Thank you very much, Chair, members, and the New Zealand electoral commissioner. Good morning from Canberra. Thanks for the invitation to appear today.
As members of the committee may be aware and as you just said, I am the commissioner of the Australian Electoral Commission, also known as the AEC, which is how I will refer to it today. I'm responsible for federal elections and referendums in Australia. In Australia we also have seven other electoral commissions at the state and territory level. I'm also responsible for certain industrial elections, but I won't be talking about that aspect today.
I understand that your mandate is to examine viable alternative voting systems, mandatory voting, and electronic voting. In my opening statement today, I'll touch on all of those in some way, but first I might just explore the Australian electoral system a little further.
As you mentioned, at the federal level we have a bicameral system, with a lower house that we call the House of Representatives and an upper house that we call the Senate. Both houses are elected by the people, but they are elected through two different systems of voting.
In the House of Representatives, we have a full preferential voting system. It requires voters to individually number and rank all candidates according to their preferences. A candidate is elected if he or she gains more than 50% of the formal vote. If a candidate doesn't gain 50% of the vote based on first preferences, the candidate with the least number of votes is excluded, and the candidate's preferences are then distributed. The process of preference distribution continues until a candidate achieves more than 50% of the vote.
Under this system, a voter must complete all boxes on the ballot paper for the vote to be formal and included in the count. Of course, some voters don't include all the boxes, and we refer to these as informal votes. There are savings provisions in the legislation that can save certain votes in certain circumstances. However, some ballot papers do remain informal.
The rate of informality is relatively low and consistent at about 5% for federal elections for the House of Representatives. For the 2016 election, which we're still in the process of conducting, the informality rate hasn't been determined, but I'm expecting it to be around the same, if not slightly lower in some cases. In some individual seats, the rate of informality remains high, and there are various factors impacting on that informality. I'm happy to expand on that later on if members are interested.
In the Senate, we use a system of proportional representation to elect candidates. Essentially, this system elects a number of candidates to represent one constituency after they receive a set proportion of the vote. For over three decades, group voting tickets were used to support a system whereby a voter's preferences would continue until all vacant positions were filled. However, in March of this year, new legislation introduced a partial preferential system that enabled voters to determine where their preferences flowed and finished. In my view, it was one of the largest changes to the Australian electoral system since 1994, and we had four months to implement it in time for a July 2 election this year.
There are various elements to this reform, including the introduction of ballot paper scanning for the first time at the federal level in Australia and the implementation of a very large national education campaign. Again, I'm happy to discuss that later if the committee is interested.
In recent weeks there's been some commentary—within Australia, at least—about the perception that the AEC is not making sufficient progress in the count of results for the federal election. I want to clarify that the speed of the AEC's count is driven by the electoral act itself and is not a result of the type of preferential and proportional voting system we have. Yes, conducting a count under a preferential system can take longer than a count under the first-past-the-post system. However, the period of time it takes to finalize the result is more due to other aspects of our legislation, most notably that voters can pretty much vote from anywhere in Australia or around the world for their home electorate. To facilitate this, the AEC undertakes a complex exchange of votes after the election.
Votes can also be received up to 13 days after election day, and the AEC is responsible for sending these votes to the correct division, where the voter's entitlement to vote is confirmed. That means you can vote on one side of Australia, but for that vote can be physically counted, I have to send the vote physically over to the other side of Australia to its home division to be counted.
All of that must occur before the counting of those votes can even commence. In addition, the number of early votes now being cast continues to rise, and this has significant logistical impacts. All of those envelopes need to go back to their home division, and this requirement creates difficulties for us.
I'll now move to the subject of mandatory voting. In Australia we refer to that as compulsory voting.
In Australia it is compulsory to enrol and to vote in federal elections. Compulsory enrolment at the federal level for Australian citizens was introduced in 1918, followed by compulsory voting in 1924.
At the last election we estimate that about 95% of all eligible electors were enrolled. That's 15.6 million people. That is the largest number of electors we've ever had enrolled and probably the most complete electoral roll we've ever had in Australia's history. It's the responsibility of every individual to update their own enrolment details; however, we also have a system of federal direct enrolment and update, and that assists the process. We use trusted third party data, such as driver's licence information, to enrol or update an elector's details.
Under current legislation there is no avenue, really, for successful prosecution of eligible electors who are not enrolled. The reason I say this is that enrolment is an absolute defence for any charge of not enrolling, so if we go down the process of taking someone to court, quite often they'll essentially enrol on the courthouse steps, which is then an absolute defence for non-enrolment.
Compulsory voting and enrolment is seen as a normal part of Australian political culture. There is lots of evidence to suggest continued support for compulsory voting: in 2013, the last time we did surveys, about 70% or thereabouts of the population indicated support for compulsory voting. At the most recent federal election, which we've just had, turnout was around 90%, but we'll have to confirm that over the coming weeks as we finish the processes with that election.
Under our system of compulsory voting, those enrolled electors who did not vote are sent a non-voter letter. It requires the electors to either respond and provide a valid excuse for not voting or pay a very small $20 fine. A small number of those voters who don't pay the fine are then prosecuted, and I think we went through a full prosecution of about 3,000 people at the last election.
In Australia our electoral matters committee has considered the issue of voluntary voting a number of times; however, the issue has never been pursued by our federal Parliament.
I might move now to electronic voting.
Electronic voting is a matter for the Australian Parliament, not for the AEC, and it would require a change to our legislation. At the federal level we do not use electronic voting, nor do we use Internet voting. In 2014 our electoral matters committee inquired into the topic of electronic voting, and I will just quote from that for one moment. It found that “irrespective of one's philosophical view about electronic voting, ...there can be no widespread introduction of electronic voting in the near term without massive costs and unacceptable security risks.”
However, in recent weeks our Prime Minister and the leader of the opposition have both pledged their support for examining some form of electronic voting, again following perceptions about the length of time it takes for us to return the result. I cannot speculate as to what model would be introduced, how it would be introduced or when, as this is a matter for Parliament.
At the state and territory level, some commissions have trialed electronic voting. In the Australian Capital Territory, where Canberra is located, electronic voting has been used in early voting centres since 2001. They use a system of personal computers. In New South Wales, Internet voting was trialed in 2011 and 2015 for particular categories of voters, those who have a disability and those who are more than 20 kilometres from a polling place. My understanding is that well over 250,000 voters availed themselves of Internet voting at the last New South Wales state election. I am aware of significant media commentary surrounding security aspects of this system, but I'm unable to comment further. I don't own that system.
We don't have electronic voting, but many aspects of our electoral process are already electronic. We've enabled voters to enrol online for a number of years, and voters can also apply for a postal vote online. The increased use of the electronic forms by the Australian community has simplified the process of these transactions without diminishing the controls in place.
We also introduced the scanning of Senate ballot papers for the first time in the 2016 election. As I'm speaking to you, we are still running very large scanning centres in each of the states, scanning Senate ballot papers for the 2016 election.
We've also deployed electronic certified lists at the last two elections. These are an electronic version of the electoral roll, and they contain a list of electors entitled to vote. I think it's a great initiative. It provides the ability to search for and mark off an elector's name, provides a real-time update to a central copy of the certified list, and, where we can, enables the printing of House of Representatives ballot papers on demand. At the recent election, we had about 1,500 devices. It is expensive but worthwhile.
There are two other aspects of our electoral system that I might touch on to provide a flavour for what we do.
Under the electoral act, the Electoral Commission is wholly responsible for undertaking the redistribution of electoral boundaries. It's an apolitical process and has no involvement from political parties or politicians. In particular, changes to divisional boundaries do not require the approval of Parliament. Submissions and objections to boundaries are invited for consideration; however, the ultimate decision rests with the Electoral Commission, a three-person commission chaired by a judge.
Another aspect of our work is managing the funding and disclosure regime of our electoral legislation. It includes registration of parties and party logos and a disclosure of expenditures and returns for an electoral campaign and ongoing expenses during the year. The scheme is designed to inform the public about financial dealings of political parties, candidates, and others involved in the process. We are responsible for managing and running this independent apolitical process. As with other areas of our electoral system, any changes to the funding and disclosure system are also a matter for Parliament.
Thank you very much for the opportunity to appear today. I'm happy to take any questions or provide any other information that the committee may find helpful.
Please, then, turn to slide 5, which deals with the electoral reform process that saw New Zealand adopt the MMP electoral system in 1993.
The process began with growing public dissatisfaction in the 1970s and early 1980s with the perceived fairness of our first-past-the-post system and its tendency to deliver Parliaments that did not reflect the nationwide vote.
The Royal Commission on the Electoral System was established in 1985. It met for 18 months, consulted widely with the public, considered a wide range of reform options, and reported in 1986. It recommended that a system of MMP be adopted and that a binding referendum be held with the 1987 general election to give the voters a choice between first past the post and MMP.
A referendum was not held with the 1987 election, but there followed a period of ongoing public and political debate, which included an inquiry by a Parliamentary select committee into reform of the electoral system in 1988.
Then, in 1991, legislation was introduced providing for a two-step reform process.
The first step was an indicative stand-alone referendum, held in September 1992, in which New Zealanders were asked two questions. The first question was on whether they wanted to change the electoral system. The second question, irrespective of whether or not they wanted to change the electoral system, was that if there were to be a change, which of four systems—MMP, a supplementary member system, a preferential voting system, or the single transferable voting system—would they support?
The referendum was preceded by a comprehensive public education campaign undertaken by an independent electoral referendum panel headed by the Chief Ombudsman.
Turnout at the referendum was 55.2%, and of those who voted, 84.7% voted for change in question 1 and 70.5% voted for MMP in question 2.
The second step in the process was a binding referendum held with the 1993 general election. Voters were asked to choose between first past the post and MMP. Again the referendum was preceded by a comprehensive public education campaign.
Turnout was 85.2%, and 54% voted for MMP. As a result, the Electoral Act 1993 provided for an MMP electoral system. It had already been enacted by the New Zealand Parliament in anticipation of this possible outcome and came into force by operation of law.
Preparations then began to deliver New Zealand's first MMP election in 1996. A substantial amount of work was required. Not only did the move to MMP require comprehensive revision of the systems and processes for the delivery of elections, but Parliament found the need to revisit the 1993 electoral legislation, and after select committee consideration and public consultations, amendments to the legislation were enacted in 1995.
Electoral boundaries had to be redrawn to meet the requirements of the new system. With regard to the mechanics of government formation, cabinet processes, and Parliamentary processes, consideration had to be given to the implications of coalition governments and the increase in the number of parliamentary parties that were likely likely under the new system, .
Political parties also needed time to adjust to the different requirements of the new system, including fewer electorates, new nomination processes, and the implications of the party vote for electioneering. Again a comprehensive public education campaign was required to prepare voters for the new system.
I now propose to move to slides 8 to 12 and to outline the features of New Zealand's system of MMP.
New Zealand's system of MMP is a moderate form of proportional representation that seeks to balance two important objectives. One is the principle of proportionality, the principle that a party's share of seats should reflect its share of the nationwide vote. The other is the need to ensure that elections deliver effective Parliaments and stable governments by avoiding the undue proliferation of very small parties in Parliament. A further objective, and one that is fundamental to the mixed member system, is to continue to have local electorate MPs. Therefore, the defining characteristics of MMP are a mix of MPs elected from single-member electorates and those elected from a party list as well as a Parliament in which parties' shares of seats roughly mirror their share of the nationwide vote.
New Zealand has now had seven MMP elections. Each election has resulted in between six and eight parties represented in Parliament. Each election has resulted in some form of coalition government or arrangement between political parties, as is to be expected under a proportional system. Each government has retained the confidence of the Parliament throughout the parliamentary term.
Moving to slide 9, please, we see that under MMP each voter has two votes. On the left-hand side of the ballot paper, they vote for the party they most want to represent them in Parliament. Only registered political parties can contest the party vote. On the right-hand side of the ballot paper, they vote for the candidate they most want to represent them in their electorate. Candidates of unregistered parties and independents can contest the electorate vote.
Let's go to slide 10, please. New Zealand is currently divided into 71 electorates. Electorate boundaries are reviewed after every five-year population census by an independent representation commission. The decisions of the representation commission on boundaries are final. Electorate seats are won, on an electorate-by-electorate basis, on a first-past-the-post basis.
Moving to slide 11, please, we see that the party vote is counted on a nationwide basis. To be eligible for an allocation of the seats, a party must win either one electorate seat or 5% or more of the party vote. These are known as the thresholds, and they are intended to avoid the undue proliferation of very small parties in Parliament. If a party wins 40% of the party vote in an election, for example, the party is entitled to 48 seats in a 120-seat Parliament. If that party's candidates won 30 electorate seats, the party would be topped up with 18 seats from its party list to bring its number of seats in Parliament up to 48. Thirty electorate seats plus 18 list seats equals the 48 seats its share of the party vote entitles it to. If another party wins 10% of the party vote and no electorate seats, it is entitled to 12 seats in a 120-seat Parliament. All those seats come from its party list.
A party's list seats are allocated to its candidates in the order they appear on the party's list, excluding those who win an electorate seat. We have included in the background papers provided for members an example of the party list provided to all voters at the 2014 election.
Moving to slide 12, we see there are a number of mathematical formulas available for the allocation of seats in proportional systems, and the particular formula adopted by New Zealand is the Sainte-Laguë formula. We have included an explanation of the method in the background papers provided for members. The explanation uses the 2014 elections results, an example that I think nicely illustrates the process.
One thing to note is that the New Zealand system does not prescribe a 120-seat Parliament. The current New Zealand Parliament, for example, has 121 members. This happens when a party wins more electorate seats than it is entitled to under the party vote. When this happens, the party keeps all its electorate seats and the number of list seats allocated to other parties is increased by the number of what we call overhang seats. Hence, the current government has 121 members.
Turning to slide 14, New Zealanders have, since the adoption of MMP in 1996, demonstrated a practical understanding of how MMP works. We can see this through low levels of informal voting and relatively high levels of split voting or strategic voting—that is, a voter casts a valid party vote and then casts a valid electorate vote for a candidate from a different party. We can see this also through the fact that the overwhelming majority of voters consider the ballot paper layout to be clear, concise, and easy to use.
As illustrated on slide 16, one of the benefits advanced for proportional systems like MMP is that it leads to a more diverse Parliament than tends to result from first past the post and from majoritarian systems. Certainly, as this graph shows, the New Zealand experience is that MMP has resulted in more women and more Maori elected to Parliament, the majority of them elected as list MPs.
For example, of all members of Parliament elected to Parliament from party lists, 43% have been women; by contrast, only 24% of MPs elected from electorates have been women. Also, 21% of all list MPs have identified as Maori, compared with 14% of all electorate MPs.
Representation of Pacific and Asian peoples in the New Zealand Parliament has also improved.
By way of conclusion, I would suggest that the key lessons from the New Zealand experience are that it is possible to successfully introduce a new electoral system, but at least in the New Zealand context, the process of reform is very important to public trust and confidence in the outcome, and the process needs to involve plenty of opportunity for public input, public consultation, and public education.
Our experience is that the design and development of different electoral systems raises many points of principle and many points of detail. These points are all important to people.
Again, at least in the New Zealand tradition, final decisions are confirmed by the voters at a referendum in which they have access to all the details necessary to make an informed decision, and this all takes time.
Finally, our experience is that once a decision is made, the time needed for careful planning and implementation to transition to the new electoral system is not to be underestimated, not just for electoral officials but also for those involved in the administration of executive and parliamentary processes and in the administration of political parties.
Mr. Chairman, that concludes the introductory comments I wanted to make. I also am very happy to take questions on any of that or any other matter.
I should put on the record that this was a legislative change that was mandated for the Electoral Commission by the Australian Parliament, not a desire for us to do that ourselves. I'm not making a judgment either pro or anti; it was just a requirement.
There was debate in Australia about the Senate voting system at the last election. Again, I'm just reflecting. There was a concern that some members of the Senate were elected with quite a small amount of the primary vote, so a system was implemented to change the way Australians vote.
The Senate paper is very long. I think it's over a metre long. There is a line, and above that line are party names and boxes, and below the party boxes are the individual candidates for each of those parties. Previously Australians could cast a valid vote just by voting “1” above the line, and then their vote was allocated according to a pre-lodged ticket by the party. We did a very complex preference allocation according to the party's wishes.
It changed at this election. For a valid vote, a citizen had to number at least six boxes above the line or 12 boxes below the line.
In the 2013 election, 97% of Australians chose the option of just putting a number “1” in their preferred party box above the line, and only 3% of Australians voted below the line. That made the count for us significantly simpler. The count is done by state. The Senate is a states House. In New South Wales, our largest state, it meant that only 100,000 people voted below the line, but each of those 100,000 votes then had to be taken to a place we call Central Senate Scrutiny in each state and individually double-data-entered to make sure that there were no errors. It was very complex. For 100,000 ballot papers, it took a lot of time.
This time, just in New South Wales, it meant that every Senate ballot paper had to go to Central Senate Scrutiny, so it meant that we had to physically count 5,000,000 ballot papers in the one spot, just in New South Wales, with a very complex series of preferences.
I had two options. We did a big project for this. Our assessment was that for me to be able to enter the data in time to return the writ to the Governor-General, we would have to have 900 data entry operators working virtually 24-hour shifts to come close to making it—just in New South Wales—or we could implement a semi-automated solution involving scanning, but still with human involvement. It has meant that right now, as I'm speaking to you, we have, just in New South Wales, 19 scanning pods with custom-made scanners and some 300 data entry operators, who at times have been working 24-hour shifts, to enable us to double-data-enter, for integrity reasons, the Senate ballot paper.
Given that there was a change to the voting system, we ran a very large public education campaign to educate the public about how to cast a formal vote. That took TV, radio, Internet, and a whole range of other forms. It was very extensive. I would have to tell you that it was very expensive. I have not yet provided a public costing for it, so I can't provide that to you today, but it was in the many, many millions of dollars. It was to educate the public to ensure that the rate of informality remained low.
The other change, for us at least, was that for the first time ever it was a requirement for us to register party logos and for party logos to appear on the ballot paper. Again, all of this, the entire project, had to be done within four months. The figures I gave you were for one state, for New South Wales. You need to multiply that around Australia, which means that we needed to scan essentially 15 million ballot papers or thereabouts, and be able to produce an outcome from that.