Individual Electoral Registration and Electoral Administration
Written evidence submitted by Dr Toby S.James,
Department of Political and Cultural Studies, Swansea University (EA 04)
1. The Government has recently published proposals to reform election administration in Britain (Deputy Prime Minister, 2011). There are numerous changes in the proposals but the principal ones are the introduction of individual electoral registration (hereafter ‘IER’) and voluntary electoral registration.
2. This briefing aims to map out the likely impact that IER might have. It does so by noting some of the findings of published research on election administration but also draws from interview evidence that the author has collected as part of an ongoing research project on election administration.
3. IER would be one of the most significant changes to election administration that Britain has seen since becoming a democracy. It will force electoral administrators to undertake significant and costly administrative changes. At a time when a number of other changes are being made to electoral law in the UK, and local government budgets are being cut, there are concerns about the funding of elections.
4. Levels of registration in the UK have been in decline for some years (Electoral Commission, 2010b). The paper suggests that IER is very likely to accelerate this decline. Although it is not considered in depth in this briefing, it is anticipated that voluntary registration is also likely to reduce the numbers on the electoral register.
5. If IER is to be introduced then it is recommended that:
· The long-term funding of election administration is duly considered, given the context of local government cuts.
· Other provisions should be put in place to boost voter registration such as enabling voter registration when citizens access other government services. Lessons can be drawn from overseas innovations.
· Issues of voter accessibility are fully considered.
· The views of citizens towards the registration process should be carefully monitored towards the registration process once during and after the implementation of IER.
Context: Change in UK Election Administration
6. There have been very few changes to electoral registration in the UK until recently. Electoral registration has been the responsibility of local government since the Representation of the People Act 1918. The Electoral Registers Act 1949 made persons of age between November and June each year to be included in the electoral register. Electoral registration has largely been unchanged until the turn of the century.
7. New Labour undertook an electoral modernisation programme while in office but most of the changes related to the procedures for casting a vote. The Representation of the People Act 2000 introduced continuous rather than annual registration, provisions for the homeless, citizens in psychiatric wards and remand prisoners to be included onto the register. The Electoral Administration Act 2006 introduced performance standards for local authorities and placed a legal requirement for them to undertake door-knocking as part of the annual canvass. The aim was on improving registration rates and targeting ‘democracy deserts’ (James, 2010b , 2011).
8. There is strong support for individual registration from many organisations. The Electoral Commission has supported Individual Registration since 2003 (Electoral Commission, 2003). The ‘Birmingham case’ in 2004, when election court uncovered fraud in local elections in two wards, led to even greater calls for its introduction (Stewart, 2006). These have come from The Committee of the Office of the Deputy Prime Minster (2004), The Office for Democratic Institutions and Human Rights (2005: 1), the Committee for Standards in Public Life (2007: 6-7), and the Association of Electoral Administrators (2010).
9. The Labour Government was initially resistant to this because it was concerned about the effect that this might have on the electoral register. However, political concerns were also important because there is evidence that they thought that it would be their voters that would be more likely to drop off the register (James, 2010b).
Diagnosing Problems with British Election Administration
10. One of the stated objectives of individual registration is concerns about fraud. According to the White Paper ‘In the past decade there have been abuses of this system which have shaken the public’s confidence in the security of our elections’ (Deputy Prime Minister, 2011: 5).
12. However, levels of voter fraud are ultimately unknown because of the problems involved in measuring them. As prominent American scholars Michael Alvarez et al. put it:
‘…are fraud accusations like airplane crashes-infrequent but focusing events that we remember; or are accusations of fraud more like car accidents, events that occur frequently but where only the most dramatic make the news?’ (Alvarez, Hall, & Hyde, 2008: 10)
14. There have been concerns about the way in which elections have been administered but these do not principally relate to fraud. The most famous cases of problems were reported in the Scottish 2007 elections (Denver, Johns, & Carmen, 2009) and the UK General Election 2010 (Electoral Commission, 2010a). Analysis of these elections pinpointed the degree of complexity in electoral law, the number of simultaneous elections and poor ballot design by election officials as sources of problems (also see: Association of Electoral Administrators, 2011; Gould, 2007).
15. Research has identified a long term decline in registration rates in the UK. This was about 95% of the voting age population in the 1950s and 1960s. Estimates based on Census records suggest that the completeness of the registers was at 93.5% in 1980, 91–3% in 1990 and 91–2% in 2000 (Electoral Commission, 2010b: 1-2).
16. There are therefore other pressing problems facing British elections in addition to fraud and perceptions of fraud.
Election Administration, Registration and Turnout: The Existing Scientific Knowledge
17. It is well established in the political science literature that different forms of election administration can affect voter participation. There is a plethora of studies from political scientists in the U.S. dating back to at least the 1930s (Harris, 1934), but this research has accelerated over the last thirty years, especially since the U.S. 2000 Presidential election.
18. This often deploys a rational choice logic that some forms of election administration create barriers to participation by increasing the ‘costs’ of registering to vote and casting a vote. Individuals will be more likely to register to vote and cast their ballot when it is more convenient to do so (Wolfinger & Rosenstone, 1980).
19. Research has differentiated between procedures which are ‘expansive’ i.e. increase participation and those which are ‘restrictive’ i.e. reduce participation. I have recently developed a continuum which categorises each of the procedures according to scientific research (Table 1).
Major restrictive effect
Minor restrictive effect
Marginal restrictive effect
Marginal expansive effect
Minor expansive effect
Major expansive effect
Excessively early registration deadlines.
Infrequent updates to the register.
Annual updates to the register.
Annual purges of the canvass.
Purge of non-voters.
15 day registration deadline.
Continuous updates to the register.
In person registration.
Citizen’s responsibility to register.
Registration office opens at weekends.
Registration possible in public service agencies and ‘Motor voter’ schemes.
House to House enquiries.
Online / mail / phone return of registration form.
State responsibility to register.
No registration requirements.
20. Individual registration is currently categorised as a restrictive procedure, although the evidence base upon which this placed is limited. Most countries operate individual rather than household registration so there have been few opportunities for researchers to assess the effects of the change from one to the other. Inferences are therefore taken from the experience in Northern Ireland.
21. Individual registration was introduced in Northern Ireland after the Electoral Fraud (Northern Ireland) Act 2002. The same legislation simultaneously ended the annual carry forward of names of individuals who did not register each year. Photographic identification was also introduced.
22. There was an immediate drop in the register from 1,192,136 to 1,072,346-a ‘loss’ of 119,790 names or approximately 10% of the electorate-and a registration rate of 86%. By the third register, published in September 2004, 1,075,439 names were included-just 82% of the eligible population (Price-Waterhouse Coopers, 2006: 3-4) (see Figure 1 below).
23. Registration rates subsequently improved after countervailing expansive provisions were introduced. The Electoral Registration (Northern Ireland) Act 2005 gave the Government the power to temporarily reinstate names on the register. The Northern Ireland (Miscellaneous Provisions) Act 2006 ended the need for citizens to re-register each year.
24. Multiple simultaneous changes make scientific analysis of the effects very difficult. There have been many different interpretations about whether the reduced numbers reflected ‘real’ people or not. However, it is very likely that individual registration leads to lower registration rates than household registration.
25. If individual registration is implemented, there are a range of other procedures that could be introduced using the continuum to offset a decline in registration levels. These would include making online registration possible, which is proposed in the White Paper. It could also include making registration possible when citizens access other government services and contact other agencies. The National Voter Registration Act was introduced in the U.S. in 1993 to enable all citizens to register to vote when they applied for a driving licence and this now accounts for a substantial amount of registrations in most states. Registration could also be made possible when citizens pay council tax bills, visit Job Centres or access other services.
The Views of Local Election Officials on the Implementation of Individual Registration
26. I am undertaking a research project on the impact of performance standards on election administration in the UK. At the time of writing, I have interviewed 33 senior elections staff across 18 local authorities in England and Wales and intend to interview staff in further authorities during 2011.  These are typically the Returning Officers, Electoral Registration Officers, Democratic Service Managers and Electoral Services Managers. The latter have comprehensive experience in managing the canvassers who collect registration information and managing teams who input the forms into databases.
27. As U.S. academics Donald P. Maoynihan and Carol L. Silva note, of U.S. electoral administrators:
‘our knowledge on the views of LEOs [local election officials] remains impoverished. LEOs are the administrators of democracy. Their actions can disenfranchise voters, subvert the political process, and damage public confidence in democracy….[there is a].. need to understand LEO attitudes toward election administration and reform, as this knowledge can help explain election outcomes and the success of mandates for change’ (Moynihan & Silva, 2008: 817)
28. Their views are also significant because academic literature from public administration suggests that ‘top-down’ implementation of policies can often face implementation problems and unforeseen consequences.
29. Many of the LEO’s suggested that they were supportive of the idea of individual registration. Some thought that it would help to alleviate concerns about electoral fraud, as proposed. According to one:
‘I think something needs to be done to reassure the electorate that there is some form of double-checking that, you know, everybody needs to produce a PIN number or a signature for most things they do nowadays so, perhaps that’s what will reassure-put a bit more confidence back in the system.’
Others suggested that it was necessary modernisation of procedures that were now out of date. IER was described as ‘overdue’ and HER as outdated. According to one LEO:
‘This idea of a household form is from a very, very old fashioned time when the head of the household filled the form in. So, from that point of view I think it's great.’
According to another:
‘Britain is way behind the rest of world. It’s archaic that you have household
30. However, three clusters of concerns or arguments against it were raised.  The first theme of concerns was about the costs and administrative burdens that they thought would fall on local government because of IER.
‘I think the biggest concern now is that all that it’s doing is adding to the bureaucracy… Because with the annual canvass, you could do it by household, with individual registration, you’ve got almost, you know, for every single person some sort of contact with them. For us, all we can see is almost all the costs, you know, in monetary terms, and also how we’re going to be managing the systems ourselves, and we’re going to have to put extra resources in to managing all this.….’
Some therefore expressed concerns that they might need additional canvassers and that every visit to households would take longer because of the need to complete a form for each person. One local authority expected that their staff would need to double to deal with the implementation. The Government’s White Paper proposes paying for the introduction of IER but LEOs expressed concern for how long this money would be available for and whether it would in practice cover the whole cost to the authority of the implementation. Moreover, they expressed concerns that IER was being implemented at a time when funding to local government was being by central government and departments were asked to make savings. Election budgets are not ring-fenced. According to one respondent:
‘We are in a world where we have got no money and they are acting as if it is endless.’
According to another:
‘I tried to persuade them [the Council] the other day that we needed an extra member of staff. When everywhere else is shrinking here is an area that is expanding and could well expand with individual registration.’
31. Related to this, concerns were expressed about late implementation. Electoral administrators have often complained that they have been put under undue pressure because of the late passage of legislation (Association of Electoral Administrators, 2011). Fears were concerned that the same could happen with individual registration. According to one participant:
‘I think it’ll be a very big change to manage, and we need very clear guidance and plenty of time to get it up and running, and not it all to be left to the last minute, because that’s my concern’
32. A second theme of concerns reported by LEOs was about data quality. This may make IER difficult to administer because it might cause costs to increase and might even cause voter disenfranchisement in some cases. On the one hand the public were often prone to make errors on their forms. One LEO worked in an authority that had piloted internet voting and reported that some citizens, especially the elderly, found difficulty in providing key identifiers that were necessary for the system to work. According to her:
‘[individual registration] is designed by these intelligent people who don’t realise how daft some of the members of the public can be. It’s a lovely idea but when I was in xxxx … [we had to] get them to supply their national insurance number. We had one woman every year would give you her national insurance number. Every year you’d write to her and say "That’s not your national insurance number." It turned out it was her gas mask number from the war and she was convinced that was her national insurance number and there’s no way you could get any other number out of her.’
33. The Electoral Administration Act 2006 required those applying for postal votes to supply a signature and their date of birth as personal identifiers. However, a number of LEOs reported that the date of births did not always match. This forced them to have to interpret whether the application was valid or not. One LEO described their predicament:
‘On postal votes the number of people that don’t know their date of birth, you wouldn’t believe it actually. We don’t reject them all because the legislation actually says "If you’re satisfied that it’s that person…." So you look at it and you think "Oh the signature’s the same, the writing or the numbers look similar, alright this one’s five years out, that one’s a month out, that one’s got something completely different." If it’s completely different you think "Yes, you know, you can’t change from the 2nd of October to the 27th of April."’
34. This highlights the importance of clear and consistent guidance for LEOs about when an application should be included or not. Much of the controversy in the U.S. Presidential election centred on whether the ‘clear intent of the voter’ was clear and the inconsistent way in which different officials applied different standards.
35. Data quality issues may also arise because of data conflicts between different government information systems, LEOs thought. This might be the case when names had been inputted by an administrator who was unfamiliar with a name because it came from a different community.
‘if somebody’s name’s, let’s say me for example, my name is spelt, first name is with a K with the national insurance people and if I’m getting benefits with the WDP, you know, it’s with a Q and when [Council A] collects that information "Oh they are two different people." Are you with me, they will have to do more investigation. These are the same person, you’ll have to write to the person and stuff like that at the moment it’s being discovered. So having a long discussion with [x] this week I think we both came to the conclusion that we would probably end up with six or seven records for each person in the Borough.’
36. One LEO therefore claimed that the key strategic problem task was deciding which unique identifier should be used for IER:
‘What's the unique identifier because I think it comes absolutely for me to that issue. Do you use NI numbers for instance as the kind of identifier for people's individual registration ID or do you use NHS numbers which are a bit more secure and less in circulation and less fraudulent than NI numbers are in practice? Or do you introduce something else? Do you use passport number? There's a whole host of identifiers and we've got to get to the bottom of that really before we then can build up a system that uses that as its prime currency, that's secure, that's trusted, and that engages with other government systems, various guises, and I see that as the fulcrum of the ability to introduce individual registration actively and securely.’
37. A third theme of concern raised in the interviews was the impact that IER would have on levels of registration. Many LEOs described problems with apathy amongst the electorate. They suggested that IER would impose an additional administrative burden on citizens which would further discourage prospective registrants:
‘Well the likelihood is that the registration rate will take a nosedive …I’ve already mentioned about the apathy which surrounds elections and the difficulty of getting people to return the forms, at the moment we’re only trying to get one form per household, in the future we’re going to try and get in every individual within that household to return a separate form so I think the problems will be amplified and that the registration rate will nosedive.’
‘We’ve got the prospect of individual registration on the horizon. I’m not sure that’s a good thing, because I want to see electors registered and I don’t want to do anything that will put them off doing so and I’m a bit concerned that what’s been proposed, if it actually comes to fruition, will put people off registering. People want things easy these days, you know, they don’t want life complicated, and asking for their national insurance number is going to complicate things.’
‘It would be a shame if we end up with registration [levels] personally affected. I think we can’t put the voter off. The voter’s got to find whatever system’s got to be easy, quick, that’s what modern life’s about, isn’t it, people only do things if it’s easy and quick.’
38. Some LEOs pinpointed the young and students as groups where registration levels would drop the most.
‘I think it's going to be very, very difficult to collect the information from all these people. I've got a 17 year old son, I can't imagine he's going to be the least bit interested in filling in a registration form to be honest. I think this voluntary element is going to mean that registration levels will drop.’
‘[I]f the parent doesn’t put that young person’s name on the form we won’t get that young person on there. And I think we’re going to lose a large chunk of registration. I think we are unless it becomes compulsory or unless they begin to register from 16 onward and allow the vote from 16 onward.’
‘[P]ersonally I think we will see a dip in registration. It’s going to be harder to get registrations. For instance, students, who aren’t always living at home, they may be living away. At the moment, the present system is, you know, the mother or father can register for them, but actually trying to capture them on an individual registration is going to be quite difficult for ourselves, and there’s quite a lot of logistical problems that we’re going to face on that, I think.’
39. Other LEOs suggested that IER may raise accessibility issues for non-English speaking citizens and argued that it needs to implemented in a way that supports qualities. Under HER, co-habitant may help them with their form and this might make registration easier:
‘We have got some issues in the sense that we have got quite a large black minority ethnic community, big chunks of which are not necessarily culturally attuned to the life in the UK. Many of whom don’t speak English as their first language; those are all a whole host of barriers that you need to get across. It is going to be made worse with individual registration although there are some benefits to individual registration which are around postal vote fraud.’
One LEO suggested that some citizens may not be comfortable providing confidential data on the doorstep to canvassers because citizens may find canvassing for such detailed information intrusive:
‘If they haven’t responded, do you get their National Insurance number off them on the door step with their date of birth? A lot of people are going to say, "I’m not going to give you my…" Somebody that’s just called round to the door, "What’s your National Insurance number, what’s your date of birth?"
40. One final strand to this theme was a concern that the new procedures may take time to embed in. According to one LEO:
‘it will take time for people to understand the difference and how to in fact do it.’
Conclusions and Recommendations:
41. This briefing has considered the likely impact of IER on British elections. It has drawn from some recently published research and interviews with LEOs.
42. The published research on election administration suggests that registration levels and voting turnout decline when additional administrative burdens are placed on citizens. This and the experience from Northern Ireland firmly suggest that there will be a decline is registration levels if IER is introduced. This would accelerate the declining rates of registration in the UK.
43. The interviews illustrate LEO views on IER and locate some possible implementation problems and effects on British elections. There is some support for the principle of IER amongst LEOs. However, there are three core concerns about its implementation.
44. Firstly, they perceive it to involve additional costs and administrative burdens. The proposals come at a time when local government budgets are being cut and election resources are not ring-fenced. Measures should be put in place to ensure sufficient long-term funding of elections.
45. Secondly, they expect data quality to be a key issue. The Government is already undertaking data-matching projects and the results of these should be carefully considered.
46. Thirdly, they expect registration levels to decline, especially amongst the young. Concerns about electoral participation have been frequently expressed in the UK, especially amongst this group (Tongue, 2009). If IER is introduced into the UK then there are strong arguments for monitoring registration levels closely. It may be necessary to offset the effects of IER with other expansive methods of voter registration perhaps drawing from overseas experience.
47. The data-matching pilots may make it easier to target those citizens who are not registered. However, it may also discourage citizens from registering to vote if they think that their name on the register will be used for other purposes. The views of citizens towards the registration process should be carefully monitored towards the registration process through survey research once during and after the implementation of IER.
48. The Coalition has introduced many other changes to elections in the UK. The Parliamentary Voting System and Constituencies Act, 2011 Act will require the re-drawing of all constituencies based on the number of citizens within a district. This is enormous change by itself. However, if registration levels are altered by the introduction of individual registration then this process may have to be undertaken again when the size of the new electorate is known. The 2020 general election will have a different map to 2015. The Coalition also proposes to introduce elections for Police Commissioners. This too will involve significant work and change. Overall, there is a risk of administrative overload.
EONI. (2011). Northern Ireland Electorate Statistics - 1976 onwards. Retrieved 17 August 2011, 2011, from http://www.eoni.org.uk/index/statistics/electorate-statistics.htm
NISRA. (2011). Mid-Year Population Estimates. Retrieved 17 August 2011, 2011, from http://www.nisra.gov.uk/demography/default.asp17.htm
 This research is generously funded by the Nuffield Foundation and McDougall Trust. Further information about the project is available here: http://www.tobysjames.com/election_laws_1.html and http://www.swan.ac.uk/artsandhumanities/riah/videos/drtobyjames/
 LEOs sometimes expressed concern about indi vidual registration in general before the White Paper was published.