Peter McKenzie: Why we need automatic voter enrolment
New Zealanders are rightly proud of our democratic history. Over 162 years of evolution, we have extended the franchise to include women, the landless and the young. Each evolution has been spurred by New Zealand’s fundamental commitment to egalitarianism.
And yet on Election Day in 2017, 282,256 adults were not allowed to cast a valid ballot.
The vast majority - 271,821 - were people whose only disqualifying sin was failing to enrol before election day. Many did not attempt to vote, either because they knew it was pointless or because of the same distrust and fear which drove them not to enrol. But 24,767 unenrolled New Zealanders did cast a ballot. All were subsequently ignored.
That our electoral laws required poll-workers to effectively throw away about 1 percent of all votes cast at the last election is a democratic disgrace. Unhappily, it has become ordinary in New Zealand.
Since the introduction of MMP in 1996, 207,585 unenrolled voters have cast ballots which were disallowed. Approximately a quarter of those were in 1996, when 50,291 ballots were disqualified. In subsequent elections the number dropped significantly, hitting a low of 13,888 in 2002. The number has been steadily increasing since then. By 2014 the number had doubled to 27,457 disqualified ballots, and it only dropped slightly in 2017.
This has often been treated as the inevitable consequence of having an electoral roll - the record of all New Zealanders who have enrolled to vote. And it is true that having a reliable electoral roll is crucial. It safeguards the integrity of our elections by preventing voter impersonation or double-voting. Its efficacy can be seen from the fact that almost no ballots in 2017 were subject to concerns about fraud.
By comparison, prior to 1975 the Māori seats suffered from significant electoral neglect, including the absence of a Māori roll. The Electoral Commission notes that there were “frequent allegations of electoral irregularities in the Māori seats” as a result.
The electoral roll is crucial for other reasons. Local councils use the national electoral roll to compile local electoral rolls. Iwi can use the electoral roll to find people who might be connected to the iwi. Academics and genealogists often use historical electoral rolls to support their inquiries. Jury service functions by selecting random individuals from the electoral roll.
The absence of 282,256 New Zealanders (7.6 percent of eligible adults) from the electoral roll is therefore incredibly destructive. Those who do not enrol are typically already marginalised from civic society. Their electoral absence compounds that marginalisation. They will be absent from local elections, invisible to their iwi and hidden from history. Far from being a benign consequence of having a reliable electoral roll, the absence of eligible adults from the roll steadily undermines its purposes.
The coalition Government appears to have recognised the problem. In late 2018, the Hon. Andrew Little announced that the Ministry of Justice would investigate allowing people to enrol on election day in the future. It is a welcome announcement. But it is not enough.
Of the 24,767 ballots disallowed in 2017 because the voter was unenrolled, approximately 18,000 were cast on election day. Had election-day enrolment been permitted, these ballots would not have been disallowed. But the other 5,800-odd ballots were cast in advance of election day, meaning that they would have been disqualified even with the reforms proposed by the government. Moreover, allowing election-day enrolment does nothing to address the civic invisibility of the 257,489 who neither enrolled nor attempted to vote.
This whole issue is slightly farcical since there is another, relatively simple solution. Automatic Voter Enrolment.
Enrolling to vote is currently a manual process done through mailing a form to the Electoral Commission or online using RealMe to enrol. It’s hardly front of mind for most people. So the Electoral Commission hires thousands of fieldworkers to approach eligible adults and help them fill out enrolment forms.
The Commission also partners with government departments like the NZ Transport Agency, Ministry of Social Development and Department of Internal Affairs. These partnerships allow the Commission to tap into the immense data collections these departments possess. For example, the partnership with NZTA allows the Commission to use data from driver licence applications and vehicle registrations to identify unenrolled adults and encourage them to enrol. In 2017, the Commission sent out 451,187 unique invitations to enrol or update details. 371,347 people did not respond. It is likely that this included the 271,821 eligible adults who were unenrolled in 2017.
In other words, because of its partnership with other government departments, the Commission knew who was unenrolled and knew enough to automatically enrol them. They just weren’t allowed to. As a result those adults remained disenfranchised.
Automatic voter enrolment (AVE) would create an opt-out system. Manual enrolment of the sort we currently have would continue to be an option. But the Commission would be able to use the data it already collects to automatically enrol eligible adults and place them on the general roll. Enrolment is compulsory under New Zealand law, so eligible voters would be able to refuse enrolment only if their details are incorrect or in truly exceptional circumstances. Enrolees with personal safety concerns could request that they be placed on the unpublished roll, as is already the case. People who have been automatically enrolled would be able to manually switch to the Māori roll outside of the Māori Electoral Option, similar to how first-time manual enrolees are able to choose the Māori roll outside of the Option.
AVE is normal in developed democracies. The United Kingdom, Canada, some states in Australia and America, and some European countries have all implemented it with great success. Oregon, a small American state with a population of approximately four million, implemented AVE in 2016. It enrolled 272,000 new people through the program, 116,000 of whom were unlikely to have enrolled otherwise (based on their demographic profile).
Another example is New South Wales, with a population of approximately 7.5 million, which introduced AVE in 2009. In its 2015 election, AVE added to the roll 150,000 eligible adults who were unlikely to have enrolled otherwise. Approximately 98 percent of eligible adults in NSW are enrolled, compared to 92.4 percent of eligible Kiwis.
All evidence indicates that AVE would increase enrolment rates here in New Zealand too, drastically reducing the number of disqualified ballots and eliminating the civic invisibility of unenrolled citizens.
Interestingly, international evidence suggests that higher enrolment also leads to higher voter turnout. Returning to the example of Oregon, voter turnout in 2016 was 4.1 percent higher than in 2012. Nationally the increase in voter turnout was just 1.6 percent. The big change in Oregon between 2012 and 2016 was the introduction of AVE. Of the seven OECD countries with higher voter turnout than New Zealand, five (Sweden, Denmark, South Korea, the Netherlands and Israel) have AVE. The reduced hassle of participation in an AVE system likely plays a significant part in that increased turnout.
AVE is also cheaper. The Electoral Commission expected to spend $105.7 million in 2017, much of which was spent on the 16,700 temporary fieldworkers it hired for the 2017 election. A large portion of these staff focused on enrolment outreach. The introduction of AVE would reduce that need. The Commission also mails out immense numbers of enrolment forms and covers the cost of their return. Reducing manual enrolment would entail huge savings in the cost of mail. AVE would also reduce the need to process paper enrolments. Officials from Arizona in the United States found that processing a paper enrolment cost $1.44 NZD on average, while enrolment via AVE and online methods cost merely $0.06 NZD.
The more money the Commission saves on enrolment, the more it can spend on outreach to increase voter turnout and public understanding of the political system - far more consequential tasks.
It is worth noting that this isn’t a partisan proposal. The groups most marginalised from politics through lack of knowledge or interest are people in rural areas, the young, men and ethnic minorities. These groups have very different partisan leanings.
The status quo cannot continue. We regularly throw away as many as 1 percent of votes cast in an election. Hundreds of thousands of eligible voters are disenfranchised. Our electoral roll is unrepresentative of the wider population, and our Electoral Commission spends much more than it needs to.
Allowing election-day enrolment would do little to address these issues. Introducing automatic voter enrolment would. It is a change which the Commission itself requested in 2018, and already has the tools to perform. It is more effective, cheaper and more reliable.
There are very rarely slam-dunks in public policy. Automatic voter enrolment is the exception.