Select Committee on Speakers Committee Second Report

APPENDIX: ELECTORAL REGISTRATION: THE LYNCHPIN OF DEMOCRACY

Summary

Electoral Registration is fundamental to the democratic process. Citizens who are not registered are unable to vote. The challenge facing those responsible for electoral registration, principally the Electoral Commission and Electoral Registration Officers, is to make sure that all who are eligible to vote are on the register, while maintaining its accuracy.[2].

The most recent estimate of the extent of non-registration puts it between eight and nine percent of the population eligible to vote: around 3.5 million people in England and Wales are missing from the registers[3]^. Reduced levels of registration are one factor in low turnouts in recent elections, augmented by an increased level of political disengagement[4]^..

3.  Our survey of Electoral Registration Officers and visits to local authorities showed that Electoral Registration Officers consider they know what works to make the registers more complete. This includes door to door canvassing; cross checking to other council records; running information campaigns; and so on. But there are no data to measure the cost effectiveness of activities nationally and there is no common, agreed basis for collecting costs data. Up to 2006, the Electoral Commission had no powers to collect these data.[5]

We also found that Electoral Registration Officers do not consider accuracy to be an issue because they think that Registers are accurate: in answer to our survey, around seventy per cent thought that the Registers were fraud free to some extent, but no work has been done nationally to establish whether or not this perception is accurate. Recent publicity over electoral fraud suggests that this may not be the case..

The Commission has investigated how complete the registers are but has not carried out similar work on accuracy, so the Commission has an estimate of the extent of under-registration, but not the extent to which there are people on electoral registers who should not be. The Commission has recently established firm definitions of accuracy and completeness in regard to electoral registers, to guide further work in this area..

6.  The Commission plans to undertake research to determine the completeness and accuracy of the electoral registers (both local government and Parliamentary) across Great Britain. A separate study is being conducted in Northern Ireland. The Commission has commissioned GfK NOP to undertake a small scale, country wide study as well as determining the most appropriate methodologies to use.

The Commission believes that using Greater London for the pilot will challenge the methodology, as London's high proportion of non-UK citizens and its relatively mobile population makes keeping accurate registers particularly challenging. Also, London is the only electoral region of the UK with no scheduled elections in 2007 making it easier for Electoral Registration Officers to facilitate access to the electoral register.7.  

The Electoral Administration Act 2006 enables the Commission to set and monitor performance standards for registration and other electoral services and to collect financial information from Electoral Registration Officers on the costs of these services. The Commission has announced that it will use these powers initially in the area of electoral registration. It should use its new powers to request information from Electoral Registration Officers on the cost of the different activities to promote registration, on numbers joining and leaving the register as a result of each of these activities, and details of any activities undertaken by Electoral Registration Officers to establish the completeness and accuracy of their registers..  

9.  These data should be used to confirm successful methods of improving the completeness and accuracy of the registers. This is likely to need further investigation of the reasons for under registration and inaccuracy in the registers, and rigorous evaluation of the steps taken to improve this. At present there is no reliable data and no reliable evaluation of the work done on the registers. The Commission has conducted research with the Office for National Statistics into the completeness of the registers, but no work has yet been done to establish how accurate the registers are, because there is no agreed methodology. Nor is there reliable data on costs, gathered on a consistent basis.

Experience in Northern Ireland, where tighter controls have been introduced, is that the apparent level of registration fell. This outcome could be fewer people registering because it was more difficult or increasing controls removing inaccuracies from the register. The Northern Ireland approach may be a way forward, and the Commission has conducted research into identifying the reasons for the fall in registration as being the abolition of the 'carry forward' which allowed names to stay on the register from one year to the next. Experience in Northern Ireland also showed that individual registration was more effective than the annual canvass in increasing registration.[6].  

Recommendations

On using the Commission's new powers:

1.  The Commission should use its new powers to request information from Electoral Registration Officers on the cost of the different activities to promote registration, numbers joining and leaving the register as a result of each of these activities, and details of any activities undertaken by Electoral Registration Officers to establish the completeness and accuracy of their registers. These data could then be used to confirm successful methods of improving the completeness and accuracy of the registers.

On improving the completeness of the register:

2.  The Commission should gather examples of initiatives taken by Electoral Registration Officers to improve response rates for registration, together with other information, such as the demographic profile of the authority, and use the data to review the effectiveness and comparative value for money, of the different methods of canvassing. This could be based on work already done by the Commission starting in 2002 with their report Making an Impact. They should share this information with Electoral Registration Officers, using among other methods, the Do Politics website (www.dopolititcs.co.uk) which acts as an ideas and resource hub for the electoral community.

On improving the accuracy of the register:

3.  The Commission should reinforce its guidance to Electoral Registration Officers on how data matching can be used to validate registers for both completeness and accuracy. Some Electoral Registration Officers have cited data protection issues as a reason for not attempting data matching[7]^, despite clear guidance from the Electoral Commission. The Commission has committed to producing a single comprehensive guide for Electoral Registration Officers by late 2007 which will cover data matching, data sharing and the intersection of electoral law and data protection. We welcome this commitment by the Commission as being a valuable step forward in making these issues clearer to all.

4.  The Commission should finalise the methodology for establishing the extent and causes of inaccuracy in the register and to identify necessary changes to systems to strengthen controls.

On performance measurement:

In finalising its performance measurement strategy, the Commission should consult with existing networks of Electoral Registration Officers (both formal such as the Association of Electoral Administrators and the Scottish Assessors Association; and informal). This will enable it to identify performance measures that are a proper indicator of success and that also can be calculated without undue expense or waste of resources.5.  

The Commission should liaise with the Audit Commission and the relevant bodies in Scotland and Wales so that the eventual performance measurement regime is aligned with changes to the Comprehensive Performance Assessment.[8] 6.  

7.  The Commission should develop benchmarks for levels of election registration expenditure along with key performance indicators, to identify where resources may be a factor in performance issues. These benchmarks could include the cost of the provision of electoral services and if possible registration; the percentage of the adult population on the register; the percentage of 16 and 17 year olds (attainers) registered as a percentage of expected population of the age group[9]; percentage and number of registered entries over 12 months old; percentage of canvass forms returned (split between initial response and after follow up); and data on activities to encourage hard to reach groups to register, along with the costs and impact of the work.

8.  The Commission should use peer review for work on performance measurement, to encourage Electoral Registration Officers to buy into standards as a way to raise performance. The Commission should also explore the potential for the relevant local government audit body to contribute to this work, as part of the work of these bodies on local authorities' performance measures.

---

2   This is normally referred to as completeness and accuracy. For the definitions of these two terms, as used by the Commission please see Annex 4 Back

3   As at October 2000 Back

4   An Audit of Political Engagement 1,2,3, and 4 2004, 2005, 2006 and 2007 The Electoral Commission and the Hansard Society Back

5   This was remedied by the Electoral Administration Act, 2006 Back

6   The Electoral Fraud (Northern Ireland) Act 2002: An Assessment of its first year of operation (December 2003); Analysis of the electoral register, September 2003 to May 2004 (August 2004); The impact of rolling registration on the electoral register (May 2005); The reinstatement effect (August 2005); The Abolition of the annual canvass (May 2006); Impact of individual registration on turnout at the May 2005 combined elections (June 2006); Analysis of the 2005 annual canvass (July 2006); Analysis of canvass non-response and non-registration and their impact on levels of registration (April 2007). Back

7   Data matching is the process of taking data from different sources and cross checking it, to ensure the accuracy of one or both of the data sets. As a key principle of data protection is that data should only be used for the purpose for which it is given, there is some concern over the legality of this. Back

8   The Comprehensive Performance Assessment is done by the Audit Commission and involves assessing all local authorities in England and Wales according to their performance across a range of indicators. The system of CPA is changing. The intention is that rather than auditing every local authority in depth every year, information and data gathered from the authority should be used to identify those with the highest levels of risk. These would then be selected for further audit. Past visits and other information from other sources would also be used as part of the risk assessment. Back

9   This is an important benchmark because if 16 and 17 year olds are included on the register, they are more likely to remain on it and then to vote. Back

---

Prepared 6 August 2007