open Secondary menu

Electoral Insight - Election Legislation Enforcement

Electoral Insight – March 2003

Electronic Voting Methods Experiments and Lessons

Daniel Guérin
Senior Analyst, National and International Research and Policy Development,
Elections Canada

Asifa Akbar
Analyst, National and International Research and Policy Development,
Elections Canada

Digital information and communication technologies tend to have an extensive, transnational reach, may be accessed from almost any location at any time of the day, and have the capacity to be used for various kinds of transactions, commercial and otherwise. In this climate of technological development, many governments have begun exploring and undertaking so-called e-(electronic) democracy initiatives.

Developments in the electronic or digitized administration of elections and electoral processes, particularly the use of on-line technologies in voting and voter registration, should be seen within the broader context of debate about public participation in the democratic process. Footnote 1 It is suggested, for example, that measures to enhance the accessibility of the electoral process could help encourage more people, including youth, to exercise their right to vote.

Voting in a secure environment by methods that protect the privacy of voter information, ensure ballot secrecy and are accessible to all eligible electors is the cornerstone of democratic elections, as articulated in the International Covenant on Civil and Political Rights and other international legal instruments. Footnote 2

In the context of the use of on-line/Internet and other electronic technologies for electoral events, the Internet as a global or transnational information and communications medium raises a unique set of concerns about security, privacy, ballot secrecy, accessibility for different socio-economic groups and other matters. Many countries and sub-national governments are currently addressing these concerns and how barriers could be surmounted so that Internet/on-line voting may be implemented as an effective alternative voting method.

This article examines recent global trends in voting and voter registration through electronic means, with a focus on developments in the use of on-line/Internet voting and voter registration to elect government representatives. The examination is based on an environmental scan of recent developments (from September 2001 to September 2002) in e-voting and e-voter registration in various countries. Our purpose is to highlight and discuss the major themes and issues emerging from these developments.

The term e-voting, according to a recent report, may encompass a range of methods, from electronic counting of paper ballots, to voting by direct recording electronic machines (DREs), to wide-scale remote voting by electronic means. DREs may include touch screen systems or PC-based technologies that use screens and keypads to register votes. The machines may be in the form of static or mobile kiosks located at, or transported to, various public sites to facilitate voting in, for example, the workplace, hospitals and seniors' homes. Footnote 3 Remote voting by electronic means (RVEM), that is, e-voting from places other than supervised polling stations, may include voting by Touch-Tone telephone (either land lines or mobile phones), SMS (short message service) text messaging, interactive digital TV (iDTV) and voting over the Internet.

Elections Canada examines on-line voter registration

Before reporting on developments elsewhere, it should be noted that the feasibility of an on-line voter registration system is being examined by Elections Canada. A study provided in November 2002 by CGI Information Systems and Management Consultants Inc. found that implementing on-line voter registration is feasible for Elections Canada, assuming legal and user authentication issues can be resolved. The feasibility study reflects Elections Canada's commitment to exploring new mechanisms to facilitate the processes by which electors add, update or confirm their elector information between and during electoral events.

The study was based on consultations with internal and external stakeholders and an environmental scan of similar initiatives in Canada and around the world. It sought to identify the operational, legal, technical and privacy considerations associated with the development of an on-line voter registration system and to recommend a strategy for implementing such a system.

United Kingdom

At the national level, many countries are now considering the possibility of implementing full-scale electronic-enabled general elections. A number of countries are first trying e-voting at the local level, in order to identify problems and potential barriers that may have to be overcome before applying e-voting on a wider electoral scale.

Two significant initiatives to examine the feasibility of electronic voting (including electronic voter registration) recently took place in the United Kingdom. The first, launched in October 2001, was a seven-month study conducted by a range of central and local governments as well as private agencies to examine the possibilities of implementing electronic voting. This study looked at the potential for, as well as the implications of, implementing various forms of electronic voting and vote counting, including via the Internet. The research findings in the report The Implementation of Electronic Voting in the UK, published by the Local Government Association of the U.K., were meant to pave the way for Britain's first "e-enabled" general election by 2008. The report concludes that the implementation of e-voting may introduce greater flexibility as well as convenience into the electoral process, and would also help modernize the electoral system. Implementing e-voting is a complex endeavour, it emphasizes, but one that ought not to compromise democratic principles of freedom and fairness in electoral processes. Footnote 4

The second and related initiative involved the piloting of alternative methods of voting, including all-postal and multi-channel/electronic voting, across 30 local councils during the May 2002 local elections. An evaluation, including recommendations for future pilots, prepared by the Electoral Commission was part of this pilot program and is contained in the report Modernising Elections: A Strategic Evaluation of the 2002 Electoral Pilot Schemes. Footnote 5 The evaluation aims to draw various implementation lessons from the pilot projects and is meant to serve as a stepping stone in the testing of on-line and other electronic and alternative voting methods.

According to the Electoral Commission, the e-voting pilots generated a great deal of positive feedback from voters, candidates, agents and polling station staff about the convenience and ease of use of new voting methods. In terms of security, in e-voting as well as other (all-postal) pilot areas, much effort was made to prevent fraud and other breaches of security. Footnote 6 In areas where electronic vote counting was adopted, ballot papers were printed in special ink, and counting machines rejected papers not printed with this ink.

The Commission states that, compared with conventional methods of voting, there was no evidence to suggest that any of the e-voting procedures led to increased rates of impersonation or other electoral offences. But the Commission does warn that, if public concerns about fraud were to grow, this could lessen public confidence in the use of e-voting mechanisms. To provide reassurance to voters, it suggests that in the future a set of technical criteria be established against which pilots may be judged for security.

The U.K. Electoral Commission asserts that remote voting is more convenient than traditional polling stations for many voters, and that, over time, remote voting is likely to become the norm for most elections. Footnote 7 "In the medium term, remote voting may be through postal voting, but over the longer term – as Internet access and digital television ownership grow – technology-based schemes are likely to increase," it observes. Nevertheless, the Commission regards it as important to retain, for the foreseeable future, the option of voting in polling stations along with "remote" and other electronic voting methods.

While the May 2002 experience appeared to be satisfactory, the Electoral Commission indicates that further pilots are definitely necessary "to tease out a number of issues and further establish the security measures necessary to protect these systems from attack and ensure public confidence." Footnote 8 In light of the success of the 2002 pilot projects, the U.K. Government has announced its intention to continue on-line voting trials by inviting local councils to submit bids to run innovative voting pilots in this year's local elections. These will include voting by the Internet, by mobile and touch phone, interactive digital television or by post. Footnote 9

The question of electronic voting was also examined by the Independent Commission on Alternative Voting Methods established by the Electoral Reform Society. Its February 2002 report, Elections in the 21st Century: From Paper Ballot to E-Voting, was based on a study that examined new methods of voting, including on-line voting. It makes recommendations on the importance of ensuring security and secrecy of the ballot as well as continued public confidence in the electoral process. Footnote 10

Recent developments in other countries

The Netherlands, the United States, Australia, New Zealand, Brazil and other governments have also begun piloting the use of on-line mechanisms to conduct referendums and legislative elections. Such experiments are being undertaken at traditional polling stations and other public sites such as libraries and shopping centres, from home and the workplace, as well as from other more remote locations overseas.

At the international and multinational levels there also have been developments in this area, involving the consideration and implementation of on-line voting for electoral events. One major initiative is the CyberVote project sponsored by the European Union (EU). It seeks to develop a universally applicable on-line/Internet voting system to facilitate e-voting within the EU. The E-Poll project is another multinational on-line and e-voting initiative in the process of being piloted in at least three European countries: Italy, France and Poland.

Australia's On-line Council is another example of a national forum that has been actively discussing major issues related to the implementation of electronic, especially on-line, voting. It is addressing such considerations as equal access to Internet technology and the so-called "digital divide."

Several studies from the United States have examined major issues and concerns about on-line and other electronic means of voting. Three of these are: the National Science Foundation report, the California Task Force on Internet Voting, and the report of the General Accounting Office. These studies concur that Internet voting at polling sites may be feasible in the near term, but that many technological and voter security concerns, such as ballot secrecy and privacy, the prevention of intrusions and accidents and the provision of equal access to user-friendly technology, must first be resolved. Footnote 11

In the Australian context, two noteworthy reports are: Electronic Voting and Electronic Counting of Votes: A Status Report, Footnote 12 a joint endeavour of the Victoria Electoral Commission and the Australian Electoral Commission (March 2001); and Electronic Voting: Benefits and Risks (April 2002), Footnote 13 published by the Australian Institute of Criminology. The former examines the status of e-voting in the U.S. and then discusses the feasibility of, and possible context-specific factors to be considered in, implementing various forms of electronic voting in Australia.

The latter report aims to test the effectiveness of electronic voting in satisfying the requirements of free and fair elections and concludes that new e-voting technologies have the potential to both facilitate and hinder electoral fraud.

Possible lessons from recent initiatives

In the quest to test and/or implement on-line and other electronic methods of voting in public elections, many countries have recognized the importance of identifying context-specific barriers to various forms of e-voting. Nevertheless, certain shared concerns and considerations seem to have emerged across countries that are experimenting with electronic voting technologies. The experiences of various countries provide useful lessons and ways of overcoming hurdles to the implementation of various e-voting systems.

One major recommendation has been that implementation ought to follow an incremental or phased-in approach that allows careful examination of particular problems and issues associated with different forms of e-voting. In other words, there should not be large-scale implementation of e-voting, especially RVEM, until issues of security, secrecy, technological penetration and voter capacity have been adequately addressed.

Most recommendations from various countries with regard to the implementation of e-voting have focused (with minor variations) on issues such as security, secrecy and privacy, accessibility, and public awareness and public confidence, in a bid to maintain the integrity of free and fair democratic elections. But the issues of the cost-effectiveness per ballot, the need to ensure the timeliness, accuracy, authentication and verifiability of e-ballots, as well as the need to review and possibly revise electoral laws, have also emerged as important concerns.

The main concerns about security have generally been that eligible voters should be allowed to cast their ballots in an unhindered, safe manner; and that, once votes are cast, they be recorded and counted accurately. To prevent major security risks posed by computerized voting, such as third-party interference through computer hacking, studies in Australia have recommended the installation of "firewalls" and other internal controls in government computer systems. Maintaining an accurate electoral roll has also been recognized as essential to an efficient voting system and a means of discouraging electoral fraud.

Effective means of voter identification, and of ensuring security and verifiability of votes, are also major concerns. The United Kingdom, the United States, Australia and other countries have begun experimenting with various technological approaches to the proper identification of voters at the time of voting, through the use of such methods as smart cards, PIN numbers, biometric (signatures, fingerprints) authentication, and/or public key cryptography and digital signatures.

The impact of e-voting methods on ballot secrecy is also a major concern. Ballot secrecy may be compromised in a number of ways in the private sphere of the home. In the U.K. it has been noted that voting by Internet or digital TV, unlike by paper ballots, means that voting is more susceptible to the "gaze of others." Footnote 14 There may also be intentional or unintentional influence or sometimes outright coercion of voters by other family members, which can compromise voter autonomy.

Another concern is the cost-effectiveness of such schemes. Countries have tended to measure the cost-effectiveness in terms of cost per vote, that is, by comparing the amount of money and other resources spent on a particular initiative with the number of voters who turned out to vote using a particular e-voting method. In the May 2002 local elections in the U.K., turnout in the pilot areas was 38.7 percent, compared to 32.8 percent in all local authority areas. The Electoral Commission's report concluded: "In general terms, the pilots appeared to provide good value for money in terms of the cost per voter as compared with previous years." As for the pilot Voting Over the Internet (VOI), a project of the U.S. Department of Defence's Federal Voting Assistance Program held during the November 2000 presidential elections, the Pentagon received severe criticism in light of the cost of US$6.2 million and the fact that only 84 voters participated.

Another important issue for countries considering e-voting, including on-line voting, is the need to make appropriate legislative changes to permit not only further testing but also the wider application of such voting methods.

In general, researchers and public bodies that have examined these issues are recommending a gradual phased-in implementation approach. The implementation of wide-scale e-voting, including remote electronic voting in general elections, is increasingly being viewed as feasible in the medium term and may even become the norm in the longer term, but not prior to rigorous and continuous pilot testing and research. It is not yet foreseeable that electronic means will completely replace conventional methods of voting. E-voting, particularly Internet voting, is considered as a complementary alternative to traditional paper ballot voting in a multi-channel system of voting that is being fairly widely advocated in the U.K., U.S., Australia and elsewhere.

Ensuring security of systems, the ability to maintain the secrecy of ballots and voter information, organizational and technological capacity, availability as well as accessibility of technology to citizens, and voter capacity to use such systems are major considerations that must be kept in mind by those responsible for the implementation of on-line/Internet voting. In the meantime, useful lessons can be learned from pilot projects and experiments such as those highlighted in this article.


Endnotes

Footnote 1 OECD. 2001. "Citizens as Partners, Information, Consultation and Public Participation in Policy-Making," OECD Publications, Paris.

Footnote 2 See International Covenant on Civil and Political Rights (http://www.hri.ca/uninfo/ treaties/3.shtml). "The right to participate in public affairs, voting rights and the right of equal access to public service" (Art. 25): December 7, 1996. CCPR General Comment 25. (General Comments – from the Committee for Civil and Political Rights of the Office of the United Nations High Commissioner for Human Rights).

Footnote 3 "Implementation of Electronic Voting in the UK" (London: Local Government Association Publications, May 2002), p. 13.

Footnote 4 Ibid. p. 71.

Footnote 5 "Modernising Elections: A Strategic Evaluation of the 2002 Electoral Pilot Schemes," United Kingdom Electoral Commission, 2002.

Footnote 6 Ibid. p. 50.

Footnote 7 Ibid. p. 78.

Footnote 8 Ibid.

Footnote 9 Office of the Deputy Prime Minister Press Release, "May 2003 Elections to Continue Online Voting Trials" (www.odpm.gov.uk/ news/0209/0086.htm).

Footnote 10 "Elections in the 21st Century: From Paper Ballot to E-Voting," by the Independent Commission on Alternative Voting Methods (London: The Electoral Reform Society, February 2002).

Footnote 11 See National Science Foundation press release of report: "Panel Calls for Further Study of Security and Societal Issues" (www.Internetpolicy.org/media/PR01-18.html). See also "Study cautions against Net voting," USA Today, February 6, 2002. See also "Do Elections Need a Technological Upgrade? Government study considers whether the Net could help restore voter confidence," Ellie Phillips, Medill News Service (pcworld.com), November 26, 2001. See also "Congressional Panel Addresses Solution for Disenfranchised Military Voters; Technology Exists Today for Secure On-line Military Voting," Business Wire, October 30, 2001.

Footnote 12 Colin Barry, Paul Dacey, Tim Pickering, Debra Byrne, "Electronic Voting and Electronic Counting of Votes: A Status Report" (Victorian Electoral Commission and Australian Electoral Commission, March 2001).

Footnote 13 Russell G. Smith, "Electronic Voting: Benefits and Risks Trends and Issues in Crime and Criminal Justice" (Canberra, ACT: Australian Institute of Criminology, April 2002), (www.aic.gov.au).

Footnote 14 "Implementation of Electronic Voting in the UK," p. 61.


Note: 

The opinions expressed are those of the authors; they do not necessarily reflect those of the Chief Electoral Officer of Canada.