Publication - Research and analysis

Electoral reform: consultation analysis

Published: 12 Oct 2018

Analysis of Electoral Reform consultation.

Contents
Electoral reform: consultation analysis
6. Electronic voting (Q12 and Q13)

6. Electronic voting (Q12 and Q13)

6.1 Chapter 2 (Section 5) of the consultation paper discussed possible options for introducing electronic voting into Scottish elections including the use of electronic voting machines at polling stations, and internet and mobile phone voting. This section of the consultation paper also discussed the possibility of voting on more than one day, and being able to vote in any polling place in Scotland (rather than at a single, assigned polling place).[8]

6.2 The consultation paper set out the potential benefits of electronic voting (in relation to increasing voter participation, reducing the number of rejected ballot papers, reducing the costs of elections, etc.); described the current use of electronic counting within local government elections; highlighted international experience of using electronic voting; and explained that the Scottish Government plans to trial innovative electronic voting methods in the future. It went on to ask for respondents’ views on a range of related issues.

6.3 Although the consultation paper asked separately about people’s views in relation to (i) using electronic voting machines rather than traditional ballot papers (Questions 12a, 12d) and (ii) using internet or mobile phone voting (Questions 13a, 13b, 13d), the issues that respondents raised in their comments at these questions overlapped to a great degree. Moreover, it was common for respondents to simply refer to ‘e-voting’ or ‘electronic voting’ as a catch-all phrase to cover issues in relation to both these topics. Therefore, these questions have been analysed together. Points which are specific to either electronic voting machines or to internet or mobile phone voting are highlighted.

6.4 The issues raised in response to the questions on voting on more than one day (Questions 12b, 13c) and on voting at any polling place in Scotland (Question 12c) provided some distinctive themes, and these are discussed later in this chapter (paragraphs 6.31–6.43).

Electronic voting (Q12a, 13a, 13b, 12d, 13d)

6.5 The consultation questions on electronic voting were as follows.

Question 12a: Would you be happy to use an electronic voting machine in a polling place instead of a traditional ballot paper? [Yes / No]

Question 13a: If internet or mobile phone voting was available, would you choose to use that rather than vote at a polling place or by post? [Yes / No]

Question 13b: If internet or mobile phone voting was available, would you be more likely to vote? [Yes / No]

6.6 In addition, Question 12d and Question 13d asked for ‘Any additional comments’.

6.7 Overall, as can be seen from the tables below:

  • 62% of those who answered the tick-box part of Question 12a said they would be happy to use an electronic voting machine in a polling place instead of a traditional ballot paper (Table 6.1)
  • 49% of those who answered the tick-box part of Question 13a said that if internet or mobile phone voting was available, they would choose to use that rather than vote at a polling place or by post (Table 6.2)
  • 35% of those who answered the tick-box part of Question 13b said that if internet or mobile phone voting was available, they would be more likely to vote (Table 6.3).

Table 6.1: Q12a – Would you be happy to use an electronic voting machine in a polling place instead of a traditional ballot paper?

Respondent type Yes No Total
n % n % n %
Organisations 11 61% 7 39% 18 100%
Individuals 454 62% 275 38% 729 100%
Total 465 62% 282 38% 747 100%

Table 6.2: Q13a – If internet or mobile phone voting was available, would you choose to use that rather than vote at a polling place or by post?

Respondent type Yes No Total
n % n % n %
Organisations 8 57% 6 43% 14 100%
Individuals 356 49% 371 51% 727 100%
Total 364 49% 377 51% 741 100%

Table 6.3: Q13b – If internet or mobile phone voting was available, would you be more likely to vote?

Respondent type Yes No Total
n % n % n %
Organisations 8 53% 7 47% 15 100%
Individuals 253 35% 473 65% 726 100%
Total 261 35% 480 65% 741 100%

6.8 A total of 396 respondents – 51 organisations and 345 individuals – provided comments at any of the questions on electronic voting (Question 12a, 13a, 13b, 12d, 13d).

6.9 It is noteworthy that in relation to Question 12a (on the use of electronic voting machines), almost all the comments (92 out of a total of 99 offered) came from individuals and organisations who did not provide a ‘yes’ or ‘no’ answer to the closed question. On the whole, these respondents can be characterised as having ‘mixed views’. This indicates that 62% (see Table 6.1 above) is likely to be an overestimate of the proportion of respondents who would be happy to use an electronic voting machine.[9]

Overview of comments on electronic voting

6.10 There was widespread agreement that public confidence in the integrity of elections was vital to protecting and preserving democracy and the democratic process. There was also widespread agreement that public confidence in elections required the population to be assured that the electoral process was verifiable, secure, and anonymous. However, respondents disagreed strongly about whether electronic voting could assist and / or guarantee these conditions – either now or in the future. The arguments about these particular aspects were at the core of respondents’ views on electronic voting.

6.11 Respondents also discussed whether increased voter participation was likely to result from the introduction of electronic voting. There was a range of views on this point, and disagreements about the evidence in relation to this issue.

6.12 Some respondents commented on the cost implications of electronic voting. The predominant view was that there would be little financial benefit in the short to medium term. This was a consequence of (i) the substantial investment required to develop and maintain any new technology and (ii) the additional support – in terms of increased numbers of properly trained election officials – that the introduction of electronic voting would require.

6.13 The main issues discussed are set out in greater detail below.

Public confidence

6.14 The importance of public confidence in the electoral process was emphasised by both organisational and individual respondents. Respondents said public confidence was ‘paramount’, that it was vital that public confidence in the electoral was ‘not undermined’ and that public confidence in the electoral process was a very basic requirement of democracy. It was common for these comments to be made in the context of current concerns about hacking and foreign interference in elections. Any system that was adopted would have to be ‘tamper proof’.

6.15 More generally, respondents set out their views that, for the electoral process to command public confidence, it would have to be verifiable (i.e. able to be independently audited and validated), secure (free from outside interference) and anonymous (in order to protect against coercion). There were also some suggestions that the standards required for public confidence in voting were higher than those for other activities such as banking or shopping.

6.16 Respondents disagreed about whether electronic voting could assist with and / or guarantee these objectives – either now or in the future. There were four main groups of respondents:

  • Those who were strongly in favour of electronic voting. These respondents believed that (there was evidence that) electronic voting had already, and could in the future, deliver a wide range of benefits. It was common for these respondents to refer to published evidence about the benefits of electronic voting.
  • Those who were against the introduction of electronic voting. Some of these respondents had principled objections and did not wish any form of electronic voting to be trialled. Others objected on more practical grounds, or thought there was no prospect in the foreseeable future of the security concerns being addressed. It was common for these respondents to refer to evidence from other countries which they argued demonstrated that electronic voting was not viable.
  • Those who thought that electronic voting should be explored / developed if and only if there was good evidence available that this could be done in a verifiable, secure and anonymous way.
  • Those who had mixed views about the potential benefits or were unsure about what effects the introduction of electronic voting might have.

6.17 Below, the main arguments made for and against the introduction of electronic voting are summarised. As can be seen there was a wide range of different arguments used to support both positions. Following these sections, other themes raised in respondents’ comments are described. These cover (i) pilots / trials of electronic voting (ii) combining electronic voting with paper-based methods and (iii) problems with the current voting system.

Arguments in favour of electronic voting

6.18 There was a range of arguments in favour of electronic voting:

  • It is important to modernise the current system of voting, and to bring it into line with other activities and behaviours which have increasingly moved online. The current system for voting is very out of date and anomalous in the context of modern life. Moreover, it was suggested that the public generally support a move in this direction. This requirement to modernise is particularly important in relation to young people who have grown up in the digital age.
  • Electronic voting improves auditability, integrity, transparency and security through its use of strong authentication procedures and highly developed security arrangements. It also increases the speed with which results can be reported.
  • Electronic voting is particularly useful in situations where the voting system is complex. For example, if voters are faced with multiple choices (e.g. under the single transferable vote system), or if there are list systems in operation, or if a randomisation process for the ordering of names is to be used (see Chapter 5). In these circumstances, electronic voting machines can help to guide voters through the task, and thereby reduce the number of inadvertently spoilt ballots.
  • Electronic voting is particularly useful for key groups for whom the current system is not satisfactory. In particular, electronic voting machines can increase accessibility for those with a range of disabilities (including visual impairments and learning disabilities), or those requiring translation services.
  • Electronic voting can reduce the requirement for postal voting and proxy voting. Given that these forms of voting are those most vulnerable to fraud, then electronic voting can bring greater integrity to the electoral process.
  • Remote or mobile voting is far more convenient for people than attending a polling station. It can be particularly helpful in situations where there is bad weather, where transport links are inadequate, or for people who find it difficult to attend in person.
  • Electronic voting may reduce costs in the longer term by removing the requirement to print and distribute large numbers of different types of ballot papers, and by reducing the numbers of polling staff required to support the electoral process.
  • Electronic voting has already been used to good effect in other countries / places. There was a particular focus on the experience of Estonia, but respondents also referred to successful pilot work in Sheffield, UK and elsewhere. Moreover, these respondents argued there is already a well-developed, secure technology solution available in the form of Blockchain.

Arguments against the introduction of electronic voting

6.19 Respondents opposed to the introduction of electronic voting highlighted a number of issues, ranging from the principled to the pragmatic as follows:

  • The technology associated with electronic voting is a ‘black box’ which is not understood by the general public. The lack of public understanding of the technology will undermine public confidence in the electoral process. This lack of understanding of the technology was contrasted with the very ‘low tech’ current system which people understand, and which it is possible to (audit and) verify.
  • No system which is currently available is safe from hacking / outside interference. There has been experience in the UK of interference in IT systems (malware) with regard to the NHS and other public bodies. No matter how sophisticated the electronic voting system is, it cannot ever be guaranteed to be fully secure. This specific point about security was made repeatedly by individual respondents who identified themselves as ‘computer scientists’ or ‘software engineers’ and who emphasised the complexity of the technology and the likelihood of errors.
  • The scale of any fraud associated with electronic voting is on a completely different scale from that associated with the current (paper-based) technology. In relation to electronic voting, a few people could do ‘untold damage’. This was contrasted with the current system in which attempts at fraud are thought to be labour intensive and to affect small numbers.
  • If electronic voting machines were to replace ballot papers at polling stations, then no additional issues – over and above current ones – in relation to voter identification arose. However, with remote or mobile voting, respondents were not convinced that there was the ability to guarantee that an individual voted once and only once.
  • There is not sufficient connectivity to allow electronic voting since this would require reliable internet access to be available in all polling stations in Scotland. Connectivity, especially in rural areas is limited. Moreover, slow or intermittent connections would be highly disruptive to the electoral process and would require major contingency arrangements to be put in place.
  • Remote or mobile voting which requires individuals to access the internet on an individual basis does not take account of the ‘digital divide’ and may disenfranchise those groups who are not digitally sophisticated (including older voters and those living in deprived circumstances). Moreover, the use of electronic voting machines may act as a barrier to the smooth running of elections by requiring a level of familiarity with electronic interfaces which is not universally held (including by election officials).
  • Developing electronic voting methods, and focusing on the technology aspects of voting, ‘trivialises’ the act of voting. Respondents argued that voting is a communal and societal event with deep significance and importance. Reducing this to a remote ‘click’ puts it on a par with other consumer activities and devalues its significance.
  • The public, and voters, are not asking for electronic voting to be introduced.
  • There are fundamental problems with the procurement arrangements for any electronic voting system. The ‘source code’ (i.e. the computer code which underpins the electronic voting system) may not be open or available; the ethics of the providers cannot be guaranteed; there may be issues of commercial confidence, non-disclosure agreements and getting ‘locked into proprietary systems’ which compromise the amount of control that the commissioner has.
  • The evidence demonstrates that electronic voting does not increase voter turnout / voter participation. This was particularly mentioned in relation to the experience in Estonia, where respondents argued that the evidence showed that the people who used electronic voting were people who were going to vote in any case. More broadly, respondents argued that people will vote if the issue under consideration is important to them (they gave the example of the high turnout for the Scottish independence referendum) and they thought that efforts to increase voter participation should focus on the salience of the election, and on other options such as voting on multiple days or at a range of polling places, rather than on the introduction of electronic voting.
  • Other places / countries that have tried electronic voting in various forms have stopped investing in it or have withdrawn from it over time due to the various problems, difficulties and lack of benefits which have been demonstrated (examples offered included Ireland’s abandoning of electronic voting machines in 2009, concerns about turnout in Norway and security in the Netherlands, France suspending its pilots and the Australian Electoral Commission questioning the security of electronic voting). The particular experience in the US of electronic voting machines was described as ‘alarming’. Respondents said that there was evidence that electronic voting machines in the US can easily be hacked and that the use of these machines has fuelled conspiracy theories at ‘both ends of the political spectrum’.
  • Electronic voting does not reduce costs in the short or medium term and may not even reduce costs in the long term. The set-up / development costs of electronic voting are likely to be very high. There will also be high maintenance costs as well as expense in training polling staff. Additional support in polling stations (in the form of election staff) will be required for the large numbers of people who are unfamiliar with this technology. The value for money of this type of investment was questioned given the relatively small number of elections which take place. In addition, it was argued that investment in electronic voting was not a priority at a time of austerity.

6.20 Finally, it was argued that although the status quo – i.e. paper based voting – was not perfect, it had many advantages. It was simple to understand, familiar, and could be implemented in a cost-effective way. Any investment should be aimed at improving the current system. Others couched this argument as ‘it isn’t broken, so why fix it?’

Pilots / trials of electronic voting

6.21 Respondents who were opposed to electronic voting in any form said that there should be no pilots or trials conducted.

6.22 However, respondents in all other groups – those in favour, those with mixed views, and those with reservations about electronic voting – emphasised the importance of undertaking proper development work and trials on a small-scale basis before adopting any solution more widely. These trials should ‘test one thing at a time’, involve a wide range of participants in any development work (in particular, electronic voting solutions should be co-designed / co-produced with those with disabilities), and be ‘open to independent scrutiny’.

Combining electronic voting with paper-based voting

6.23 There was substantial comment on the need to run electronic voting systems in tandem with traditional paper-based voting methods. Most often, this kind of comment came from those who did not favour the introduction of electronic voting; however, it was also mentioned as a potentially important and valuable feature by those who were in favour of electronic voting.

6.24 The main argument for running systems in tandem was that not everyone would favour using electronic methods, or indeed would be comfortable using electronic voting; thus, to ensure that no-one was disenfranchised, a choice would have to be available. This argument was made by disability organisations who wished electronic voting to be developed as an alternative to – but not as a replacement for – current systems.

6.25 A subsidiary argument, expressed less often, was that electronic voting should be used only as an ‘assistive technology’, i.e. in a context where an individual was not able to participate using the traditional ballot paper method, but would be able to participate (with appropriate support) using an electronic voting approach.

6.26 Linked to this, there was also discussion about the importance of there being a ‘paper trail’ for every vote cast if electronic voting was adopted. This was sometimes referred to as a ‘VVPAT’ (Voter Verified Paper Audit Trail). Essentially this would mean that the voter would (i) register their vote using an electronic machine (ii) receive a print out with confirmation of their choice(s) and then (iii) place this print out in a ballot box. This would recreate the current paper system and allow for full and transparent auditing and verification.

Problems with the current voting system

6.27 There were three main issues discussed in relation to the current voting system. Respondents disagreed about whether, and the extent to which, electronic voting might provide a solution to these issues.

6.28 First, the question of spoilt ballot papers was raised. There was concern that these did not necessarily always indicate (as suggested in the consultation paper) that they had been completed ‘in error’. Indeed, respondents emphasised the importance of there always being an opportunity for voters to register that ‘none of the above’ candidates were suitable; this was not the same as a ‘spoilt’ ballot paper.

6.29 Second, respondents discussed the issue of voter identification. In general, those who raised this thought that the current system for voter identification was inadequate. They wanted more thorough and rigorous identification procedures to be implemented at polling stations.

6.30 Third, respondents were not confident about the current system for postal and / or proxy voting. Whilst those in favour of electronic voting suggested that the current concerns with postal / proxy votes regarding electoral fraud would be reduced if electronic voting came into effect, those against thought any current issues would simply be magnified.

Voting on more than one day (Q12b and Q13c)

6.31 Question 12b, which followed the question on using electronic voting machines, asked for respondents’ views on the possibility of voting on more than one day. An identical question was also asked at Question 13c, following the questions on internet and mobile phone voting.

Question 12b / Question 13c: Would you like voting to be possible on more than one day?

6.32 The tables below (Table 6.4 and Table 6.5) show that in both cases, just over one-third of individual respondents (37% at Question 12b and 38% at Question 13c) and around two-thirds of organisations (63% and 67% respectively) supported the idea of being able to vote on more than one day.

Table 6.4: Q12b – Would you like voting to be possible on more than one day?

Respondent type Yes No Total
n % n % n %
Organisations 16 67% 8 33% 24 100%
Individuals 269 37% 454 63% 723 100%
Total 285 38% 462 62% 747 100%

Table 6.5: Q13c – Would you like voting to be possible on more than one day?

Respondent type Yes No Total
n % n % n %
Organisations 12 63% 7 37% 19 100%
Individuals 269 38% 446 62% 715 100%
Total 281 38% 453 62% 734 100%

6.33 A total of 146 respondents – 31 organisations and 115 individuals – provided comments on this topic (either at Question 12b or 13c, or at Question 12d or 13d). The views offered are discussed below.

6.34 Respondents commenting favourably on this proposal said that allowing for voting on more than one day could (i) mitigate the effects of bad weather and / or transport delays (ii) reduce or remove the requirement to close schools if weekends were chosen (note that this objective could also be achieved with voting on a single day if a weekend day was used) (iii) be popular with some particular groups (e.g. young people, those dependent on a carer) (iv) increase accessibility and (v) reduce the requirement for postal voting.

6.35 Respondents who were generally against this proposal said it (i) could be ‘confusing’ (ii) would increase administration costs due to the need for more polling staff / election officials over a longer period (iii) ‘devalues’ the act of voting (iv) might depress turnout, as people might ‘put off’ voting and the momentum generated by a single day event would be lost (v) would require careful consideration as to how this proposal interacts with the existing registration timetable (e.g. last date of registration, postal and proxy applications, etc.) (vi) might have potential to enhance accessibility but the evidence from pilots was that any effects were very small and (vii) was not required given the existence of postal voting. In addition, respondents suggested that if this were to happen, there would need to be a system which allowed individuals to review and alter their votes if significant events took place during the period the polls were open.

6.36 There were different suggestions about how many days / how long a period would be appropriate. At least two days, ‘a weekend’, ‘several days’, ‘not too long’ and ‘up to one week’ were all suggested.

6.37 A small number of respondents asked why the question on voting on more than one day had been repeated.

Voting at multiple polling stations (Q12c)

6.38 Question 12c, which followed the question on extending the voting period, asked for respondents’ views on the possibility of voting at any polling place in Scotland.

Question 12c: Would you like to be able to vote at any polling place in Scotland?

6.39 Table 6.6 shows that there were mixed views among individuals on the desirability of being able to vote at any polling place (53% said ‘yes’ and 47% said ‘no’). Organisational respondents, by contrast, were generally in favour of the proposal (85% said ‘yes’).

Table 6.6: Q12c – Would you like to be able to vote at any polling place in Scotland?

Respondent type Yes No Total
n % n % n %
Organisations 17 85% 3 15% 20 100%
Individuals 386 53% 340 47% 726 100%
Total 403 54% 343 46% 746 100%

6.40 A total of 116 respondents – 26 organisations and 90 individuals – provided comments on this topic (at Question 12c, 12d and / or 13d). Respondents who supported electronic voting in general tended to comment favourably on this proposal, while those who did not support electronic voting did not.

6.41 There was comment from respondents on both sides of the argument that this facility could only be available, and made to work if there were a single, centralised version of the electoral register, kept up-to-date in real time, fully networked, and available in every polling station. This would require major investment. Respondents who did not support being able to vote at any polling station thought the expense of such a system could not be justified.

6.42 Those who were in favour said such a system made practical sense, and would reduce or remove the need for postal voting, reduce fraud, and help to make voting more accessible. These respondents also argued that this is already done successfully in other countries.

6.43 By contrast, those who were against such a development argued that such a system was unnecessary (given the existence of postal voting), that it would increase the risk of electoral fraud, that the transporting of votes to the correct counting venue would be problematic, that it would require wholesale changes to the design of the ballot paper, and that it would remove the local link to the activity of voting.


Contact

ElectionsTeam@gov.scot