At the federal level, the Fair Elections Act was passed after much discussion. Although the legislation does not address the issue of e-voting, municipalities are already experimenting with it.
Canada has been among the more active countries in conducting trials in e-voting. To date, e-voting has been used only for municipal elections in Canada, although it has also been used to elect leaders of political parties, both provincially and federally. The current leaders of the federal Liberal and New Democratic parties were both elected using Internet voting.
In 2010, 44 Ontario municipalities with a total population of just under one million provided e-voting during their municipal elections. In most, e-voting was offered as an alternative during an advanced polling period; however, some (e.g., Stratford) have eliminated paper ballots, relying entirely on Internet and telephone voting.
Markham’s experience with e-voting has been well documented. The 2010 election was their third using e-voting, having used it previously in 2003 and 2006. Like many jurisdictions, Markham required pre-registration for Internet voting and offered e-voting only in a defined advanced polling period. Voters selecting the e-voting option were removed from the lists of those eligible to vote at polling booths. Such voters were required to create a unique security question and were mailed a unique PIN.
In Peterborough, all voters were mailed a registration card with a unique elector ID. Those wishing to vote remotely were again required to register online using their elector ID and were sent a unique PIN by email or regular mail.
Halifax Regional Municipality, another e-voting pioneer, provided both Internet and telephone voting in 2008, a 2009 by-election and again in 2012. Halifax does not require pre-registration for e-voting. A letter, including PIN, is sent to all voters, explaining how to vote electronically. In 2008, e-voting was accepted only during a three-day advanced poll window. However, in 2009 and 2012, voters were able to vote electronically during the entire election period.
Based on its earlier success, Halifax was able to eliminate a number of polling stations for its 2012 municipal election, when over 22 percent of votes were cast online.
Halifax allowed candidates to track voters by name or address to see if they had voted, supporting efforts to “get out the vote.” Halifax is also one of the few jurisdictions to specifically allow a “spoiled” ballot (to register a protest vote), considered an important option by some critics of e-voting.
Internationally, Estonia has used Internet voting for national elections since 2005. Although Internet voting accounted for less than two percent of votes cast in 2005, almost 25 percent of votes in the 2011 parliamentary elections were Internet votes.
Estonia is one of the most electronically enabled countries in the world and the only country to have legislated Internet access as a social right. All Estonians have a government-issued identity card, which is used for e-voting, as well as for accessing other government services.
In Estonia (and also in Norway), voters can e-vote multiple times (and also cast a paper ballot on election day) with only the last ballot being counted. Allowing multiple votes by the same voter provides a safeguard against vote buying or coercion, an increased risk where someone can be watching as the vote is cast. Allowing voters to submit a second, later vote can partially address this risk.
Elsewhere, Norway allowed Internet voting in their 2011 and 2013 local government elections. In Switzerland, both Geneva and Zurich have used Internet voting since 2003. E-voting trials have also been undertaken in U.K. local government elections, although these have been discontinued. The United States has piloted Internet voting for overseas military personnel and Americans living abroad.
By allowing voting at any time of day from the comfort of home, Internet voting may provide one possible solution to declining voter participation. While this convenience might be particularly important for attracting younger voters, it also benefits those with physical handicaps, living in remote areas or away from home during an election (attending school, work assignments, vacations).
Notwithstanding this assumed benefit, actual experience has been mixed. Some jurisdictions have reported modest increases after introducing e-voting, but others have actually experienced decreases, indicating that e-voting is just one of many factors (e.g., weather, close races) that impact turnout. One surprising result has been that it is often baby boomers (ages 55-65), rather than younger voters, who take most advantage of e-voting, suggesting that e-voting is a convenience for those who would be voting anyway, but may not increase participation in groups with traditionally lower turnout.
E-voting may also provide cost savings by eliminating or reducing reliance on polling stations and associated staffing. However, cost savings are unlikely to be fully realized until e-voting becomes more widely used and standardized. Trials to date require significant investment in developing new processes and custom software for each implementation.
Finally, e-voting may provide faster election results, but only if there are no technology issues, which may actually delay results.
Notwithstanding the potential benefits, recent studies by Elections Canada, Elections Ontario, Elections BC and others have been generally cautious or negative on e-voting.
The studies have all emphasized the risks:
• Security: protection of the servers, exploiting potential vulnerabilities in the voters’ devices, denial of service attacks (experienced in the 2012 NDP leadership vote), etc.
• Voter authentication: ensuring that only authorized voters can vote. It is worth noting that manual processes are also imperfect in this regard. E-voting could easily reach the level of assurance already accepted for mail-in ballots.
• Secret ballot: ensuring that no one can know how any individual voted, replicating the polling station situation in which there is, in theory, no way to associate a specific paper ballot with the individual who cast it. For e-voting, one can likewise immediately separate the vote “content” from information identifying the voter.
• Equality of access: addressing the ”digital divide,” especially if e-voting is the only option, rather than an additional option to traditional voting.
Underlying these issues is the need for public trust in the process. While there may be technology solutions to the above issues, many are complex to explain to a non-technical audience that may be pre-disposed to suspicion. On the other hand, as more people use the Internet for a broader range of purposes, they may be more likely to accept (or demand) its use for voting.
While not wishing to downplay the risks, which could have extremely serious consequences, no serious disruptions have occurred in any of the trials to date.
Yes, things can go wrong and we need protocols to address such situations when they do occur. However, we now have enough experience in Canada and elsewhere to design e-voting systems and processes that minimize the risks, while preserving the essential characteristics of free and fair elections.
To quote Elizabeth May, leader of the Green Party of Canada: “The crisis in Canadian democracy is not that Canadians are voting more than once but that they are voting less than once.” If e-voting can partly address this, it is worth trying.
This article is adapted from one originally published in the April 2014 issue of Municipal Interface.