to reach groups young people and those living in multi-occupancy accommodation (such as students, over 65's in sheltered housing or care homes, and service personnel) and those living in metropolitan areas, with more mobile populations. 2.9

Our review of responses by local authority type showed that non-registration was a bigger issue in London Boroughs and large cities. Electoral Registration Officers noted that non-registration was more likely where the "household" being registered was, for example, a multi-occupancy home or block of flats rather than the "traditional" household. Disengagement, or lack of interest in the process (sometimes even a distrust of authority) were also cited by interviewees as reasons for non-registration.2.10

There are a number of barriers to registration

Our survey, interviews with Electoral Registration Officers and interviews with other stakeholders, together with the Commission's own research, highlight a number of barriers which need to be addressed if registration is to increase:2.11

• Indifference or hostility to registration in particular or authority in general;
• Limited penalties to force registration, and a reluctance to enforce the penalties[18] that exist;
• A lack of understanding of the benefits of registration;
• The complexity of the registration form;
• Literacy issues amongst those completing the form;
• English as a second language;
• Difficulties faced by canvassers;
• The number of bodies involved or with an interest in the registration process;
• Confusion amongst Electoral Registration Officers and their staff over rights to use data matching to identify under registration as well as inaccuracies;
• Scarce or limited resources available to Electoral Registration Officers.

Non-registrants see no benefits to being registered and are sometimes actively hostile

2.12 Indifference or hostility towards registration, because of peoples' misunderstanding of the purpose of the register or from suspicion that the register would be used for other purposes was an issue raised by Electoral Services staff, and senior officers; and amongst others, by canvassers we spoke with at one local authority. This reaction was especially the case in metropolitan areas with a highly mobile population, a large number of young people and people from ethnic minorities.

2.13 The Electoral Commission and Electoral Registration Officers run information campaigns aimed at overcoming resistance to registration. However, both Electoral Registration Officers and the Commission recognise that there may be a point beyond which further expenditure ceases to provide value for money, because fewer new registrations will result from the activity. For example one Electoral Registration Officer noted that it was a commonly held belief amongst his peers that a lot of extra money was now needed to overcome resistance to registration sufficiently to gain a significant number of extra registrants.

Stakeholders believed sanctions, such as fines, may act as a deterrent to people registering

Electoral Registration Officers and the Department for Constitutional Affairs noted that legislation provides for fines of £1000 for non-registration, but some stakeholders we interviewed, including some Electoral Registration Officers thought that fines were not worth pursuing and may even act as a deterrent to registration. Electoral Registration Officers were not aware of fines being imposed. Our field work in Northern Ireland found that some prosecutions are being considered to encourage people to maintain their personal details accurately on the register in the absence of an annual canvass. 2.14

People's understanding of the benefits of registration or risks of not registering is limited

Few non-registrants see benefits to being registered, [19] even though unless registered they would not be able to vote. There are, however, a number of other risks, in addition to the possible fine. It may be difficult to obtain credit, because credit reference companies make checks against the register. Some Electoral Registration Officers highlight this with potential registrants, but others are reluctant to do so because some members of the public see this as a disbenefit and a reason for hostility to the registration process.2.15

2.16 Electoral Registration Officers, staff at Department for Constitutional Affairs and other stakeholders (including the Commission) we interviewed agreed that persuading people of the benefits of registration was key to increasing engagement, and translating registration into votes. This view guides both Electoral Registration Officers and the Commission's approach based on information campaigns and partnering arrangements with groups working with those 'hard to reach'.

Electoral Registration Officers we surveyed felt that the complexity of the registration form and eligibility rules make it difficult for some people to complete the form.

Canvassers said that part of their role was to answer questions from those wishing to register, who found the registration form difficult. The Electoral Commission's rolling registration form model has been checked for plain English and Electoral Registration Officers should be encouraged to use this version. In practice, our visits identified some differences in the forms used. Most Electoral Registration Officers surveyed use information campaigns to inform the public but said that personal canvassing was an important element to explain the form or answer individual's queries. 2.17

2.18 Complexity of the eligibility rules can be exacerbated by literacy problems. One of the Electoral Registration Officers stated that literacy was an issue in their authority and is a contributory factor to fears over electoral fraud, as the canvass form allows the voter to ask for a postal or proxy form to be sent to them. Amendments to the electoral process following the Electoral Administration Act 2006 will require those applying for an absent vote to submit a signature and date of birth to their Electoral Registration Officer. While this requirement acts as a safeguard, it may also deter those with literacy problems from registering.

The complexity of the rules over eligibility is even more of a problem for those with English as a second language. Electoral Registration Officers in our survey identified non-English speakers as hardest to register: 40 per cent of our respondents said that it was hard or very hard for non-English speakers to register and they are a growing group who are often young and mobile as well. 2.19

2.20 The Electoral Commission and Electoral Registration Officers publish leaflets in relevant languages to encourage registration, but registration may be further complicated depending on the nationality of the non-English speaker concerned. Some non-English speakers will be ineligible and some may be eligible to register but not to vote in all elections. Eligibility rules are different for local, national, and European elections. Any leaflet published or other guidance given (for example by telephone advice lines or on websites) needs to be clear to avoid confusion. Our review of forms from a sample of authorities visited suggests that forms are generally clear and reasonably easy to understand, but this is a risk that needs to be managed by authorities.

2.21 Once registered, providing there are no changes, people can sign and return the form year-on-year, or confirm by telephone or on-line. On-line and phone declarations can free up resources to be redeployed on hard to reach groups. One Electoral Registration Officer considered that simplified re­registration reduced the need to send as many reminders to those already on the register and had increased re-registration rates by 25 per cent.

2.22 The case study below on increasing registration rates amongst service personnel illustrates the progress that can be made by a coordinated and thought out approach.

The Electoral Commission worked with the Ministry of Defence on an information campaign aimed at increasing registration among service personnel 1)  After the 2005 General Election the Ministry of Defence (MoD) ran an information campaign to coincide with the annual canvass. They also issued a questionnaire to a random sample of service personnel asking a range of questions about electoral registration: how many people were registered before the general election; how many voted; and how many were currently registered. The reasons for non-registration were also explored and questions asked about the information sent out regarding electoral matters. 2)  60% of respondents were registered to vote at the last general election (13% did not know whether they were or not). 42% of those not registered claimed they had not received a form. 25% were "not interested in voting", 23%"did not know how to register" and 15% intended to but did not get round to it. There were a number of reasons for the low levels of registration 3)  The overall rate of registration - at 60% - is significantly lower than the national average of around 90%. The extent of non-registration should be considered in the context that the prevalent demographic of members of the armed forces is young and mobile. But 40% non-registration level is significant. 4)  At one time Service Personnel could only register as Service voters and, once they had, this remained extant throughout their time in the Services. Changes to legislation allowed service personnel to register as ordinary voters if they wished. Those continuing to register as service voters had to register annually. However, only 39% of respondents to the survey were aware of this requirement. 5)  The Electoral Administration Act 2006 contains a power enabling extension of the service declaration from 1 year up to 5. As the length of a posting is typically 2-3 years, this would benefit personnel abroad. As a consequence, the service declaration has been extended to 3 years. This provision came into effect in January 2007. The MOD has publicised this change on its intranet. 6)  Reasons for non-registration are similar to the rest of the population: disengagement; mobility; lack of information or awareness that they needed to register, complicated by legislative change. The Electoral Commission has worked with the MoD to address registration of service personnel The MoD holds monthly operational meetings with the Commission. Relationships are good and a combined information and registration form leaflet, "Register to Vote" was jointly designed and published by the Commission and distributed by the MoD. They also ran a major joint information campaign to coincide with the annual canvass. The MoD has undertaken to repeat this each year. 7)   The MoD repeated their survey after this year's information campaign, to assess the impact since the 2005 survey. The survey will determine registration levels for service personnel in the UK and abroad, and further explore some of the reasons that cause Service personnel not to register.8)

Our survey of Electoral Registration Officers found that lack or scarcity of resources is seen by them as a barrier to fulfilling their role as effectively as they would like

2.23 In our survey, we asked Electoral Registration Officers for the levels of resources they received and for their views. The majority of Electoral Registration Officers responded that funding from the local authority and from central government was just adequate, or not quite enough (Figure 6). Electoral Registration Officers felt that only a basic service could be provided (essentially only the annual canvass) and that focused campaigns, and in some cases additional responsibilities under the Electoral Administration Act 2006, could suffer.

Source: NAO Survey on Electoral Registration

2.24 We asked Electoral Registration Officers to provide budget figures in response to our survey. We also asked for the number of electors on their register, to the nearest thousand.

Figure 7: There is a wide range of expenditure and a wide range of electoral population among the Electoral Registration Officers surveyed

Electorate	Average Cost	High[20]	Low[21]	Average[22]
Over 200,000	£518,769	£3.28	£0.38	£1.72
100,000 to 200,000	£171,445	£3.39	£0.27	£1.34
Under 100,000	£134,260	£25.36	£0.27	£2.18

Source: NAO Survey on Electoral Registration

2.25 Twenty three per cent of Electoral Registration Officers did not provide data in responding to our survey. Of the 23 per cent, one percent was unwilling to share data; three per cent specifically stated that they did not have the figures and 19 per cent provided no reason for not providing figures.

2.26 The amounts varied widely in terms of what was included in costs, making meaningful comparisons difficult. In some cases the amounts included all electoral services costs and in other cases amounts were estimates or the actual costs of registration alone. Some amounts included the department's share of overhead. Without a common basis for reporting costs, the Commission cannot establish robust performance measures, taking account of the substantial differences in the size of populations for which Electoral Registration Officers are responsible, and demographic differences between areas.

Figure 8: There is a wide variation in expenditure per elector from authority to authority

Source: NAO analysis of survey results

2.27 Based on our sample, there is no correlation between spend and levels of registration. Of the six authorities with completeness levels estimated by Electoral Registration Officers as below 80%; two had expenditure levels above the average spend per elector and the other four had below average spend per elector, although two of these were only just below average. The authority with the lowest reported levels of registration did however have the lowest expenditure per elector. Without more detailed data, both in terms of how the costs are made up and other factors affecting the authority (e.g. in terms of the electorate's demographic), it is not possible to draw firm conclusions from this data. At present this data is not always available and is not prepared on a consistent or comparable basis.

2.28 The Department for Constitutional Affairs provided local authorities with an additional £2.1 million funding, equivalent to an estimated 53 pence per elector, to support increased duties following the Electoral Administration Act 2006. Compared to the average of £1.39 total spending estimated above, the addition was significant. Approximately one third of the money, which was distributed through the Rates Support Grant, is estimated to have been used for electoral registration.[23] Electoral Registration Officers we surveyed noted that the monies were not ring fenced for registration and in their view Electoral Registration teams had not received the extra money, or only in part. One Electoral Registration Officer said "I have enough funding to do what is required of me and no more. I did not receive any of the money provided by Department of Constitutional Affairs".

Electoral Registration Officers can use other sources of information to find people who are not on the register and then canvass them but not all are aware of the extent of their powers

Our interviews and fieldwork showed that there was confusion among Electoral Registration Officers over using other data to find people not on the register. In Northern Ireland there is specific legislation to enable use of relevant data sources such as the National Insurance Number database, for checking the register. The Commission have published guidance on the use of data matching but some Electoral Registration Officers do not seem to be aware of this. 2.29

Local authorities' campaigns use both Electoral Commission and local material

Electoral Registration Officers carry out information campaigns, especially around the time of the annual canvass using Electoral Commission material, or their own information and advertising. In our survey and visits, we asked Electoral Registration Officers about the effectiveness of their campaigns and different methods of campaigning. They had views on what they thought worked but our visits revealed that there was little formal evaluation to gauge efficiency or cost-effectiveness. Figure 7 shows in order of effectiveness, Electoral Registration Officers' views on methods of canvassing Electoral Registration Officers recognise that certain groups are under-registered, but information based campaigns do not appear to be reaching them.2.30

Figure 7: Electoral Registration Officers believe the annual canvass the most effective, and focus on under represented groups the least effective method of campaigning

The Commission conducts information campaigns and provides information to the public

The Commission targets its information campaigns using a variety of methods chosen according to the intended audience. For example, home movers are targeted using Royal Mail information on those changing address. Hard to reach groups often require more resource intensive methods. Through its Partnership Grants the Commission is trying to coordinate with other organisations already working with hard to reach groups. The first grants, in October 2006, have been made to groups working with young people outside formal education, ethnic minorities and disabled people. 2.31

THE COMMISSION PROVIDES ADVICE AND GUIDANCE TO HELP ELECTORAL REGISTRATION OFFICERS OPERATE EFFICIENTLY

2.32 Following the Electoral Administration Act 2006, electoral officers can do work to promote awareness, but should have regard to Electoral Commission guidance. Central government funding for this work will be available from the beginning of the 2007-08 financial year. Electoral Registration Officers surveyed believed that the Commission had made an impact, including through their awareness campaigns, but they expressed some reservations about the Commission's understanding of local issues and priorities. The results of our survey and anecdotal evidence from our fieldwork suggest that the use of the Commission's materials is patchy. The current restructuring of the Commission, to include regional offices in England should be seen as an opportunity to build relationships and develop a greater understanding of local issues, whilst retaining a national perspective as appropriate.

There is little monitoring of the impact of Electoral Registration Officers on registration locally, and it is difficult to assess the impact of the Commission nationally on registration

Local authorities have not monitored or evaluated the impact of their work on registration levels systematically. We found that some authorities have prepared reports, and tried to gather and share information with peers. This is not, however, universal practice among Electoral Registration Officers. The Commission have undertaken tracking research into the effectiveness of various types of advertising. As part of building relationships with Electoral Registration Officers the Commission should share its research with Electoral Registration Officers. 2.33

2.34 The Commission has done this in the past through the 2002 report Making an Impact which reviewed the communication methods used by local authorities to promote awareness of electoral issues. A cope of this report was sent to all local authorities in the UK. The Commission also has case studies of local authority work to encourage participation in lections on its website www.dopolitics.org.uk. As well as tracking research, the Commission evaluates the impact of particular activities by asking Electoral Registration Officers to record the numbers of forms returned and levels of duplication. For example in 2006, 272 Electoral Registration Officers (67%) reported that they had received 46,000 new registrations. From these results the Commission was able to extrapolate that approximately 69,000 people had registered as a result of the home mover mailing.

2.35 In the longer term, a key impact of the performance measurement and standard setting role of the Commission under the Electoral Administration Act 2006, should be the gathering of data to enable assessments of the effectiveness of different approaches to enable Electoral Registration Officers to better focus resources. There should also be post campaign evaluation. This approach would require the Electoral Registration Officers to identify how a successful outcome should be measured prior to the campaign. The results should be shared with the Commission to enable benchmarking and sharing of best practice between Electoral Registration Officers. Some Electoral Registration Officers already produce information and share this with others. As part of the preparation for the introduction of performance measurement, the Commission should identify those Electoral Registration Officers already preparing data and sharing information with each other.

16   See Appendix 3 for a Summary of Key Legislation Back

17   Steps include: sending more than once to any address the form to be used for the canvass; making on one or more occasions house to house inquiries; making contact by other means with persons who do not have an entry in a register; and inspecting any records held by any person. Back

18   The penalty for non-completion or false declaration on a Registration form is a fine of up to £1,000. Back

19   Understanding Electoral Registration, September 2005, published by the Electoral Commission. Back

20   The highest spend per elector by an authority in that range Back

21   The lowest spend per elector in that range Back

22   The average figure is found by adding the individual averages for each authority in the range and dividing it by the number of authorities in that range Back

23   Research conducted by the Association of Electoral Administrators. Back