Written evidence from Professor Toby S. James[1] (TEB 35)
Public Administration and Constitutional Affairs Committee
The Elections Bill inquiry
1. I am the current Deputy Director of the Electoral Integrity Project. The Electoral Integrity Project is a world leading project which produces innovative and policy-relevant research comparing elections worldwide. My individual research expertise focusses on the administration and management of elections, as well as electoral governance. I am giving evidence in a personal capacity in response to the committee’s call for evidence on the Elections Bill.
2. This is a welcome and timely inquiry on one of the most significant set of reforms to elections since the UK became a democracy. In summary:
Electoral Integrity in the UK: An Overview
3. The Electoral Integrity Project collects and publishes data on the quality of national elections held around the world. The Perceptions of Electoral Integrity Index dataset is based on a survey of academic experts and provides a useful overview of the areas of the electoral cycle where there are strengths and weaknesses.[2] Figure 1 (at the end of the document) lists potential problems that countries may experience with the electoral process and the extent to which they are found in UK parliamentary elections. Data is based on the 2015 and 2017 elections, with data on the 2019 election being collected at the moment.
4. The data presented in Figure suggests that the most significant problems with the electoral process as a whole lies in the electoral system and drawing of boundaries, which tends to be unfair to smaller parties. Problems with eligible citizens not appearing on the electoral register are also highly ranked. Problems with newspapers not providing balanced coverage are also high. By contrast the PEI suggests that the electoral authorities perform very well in the UK. Responses suggest no serious problems with the counting process, announcement of results and security of ballot boxes, for example.
5. A second source of evidence that provides an overview of the presence of potential problems are surveys of poll workers – the officials or run polling stations on the day of the election. We assessed the problems that occurred in polling stations at the 2018 and 2019 local elections in England, based on responses from 5,659 poll workers. [3] This is the largest volume of electoral officials ever to take part in such a study in the UK and it therefore provides a very comprehensive picture of the frequency of problems experienced at polling stations.
6. Table 1 demonstrates that people asking to vote, who were not on the electoral register was by far the most common problem, reported by 47.3 percent of poll workers. This is likely to be because citizens were confused about the process and thought that they were on the register when they were not. By contrast, electoral fraud and suspected cases of personation are exceptionally rare, with less than one percent of poll workers expressing concern that there had been suspected cases of electoral fraud in their polling stations.
Potential problem | Percentage of respondents reporting at least one problem in their polling station |
People asking to vote, who were not on the electoral register | 47.3 |
Disabled voters having problems completing ballot papers | 14.3 |
Disabled voters having problems with access to the polling station | 8.6 |
Members of political parties being where they should not be | 8.4 |
7.3 | |
People asking to vote whose identity I was unsure of | 6.0 |
Suspected cases of electoral fraud | 0.7 |
7. Overall, it is therefore clear that the Elections Bill does not seek to redress the main problems with the electoral process that has been flagged by academic research.[4] While there is neither time nor scope to introduce amendments to address these areas, a wider review of the quality of elections should be encouraged. The Bill does provide an opportunity to redress some problems, and these are set out at the end of this evidence.
Recommendation #1: the government should undertake a wider review of the quality of elections.
The Need for Cross-Party Approaches to Changing Electoral Law
8. Laws regarding elections can often benefit some political parties and disadvantage others. For this reason, a cross-party approach to reforming electoral laws is important. There is otherwise a risk that the public will perceive that the incumbent government is seeking to pass laws to give it political advantage and this may undermine their confidence in the democratic process. There is also the risk that this perception matches reality: that laws are being amended to give the incumbent governing party a better chance of electoral success.[5]
9. A cross-party approach therefore has traditionally been adopted in reforming electoral laws in the UK. The Prime Minister of the day would ask the Speaker of the House of Commons to initiate a Speakers Commission in order to invite opinions from civil society and broker compromises between parties.[6]
10. Sadly, the government has not taken this approach and there is no cross-party consensus. The Elections Bill is a major piece of electoral law covering much of the electoral cycle. Yet, the prime minister has not initiated a Speaker’s Conference. The Conservative Party manifesto was committed to introducing voter identification and making it easier for expats to vote – measures which are included in the Elections Bill.[7] Most political parties have publicly criticised these reforms. The Labour Party, for example, campaigned in 2019 to abandon voter identification and instead proposed automatic voter registration. A Speaker’s Conference would be an important opportunity to help to build consensus and reduce partisan rhetoric with a more evidence-based form of decision making.
Recommendation #2: Speaker’s Conferences should be called when major reforms to electoral law are proposed to bring about greater consensus, less inflammatory parliamentary rhetoric and scrutiny of the legislation.
Provisions on voter ID
Personation is not widespread in the UK
11. The Bill will introduce a new requirement for voters across the whole of the UK to provide photographic identification at polling stations when voting at UK Parliamentary elections. The government has frequently advocated this on the basis that it wishes to reduce personation in polling stations.
12. Research has consistently shown that personation is not a widespread problem at polling stations. The poll worker studies described above found that only 0.7 per cent of poll workers were concerned that electoral fraud might have happened in their polling stations. Similar findings have been found at general elections.[8] Where concerns about ‘fraud’ were raised by poll workers, these were often the result of misunderstandings about the electoral process by voters. For example, some citizens were confused about the differences in eligibility between parliamentary and local registers and had in advertently registered on a register where they might not be eligible.
13. The Electoral Integrity Project has produced an index of electoral integrity worldwide based on expert perceptions 2012-8. This supports findings from the poll worker surveys, demonstrating that problems with the completeness and accuracy of the electoral register are much more common than those with electoral fraud.
Voter identification requirements will reduce turnout
14. Voter identification requirements will also lead to citizens not exercising their right to vote. Evidence for this comes in at least three forms:
i) The English pilots of voter identification. A study of the voter identification pilots in 2018 and 2019 demonstrated that many citizens were unable to vote because of the requirements. Table 2 shows that over half of poll workers experienced an issue with a voter who was unable to vote because they did not have the appropriate identification. However, many citizens also declined to provide identification because they did not want to. This represents a major problem with the voter identification requirements.
Potential problem | Percentage of poll workers reporting at least one problem in their polling station |
People being turned away because they did not have the appropriate identification | 52.4 |
People coming to the polling station but deciding not to vote as they did not want to comply with the ID verification requirements | 23.3 |
Table 2: Source: author, based on research by James and Clark.[9]
ii) Evidence from Northern Ireland. Photographic voter identification requirements have been in place in Northern Ireland since the early 2000s and was first in place at the Northern Ireland Assembly Elections in 2004. It was estimated that around 25,000 voters did not vote in that election because they did not have the required form of identification. Almost 3,500 people were initially refused a vote for not presenting the required identification requirements.[10] This was 2.3% of the electorate.[11]
iii) Evidence from overseas. Studies undertaken of the effects of voter identification requirement in other countries shows that it commonly leads to a decline in turnout.[12] This effect can be masked in the short term by a ‘mobilisation effect’ where political parties and pressure groups exert additional efforts to encourage citizens to vote because of fears about the voter identification requirements. In the long term, however, declines are often seen.
iv) The government’s research. The government’s own research suggests that 9% of the public do not have in-date and recognisable identification.[13] The availability of identification is lower amongst those with a severely limiting disability, the unemployed and those without educational qualifications. It is also substantially lower for the - trans and gender non-conforming (GNC) individuals, a group who have protected characteristics under the Equality Act 2010, but stand to be adversely affected by the reforms.
Devolved issues
15. It is important to note that there are no plans to introduce voter identification requirements in Scotland and Wales. The different identification requirements for different elections is likely to lead to considerable confusion amongst the public, which may also affect turnout and confidence in the process.
16. It has commonly been stated that Northern Ireland provides a good model for introducing voter identification in Britain. However, it should be noted that:
Recommendation #3: Voter identification requirements should not be introduced.
Options for amending voter identification requirements
17. The government has persistently stated that it wishes to introduce voter identification requirements. This is discouraged for the reasons stated above. However, assuming that the government persists then the Bill should include additional measures to ensure that voters are still able to participate. International best practices and academic research can be drawn from. Four proposals are suggested here:
‘Vouching’
18. Elections in Canada have commonly used a ‘vouching system.’ Voter identification is required, but if a citizen does not have their identification available then they can still vote if they declare their identity and address in writing and have someone who knows them (and who is assigned to their polling station) vouch for them. The person who vouches for the citizen is required to provide their identity and address and can only vouch for one other person.[14] This was previously repealed in Canada but reinstated. This would be a simple and effective way of preventing citizens who do not have voter identification on the day still being able to vote but retains security measures.
Recommendation: a Canadian system of vouching should be introduced to the Bill. Citizens who do not have identification should be able to vote if another registered elector can verify their identity in a polling station.
Provisional ballots
19. Citizens could be allowed to cast ‘provisional ballots’ if they do not have suitable identification at hand at polling stations. These ballots could then be put aside and not included in the provisional count. Citizens could then be provided with the opportunity to present identification at a later point for their vote to still be included. This process is used in many states within the US to ensure that citizens are still able to have their vote cast.[15]
20. Given that a large volume of citizens do not vote because their name is not on the electoral register, as noted above, the Bill could also be revised to enable citizens who are not registered to cast a provisional ballot. Electoral registration officers could then be given a short period of time to verify their registration status before including the vote into the final vote tally.
21. The downside of introducing provisional ballots is that there would be an additional administrative investment needed. It may also mean that final results would be slower, as has been the experience in the USA. However, it would ensure that the election is more inclusive and more citizens would have their vote included.
Recommendation #4: citizens should be enabled to cast provisional ballots if they are unable to provide identification or do not have their name on the electoral register.
Poll cards
22. One way to reduce the number of citizens who are unable to vote is to increase the range of forms of identification that could be presented. Given that all registered electors are provided with poll cards, electors could be allowed to present these as an acceptable form of identification.
Recommendation #5: electors should be allowed to present their poll card as one of many possible forms of identification.
Systematic monitoring
23. Information about the number of people unable to vote because of voter identification requirements is vitally important to enable the problem to be addressed in the future. The Bill could introduce a requirement for polling clerks and presiding officers to keep of log of the number of voters who were unable to vote and the reason why. For example, that they were not registered, needed to vote at another polling station, did not have ID or any other reason. There should be a statutory requirement for the Returning Officer to publish this information within a specified period of the election.
Recommendation #6: Returning Officers should be required to publish information about the number of citizens who were unable to vote at each polling station and the reason why.
Changes to Postal and Proxy Voting
24. The bill proposes to abolish permanent postal and proxy votes, requiring them to be re-applied for every three years.
25. Postal voting and proxy voting are important measures for ensuring inclusive elections. They particularly help citizens with disabilities which means that attending polling stations are difficult.[16] Many citizens who request them are likely to permanently need them and asking them to reapply frequently may cause them to not take part in the electoral process.
26. They have been vitally important during the covid pandemic and it has been recommended that all countries ensure that they have postal/proxy facilities place as a risk management solution should an election take place during an emergency situation.[17]
27. If citizens are required to reapply for the postal and proxy vote then this needs to be clearly communicated to the elector ahead of the expiry of their postal/proxy vote. There is a high risk that they would otherwise assume that they still had this in place and might miss a later postal vote deadline. A requirement to notify electors that their postal/proxy vote has expired should be set out in law.
28. Given that so many citizens will need them on a permanent basis it is recommend that postal and proxy votes are in place for a period of five years rather than three. Five years is the length of a full parliamentary cycle and would still mean that they would be reapplying sufficiently regularly, but would lessen the administrative burden on the voter and Electoral Registration Officers.
29. It should be noted that indefinite postal and proxy votes will be possible in Wales and Scotland for elections other than UK parliamentary elections. A citizen may therefore be registered for a postal vote for a Scottish Parliament election, but need to reapply for a UK election. This will lead to considerable confusion for the voter and a headache for the administrator. On balance, a UK wide solution should be found or the proposal should be withdrawn.
Recommendation #7: A requirement to notify electors that their postal/proxy vote has expired should be set out in law.
Recommendation #8: A UK-wide approach for postal/proxy votes should be encouraged.
Recommendation #9: Postal and proxy votes should remain valid for five years.
Accessibility
30. The Bill seeks to improve accessibility for voters, which is obviously very welcome. Research suggests that nearly 15 percent of poll workers encounter disabled voters who experience problems completing ballot papers (Table 1).
31. There is an information gap in the election process which prevents problems being reported, however. Procedures and laws can be prescribed in Parliament but not implemented in practice and voters have no real mechanism for redress. It is therefore proposed that the Bill is amended so that a centralised complaints mechanism is introduced so that voters can report problems on election day (or during the voter registration process) to a central authority. The most obvious body to undertake this role would be the Electoral Commission. A simple web form could be set up so that electors (but also candidates and party agents) could report problems. The Commission should then be required to publish summary statistics of the number of cases by ERO and RO and record the outcomes. They should also be required to pass these on to the appropriate ERO and RO for their responses.
Recommendation #10: The Electoral Commission should run a central complaints service for voters, candidates and party agents.
The Role of the Electoral Commission
32. Independent electoral authorities are essential components of democracy and this is widely established in international best practices and by academic research. Independence from the government of the day is important because it prevents an incumbent changing laws or practices to suit their political interests. It can also strengthen public trust in the political process. Just as the judiciary should be independent, electoral officials should be non-partisan. [18]
33. The Bill, in contrast, proposes to weaken the Commission’s independence. It proposes to give the government greater power by allowing it to designate a Strategy and Policy Statement. It gives Parliament (but in practice government, assuming that it has a majority) the power to examine the Electoral Commission’s compliance with this. It also seeks to amend the Electoral Commission’s powers to expressly prevent it from bringing prosecutions in England, Wales and Northern Ireland.
34. This is therefore a direct violation of international best practices and would constitute democratic backsliding because it is giving the government and future governments greater control over the conduct of elections - the process through which citizens are enabled to hold government to account.
35. It should be noted that democratic backsliding is an important theme in many other countries as governments have sought to exert control over the electoral process.[19]
36. It is therefore recommended that the proposed changes to the Electoral Commission are removed in the strongest possible terms. The Electoral Commission and conduct of elections is highly regarded and the Bill would only jeopardise this needlessly.
Recommendation #11: No changes should be made to the Electoral Commission’s role and functions.
The Speakers Committee
37. The Bill proposes amendments to the composition of the Speaker’s Committee. The committee plays an important role in linking Parliament with the Commission.
38. A situation has arisen recently that has meant that the Committee has had a government majority in control of the committee. This is problematic because it could lead to a single party-controlled committee making partisan appointments to the Electoral Commission. This would undermine the independence of the Commission and voters’ confidence in the process.
39. The solution would be for a two-third majority to be required on all decisions. Lay members could also be appointed to ensure that there was no party with an overall control.
Recommendation #12: a two-third majority to be required on all decisions made by the Speaker’s Committee.
Overseas voting
40. The Bill proposes abolishing the 15-year limitation on eligible British citizens living overseas to be registered to vote in relevant elections in the UK.
41. It should be noted that the existing practice for overseas electors casting their vote is already problematic. This relies on overseas electors being sent their vote via the international post, and the elector being able to return it in time for the count. An evaluation of the EU Referendum showed that this was not possible within the narrow electoral timetable. Electoral officials reported many instances of voters receiving their ballot too late to be returned.[20] The Bill will substantially increase the number of electors living overseas who are eligible. The case for a mix of telephone/internet voting should therefore be considered in the long term.
Recommendation #13: the case for telephone/internet voting should therefore be considered for overseas electors in the future.
Undue influence and electoral intimidation
42. New sanctions for intimidation during the electoral process are proposed. These are welcome. Cases of intimidation are thought to be rare, but as Table 1 illustrates, 7.3 percent of poll workers reported members of political parties intimidating the public at the 2018 and 2019 local elections. A sanction of disqualification for standing for office for five years is reasonable. The implementation of this should be carefully evaluated to ensure that there are no unforeseen consequences.
Information to be included in digital election material
43. The rapid transformation in the development of technology has meant that electoral laws are often in need of updating. The regulation of political advertising is one such area, where there are insufficient safeguards for i) misinformation and ii) some political parties to be able to outspend their opponents.[21] The introduction of digital imprints is well overdue and it is essential that it is included in the bill.
Further amendments
44. As noted above, the Bill does not cover all areas where problems have been identified in the electoral process. It would not be possible to address all of these within the Bill, but some have already been considered by this Committee, or by other committees looking at the electoral process, and could therefore be taken forward as amendments.
Funding elections
45. There has been an increasing strain on the funding of the electoral process in recent years, with many EROs and ROs reporting a lack of funding to run elections or compile the electoral register effectively.[22] This has led to some compromises in service. For example, voter outreach work has been reduced because of limited funds.[23]
46. The transparent and timely reporting of the expenditure of elections could help to ensure cost efficiency and enable best practices to be identified. There is currently no obligation, however, for EROs and ROs to publish their accounts. There have also been (not necessarily justified) concerns raised that ROs have received excessively high fees for running elections.[24] In addition, there has been concern that EROs and ROs may not be covered by Freedom of Information requests like many public bodies are. To ensure greater transparency it is therefore recommended that:
Recommendation #14: The Bill is amended to clarify that EROs and ROs should be subject to Freedom of Information Requests.
Recommendation #15: EROs and ROs should be required to publish annual accounts and expenditure in a standard reporting format specified by the Electoral Commission.
Consolidating electoral law
47. I have previously given evidence to the Committee in support of the consolidation of electoral law. Complex electoral law makes elections difficult to administer and adds to the risk that errors might be made.[25] The Committee has stressed the importance of legal consolidation.[26] The Elections Bill would be another layer of legislation which would add to the complexity of electoral law. The government should therefore pursue the long-overdue consolidation of electoral law.
Recommendation #16: The government should therefore pursue the long-overdue consolidation of electoral law.
Voter registration
48. The electoral register has seen a long-term decline in levels of completeness in the UK. My research has shown that this was accelerated by the introduction of individual electoral registration. This had a particularly negative effect on the completeness of the register of young people and students who would have previously had their parents or university register them on their behalf.[27] The latest estimates from the Electoral Commission were that there was between 8.3 and 9.4 million people in Great Britain who were eligible to be on the local government registers were not correctly registered on the December 2018 registers.[28] A similar number will be missing from the parliamentary register.
49. A recent report commissioned by the Joseph Rowntree Reform Trust set out measures that could introduce this long-term accuracy.[29] These include:
50. More recent research shows that automatic voter registration increases the completeness of electoral registers and does not compromise completeness.[30]
51. The JRRT report also recommended that the open/edited electoral register (which can be bought by anyone) and this register should be abolished. This register is not used for electoral purposes, but by commercial organisations.
Recommendation #17: The automatic registration of citizens when they receive their National Insurance Number ahead of their 16th birthday
Recommendation #18 Providing citizens opportunities to register to vote when they access other government service – such as the DVLA, Universal Credit or the Student Loan Company.
Recommendation #19: The Act abolishes the open/edited electoral register.
Summary of recommendations
Figure 1: The extent that potential defects in the electoral process are found at UK General Elections. Source: author, using data from the Electoral Integrity Project. Scale 1 to 5 where 1=low, 5=high. Based on the 2015 and 2017 general elections.
August 2021
[1] Professor Toby S. James, Professor of Politics and Public Policy, University of East Anglia, Deputy Director of the Electoral Integrity Project
[2] For data, please see: https://www.electoralintegrityproject.com/data-1
[3] Toby S. James & Alistair Clark (2020) Electoral integrity, voter fraud and voter ID in polling stations: lessons from English local elections, Policy Studies, 41:2-3, 190-209.
[4] For further examples, also see: Patrick Dunleavy, Alice Park and Ros Taylor (eds) The UK’s Changing Democracy: The 2018 Democratic Audit (London School of Economics Press).
[5] Toby S. James (2012) Elite Statecraft and Election Administration (Palgrave: Basingstoke).
[6] Toby S. James (2021). ‘Who decides how to run elections? The electoral governance theory approach’, Paper presented at the Political Studies Association Conference, April 2021
[7] The Conservative Party. 2019. Get Brexit Done. Unleash Britain’s Potential. London: The Conservative and Unionist Party, p.48.
[8] Toby S. James (2014) ‘Electoral Management in Britain‘ in Pippa Norris, Richard Frank and Ferran Matinez I Coma (eds) Advancing Electoral Integrity (New York: Oxford University Press).
[9] Toby S. James & Alistair Clark (2020) Electoral integrity, voter fraud and voter ID in polling stations: lessons from English local elections, Policy Studies, 41:2-3, 190-209.
[10] Electoral Commission (2004) The Electoral Fraud (Northern Ireland) Act 2002: An Assessment of its First Year in Operation. Electoral Commission: London, pp/61-102.
[11] Electorate of 1,097,526, based on Colin Rallings and Michael Thrasher (2003) Voting at the Northern Ireland Assembly Election 2003, p. 13.
[12] Toby S. James (2010) ‘Electoral Administration and Voter Turnout: Towards an International Public Policy Continuum‘, Representation, 46(4), p.369-89; Toby S. James (2012) Elite Statecraft and Election Administration (Palgrave: Basingstoke).
[13] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/984918/Photographic_ID_research-_headline_findings_report.pdf
[14] https://www.elections.ca/content2.aspx?section=id&document=index&lang=e. Also See: Toby S. James (2020) Comparative Electoral Management (London and New York: Routledge).
[15] https://www.ncsl.org/research/elections-and-campaigns/provisional-ballots.aspx
[16] Toby S. James and Holly Ann Garnett (eds) (2020) Building Inclusive Elections (Routledge: London and New York).
[17] Toby S. James and Sead Alihodzic (2020) ‘When is it democratic to postpone an election? Elections during natural disasters, COVID-19 and emergency situations’, Election Law Journal, 19(3), pp. 344-362. Also see: https://www.electoralintegrityproject.com/elections-and-covid19
[18] Toby S. James (2020) Comparative Electoral Management (London and New York: Routledge)
[19] VDEM (2021) Autocratization Turns Viral: Democracy Report 2021.
[20] Alistair Clark and Toby S. James (2016) ‘An Evaluation of Electoral Administration at the EU Referendum,’ Electoral Commission, September 2016.
[21] Holly Ann Garnett and Toby S. James (2020) ‘Cyber elections: the threats and opportunities of using technology for electoral integrity’, Election Law Journal, 19(2), p.111-126.
[22] See: Toby S. James and Tyrone Jervier (2017) The Cost of Elections: Funding Electoral Services in England and Wales, ClearView Research: London. Toby S. James and Alistair Clark (2020) ‘Delivering Electoral Integrity Under Pressure: Local Government, Electoral Administration and the 2016 EU Referendum in the UK‘, Local Government Studies, 47(2), 186-207.
[23] Toby S. James and Tyrone Jervier (2017) ‘The cost of elections: The effects of public sector austerity on electoral integrity and voter engagement,’ Public Money and Management, volume 37(7), pp. 461-468
[24] Scottish Parliament Local Government and Communities Committee (2017) Payments to Returning Officers
[25] Toby S. James (2014) ‘Electoral Management in Britain‘ in Pippa Norris, Richard Frank and Ferran Matinez I Coma (eds) Advancing Electoral Integrity (New York: Oxford University Press).
[26] House of Commons Public Administration and Constitutional Affairs Committee Electoral law: The Urgent Need for Review: Government Response to the Committee’s First Report of Session 2019.
[27] Toby S. James (2020), Comparative Electoral Management: Performance, Networks and Instruments (Routledge: London and New York)
[28] https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/our-views-and-research/our-research/accuracy-and-completeness-electoral-registers/2019-report-2018-electoral-registers-great-britain/completeness-great-britain
[29] https://www.jrrt.org.uk/wp-content/uploads/2020/04/Is_it_time_for_AVR_in_the_UK.pdf
[30] Toby S. James and Holly Ann Garnett (2021) ‘The Determinants of Electoral Register Quality’, Political Studies Association Annual Conference, April 2021.