Electoral Insight - Technology in the electoral process
The Feasibility of Electronic Voting in Canada
A Summary Of A 1998 KPMG/Sussex Circle Report
Parliament recently passed Bill C-2 that would allow the Chief Electoral Officer to study alternative means of voting and to devise and test an electronic voting process for future use. The actual use of such a new process in an official election would require the prior approval of the committee of the House of Commons that normally considers electoral matters.
In 1998, Elections Canada commissioned KPMG/Sussex Circle to examine the implications of information technology for the voting process in Canada. The following is condensed from their report.
Technology in the voting process is a subject that legislators and citizens are beginning to explore seriously both in Canada and abroad. The electoral process at the federal level in Canada is one of the most efficient and respected in the world, and Canada is among the most technologically advanced countries. It is appropriate, therefore, that Canadian parliamentarians be in a good position to consider the issues raised by the new voting technologies and to assess their potential for improving the accessibility and efficiency of the voting process.
The election environment
KPMG/Sussex Circle's first task was to identify the forces and factors that have changed, or may imminently change, the environment in which elections are conducted in Canada. These forces include changes in the attitudes, perceptions, expectations and voting habits of Canadians, and the effects of information and communications technology on our electoral environment.
It was noted that Canadians are increasingly using new information technologies in many dimensions of their lives, from banking to shopping to gathering information and expressing their views. Moreover, one-third to one-half of Canadians surveyed at the time of the last federal election in June 1997 indicated they would be prepared to use one or more new voting technologies. This finding was borne out in focus group discussions conducted in conjunction with the study. KPMG/Sussex Circle also found that Canadians see the new technologies as potentially increasing their choices in the timing and method of voting. That is, they regard electronic voting as a means of increasing the ease and accessibility of voting, rather than as a wholesale substitute for the traditional method of balloting.
KPMG/Sussex Circle also examined the opportunities offered by new voting technologies to enhance the accessibility of the voting process to Canadians. In the course of assessing those technologies, it reviewed Canadian and international experiences and found that the rhetoric of innovation in this area has so far greatly exceeded the results. Despite the talk, few jurisdictions have actually gone very far in implementing new voting technologies, though a number have launched pilot projects or at least made it legally possible to begin such experiments. The key point here is that none of the new technologies has yet been adequately tested in a way that would satisfy the requirements of electoral democracy in Canada.
In its review, KPMG/Sussex Circle examined a range of current voting technologies, including telephone, Internet, cable, kiosk and ATM (automatic teller machine) devices, portable data capture devices (both digital and cellular), smart cards and other personal identifier devices. It concluded that three technologies offer the greatest potential utility to Canadians because of their wide accessibility and public acceptance. These are the electronic kiosk, the telephone and the Internet. It modelled those three technologies against a detailed model of the current manual voting process.
Voting by telephone
Many technologically assisted voting options require the elector to have some unique identifier recognizable by the voting system. A personal identification number (PIN) similar to that issued by financial institutions for the use of debit or credit cards has been used in some past political party leadership contests. PINs can be assigned to registered electors on a random basis and sent in secure mail envelopes. There are well-established PIN print and mail routines generally accepted and used by all major financial institutions and others.
Use of the telephone to increase access to the voting process is an attractive option for a number of reasons, including the near universal presence of telephones in Canadian households, public familiarity with the device, and the fact that an elector would not need to go to a polling station to vote. For these reasons, telephone voting is the most viable of the three voting options assessed.
The main challenges to telephone voting include system limitations, providing access for electors with disabilities, electors whose language is not English or French, electors with rotary dial telephones who could not take advantage of this option, and the issuance of PINs to electors.
Canadian telephone companies are updating analog telephone switches across the country to digital switches, which have much greater capacity to handle the large volume of election day calls.
A number of ballot options are available. For example, electors could listen to a list of candidates and their respective selection codes, or electors could be prompted to enter the selection code of their chosen candidate from a paper ballot or other information provided to them before election day.
Voting by kiosk
The experts interviewed by KPMG/Sussex Circle regarded this option as technologically viable, given the maturity of the technology and the availability of public networks (such as the Interac network of banking machines), which could be used for voting purposes. While the use of publicly available networks is attractive from a cost-effectiveness perspective, it opens up a host of issues associated with security and secrecy. As a result, KPMG/Sussex Circle focused its efforts on assessing terminals that could be placed in polling stations, and portable terminals that could be used in acute care and mobile settings. Its study concluded that the cost of deploying this technology will result in only selective and limited use in Canada.
As with the telephone, there are a number of ballot options available for voting by kiosk or Internet. For example, electors could be prompted to enter the selection code of their chosen candidate from a paper ballot provided to them before election day, view a list of candidates and their respective selection codes, or view a combination of written text and pictures for each candidate.
Voting by Internet
According to KPMG/Sussex Circle, Internet-based voting is the least viable of the three technologies reviewed, because of shortcomings in both accessibility and security. Despite the dramatic increase in the number of Canadians who are "on the Net", this voting option would not be available to all electors, unless computers were provided in polling stations or public buildings, such as libraries. After prompting the elector to input his or her chosen language, the system would prompt the elector to input his or her PIN.
Security concerns associated with Internet-based voting relate to the link between the elector's computer and the Internet service provider. Once the two computers are linked, there is an increased possibility that computer "hackers" could reach and manipulate election results. The experts questioned were confident that security challenges would be surmountable, but not immediately.
The most important general finding from this study is that the new technologies – and especially the selected three of electronic kiosk, telephone and Internet – offer the prospect of significantly improving both the accessibility and the efficiency of the electoral process in Canada. As well, the study concluded that all three new voting technologies are sufficiently evolved to support testing in a fully functional pilot. However, none of the technologies examined or options available in the near future presents a universal solution.
It was noted that a number of potential pitfalls and obstacles, including issues of security, cost, privacy and public acceptance obstruct the route toward adoption of any of these new technologies. Nevertheless, there is also good reason to believe that the technological challenges posed by electronic balloting can be overcome, and there is no reason in principle why Canada's stringent criteria for effective electoral administration cannot be met by at least some of the new voting technologies.
The integrity of the voting process
Central to the whole issue of technology and the voting process is whether Canadians can – and would – have confidence that use of the new technologies would preserve the integrity of the voting process in Canada, in all its dimensions. Electronic voting options must be considered against many criteria, including:
- Democracy: one elector can cast one vote
- Accuracy: the final vote count reflects the intent of electors
- Security: measures are in place to protect the integrity of the process
- Secrecy: no vote can be traced to the elector
- Verifiability/auditability: voting results can be verified after the initial count
- Privacy/confidentiality: elector information is used for election purposes only, and within the purpose for which it was collected
- Transparency: the process is open to outside scrutiny
- Accessibility: the reasonable, specific needs of electors are taken into account so that none are disenfranchised
- Neutrality: electoral procedures or materials do not favour one candidate or party over another
- Simplicity: voting procedures do not make voting unduly complicated
It would be difficult to imagine a jurisdiction adopting a new technology for voting if it were not satisfied that the proposed innovation met every one of these criteria at least as well as the present system.
It is also important to bear in mind that no voting process – whether the present one or a new, electronic form of balloting – can be perfectly secure. After all, the current electoral process in Canada is a complex system of law, procedures, practices and dedicated administration that involves not just the Chief Electoral Officer and his staff, but also the many thousands of Canadians who are involved in the conduct of every election.
The integrity of the present system is something to which Canadians attach a high value. But our electoral system also reflects what Canadians are prepared to accept as a reasonable standard of security and integrity. Canadians trust our present systems and procedures – what could be called our present "technologies" – because they are used to them, because they see that they work and because people accept that they will produce an honest result. Canadians also have taken for granted that our present system, as updated with such measures as voting by special ballot, is reasonably accessible to electors. Whether this degree of accessibility will continue to be satisfactory to Canadians in the future, however, is an open question.
Similarly, the attitude of Canadians to the new technologies is changing, as those technologies become ever more present in our lives, whether in the form of banking machines, or scanning devices at the checkout counter, or Internet commerce. People see that these electronic systems work, and they develop a reasonable degree of trust in them, despite the fact that any of these systems is open, both in principle and in fact, to some form of compromise.
Public acceptance of new voting technologies, therefore, will depend ultimately on the kinds of technologies the members of the public use in their daily lives. It will also depend on people having seen the new voting methods tried and tested. It will depend on predictable reductions in cost, and the acceptability of various means of assuring security and integrity in elector identification, a key issue for any proposal that involves "voting at a distance." It should also reflect the recognition that different technologies are best applied to different subsets of the population, whether the group in question comprises rural Canadians, or persons with disabilities, or those, such as young people, for whom flexibility and accessibility is a high priority.
Possible next steps
Elections Canada already has done much to apply information technology to the "background" processes that support electoral events. Parliamentarians may now wish to explore ways of using the new technologies to make the act of voting itself more accessible to Canadians. This exploration can be undertaken in several ways.
The study stated that Parliament could make the necessary legislative changes to allow Elections Canada to test some of the promising technologies in controlled, pilot situations, where electoral administrators can learn from experience, where Canadians can observe the new methods in action, and from which parliamentarians themselves can draw conclusions about the directions in which they wish to proceed and at what pace. One simple way to do this would be for Elections Canada to commission a "pilot" system, using a particular technology, so that parliamentarians, and Canadians generally, could observe how it worked in a controlled environment, such as a student election at a secondary school.
Second, it was noted that Elections Canada should continue a dialogue on these issues with interested Canadians, including those with a professional interest in elections, those who supply technology, and those who speak for sectors of our society for whom technology offers particular benefits in terms of access to voting.
Third, there is probably work to be done in educating the public at large on the benefits of the new technologies and their application to the electoral process. Only if the public is fully informed, will it be prepared to support changes to something as important as the process of voting in a federal election.
Fourth, there would be merit in Elections Canada continuing to monitor technological developments in this area, and perhaps to fund appropriate research in electoral technologies and their application to voting processes in Canada and abroad. Elections Canada is Canada's "centre of excellence" in these matters; it should continue to invest in its knowledge base and expertise in technology and the voting process so that parliamentarians, and Canadians generally, have the benefit of up-to-date information and advice.
In addition, the study suggests that the new technologies are unlikely to replace our current methods of voting in the near future. Canadians appear to want choice, not a dramatic change, in how elections are conducted or votes are cast. But as the information revolution permeates more and more aspects of our daily lives, and as Elections Canada strives to ensure the electoral process remains relevant and accessible to all Canadians, it is reasonable to assume that some steps in the direction of electronic voting are inevitable.
The challenge and the opportunity is to ensure that the potential benefits of the new voting technologies are secured for Canadians, without in any way compromising the integrity of the voting process or the confidence of Canadians in their electoral system. The findings in the study suggest clearly that this objective can be achieved, provided it is pursued with care and prudence, on a controlled basis, under the direction of Parliament.
The opinions expressed are those of the authors; they do not necessarily reflect those of the Chief Electoral Officer of Canada.