No Silver Bullet: Online voting and local election turnout

Local body election time is over for another three years, and even before polls closed, there were laments over low turnout. A low turnout undermines the legitimacy of the winners and can point to wider problems: disillusionment with democratic processes, institutions and actors. It is also problematic because some groups are less likely to vote than others, and so candidates appeal to the interests of those who vote over those who don’t. Older people and home owners are more likely to vote in local body elections, which may explain the prevalence of ‘controlling rates’ as a campaign slogan.

In the lead up to the 2016 local body elections, a trial of electronic voting was proposed and was some way towards implementation before being abandoned, because of security concerns. A number of commentators have argued the online voting will help turn around declining local body election turnouts, but I want to argue this is not necessarily the solution to the problem. I ask two simple questions: will the proposed solution solve the problem, and what new problems will it create? Not only should the solution work, but, when balancing all effects, it should be worthwhile.

Will electronic voting solve the problem of turnout?

The first question to ask is, would online voting actually solve the problem? Is the means of voting the cause of non-voting, or are the problems ones of political engagement? Research in the United States showed that easing voter registration requirements did not lead to higher turnout; the ‘cost’ of voting to electors was not voting mechanics as such, but getting informed about politics. Making voting easier helped the already engaged vote, but did not increase engagement. Solving low turnout is more complex than just making the mechanics of voting easier.

Local Government New Zealand’s surveys of electors reveal a range of reasons for not voting, including 24% of non-voters saying they forgot or left it too late. An App that reminded people to vote may increase voting among this group (and worked when used by the Electoral Commission in 2008), as may a return to election days. But non-voters also reported engagement issues such as ‘not knowing enough about the candidates’ (31%) or not being interested (14%).

A number of studies including by the Electoral Commission (2008), Statistics New Zealand and the New Zealand Election Study (2014) (with data analysedhere), have asked New Zealanders whether they want online voting and whether it would make them more or less likely to vote. While some respondents like the idea of online voting or were more likely to vote as a result, a significant number of electors reported the opposite. At present we know that the people least likely to vote are younger, people without work, have lower income, and less education, those who don’t own their own home, and are recent migrants. In terms of ethnicity, Asian, Māori and Pacific peoples report higher levels of non-voting than European New Zealanders.

Young people, who are under-represented in voter turnout, report they would be more likely to vote if there was an online voting option. But other groups supportive of online voting were the groups who are already well represented in present turnout: those with higher incomes and higher status jobs, and European ethnicity. It is hard to see how online voting – with no other changes – would increase turnout among existing low-voting groups, who feel unengaged from the process, uninformed about candidates or can’t see how the outcome affects them.

The new problem: Security

New solutions create new problems. In the case of online voting, the most intractable problem relates to the security of the system. If the voting system is not secure, the whole process risks losing public confidence, creating a downwards spiral of even more disengagement and non-voting.

Voting in democracies has important features, including anonymity. The privacy and anonymity of voting ensures that people are not able to be bullied or bribed to vote a particular way; the secrecy of the ballot is considered such an important feature of elections that it is included in Article 21 of the Universal Declaration of Human Rights. Another important feature is verifiability – knowing that the votes cast are the votes counted. These two features are essential to trust in the system, yet with online voting they pull against each other. In a paper ballot system – and this holds particularly where people vote at a polling booth and less so for postal voting – the vote you cast is the vote that gets counted. The scrutineers and the electoral staff can tell your intention (unless your ballot paper is very messy) and count it accordingly. They do not trace the vote back to you unless there is suspicion the vote was tampered with, and so anonymity is retained.

The security concern with online voting includes the inability to prove that the vote counted was the same as the vote lodged, given anonymity. What happens to the vote between the citizen making their vote, and it being counted? Can it be tampered with or hacked in any way – and if it was, how would anyone know? This concern just does not happen with paper ballots; voting papers are kept secure until counting, at which point the vote on the paper is witnessed by multiple people during the counting process who know what your intention was. But how do scrutineers verify that the online votes that arrived in the “in” box for counting are identical to the votes cast? This particular security concern does not exist with other important online transactions such as electronic banking, where customers can see their bank activity and balances and contact the bank if anything is remiss. As voters, we will have no idea whether the vote we make is the vote that gets counted. Further, all online banks and retailers expect a certain ‘loss’ from things going wrong: is this acceptable for our democratic processes? There are currently multiple security weaknesses with online voting as outlined here and here, including the inability to guarantee both anonymity and verifiability.

Conclusion

There is no silver bullet to improving turn-out for local body elections. A number of suggestions have been made which may make small differences:more autonomy for local government from central government; increasing civics education in schools; and a call for the use of political tickets so voters know how candidates are aligned. We need to understand why different groups do not vote – new migrants, for example, will have different reasons to the unemployed – and adopt different strategies to encourage civic engagement. Councils themselves need to work to improve the trust of their communities. A greater diversity of candidates who better reflect the general population, may encourage more engagement. People are more likely to vote when they understand the voting system, and the complexity of local body voting may be an issue for some: in some localities, electors are asked to vote for a regional council, a local mayor, local councillor, local or community board, licensing trusts and DHB members. Further, these votes are executed using different voting systems. The cost of voting – of being informed about all these candidates, structures of local government, and voting methods – is very high.

Central government will continue to investigate online voting and local authorities will run their own reviews of the 2016 elections. Online voting will not solve the problem of civic engagement by itself, and it creates real security risks. It is no silver bullet.

This was first published as a Briefing Paper on 11 October 2016 http://briefingpapers.co.nz/2016/10/no-silver-bullet/

© Scoop Media