Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Conduct of the 2010 federal election and matters related thereto

CHAIR (Mr Melham) —I declare open this public hearing of the Joint Standing Committee on Electoral Matters for its inquiry into the conduct of the 2010 federal election and related matters. To date the committee has received 92 submissions that have raised a number of issues that the committee will consider during the course of the inquiry. The committee held its first public hearing on Wednesday this week in Canberra. I can indicate that further hearings are planned for other major cities. Subject to who submits to the committee some of those might be teleconferences. Today the committee will hear from representatives of the Australian Electoral Commission followed by representatives of the Democratic Audit of Australia.

I now welcome to today’s hearing the Electoral Commissioner, Mr Ed Killesteyn, and other representatives of the Australian Electoral Commission. Although the committee does not require you to give evidence on oath, I should advise you that these hearings are legal proceedings of the parliament and therefore have the same standing as proceedings of the respective houses. We have received a submission from the Australian Electoral Commission and I understand that the commissioner wishes to make a presentation and some opening remarks.

Mr Killesteyn —Thank you, Chair, and thank you for allowing me to make this presentation and take some of your time at the outset of the hearings. The presentation tries to put in place our perspective on the issues that we faced in the 2010 election and some of the recommendations that might flow from that. Before commencing my presentation, I want to acknowledge with some sadness the recent passing of Mr Dick Klugman, a former member and chair of the Joint Select Committee on Electoral Reform, as it was known in those days. It was of course an important part of the process which led to the establishment of the independent Australian Electoral Commission. Mr Klugman is fondly remembered by those in the commission who were here at the time and certainly his passing is sadly mourned. Coincidentally, the date of his passing was on the 27th anniversary of the establishment of the Australian Electoral Commission.

The AEC’s submission, as you have referred to, Chair, sets out two broad themes. The first is that the nature and volumes of workloads encountered by the AEC during the election period are changing and increasing, driven by elector behaviour, preferences and overall numbers. Our submission suggests that these drivers of change need to be accommodated in the legislation in some fashion to maintain good electoral practice.

Secondly, our submission deals with the challenge to the health of Australia’s electoral democracy posed by declining levels of engagement, and that remains to be dealt with. The presentation that I have for you is a series of charts, most of which are set out in our submission but we have made them separately available to you just in case, like me, you are optically challenged by watching the screen.

To go to the first slide, the announcement of the election led to a dramatic reaction by the electorate. While that is normal at the time of calling an election, the scale of the reaction in 2010 was unprecedented. On Monday, 19 July, the AEC call centre received in excess of 112,000 telephone calls, compared with 33,000 on the first Monday following the announcement of the 2007 election. So that was nearly a fourfold increase in telephone calls on the first working day after the announcement of the election.

CHAIR —Before we go any further I might actually interrupt there. I do not want to break your presentation but I think it is worthwhile raising matters at the time. Have you managed to work out or discover why that was?

Mr Killesteyn —Yes, and it will be covered in my later charts. Do you want to wait until I get to it?

CHAIR —It is up to you.

Mr Killesteyn —By day 7 of the election period, the total calls received at that point was 261,000 in 2010 compared to 131,000 in 2007. The total number of calls received through to polling day was 730,000 in 2010 compared to 580,000 in 2007, an increase of over 25 per cent but in a six-day shorter period. Over 50 per cent of the increase in telephone calls over the full period came on the first working day. That is my point: it was an unprecedented reaction by the electorate on that working day.

Telephone calls were not the only medium used by electors to contact the AEC. Indeed, the AEC’s enrolment check facility received nearly 9.5 million page views compared with 2.6 million in 2007. Moving to the second slide, in a similar fashion the 2010 election saw a significant shift in the manner in which electors wished to interact with the AEC. In response to deliberate policies to make the AEC more accessible to electors, emails received by the AEC numbered 170,000 over the 2010 election period compared with 2007 when only 37,000 emails were received. Taken together, that is telephone calls, online enrolment checks and emails received, all these indicators suggest an electorate keen to interact with the AEC electronically in the same way as they do with many other government agencies.

Turning to slide 3—and this is addressing your question, Chair—it is clear that most of the electorate was contacting the AEC in the early part of the election period because of concern about their enrolment status. That is shown by the analysis of the telephone call centre, which is managed for us by Centrelink. As you can see in this slide, on the first Monday after the announcement of the election—that is day 3 of the 2010 election period—there was a significant surge of applications or transactions for enrolment. In fact, that Monday, 190,000 enrolment transactions were received compared with 17,000 and 15,000 in 2007 and 2004 respectively.

CHAIR —That is why I am interested in why—

Mr Killesteyn —It is in the next slide, we are getting closer! Many of the 170,000 emails that I mentioned earlier also had attachments which were PDFs of enrolment applications. Overall, if you look at the final figure across the total number of transactions, the AEC processed in excess of 560,000 enrolment transactions, eclipsing the record set in 2004—nearly 150,000 more than in 2004 and a massive increase of 280,000 on the 2007 election.

These enrolment workloads, to be completed against the immutable deadlines of an election, are testimony to the hard work and dedication of all AEC staff in national office, in our state offices and most particularly in our divisional offices. But two issues arise from this analysis. The first is that I want to assure the committee that each of these enrolment transactions was processed with the same enrolment fraud detection procedures as any other enrolment transaction processed outside of an election period. Secondly, while the surge in transactions on the first working day after the announcement of the election is arguably related to the close of rolls period which was in place at that time for new enrollees, the overall volume in enrolment transactions is clearly linked to the success of measures employed to build the roll in the intervening period between elections. We are getting closer to your question, Chair, and I think this will go the answer.

As this slide shows, the relatively low volume of enrolment transactions processed at the 2007 election—we see here that we processed some 280,000 enrolment transactions in 2007 in comparison to 2010 where the number  was 560,00—is clearly linked to the AEC’s success in building the roll between the 2004 election and the 2007 election. You see here a very large increase in the roll between 2004 and 2007. You see a smaller increase in the roll between 2007 and 2010. Correspondingly, the number of transactions that come in in the close of rolls period and is done during the close of rolls period is inversely related. The more successful we are with building the roll the less they need to send in transactions at the last minute; the less successful we are the more transactions come in at the close of rolls period.

CHAIR —I will explore that now. Is that because they are on the roll, or you have successfully got them on the roll, but they have changed address and a lot are change of address transactions?

Mr Killesteyn —That is correct.

CHAIR —And I will come to that later. In your submission you have got different figures—282,000 to 563,000 is a huge figure.

Mr Killesteyn —It is a massive increase.

CHAIR —I am wondering whether GetUp! played a role in that. I am not giving them exclusive support, but they have got a large network. They obviously had the High Court case. Is it as a result of a lot more awareness that there was less time from when an election is called to get on the roll or make a change? It is just double in a three-year period; it is not a gradual increase. I am just asking the question, I am not—

Mr Dacey —We got nearly 200,000 on the first day, Monday, and there was not a lot of time between the announcement on the Saturday and the Monday to do a lot of advertising. But nearly 200,000 of those 563,000 came in on that first day.

CHAIR —But overall there is a doubling in a less than three-year period.

Mr Killesteyn —Could I take you to table 3.4 on page 28 of our submission.

CHAIR —I think it is a successful campaign by the commission—do not get me wrong.

Mr Killesteyn —It is a successful processing.


Mr SOMLYAY —Do you think there was any influence on awareness by the political parties sending out postal vote applications to everybody?

Mr Killesteyn —Postal vote applications will generally start to flow sometime after that close of rolls period.

Mr SOMLYAY —Yes, but they would have received their political party stuff on day 1 or day 2.

Mr Killesteyn —They may have. I suspect it has something to do with the close of rolls period. We had the announcement on the Saturday, we had the close of rolls period taking effect immediately on the first working day—that is the Monday that the writ was issued—and our advertising kicked into place immediately on the Saturday, or the Sunday, where we said ‘You have only got till 8 pm on Monday to get your transactions in’.

CHAIR —That is right. There were an extra couple of days in effect because of the weekend and the way it was announced, wasn’t there? And that is when your advertising campaign kicked in.

Mr Killesteyn —This is why I suggest you go to that table 3.4.

CHAIR —Yes, I am looking at it.

Mr Killesteyn —What you see there is an interesting picture between 2007 and 2004; where in 2007 the announcement was on the Sunday, 14 October. The writ was not issued until day 4, and you see there a rather large surge of 107,000. You see exactly the same pattern in 2010, albeit that the writ is issued one day earlier, again a large surge. So it is quite clear that the pattern of people simply responding with their enrolment transactions at the time that the election is announced is the same that you get every single election.

CHAIR —I am making the point that the announcement of the election occurred a couple of days before the writs were issued. If the election were announced and the writs were issued on the same day, then a lot of people would have been disenfranchised. In 2007 I think we had a couple of public holidays in different states—

Mr Killesteyn —That is correct.

CHAIR —which stopped the government from—

Mr Dacey —That extended the close of rolls period for people wishing to update their address.

CHAIR —So that critical 24-48 hours or 72-hour period seems to produce the surge as a result of advertising by you and heightened awareness?

Mr Killesteyn —Absolutely. The advertising is focused on making sure that people understand what the deadlines are. If you contrast that same table in 2004, where the seven-day period applied, you see a much more gradual, flatter curve. You do not get that huge surge, albeit there is a surge on the last day of the 2004 close of rolls period.

CHAIR —But you say the biggest surge in 2010 as against 2007 is due to the fact that you have had more people on the rolls.

Mr Killesteyn —The comparison between 2010 and 2007 is that there were fewer people added to the roll between 2007 and 2010.

Mr Dacey —Yes, from one event to the other there were fewer people.

Mr Killesteyn —As this slide shows, the relatively low volume of enrolment transactions processed at 2007 is linked to the AEC success in building the roll between 2004 and 2007. One immediate conclusion that people might jump to is that the AEC did not do enough since 2007 to build the roll. That is not a conclusion that is supported by the facts. In chapter 3 of our submission we outline the comprehensive array of activities undertaken to enforce the obligation to enrol and to facilitate the entitlement to enrol. For example, the continuous roll update program was scaled up significantly with 4½ million continuous roll update letters sent out in 2009-10, an increase of 84 per cent on the 2008-09 mailing of around 2½ million letters. Almost 2.6 million of these letters were sent between January 2010 and July 2010 in the lead-up to the election.

It is also worth noting the conclusions of the ANAO in its 2009-10 audit on the AEC’s preparation for the conduct of the 2007 federal general election. The ANAO also looked at the successful increase in the roll in 2007. In its report the ANAO said about this successful increase:

These gains were, however, achieved at considerable extra expense. Including the costs of  TES for 2007, the AEC spent almost $36 million on enrolment activities, promotion and advertising prior to the 2007 election, compared to a little over $10 million prior to the 2004 election.

It continued:

On this basis:

  • the unit cost of each additional enrolment transaction rose to more than $14, more than three times the unit transaction cost in 2004; and
  • the unit cost of each new enrolment rose to more than $83 (two and a half times the unit cost in 2004).

The overall result was that the marginal unit cost of each of the net additional 118 885 enrolments achieved in 2007 compared to 2004 was over $216, or nearly seven times the unit cost per additional enrolment in 2004.

In other words, this increase here was a significant success in terms of building the roll but it came at a huge expense. Of the $36 million that was spent in building the roll in the 2007 election year, $29 million was solely on advertising—$29 million out of $36 million was on advertising.

CHAIR —But is that not because the government at the time, in conjunction with the Electoral Commission obviously, were of the view that with the change of close of roll that advertising was important to alert people to the need to get on the roll?

Mr Killesteyn —That is correct. It was a deliberate policy of the government of the time. But I should say—and I can confirm the figures—the additional funds that were provided by government in relation to that particular policy change were around $9 million. The rest was spent by the AEC from its own reserves.

CHAIR —I understand that. I suppose the point I am making is that we learn from that experience. In other words, what you are saying is that that was not an efficient way of spending money to get people on the roll in an effective fashion compared to the old-fashioned ways of what we now have. Hindsight is a great thing. I was critical of the early close of the roll, but I cannot say that I am critical of the government or yourselves spending extra money because of the early cut-off. But we now have moved on from that, particularly in relation to the High Court decision and the government’s legislation. Even if it does not pass, the current situation is seven days.

Mr Killesteyn —Indeed. And I guess the point about learning leads me to make the following statements. Firstly, that sort of funding was not available to the AEC in 2010 nor is it, in my view, sustainable. If we have to spend $36 million every election year then that is a significant amount of money over successive elections. But, perhaps more importantly, it does not lead to a permanent or long-lasting improvement in the electoral roll. These gains are quickly dissipated months after the close of rolls as people start to move again and the same challenges are faced by the AEC to encourage people to update their enrolment and to get on the roll in the first place.

CHAIR —This is where the automatic enrolment changes come in, isn’t it?

Mr Killesteyn —Our submission recommends that further enrolment measures, which we believe will not only lead to more long-lasting benefits to the accuracy, integrity and completeness of the electoral roll but also will assist in significantly shifting the timing of enrolment transactions away from the election periods, when workloads are at their most intense, to an earlier part of the overall electoral cycle. The implementation of online update, which the joint standing committee recommended, the government has now supported and parliament has passed, is a significant first step. But our view is that we now need to go further with our recommendation suggesting direct update of the electoral roll based on third party information as well as direct enrolment of new enrollees. We are also recommending that the joint standing committee give further consideration to the potential for online enrolment, moving away from the current paper process for new enrollees. We would be more than happy to provide a further submission on how that might work.

CHAIR —You are going to need to, because I think it is very important for an open and transparent process that it is all out there in the public arena and people can comment on it. I am a great believer in doing these things in that way because they will withstand scrutiny or the shortfalls will come. We will ask a few questions on that shortly.

Mr Killesteyn —In summary, big transaction workloads faced by the AEC—as time marches on those workloads are continuing to increase.

CHAIR —But I think this time it is also fair to say that we dodged a bullet with the High Court’s decision, because that coming at a time early in an election period meant that you were going to be doing transactions that you did not anticipate when the process was first called. So that would have stretched, I would have thought, the resources at every level of the Electoral Commission to do the processing.

Mr Killesteyn —We may have dodged a bullet, but we got hit with some shrapnel. There was a lot of hard work in processing not only the amount of transactions that we had to at the original close of rolls but also the extra 100,000-odd transactions that we processed as a consequence of the High Court action. My point is simply that those large workloads will continue to grow. We have a fixed period of election. We need to try and shift the work out of the election period as much as possible, as well as achieve long-lasting gains to the electoral roll. We think that the recommendations that we have made about the enrolment measures would help us do that.

Moving on to the next slide, this slide focuses on the processing of postal vote applications. What we see from this slide is a growing pre-polling-pay workload that needs to be completed by AEC staff. As the graph demonstrates, in 2010 the AEC manually entered over 800,000 postal vote applications into the system during the election period, with around two-thirds of those generated by the political parties. While the percentage of political party generated postal vote applications is similar between 2007 and 2010, the sheer volume of postal vote applications represents a growing workload that presents risks to electors, who need to get postal votes early enough to be able to cast a ballot before polling day. New legislation, which came into effect 14 January this year and which will facilitate postal vote applications being made online, represents a good first step in improving this service for electors.

We are also recommending in our submission that postal vote applications be returned three days before polling day, as well as political—

CHAIR —Sorry: when you say be returned three days, you mean issued, don’t you?

Mr Dacey —The cut-off for receipt.

CHAIR —The cut-off for receipt.

Mr Dacey —That is right.

Mr Killesteyn —As well, we are recommending that postal vote applications from political parties be returned directly to the AEC. These are both aimed at giving greater surety that postal voters will not experience unnecessary delays in receiving their postal vote packs. I should note that the last point is also a recommendation of the ANAO.

CHAIR —So what you are in effect saying—and correct me if I am wrong—is that political parties can process the applications in the sense of sending them out but, if they do that, the return address has to be the Australian Electoral Commission. They should not come back to the political parties for passing on.

Mr Killesteyn —That is correct. That is what we are saying.

CHAIR —In other words, as a member of parliament, I can provide the service in terms of sending out the applications—as I did at the last election—but the return address should not be my office or the head office of the party but the AEC.

Mr Killesteyn —We are not suggesting in any way that we interfere with the political party process of sending out postal vote applications. But, to ensure that there is greater certainty, they should come directly back to the AEC. One of the issues that we will confront is that as we provide an online facility then potentially people will not need the postal vote applications from political parties. We are currently designing the online postal vote application facility and will be working very closely with the secretariats of all the political parties on that design so that the data that the political parties use for their own purposes—which is allowed under the legislation—is not in any way compromised.

Moving on to the next slide, which focuses on early voting, obviously the growth in postal votes continues. The growth in postal votes is consistent with overall early voting trends. As this slide shows, nearly 2.4 million early votes were received in 2010, which is up from about 1.8 million in 2007, which is a growth of some 500,000 or 28 per cent on the 2007 figures. Of the 2.4 million early votes, nearly one million were pre-poll ordinary votes—that is, pre-poll votes cast in the elector’s home division. The initiative of the 2007 joint standing committee to recommend that home division pre-polls be counted as ordinary votes has, in our view, been a great success. It has allowed many of those votes to be counted on election night with other ordinary votes cast on polling day.

CHAIR —I know that you want us to remove the requirement of a signature in relation to those that vote within their own division. We just had this preliminary discussion in relation to postal votes and online postal votes. For the last election, we actually expanded the criteria under which you could do a pre-poll vote if you were absent from your division; it is a lot more flexible than earlier conditions. I am not wedded to this. I am now at the stage where I cannot see why people should even have that condition attached to them if they want to do a pre-poll vote, which may well relieve postal votes and a whole lot of other things. If we have an election date and people want to vote early, why should they not be allowed to roll up to a pre-poll place early. I am putting it in the public domain so that it can be tested and the doomsayers can tell me what is wrong with the idea of someone having to vote on a particular day when they could vote beforehand, whether it is because of work hours or whatever.

Mr Killesteyn —I think you can approach that debate from a number of perspectives. You can approach from the perspective of administrative issues around taking large numbers of votes. From our perspective, if the votes are spread across a longer period then obviously the intensity of election day and the intensity of all the administrative tasks associated with that can be spread across a larger period.

CHAIR —I am just asking the question.

Mr Killesteyn —What I am also indicating is the philosophical perspective, where the tradition of the nation coming together on a single day to cast its preference for a particular government is one that has been part of our democracy for many centuries. So, the notion of a polling period as distinct from a polling day is quite significant philosophical change.

CHAIR —In response to that, and being the devil’s advocate, what is the percentage of people who cast their vote before election day? Pre-poll and postal votes are over 20 per cent aren’t they?

Mr Killesteyn —It is about 25 per cent.

CHAIR —The reason I am putting it in is just to test the waters. I think it was successful at the last election to loosen it. There were no integrity questions. With a mobile electorate, I am interested in the arguments for and against and whether it is worth exploring.

Mr Killesteyn —If there is a significant increase in early voting and most of those early votes were still in declaration envelopes our concern would then be the post-election process that we would need to go through with them.

CHAIR —I am talking about every ordinary poll within the division. The reason I say that is that people are not being held to the reasons for which they are exercising a pre-poll vote. The practicality and the reality is that if they rock up they get a pre-poll vote whether they are going to be in their electorate or not. I do not believe in having red tape restrictions that do not really apply. What polling official at present question a person on the reasons for their pre-poll vote? I am talking about votes within the division; I am not talking externally.

Mr Dacey —We accept the reasons and if the person signs a declaration they are entitled to an early vote, currently.

CHAIR —You accept it on the fact of it. Anyway, it is out there for people to provide their views on it.

—I think there is a philosophical debate as well as a practical debate. We can point to many examples of countries overseas where there is a polling period as distinct from a polling day. Is it time for Australia to go that path? That is ultimately a matter for the community and parliament to debate. There are questions that at some point when the electorate is so large, at some time in the future, that I may be beyond the capacity of even the Australian Electoral Commission to conduct a ballot on a single day. It is a debate that is perhaps timely and maybe a debate for consideration for future—

CHAIR —The converse argument is that it is not onerous in terms of the conditions for pre-poll voting. Most people can pick it up.

Mr Killesteyn —Going to your point about the need for the declaration, we are now of the view that the JSCEM can feel confident to proceed now on the basis of the 2010 experience in recommending the removal of the requirement for voters casting a pre-poll vote in their home division to make a declaration.

CHAIR —Can you flesh that out a bit at this point to save me coming back to it. Did you do some studies of the 2010—

Mr Killesteyn —Essentially it is on the basis of the experience that the pre-poll home division process did not cause us any issues. There were no examples that we have detected at this point of multiple voting by those people who were casting pre-poll home division votes. Our view is that this is a much needed reform and we can now complete it.

Turning now to slide 7, at the same time it is worth noting that but for the change in treating home division pre-polls as ordinary votes. The number of declaration votes would have been nearly 3½ million but for that change. That is one indicator of success of the measure. If we had not made that change, such large numbers of declaration envelopes represent, in my view and in the view of the AEC, a high risk to good electoral practice, because it increases the time associated with including such votes in the count, with consequential delays in the provision of election results, and the manual handling of such votes. The more votes you have to manually handle, the more risk there is that something will go wrong.

It is also inevitable in our view that, notwithstanding that there were fewer declaration votes in 2010 than there were in 2007 because of the change for home division pre-polls, the growth in early voting is likely to start to increase again at the next election. That is consistent with all of the trends.

Turning to the next slide, our submission is therefore suggesting that, in anticipation of this future growth in declaration voting, we start now to explore measures to limit the number of votes in declaration envelopes, possibly focused on absentee votes, which, as you can see in this slide, are the next biggest proportion of declaration envelopes. The numbers are staying roughly the same from year to year, but in terms of looking at the inevitable growth in declaration envelopes that will start again, we think it is time to start looking at whether absentee votes might be able to be cast without a declaration.

CHAIR —What is required to allow that to happen?

Mr Killesteyn —It is connected to our implementation. Again, one of the great outcomes of the joint standing committee of 2007 was the provisions to allow certified lists to be produced electronically.

CHAIR —You are saying that a certified list can virtually turn it into an ordinary vote.

Mr Killesteyn —Indeed. You can have an electronic certified list on one of these items that I am indicating—even a smaller footprint than these—anywhere in Australia so that anybody who fronts up at any polling station could be marked off an electronic roll and the vote cast as an absent ordinary. It would still go in an envelope, because it has to be transported to the relevant division for counting, but it would not require the declaration.

CHAIR —But the integrity would be maintained because of the nature of the roll you are using. There could not be multiple absentee votes because that would show up on the roll—would it, or not? I am interested in overcoming this furphy that people go from polling booth to polling booth and vote absentee and you have all these multiple votes.

Mr Killesteyn —It is an interesting debate. I know Mrs Bishop has put this issue on the table. Technically, it is possible. There is no doubt that we could have an electronic certified list in every single polling station around Australia. There is no doubt, with the technology that is available, that that could be linked back to a central database and the electoral roll updated almost instantly as people’s names are marked off the roll. Technically, it is possible. The cost would be rather large, however. It would be an extreme cost that I am not sure the government would be willing to invest in. However, in saying that, I will also say that it is much more feasible and perhaps sensible to look at where there are potential problems with close seats. You might roll out that sort of technology in close seats to give you much more confidence about multiple votes not being cast in close seats, and that would limit the technology investment that would need to be made.

Mr Dacey —Perhaps I could give you another example of where it might work in a smaller geographic unit. In the ACT, for example, with the divisions of Fraser and Canberra, if all the polling places had electronic certified lists and you established the entitlement to vote as the people came in, not post, people could vote anywhere across the ACT and you would do away with all those absent votes from Fraser to Canberra and vice versa.

CHAIR —I am interested in how this would dovetail into automatic enrolment, where people can roll up and vote on the day itself. I am just putting that out there.

Mr Killesteyn —Again, anything is possible with technology. The models for election day enrolment, as far as we were concerned, would be that a person would provide some form of evidence of identity. The roll would be relevant in that particular polling station where they had just voted.

CHAIR —But you would also be giving them a declaration vote. They would not be getting a vote without a signature.

Mr Killesteyn —The model that has been applied in both Victoria and New South Wales is that they would be provisional votes, subject to further confirmation of the eligibility of the person.

Senator RYAN —Firstly, I want to flag a concern and see if it is a concern you share. I have heard from other witnesses in this inquiry that in close seats there can somehow be a different sort of system in place. For example, if we only wanted to invest a certain amount in technology in a seat, we would only invest in, say, automatic updating of a central list from an iPad or a laptop in close seats, because of the cost. I am concerned that to start applying different standards of investment in maintenance issues or integrity issues in marginal seats versus other seats would pose a real risk. It is a philosophical and in-principle objection I have, and I would be interested to know whether that is a concern you share. I would not want the voters in Maribyrnong to be treated differently in any way—through automatic updating of central registers or the way we treat absentee ballots—to the people of McEwen.

Mr Killesteyn —Those differences already exist between state and federal elections. In a state election in Tasmania you can have a voter who is—

Senator RYAN —That is true but, with all due respect, what happens at a state election is not within the purview of this parliament.

Mr Killesteyn —What I am pointing to are practical issues. We have 7,760 polling stations in a federal election. Each of those polling stations contains a number of issuing points—let us say 10 issuing points per polling station. That brings me up to 77,000 pieces of equipment that need to be purchased, stored and then distributed during an election if you want absolute equity and uniformity across the electorate.

Senator RYAN —I appreciate all that; it would be a huge investment. My concern is that if we cannot do it everywhere it is a worry doing it anywhere.

Mr Killesteyn —I do not share that concern. It seems to me that to the extent that we can progressively improve elector services by using technology to facilitate the process of marking off the roll and potentially reducing the number of absentee votes that go into declaration envelopes, then we should be doing that.

Senator RYAN —But if you were doing it on that basis, wouldn’t you do it on the basis of where the volume of absentee ballots is, rather than choosing on the basis of the narrowness of the margin in the seat?

Mr Killesteyn —That could be one criterion that we use. There are multiple criteria. What I was responding to was some of the concerns about multiple voting and the use of technology to try to give greater assurance to you as well as to the electorate that technology can be used to minimise the risk of multiple voting. One approach to that is to look at close seats and apply the technology in those seats. It is an approach which has been used in Western Australia, most recently in one of their by-elections.

Senator RYAN —I appreciate the frankness of your answer. I suppose I wanted to flag my concern that Maribyrnong and McEwen should be treated the same, not differently. The voters of Maribyrnong have just as much legitimate interest in whether someone is multiple voting as the voters of McEwen do. One of the issues you raised there was the handling issue—every time we handle envelopes there is a risk. If we are still putting all these votes in envelopes, albeit a different envelope because they are not a declaration vote, we have not really mitigated a great deal of the risk, have we? We are still actually double-handling all these votes.

Mr Killesteyn —What you do is take out one part of the process. They have to be moved  from one division to another division, but at the point of receipt the amount of handling that goes into the further scrutiny process is eliminated.

Senator RYAN —They have already been packed and shipped and counted; we are not just matching signatures?

Mr Killesteyn —That is true.

CHAIR —But the issue of not matching signatures is more about quickening up the count and not about those other issues.

Senator RYAN —I appreciate that.

Mr Dacey —One of the other issues in relation to declaration voting is the risk of error, where the elector actually has to fill in the declaration envelope rather than having an ordinary vote and it goes into an envelope marked with that division name. Declaration votes have a higher percentage of rejection, obviously, because of polling official error, elector error or whatever.

CHAIR —Can I just throw into the ring this idea of selecting seats. The last election was a very interesting election—there were results that were never foretold. In my seat the Liberal Party are now bleating and blowing up that not enough resources were put into the seat—given the swing that occurred they could have won it. Banks was never on the radar. I can remember in the past the Electoral Commission doing an assessment as to the last two candidates remaining, and in the contest something else happened. I am a bit concerned about picking and choosing, so to speak. I put that on the table because I am attracted to a consistency argument but I also want to increase the count. I just put that there; there is no answer.

Mr Killesteyn —My position is simply that there is no evidence anywhere in 2010, 2007 or any other election, in the context of multiple voting in particular, that even the incidence of multiple voting would have changed the outcome in any of the seats. There is no evidence at all. However, if there is a perception about the possibility of multiple voting changing the outcome in a particular seat, then one way of dealing with that perception is to apply technology in a way which is available to us now. I am not suggesting that it is a must because we are trying to prevent a problem with a seat—there is no evidence of that—but if there is a perception we can deal with it.

CHAIR —My problem is that I do not want to use a sledgehammer to crack a walnut. That has happened in the past. The point is that if there is electoral fraud, if there is multiple voting and it is proved and the result is close, there are provisions to allow challenges to the result and for the result to be overturned. That is the reality of it.

Mr Killesteyn —And our existing processes are already very well refined to be able to detect multiple voting before the results are declared.

CHAIR —My concern is that I do not want to spend all this money if it is not really necessary. That is why I want evidence from you in a whole range of areas, to put the evidence on the table so that we can argue on the facts and not on the myths.

Senator RYAN —I agree with what you said there; we do have good procedures in place. I have a concern, which I mentioned in the minority report in respect of automatic enrolment and which we will come to later. It is that every time the commission makes a choice, particularly if it is recommending a policy change or picking electorates that might have different technological means, I genuinely believe that if something goes wrong it puts the commission in a position where it is not going to necessarily be—where it has historically been—completely beyond any issue of one side thinking the ball has bounced the right way or not. I am using my words very carefully here. But the AEC has a very impressive record, which other countries look to, of being completely above the political fray. When we talk about choices being made about recommendations on the Electoral Act or about different electorates being managed in different ways my fear is that we are getting into territory where that will become slightly more contestable.

Mr Killesteyn —My point, and I am not disagreeing with you, is that if there is a view that technology can help the perception, then it is open for us to take that role. Guessing which the close seats are is always going to be a guess, in some sense. I am confident in our current procedures, but we can try to accommodate what the parliament wants.

CHAIR —Would you like to continue with the slides?

Mr Killesteyn —The next slide talks about provisional votes in particular. For provisional votes our submission recommends that the requirement to provide evidence of identity be repealed, primarily because its impact is random. Correctly enrolled voters who cast provisional votes and whose provisional votes are rejected at preliminary scrutiny only because they did not provide evidence of identity would, in general terms, have had their vote counted without providing evidence of identity if they had cast an absent postal or pre-poll vote rather than a provisional vote.

CHAIR —This is the part of your submission that basically says that these provisional voters were actually on the rolls; they were not identified as on the rolls on the day of the election, so they filled out provisional votes which then precluded them from being included in the count because they did not come back and provide proof of identity.

Mr Killesteyn —That is part of the argument. If you look at—

CHAIR —Sorry; I had the wrong page. I had page nine.

Mr Killesteyn —It is part of the argument. If you look at the darker shading, they are provisional voters who, out of the 27,000 in 2007 and 28,000 in 2010, did not provide POI at the time that they voted and did not then provide POI by the following Friday. They were, however, people who were subsequently found by us to be on the roll. In other words, the polling official potentially made an error. They could not be found on the roll; they were therefore asked to cast a provisional vote and, because of our error—

CHAIR —They lost their vote.

Mr Killesteyn —They lost their vote.

Senator RYAN —Excuse me if this is a silly question. Have you checked how many of these physically were on the roll in the polling place where they cast the ballot? Can you tell if it was a physical misreading or a misprint of the roll in some way?

Mr Killesteyn —No, it is a physical misreading. It is probably associated with language issues; probably associated with non-English-speaking names, spelling mistakes, those sorts of things. Our point is simply that these people should not have been disenfranchised and we think that, as a minimum, where a person casts a provisional vote and they are subsequently found to be on the roll and they do not also subsequently provide EOI that we should be able to admit the vote to scrutiny.

CHAIR —Under the old rule they would have been readmitted.

Mr Killesteyn —Yes. However, we are also suggesting that overall that section for the requirement of POI be removed, because, as I said, it acts randomly. If a person casts an absent vote, then their vote would have been counted. I should add that this is a position that the AEC has taken publicly for some time. The AEC has had this position at least since early 2008 from a public position. It was included in evidence that we provided to Senate estimates back in February of 2008. It was also included in our submission on the 2007 election. This is not an issue that I, as the commissioner, have suddenly thought was important to put forward on the agenda. It has been our public position for many years.

CHAIR —The truth is, Mr Killesteyn, there is an availability for these votes to be checked by way of signature. The previous way of doing it was that the signature on the declaration vote could be compared with the signature that the Electoral Commission already had by way of an enrolment form or a transfer of enrolment form. If there was any doubt, a scrutineer in that election could have challenged it and, if in doubt, the vote would not have been counted. So there was an identifying signature which could form the basis of proof of identity.

Mr Killesteyn —That is correct—

CHAIR —These are not votes that have gone in an envelope without a signature. There is a signature attached to them and that signature purports to be the signature of the voter. If someone is wary about whether it is the person, the vote can be checked during the checking period.

Mr Killesteyn —There is a process by which those votes can be admitted with absolute confidence about the identity of the elector.

CHAIR —There were plenty of them that were removed from the count or did not go into the count because of doubt over the signature under the previous rules.

Mr Killesteyn —I just wanted to add—and this goes to your concern, Senator Ryan, about electors being treated equitably—that here is an example, on the basis of equity, where two different electors are treated differently simply on the basis of one casting a provisional vote and one casting an absent vote.

Senator RYAN —This is what I wanted to clarify. I joined the Senate in July 2008, so my apologies if I missed it. Basically if someone walked over the road, crossed an electoral boundary and voted in a polling place that was in a neighbouring electorate by casting an absentee ballot, they would be treated differently.

Mr Killesteyn —That is right.

Mr Dacey —They would not have had to provide POI.

CHAIR —That is the point. They do not have to provide proof of identity for an absentee vote.

Mr Dacey —Correct.

CHAIR —The signature suffices, and the signature actually suffices in relation to postal votes, if I am not wrong.

Mr Killesteyn —That is correct.

CHAIR —We have a higher threshold that is knocking people out.

Senator RYAN —For provisionals cast inside an electorate.

Mr Killesteyn —So our suggestion is to repeal the provision entirely, but if that is not agreed then we think as a minimum we should be dealing with these electors who we find to be on the roll.

Senator RYAN —I am not proposing this but you could achieve consistency by heading the other way and applying those standards elsewhere.

Mr Killesteyn —We would not be recommending that.

CHAIR —If you did, the history is such that more people would get knocked out.

Mr Killesteyn —That is correct.

Senator RYAN —Why would you not recommend that?

Mr Killesteyn —Precisely because we would find that with the requirement to submit evidence of identity by the end of the following week, the history is that people do not do it. So a lot of those votes are lost.

Senator RYAN —So you would have more of the purple box. We are not worried about the blue box because that is the consistency issue and consistency can be dealt with by a set of rules that applies equally, but the purple box, which is those due to AEC error—and that is actually an impressively low rate—would go up.

Mr Killesteyn —Not the error rate, the rejection rate.

Senator RYAN —Yes, but that is the rejection rate that started with an error at the polling place.

CHAIR —But if you required proof of identity as an addition on an absentee vote then that would knock you out. That is the point.

Mr Dacey —Yes. There are 832,000 absentee votes. If you applied the same sorts of standards—

Senator RYAN —I am just making the point that consistency does not necessarily mean that there is always one proposal.

Mr Killesteyn —But you then also load the requirement on the Australian Electoral Commission to check all of those.

Senator RYAN —I appreciate that.

Mr Killesteyn —I will move on to the next slide. To close the loop on election related workloads. This graph shows the total votes counted and preliminary two-candidate preferred count conducted on election night. As you can see, more than 11 million House of Representatives votes were counted on election night in 2010, some 600,000 more than 2007. While our submission makes no recommendation in this regard, there will come a time as voter numbers continue to grow when physical limitations will prevent completion of the preliminary count of ordinary votes on election night. It may be appropriate to commence investigation of machine assisted counting of House of Representatives votes in the same way we do for Senate below-the-line votes. I note here that that growth is inevitable. If you compare 2004 to 2010, we now have one million more voters on the electoral roll. So I could expect by 2016 there would probably be another million-plus voters on the electoral roll.

Senator RYAN —Can I ask what you mean by machine assisted counting? I know that for the Senate it is double-data entry. What do you mean by machine assisted counting for the House of Representatives ballots?

Mr Dacey —One example is where they could be machine-read through scanners—House of Representatives ballot practice. There are some jurisdictions—I do not want to refer particularly to the US model—that have scanners in the lid of the ballot box and the vote is actually read as it goes into the ballot box. But there is still a hard copy of the ballot paper in the ballot box as an audit trail so that is one of the possibilities.

Mr Killesteyn —I am not suggesting electronic voting.

Mr Dacey —No, of course not.

CHAIR —I suppose we could always delay the vote of the Senate on the night—they are not declared on the night.

Mr Dacey —We could count the Senate votes in the next week. Just one of the advantages of scanning is that all the rechecks that we do post election on votes from polling places, the distribution of preferences and all those tasks that we currently perform could be done by the system. It would be a matter of a significant change for parties and scrutineers, but that change was accepted for the Senate some 13 or 14 years ago.

Mr Killesteyn —Senator, I am being deliberately general in the description because I do not want to give any impression of a particular technology. I am putting up a suggestion that we start to explore what is available. I am not necessarily suggesting that we need to have this in place by the next election, but I think at some point we need to spend time looking at this issue and finding ways in which we not only satisfy the issue of the volume that has to be done in the count but also recognise the interest of the political parties, particularly the scrutiny process, which is one which is important to preserve.

CHAIR —Can I suggest in relation to that and other suggestions you have got, which do not require recommendations for the next election, that you look at an issues paper that we can maybe release to those interested in electoral reform and electoral processes. I think all of these things should be put out there well in advance of when they may be necessary but, by doing it through an issues paper—I am sure that the committee would be happy to assist in any way it can to get it on the table—we can get to the discussions, so we can have a long lead time. I think the defence of that suggestion is that people do not feel as though they have been ambushed or it has been put on them late in a particular process.

Mr Killesteyn —That is a very good suggestion.

Senator RYAN —Is it easy for you to get the number of people—the number of officials you have, temporary and permanent—who were employed on polling day, working from 2004  to 2010?

Mr Killesteyn —Yes.

Senator RYAN —I wonder because you had an 850,000 increase in the number of votes being cast on the day. That meant you had to scale up in terms of employment numbers.

Mr Dacey —There are some pockets where recruitment is a little more difficult but, in the end, we usually manage to get the number of people we want.

Mr SOMLYAY —What does the commission do to keep up with international best practice?

Mr Killesteyn —In relation to counting?

Mr SOMLYAY —And technology?

Mr Killesteyn —Particularly in relation to technology, we usually conduct field trips overseas. There have been a number of field trips over the last couple of years to look at different approaches taken. We have a four-country conference between New Zealand, Canada and the UK, which meets once every two years, where we compare approaches to conducting elections. There are plenty of information technology suppliers who are constantly knocking on our door. We have an open-door policy. We certainly look at things but, at the same time, they are generally very expensive. So my approach here is simply to throw out the general issue and I think your suggestion, Chair, of an issues paper is a good one, which will also pick up some of the international practices.

Mr Killesteyn —The final slide moves away from election period activities to the broader indicators of electoral participation in Australia. I think this slide clearly illustrates that the long-term trend of declining electoral participation persists in the 2010 election. Taken together, the number of people who are not on the roll, the number of people on the roll who did not vote and then the number of people who voted but for one reason or another had their vote counted as informal, we see around 3 million people who are not participating in our democracy.

Senator RYAN —It is more a reflection on us than it is on you, Mr Killesteyn.

Mr Killesteyn —The causes of that are open to debate. The AEC will continue to focus on what we can do to get people on the roll and to get them motivated sufficiently through advertising and other mechanisms—particularly our education processes— to participate, and then we will do as much as we can to eliminate the unintentional informality. But I agree with you, Senator, that there are many other complicated issues around this particular set of indicators. I simply leave that for the debate of the commission and the joint standing committee at some appropriate time. Thank you, Chair. That is the end of my presentation.

CHAIR —The first thing I want to do is to ask for some further information and clarify some stuff arising from some earlier evidence. We heard from Antony Green on Wednesday in relation to some options as far as lowering the formal vote is concerned. He put three options to us: one was optional preferential, which I do not think it on anyone’s radar; the second was what is called ‘progressive informality’; and the third was in relation to the South Australian system of voting, which is registered preference tickets. I understand that you are doing an analysis of the informal voting and that that will be part of a supplementary submission. I would like from you an analysis of, if the South Australian position had applied to the Commonwealth, how many votes would have been captured by that in terms of the count. I am also interested, so that the committee can compare, in your doing that also for progressive informality. I am interested in the difference between them. We work off statistics—

Mr Killesteyn —That is fine. The informality survey and its report should be public by the end of this month, so we will include that in a supplementary submission to the committee. On the basis of the survey, we should be able to do an analysis against the South Australian model and Mr Green’s progressive informality model. We may have to have some discussions with Mr Green about how his model applies in particular instances in order to be able to give you a reasonable assessment of what is the informality reduction would have been.

CHAIR —I took his model to be about a situation where you had a number of options in the square but they did not follow the ticket, like a Langer vote. A Langer vote would be accepted under progressive informality to the point of informality, but it would not be accepted under the ticket voting system of South Australia because it is not similar to the ticket. But you are right—you need to talk to him. I just want us to have accurate figures so that we know what we are talking about. Is that not a problem, Mr Maley, in relation to that?

Mr Maley —I think one point that needs to be mentioned is that, in relation to the postulation of the application of the South Australian ticket voting system at the federal level, one needs to make assumptions about what would have been the behaviour of the candidates in lodging tickets and also the behaviour of the voters. One can probably plausibly assume that all major party candidates would lodge tickets, and therefore one could say that any No. 1 vote which had been cast for those candidates would have been saved by the tickets. But, if, for example, some of the voters had expressed further preferences that were not consistent with the ticket, they would not be saved. There may be other candidates who had received No. 1 votes but who had not lodged a ticket, and therefore those votes would not be saved.

CHAIR —Yes, if a ticket is not lodged they are not saved, but the assumption I would be making is: I had a ticket which was what was handed out on election day and my opponent had a ticket; that is the ticket you would lodge. That is the assumption upon which I would ask you to proceed. We know that you can lodge two tickets. Technically, if you were the Democrats you could have split your ticket, but I am asking you, if possible, to do the analysis of the tickets that were actually lodged by the respective candidates.

Mr Maley —We will have to make some plausible assumptions about certain—

CHAIR —That is okay. As long as those assumptions are detailed, we can work off them. I made a comment on Wednesday which got a run in today’s newspapers, on multiple votes. I pointed out that a lot of error involved aged people and people in nursing homes due to confusion, poor comprehension or whatever. We had a table to that effect in our last report, table 2.3 on page 18, which was on the basis of something the commission submitted to us. It stated that ‘of those in the aged category, 98 per cent were aged 70 or over’. Is the commission doing a similar table in relation to multiple votes that is able to extract the number in nursing homes and so on?

Mr Killesteyn —Yes. The process for pursuing all apparent and alleged multiple voters is still continuing, so we have not completed that but at the end of that process we will do the table. However, the preliminary information that we are getting suggests that those patterns that we have encountered in previous elections are almost exactly the same.

CHAIR —Would you just repeat what those patterns are, if you can remember them.

Mr Killesteyn —Essentially, most dual voting is in relation to people either with language difficulties or who are elderly who get confused about whether they have voted or not. I think during the last joint standing committee inquiry I made a commitment that we would seek to eliminate that level of confusion by providing people, particularly elderly people in aged-care facilities and so forth, with a letter that they could present to their family members. We did that. I do not know whether that has had a significant impact, however, but we will look at that again. In terms of dual votes, as I said, it is primarily through confusion.

However, we do have a number of instances—over 100, which I think I have reported to the committee as well—of multiple voting. These are instances of three or more votes. We continue to work through the investigation of those cases and as necessary refer them to the Australian Federal Police for investigation as to whether any action can be taken by way of prosecution.

CHAIR —One of the criticisms in the past, I think, has been the police have been reluctant to prosecute in these cases because the argument is an evidentiary argument. Obviously I am not a fan of strict liability offences per se, particularly where it involves imprisonment, but it would seem to me that multiple voting is something that we need to send a message on. I have to say I am very open to the suggestion of strict liability with a civil penalty if, prima facie, people have been shown to do multiple votes. A defence would be available to them if they went to court.

I am also concerned about the suggestion—and I put this on the record—that police are not interested in prosecuting this stuff because it is not really in the ‘serious’ category, in the sense that their workload in relation to other criminal matters does not give them the resources to spend on this. I am just saying that I do not think anyone on the committee would accept that it has to stay there. We need to protect our electoral system, and that might require some changes. I am open to suggestions from the commission, or reasonable suggestions from reasonable persons, as to how we can address this problem if it is there.

Mr Killesteyn —I think it is there. What the AEC faces is a relatively low pecuniary penalty such that, when the AFP examine that level of penalty against other matters they are pursuing, we will always rate as a very low priority.

CHAIR —How do you fine people who do not vote?

Mr Dacey —It is an administrative penalty for failure to vote, and if you choose not to pay the administrative penalty it can proceed to court.

CHAIR —Why do we not look at administrative penalties? I am raising this to test it because, to me, doing nothing—it is about confidence in the system. I would rather an administrative penalty in these matters which is substantial for gradation of the offence, and if people want to contest it they can take to court.

Mr Killesteyn —Indeed.

CHAIR —I am just raising it as an option.

Senator RYAN —Voting twice is more of an offence than not turning up to vote.

CHAIR —I understand, but what it might mean is that you have an administrative penalty that carries a bit more than $50. I am raising it as an option. I am raising these issues because it is about confidence in the system. If you are saying to us that, over a period of time, you have referred a number of matters to the Federal Police on which you have established a brief but they are not getting to court, I am asking whether we should look at changing the system. If you set administrative penalties for failing to vote, I ask why we cannot have a bigger administrative penalty for multiple voting, which at least gets it into the system?

Mr Pirani —I would just remind the committee that the relevant offence we are talking about is in section 339. There are two penalties: one in 339 1A, which is merely that a person is guilty of an offence if they vote more than once, and one in 339 1B, which makes an offence a strict liability, and the current penalty is 10 penalty units. A penalty unit, of course, is currently $110. That is for that offence. Then there is the more serious offence, which is in section 339 1C, where, if there is evidence that the person intentionally voted more than once in the same election, it is 60 penalty units or imprisonment for 12 months, or both.

CHAIR —There is a penalty of imprisonment there; you should have intent. I am not suggesting to you that you have an administrative—

Mr Pirani —We did make a similar submission to the 2007 committee inquiry; it was submission number 169.19. We had an attachment there setting out the figures from 1998, 2001, 2004 and 2007.

CHAIR —Did you offer the option of an administrative penalty with a substantial fine in that submission?

Mr Pirani —That was in the earlier submission, in 2004, that the AEC made a submission to that effect.

Mr Killesteyn —We are looking at voting but also at enrolment, and we are of the view that non-enrolment should also attract an administrative penalty.

CHAIR —I have a different view to you on that, because of language difficulties and other difficulties it is not as simple.

Mr Killesteyn —Yes. Do not get me wrong, I am not suggesting that it would be a strict liability, either. There would be some discretion open to the commission in the same way that there is a discretion to the tax office in terms of its administrative penalties. Matters of policy will be publicly exposed in relation to how those penalties would be applied. I will give you one example in relation to the most recent election. We had one case in one state where there was apparently a casting of 10 ordinary votes. The polling places at which the votes were cast are all within a single electorate, all within a relatively short distance from each other and all within a relatively short distance of the enrolled address of the apparent multiple voter.

CHAIR —And you have identified that person?

Mr Killesteyn —Yes. I am not giving you any information on that, but prior to—

CHAIR —I do not want it.

Mr Killesteyn —No, good.

Mr Pirani —Yes, we have identified who it is.

CHAIR —It is privacy; I would not ask you for it.

Mr Killesteyn —Prima facie, we have evidence here that should be pursued. That matter cannot get over the priorities within AFP to take it further.

CHAIR —But isn’t that where the suggestion that I just made comes in? Does that at least take it somewhere? From what you are telling me, that person would have committed nine separate offences—

Mr Killesteyn —Allegedly.

CHAIR —Yes, that person would have allegedly committed nine separate offences, if you can prove it, which you could then administratively deal with him or her on. If they contest it, they take it elsewhere. So you understand where I am coming from: I have a defence background not a prosecutor’s background; but as a legislator, I have an obligation to maintain integrity and confidence in the system. And, in relation to this, I have now heard you pass the ball a number of times, and the ball is going nowhere. The try line is wide open, but they do not have the capacity to get over the line.

Mr Killesteyn —And that is only the first line of defence, because even if we get over the AFP—and I should also note that in other cases the AFP have taken on the referral—

CHAIR —I understand that.

Mr Killesteyn —and that is continuing to work. But, even if we get through the AFP, there is also the DPP and whether it believes that the prosecution is in the public interest. Again, given the smallness of the penalties, it is hard to get priority.

CHAIR —And that is the whole point. That is why the ministry penalty system probably is more appropriate. If I were defending the person, I would imagine that in most instances the fine would be minimal, and I would very surprised if convictions were recorded, unless it was in the category of what you just said. That is why I am throwing the ball back to you to say: we should look at an alternative way of doing this, a way that gives it credence. It may well involve, Mr Killesteyn, your officers having a specialist division within the AEC who take these matters on. I am sure that there are people there with legal qualifications.

Mr Killesteyn —As Mr Pirani has said, these issues were part of our submission to the joint standing committee—

CHAIR —In 2004?

Mr Killesteyn —Yes.

CHAIR —I am asking—

Mr Killesteyn —But also of the Lindsay review.

CHAIR —In Lindsay we made a recommendation; we are awaiting government.

Mr Killesteyn —So are we.

CHAIR —We have made a recommendation. I will follow that through. We do not want you to be a toothless tiger. Senator Ryan, would you like to go first, and I will go after you.

Senator RYAN —I would like to continue our discussions around your views on automatic enrolment, Mr Killesteyn. You are again proposing—recommending—that we move to adopt a legislative model more akin or to match New South Wales’s and Victoria’s. As you know, we had a minority report on this issue when this bill was proposed. There are a number of issues raised in that, and this is the first opportunity I have had to explore those with you. I am not sure if the New South Wales Electoral Commission has at the moment published criteria around its use of data sources. It did not at the time we had the inquiry into that bill. Do you know if it has published such criteria yet?

Mr Killesteyn —As far as I am aware, it has not.

Senator RYAN —But it has chosen to use certain forms of data as I understand—New South Wales school leavers’ data or something like that?

Mr Killesteyn —Yes. The New South Wales Electoral Commission uses a range of data, including Roads and Traffic Authority and school education information. It checks against citizenship records held by the Department of Immigration and Citizenship to ensure that the person is eligible from a citizenship perspective. So it has a range of data sources as well as the Registry of Births, Deaths and Marriages—

Senator RYAN —Which it has state access to.

Mr Killesteyn —Yes.

Senator RYAN —If we adopted the approach that you are recommending, as I understand it, when the New South Wales Electoral Commission put someone on the roll, would they be automatically added to the joint roll and therefore be able to vote in Commonwealth elections?

Mr Killesteyn —No. At the moment I have no legislative authority.

Senator RYAN —No, but you are recommending that we move to a situation like that. I am asking if your recommendation was delivered. I understand that you cannot do that currently and I have seen the numbers about the mismatching—that is, there are fewer people subsequently joining the Commonwealth roll. If we went down this path would that mean that people added by the New South Wales Electoral Commission would be added to the Commonwealth roles?

Mr Killesteyn —No, what we are recommending is a model of direct enrolment which, from a process perspective, looks similar to the New South Wales Electoral Commission but would have all of the checks and balances that I think you are referring to. For example, it would include that the data sources that we use are public and that the data sources to be used would be approved by the minister before it is used.

Senator RYAN —That is one of the issues that has not been flagged to me, but please go on.

Mr Killesteyn —All I am saying is that your concern about the transparency of the process is one that we agree with and we would be looking to adopt a model which is simply around direct enrolment but is based on a highly transparent and cautious approach to whether people are directly enrolled.

CHAIR —So what would you do then? If New South Wales gets on the roll, you would then inquire into those? They would not automatically go onto the Commonwealth’s roll? That is not what you are seeking.

Mr Killesteyn —There are various models that you can adopt here. One model is that you could automatically put an elector who was enrolled in the New South Wales, Victorian or Queensland electoral commissions automatically on the Commonwealth roll. If you did that you would be doing it on the basis of being satisfied about the process as well as the data sources that are being used. That is our point: we would want to be satisfied in our own right about the data sources and not simply take the information provided by another agency.

Senator RYAN —If a bill proposed that the Commonwealth roll would reflect the automatic enrolment activities of the New South Wales or Victorian electoral commissions without individual data matching being done by the AEC, you would not be supportive of that?

Mr Killesteyn —I do not think that would be the way you would frame the legislation. I think you would frame the legislation to give authority to the Australian Electoral Commission to directly enrol people on the basis of third-party data. You would not frame the legislation to say that the AEC can simply pick up the New South Wales stuff or the Victorian stuff or whoever.

Senator RYAN —That is one important point.

Mr Killesteyn —So that gives us the basis for then still having the discretion about the credibility of the information and subjecting it—

Senator RYAN —That would then also mean that you could still have mismatched rolls if you, for example, were not happy with a data source; whereas a Victorian, Queensland or New South Wales electoral commission was happy with a data source. That is possible, isn’t it?

Mr Killesteyn —That is possible, yes. But the question is the extent of the discrepancy. At the moment, if the Commonwealth does not act at all then, in my view, within a matter of years the discrepancy could be hundreds of thousands.

Senator RYAN —With all due respect, the discrepancy is not my concern. I personally believe the AEC is placing too great a concern on discrepancy. The integrity of the roll is the first and foremost issue that will ensure as wide a franchise as possible.

Mr Killesteyn —I will respond to that point. We do not disagree with the notion of the integrity of the roll. It is an important hallmark of the way in which we manage the roll. The other perspective is from the electors’ perspective.

Senator RYAN —No-one is denied a vote here; they are just not filling out a form. Let’s be honest.

Mr Killesteyn —What I am suggesting is that to the extent that there are discrepancies between the roll and an elector believing that because they have been enrolled by one arm of government that they are automatically enrolled in the other arm of government then you have an elector issue. The elector is confused. I can turn to one bit of evidence here. If you look at the number of provisional votes that were taken both in 2010 and 2007, 203,000 provisional votes were taken in this election. Of that number that were rejected, 90,000 people were rejected because they were not on the roll—that is, they turned up at a polling place because they believed they were on the roll. I would suggest that with the confusion of the electorates, through the discrepancy, that that number could increase.

Senator RYAN —This is an issue that divides participants in Australian politics. I will put to you a frank view that in the end there is a burden on the individual to enrol under the act. There is no lack of attempt by the AEC to contact people by written letters. I appreciate that we need to update those. We have had conversations about that before, whether or not the purple people eater—as I understand some of our witnesses have called it—is the most enfranchising or easy way to go about it. I take those points; but, please Mr Killesteyn, do not try and put the view that somehow anyone in New South Wales who is automatically enrolled is disenfranchised at the Commonwealth level, because you do not give them that impression. You try to contact them; in fact you contact them on multiple occasions and the New South Wales Electoral Commission does not tell them that. At some point, surely, there is a burden upon the individual to comply with the law?

Mr Killesteyn —I am not suggesting that simply because of the potential discrepancy between the rolls that there is a lack of enfranchisement. I am suggesting that there is potential elector confusion, and that is an issue that I have to be worried about.

This is a philosophical issue, I grant you, but I should also add that the act is quite clear. The act not only imposes an obligation on a person to enrol it also confers a right—an entitlement.

Senator RYAN —Quite right.

Mr Killesteyn —From my perspective as an administrator faced with the sorts of figures that we are now being confronted with for the lack of electoral participation, I not only have a duty as an administrator enforcing the act to enforce the obligation but I also have a duty to facilitate the entitlement. All of the processes we are putting out there for debate are both about enforcement and facilitation.

Senator RYAN —Then let us be honest about what it is, and stop using Orwellian terms like ‘smart roll’ and ‘direct enrolment’: it is a way to use data sources to compel enrolment where people are not complying with the law. That is what is happening, and—

Mr Killesteyn —It could also be argued philosophically that it is a way to facilitate enrolment.

Senator RYAN —No; when the government puts you on a roll, facilitation does not involve compulsion. This involves conscription. If someone says to you—

CHAIR —Hang on—my understanding is that nowhere is there conscription. At every level, even in New South Wales and Victoria, it has to come from the individual. We had this debate in the last parliament about conscription, and—

Senator RYAN —But if you are automatically on—

CHAIR —I do not say ‘conscription’ is the right word—

Senator RYAN —If you are automatically enrolled, the only way you can object to it is by saying, ‘The details are incorrect,’ as I understand it. There is no right to say, ‘I don’t want to be on the electoral roll.’ There is no right in the Commonwealth act at the moment, there is no right under these state acts—

Mr Killesteyn —That is right. In New South Wales, if you do not respond—

Senator RYAN —It is compulsory enrolment. It is using Commonwealth data sources, as we do with tax and other things, to data match and to compel enrolment because technologically we can do now what we could not do 30 years ago.

Mr Killesteyn —Can I just make one other point? The AEC is not suggesting that direct enrolment is the panacea for the sorts of trends we are seeing in relation to nonenrolment. What we are simply suggesting is that direct enrolment is one part of a set of tools that we have, and should have, at our disposal to try to deal with that large number.

We do not think this will solve the problem. We think it is a range of measures, which includes our continuing processes—particularly around continuous roll update, mechanisms and strategies which we are about to unfold this year—to be closer to the point at which the person is changing their address with other agencies—and I can go into the detail of some of that if you like—that direct update using third party data—that is, when the person is already on the roll we use third party data to change their address—and then, finally, direct enrolment. It is a whole suite of programs—

Senator RYAN —I will be honest: I have not hidden my objection to most of those, and I want to explore these because this is a highly charged issue. I do not want the commission leaving thinking that this is not a highly partisan issue.

Let us go to the issue of data. One of the data sources used by New South Wales is data sources for first home owners grants. Other data sources which have been proposed have been land tax databases and things that states have. How on earth do we take into account that we have, as Mr Green outlined the other day, spouses living at different addresses for land tax purposes? People can have one place listed as a principle place of residence but which may not be where they live because they live in another area for a particular reason.

None of these databases were designed for the purpose for which the electoral roll was. We cannot even manage Medicare and tax file number databases well in this country, and they manage something the government usually views pretty importantly: money. How on earth are we going to data match and maintain a degree of confidence in an electoral roll where we cannot necessarily do it for a tax system?

Mr Killesteyn —I am not trying to design a system here. What I have put forward and what the commission is putting forward is that before any databases are used to directly enrol individuals, they ought to be subjected to some form of testing.

Senator RYAN —Does it have to be a perfect database? Are we going to accept 95 per cent confidence?

Mr Killesteyn —These are the rules that have to be worked through.

CHAIR —But you are not accepting the databases currently for New South Wales or Victoria?

Mr Killesteyn —That is correct.

CHAIR —You will formulate your own.

Mr Killesteyn —We are not indicating that any particular database should be adopted right now; it should go through a process of determining whether it is a credible database which gives us the confidence that a person can be enrolled. That would be a process which would be transparent, it would have rules and it would be subject to ministerial approval.

Senator RYAN —Ministerial approval of this concerns me. What worries me here is that you are going to make some choices about databases. Humans are not perfect. A mistake is going to be made that is then going to drag the AEC into an argument about electoral roll integrity. You cannot guarantee that a mistake will not be made. My view on this recommendation is that this is a highly charged political policy issue over which you have admitted there is a philosophical difference. I do not necessarily think it is appropriate the AEC is making policy recommendations on this as opposed to administrative ones.

Mr Killesteyn —If you are suggesting in any way that the AEC is being political in this process—

Senator RYAN —No, I am not at all. I am saying—

Mr Killesteyn —Thank you for that clarification.

Senator RYAN —Sorry, I did not mean to convey that. My point is that by making recommendations in spaces that are politically charged, just like making decisions that could end up being politically charged about data sources, you might actually inadvertently drag the AEC into a political debate that it is not currently in.

CHAIR —I suppose the commission’s motivation is an accurate roll, is it not?

Senator RYAN —I agreed with the point you made earlier.

Mr Killesteyn —I have a statutory role in advising the joint standing committee in parliament on issues around the administration of the Commonwealth Electorate Act. The Commonwealth Electoral Act requires that people are put on the roll.

Senator RYAN —This is not about administration of the Commonwealth Electoral Act.

Mr Killesteyn —No, I am sorry, it is—

CHAIR —Let me just intervene, Mr Killesteyn. The problem you have is that what has happened in New South Wales and in Victoria impacts on the Commonwealth electoral roll. On the evidence you have just given, 90,000 people thought they were on the roll and that is why you have to come to us and say: ‘There is an impact here. We have to act. We just can’t remain mute.’ Is that fair?

Senator RYAN —That is a fair point. I will accept that.

Mr Killesteyn —That is fair. I am sure there would be people who look at my accountability in that last slide and say, ‘What is the AEC doing because there are 1.4 million people not on the roll?’ That is my accountability because that is what the act requires. In terms of my accountability to deliver that sort of result that the parliament has expressed through making enrolment compulsory since 1911, I am suggesting that I need other tools. I am suggesting a suite of tools and this is just one of them. I am not entering a policy debate; I am entering a debate around what the act requires and the tool set that we have.

Senator RYAN —I think the point the chair made is fair. I can see indirect impact of what is happening in New South Wales. There is always a grey zone about what is politically charged and what is not. I do not like compulsory voting, but I am not going to say you should not be enforcing that because that is clearly what the act requires. But this is an issue about which you have admitted there is a philosophical difference. I have stated in every hearing where we have had a discussion around this that I do not want to see the AEC dragged into a political debate.

CHAIR —We have been through this.

Senator RYAN —If a decision is made and we have another McEwen type result, and we find the utilities bills for TRUenergy in Victoria are out, which they are, then that is going to cause a problem for our electorate system that we do not currently have.

CHAIR —This is a live issue. This is an issue about which I am personally interested in the Electoral Commission’s views as to what it is that you want to use to assist you. My concern is that just because it is happening in New South Wales, should it happen federally? Just because it is happening in Victoria, should it happen federally? I am interested in the strengths and weaknesses of each of the systems; I am not interested in a column of smoke. At the moment, Senator Ryan’s frustration is that he does not want the commission adopting New South Wales’ provisions.

Senator RYAN —I have seen a proposal for this. I have seen your arguments for this. I have seen no risk assessment. Your proposal previously and now has been to recommend this Smart Roll sort of activity. You have put the case quite well, but I happen to disagree with it. I have not seen an AEC assessment of the risk of adopting a smart roll. I have to find that out myself.

Mr Killesteyn —We have done that risk assessment. I am more than happy to table that as part of the commission’s evidence.

Senator RYAN —That would give me some comfort.

CHAIR —Obviously there is going to be an analysis of what has happened in Victoria and we have the New South Wales election happening this month with its system. I am interested in the feedback, as I said, that we might be able to bring into our deliberations as to the strengths and weaknesses of that legislation and where it might be improved. What is it that you are recommending to us that we might recommend? The problem we have, as I said earlier, is that I do not think we have any option because New South Wales or Victoria have basically meant that we at least have to look at it, because it has an impact—

Senator RYAN —There is an assumption there that we will continue arguing. Because the states do something does not mean that we can let the lowest—

CHAIR —I am not interested in the lowest common denominator. My policy is very clear: I am interested in confidence in the system and integrity in the roll. I do not want to go down the low road. But there is also the question: if there is not uniformity, what does it mean for us?

Mr Killesteyn —In 1984 that was the point at which the joint roll arrangements between the Commonwealth and the states were first put together. At that time it was seen to be a good idea to have a single electoral roll for Commonwealth, state and local government elections. One of the potential implications, notwithstanding all the other arguments that we have just traversed, is the whole notion of a joint roll dissipates or disappears. These are debates we need to have. That is my role. My role under section 7 of the Commonwealth Electoral Act is to find and bring these issues to the joint standing committee so they can be fully explored.

CHAIR —I have a few questions and then we might go to your submission. On page 27 of your submission there is a table 3.3. I am just interested again in the number of electors enrolled by age at the close of rolls. The number 57,732 is new electors, isn’t it?

Mr Killesteyn —Sorry, no—57,732 is the number of people who were added to the role as a consequence of the High Court decision. The original close of rolls was 22 July, which allowed the close of rolls for updates. Then the High Court extended that back to seven days, which brought it out to 26 July.

CHAIR —But for the High Court, that 57,732 would have missed the opportunity.

Mr Killesteyn —That is correct.

CHAIR —Those are not the only ones who would have missed out, are they? Let’s go to table 3.5. I just want to make sure that I am reading it correctly. You have at the end on both certified lists 40,408. Is that people who changed addresses?

Mr Killesteyn —Yes. They were updates.

CHAIR —There would have been a problem for those people as well but for the High Court decision, wouldn’t there?

Mr Killesteyn —It depends on the way you look at it. They could have voted but in their old division.

CHAIR —Correct, which would not—

Mr Killesteyn —So it is a distortion.

CHAIR —It is a distortion. Let’s say that they filled in their new division on the day, but for the High Court decision their vote would have been knocked out because—

Mr Killesteyn —They could have voted provisionally.

CHAIR —Unless they, in effect, said their old address—

Mr Dacey —If they had said their old address or given their new address and had a provisional vote for that they would have been okay.

CHAIR —So would it be fair to say that there were in effect about 90,000 voters who were new and regulated as a result of that High Court decision?

Mr Killesteyn —There were 57,732 new voters, and 40,408 were updated addresses.

CHAIR —That is what I am trying to establish—that they would not have updated at the close of the rolls before the High Court made the determination.

Mr Killesteyn —That is correct.

CHAIR —They would have missed out.

Mr Carpay —It is worth noting, too, that there is a significant change within divisions. So not everyone would move between divisions, and that has an impact—

CHAIR —I will come to that in a second. But that is the whole point. I made comment on this yesterday in the parliament. Changes within divisions might not have saved them if they had had to show POI.

Mr Killesteyn —One would assume that, if they had simply changed address within the division, their name would be on the certified list at the polling place that they turned up to.

CHAIR —Are you saying that is an extra one? It is changes within divisions.

Mr Carpay —I am simply saying that, of the 40,408, there is a portion who—

CHAIR —Within the division?

Mr Carpay —moved within the division and there is a portion who moved between divisions and possibly interstate. So the 40,000 is all of the changes in address that occurred—

CHAIR —Within divisions and external to divisions. So those who moved external to the divisions, if they had stayed in the state, would maybe have kept a Senate vote but not a House of Representatives vote. Those who moved internally would have kept a House of Representatives vote.

Mr Pirani —That is correct.

Mr Dacey —If they were identified on the roll.

CHAIR —If they were identified on the roll. Paragraph 3.43 refers to 67,000 polling officials. Is that what we are talking about on election day, is it?

Mr Dacey —Yes, that is election day.

CHAIR —So they are additional people other than permanent staff?

Mr Killesteyn —That is correct. They are casuals employed under the Commonwealth Electoral Act.

CHAIR —Are you able to say how many of those are people who have done it before and how many are new ones?

Mr Killesteyn —We can do that.

Mr Mason —We classify previous electoral experience, so if you are after information specifically on—

CHAIR —Yes, because I am interested in previous electoral experience.

Mr Mason —The number of those who worked at the 2010 election with previous electoral experience was just over 54 per cent, so 46.54 per cent of our officials at the 2010 election had no prior experience.

CHAIR  —Those figures do not include pre-polling?

Mr Pirani —I think they include everything to do with elections.

Mr Dacey —It would include pre-poll staff as well.

Mr Pirani —That is correct.

CHAIR —I am interested in what level of coaching and training they get, both those with previous experience and new staff. I assume they all get the same refresher course?

Mr Mason —That is correct.

Mr Dacey —There is no short course for experienced staff.

CHAIR —That is what I am asking. So they all get that. How long is the course? What does it entail?

Mr Mason —Training varies depending on the role of the polling official. Dependent on that role there is a combination of training that was undertaken prior to election day, and that was also supplemented by briefing sessions for certain roles, also prior to election day. The varying times of those briefing sessions depended on the type of role that was being performed. Obviously, an officer in charge receives more extensive training than a polling official who is just an ordinary issuing officer.

CHAIR —How long before an election does that kick in? Do these people stay on your list of recruits, or is it six months before an anticipated election, or 12 months?

Mr Mason —We have quite an extensive program prior to the lead-up to an election where we refresh the potential polling officials. We do not actually offer employment to those polling officials until the announcement of the election. Following the announcement of the election, we conduct an extensive training program for those polling officials in the lead-up to polling day. We do quite a large amount of training with our casual workforce, however, in the lead-up to the announcement. That is to assist us in things like covering off the close of rolls and the processing of those close of rolls—

CHAIR —Do divisional returning officers play a role in any of this, or is this something that is done by the state offices?

Mr Mason —Divisional returning officers are the ones who facilitate the briefing sessions face to face and facilitate liaison with polling officials on the training that is undertaken. The coordination of training is done through our national office and flows through all of our state and divisional offices.

Mr Killesteyn —Invariably the training does not kick in until the election is announced, essentially because we do not know when the election is going to be. I guess it is a balance that you can make about whether you kick off training and if the election is not for another six or nine months you might have lost all of the value of the training. One of the things we are now exploring is the notion of some form of what we are calling soft contact with these polling officials between elections.

CHAIR —That is what I am interested in.

Mr Killesteyn —Yes. We think there is value in having ongoing contact with polling officials both in terms of potential training but also just in terms of keeping them informed about things that are happening in the electoral world. This not only helps us at the time an election is called but would also give us some advantages in relation to their knowledge, if you like, of electoral principles. There are some issues with it, particularly around whether these people could from an industrial law perspective be regarded as employees. We have to manage this fairly carefully. But the notion of soft contact is one that we now think it is worth while investing in.

CHAIR —Would that soft contact in of remuneration to them?

Mr Killesteyn —No. Unless you are going to recommend to government to give us more money to do it!

CHAIR —I am on a fishing expedition, just following the principle through.

Mr Dacey —For those who we have email addresses for, which is quite a few, it could be, for example, an email newsletter.

CHAIR —Basically by going through that there is a likelihood that when an election comes along they will be picked up. There is the incentive for them.

Mr Dacey —Just to add to Mr Mason’s comments, we did an expression of interest, which is not an offer of employment. We did that eight months before the election, the December-January period.

CHAIR —Paragraph 3.48 says that the large volumes of applications received reveal that Genesis throughput times need to be improved before the next event. Is that as a result of information overload and the High Court decision? Can you elaborate on that?

Mr Killesteyn —Genesis is a new role management system that we introduced in September 2009, so well before the election was called. Genesis is a system that gives us much more assurance about the integrity of an individual enrolment transaction. It has a range of checks and balances which do comparisons with addresses, with names, with evidence of identity and so forth. Its first baptism of fire, if I can put it that way, was the 2010 election. Our post-implementation review evaluation of Genesis has found that as a mechanism for entering data it is comparatively slower than the former system that we used, the role management system. So what we have been doing since then is to tweak, but it is a lot more complicated than that word sounds, Genesis so that its capacity to take quick data entry is improved.

CHAIR —What does your evaluation involve? Does that involve your divisional returning officers?

Mr Killesteyn —Indeed it has. We have had extensive evaluation. In fact, we had a post-implementation review team, we appointed some independent experts in relation to systems and we had extensive workshops right across every single state. My recollection is that the post-implementation review of our systems involved more than 100 people across the state, including many divisional returning officers. There was one divisional returning officer who was part of the post-implementation review team. So they were very closely involved in the whole process.

CHAIR —You say in paragraph 3.4.15 that over 21,000 already enrolled electors had changes to their enrolment details effected without having to provide a signed enrolment form. My concern is to make sure that there is always a signature there that can be compared to a declaration vote. Are you suggesting that new online enrolments be done without a signature being received by the commission or retained by them?

Mr Killestyn —The signature has been the hallmark of the identity issue for many years. The online update facility is still based—

CHAIR —I am not worried about the update—

Mr Killestyn —I understand that, but it still gives us the signature. What we are suggesting is that we need to start to explore whether online enrolment is open to us and look for alternative means of being able to establish identity both at the time of enrolment and also in subsequent processes. It just seems to me to be the way in which our future electorate is going to be transacting its business with all government agencies.

CHAIR —I understand that. We have had this debate about provisional votes and a whole range of other things. To me what is vital is to retain integrity in the system and confidence in the system. As you say, the signature has been a hallmark of that.

Mr Killestyn —And I am not denying that, either. What we are putting on the table is the need to explore whether there are other models.

CHAIR —In 3.6.16 you say that Australia Post is, obviously, an important source of enrolments. Have you had any experience that you can give us, or is it still too soon, on the provision that allows provisional enrolment for 16-year-olds? What has that meant in terms of capturing kids at school who are 16 and above, and what are the indications for the future?

Mr Carpay —We have not specifically explored at this stage exactly how we will engage. Obviously the argument around the enrolment of 16-year-olds was the majority are in school and therefore the best way to actually engage them is in that context. The School and Community Visits Program and Enrol to Vote Week are programs we have run for a number of years, and in that context we would have to work out how to include 16-year-olds. Our ongoing issue I think is access to schools and schools allowing us in at appropriate times, and that is the predominant mechanism through which we would actually pursue those people.

CHAIR —It also has not been in place for a long time; I accept that.

Mr Killestyn —All of this legislation, including the online update, was only passed days before the election was announced.

CHAIR —I am just saying to you that from the committee’s point of view if we make amendments that are picked up by the government and the parliament, I am interested intermittently to see what the impacts of some of those amendments are. It is just smart to see whether we are on the money.

Mr Killestyn —We have reached a view that our current School and Community Visits Program is not as effective as it should be. As Mr Carpay has outlined, a lot of those issues are around getting access to schools. One of the things we have learnt from the research that we have now done is that perhaps it might be better for the AEC to assist schools with their elections of their own prefects and so forth.

CHAIR —That is something you would get the DROs in their state to do?

Mr Killestyn —Exactly. They are the ideal group to run elections on behalf of schools. Our research tells us that it is more likely that the kids will take notice of subsequent obligations to vote if they participate in a real election conducted by the AEC.

CHAIR —Can I also go to paragraph 3.6.25, which is table 3.6. I am particularly interested in the citizenship ceremonies. I think there was legislation passed in the dying days of the last parliament in relation to provisional enrolment for citizens, or am I dreaming?

Mr Pirani —Section 99A and 99B are still there which have provisional enrolment of new citizens. The previous JSCEM made recommendations. The government has responded and accepted those recommendations but we do not have legislation yet.

CHAIR —Again, it would capture most people in a citizenship ceremony rather than having them go up to the back of the room and fill out a form. I will go to paragraph 3.713. It is a paragraph in relation to the integrity of the roll where you did a rolls transactional analysis for nine seats. Paragraph 3.713 states:

The results from this analysis indicate a very low incidence of this type of movement and give no cause for concern. The maximum number of incidences identified for any particular division was seven.

The incidence is detailed in paragraph 3.712:

This analysis was to identify any instances where electors enrolled for a division in the three month period leading up to the close of rolls and then transferred back to their previously enrolled address in a three month period following the election.

Why I want to highlight that in public hearings is that the fear that you are going to get mass movement in and out of electorates to try and manipulate results is not based on any evidence that you have picked up, is it?

Mr Carpay —No. We analysed for a period leading in to the election which had roughly a million transactions where people moved, which is comparable to 2007, and then looked at the specifics of people as individuals moving into an electorate and moving out of that electorate. I think there were 33 instances where people moved in and out. The majority were younger people.

CHAIR —So it could be kids going to uni or boarding with someone.

Mr Carpay —On the first analysis it looks like those people are in fact returning home because if you look at the address to which they return and you look at the names of the people in those addresses, they are the same, which would lead to that conclusion. The step we have not done yet but are doing shortly is to ring those individuals just to confirm the fact of their circumstances. All the evidence suggests the majority of those 33 people had gone to universities or other accommodation and are returning home.

CHAIR —So the fear that there are going to be hordes of people who change enrolment to effect another result is just not there on the evidence.

Mr Carpay —There is a large number of people who do update in the lead-up to an election because they have moved and have not got around to it yet as some of our research evidence is suggesting. It remains at roughly the same proportion.

CHAIR —Why did you particularly pick this analysis of people moving to a new address and going back to the old address?

Mr Carpay —On the basis of concerns around people flocking into a particular division to change an outcome in that location. If you were going to do that, you would not know when the election was going to be called so you would do it presumably shortly before you think the election will be called or at the time it is called, and then presumably you would return to your other address notionally a short time afterwards. That is the methodology we have applied.

CHAIR —Okay.

Senator RYAN —I commend you on that analysis. To be honest, I have not had a chance to finish all of your submission but constant little activities like that are what maintain a strong degree of faith. Congratulations on undertaking that. I want to quickly turn to the number of pre-poll places. It is the last issue I want to raise. It relates to an answer to a previous question. Mr Killesteyn, it could have been a question on notice in estimates. I have been looking for it but I cannot find it. It indicated there was roughly a 20 per cent increase in the number of pre-poll places. It might have been in my home state of Victoria or more generally. I thought there was a significant increase from 2007 to 2010. If you do not have it handy or you cannot recall it, I am happy to defer the discussion. There has been an explosion obviously due to legislative change and behaviour change in pre-poll voting and we have reflected that in updating the way we count pre-poll votes. But that struck me as a significant increase in the number of locations between 2007 and 2010. What was driving that?

Mr Killesteyn —It is relatively simple. It is about the demand. I am not sure whether you missed that part of the presentation, but the growth of early voting continues. We have seen another 20 per cent plus in early voting. The joint standing committee, in its number of previous reports in both 2004 and 2007, has been encouraging us to meet the demand. With that growth, we think we have to add more prepoll voting centres.

Senator RYAN —It strikes me that there is a bit of a feedback loop in operation here. The number of prepoll centres does mean that you tend to get a higher demand. I am not putting this on the commission in any way, because I know that it has been pushed from this side of the table over many years. Do you foresee there to be another, similar sized increase in the next two or three years? How many prepoll centres are we going to get to before you think we have enough with this sort of growth? We have three weeks now. It is not about location, I imagine; it is about manpower as much as anything else.

Mr Killesteyn —I wish I knew just how much the growth in prepoll voting centres would be. If the growth in early voting continues and we simply hold the number of prepoll voting centres then we will face at some point the same sorts of complaints as we currently get about polling day. We will start seeing large numbers of queues and people waiting for a long time to cast their ballots. So there is a balance here.

Senator RYAN —I appreciate that. I am a bit of a romantic when it comes to polling day voting.

Mr Killesteyn —We have already had the debate about whether we were having a polling period as distinct from a polling day.

Senator RYAN —It could have been impossible, but have you done a geographic analysis? I remember there days where the only place you could prepoll vote was the AEC divisional office. Then we went to two centres per electorate. Some electorates now have three or four. Have you done a geographic analysis? If you pop a centre in one part of an electorate—it would obviously be a larger electorate—you would then tend to get a higher proportion of people prepolling from that geographic part of the electorate?

Mr Killesteyn —I do not think we have done that cause and effect type of analysis. The reason we encouraged prepoll voting away from the divisional office is that on election day the divisional office is busy doing 101,000 other things, so it was important to try to alleviate the stress in divisional offices. But we have not done analysis of cause and effect and whether you put something in place and it generates the demand, as you are suggesting.

CHAIR —There is a general question I have not taken up with you. In your submission there is a suggestion of nomination periods being reduced in prepolling. There are some recommendations. I think you wanted to change some time frames.

Mr Killesteyn —That is correct: to bring back the nomination period by a day.

CHAIR —Can you tell us publicly what that is?

Mr Killesteyn —Essentially, the issue that we are facing is that nominations close on the Thursday or Friday.

Mr SOMLYAY —The draw is on the Friday.

Mr Killesteyn —So the issue is around producing ballot papers that can get into postal vote certificates to be mailed out.

CHAIR —So what is it you want?

Mr Killesteyn —We want an extra day to be able to produce the 40 million ballot papers over a concentrated period.

CHAIR —So you want nominations to close a day earlier—

Mr Killesteyn —On the Wednesday.

Mr Dacey —But we would not start, as we do now, the issuing of prepolls until the Monday, which gives us one more day to print.

CHAIR —But if nominations close a day earlier you then have the public draw a day earlier?

Mr Dacey —That is correct—and the lodgement of group voting tickets for Senate candidates has to come a day earlier.

CHAIR —But, in effect, what will happen is the draw will occur on the Thursday but you will be in a better position to have the ballot papers for the Monday. Is that what it is about?

Mr Killesteyn —That is correct.

Mr Dacey —That is all it is.

CHAIR —So there is no conspiracy theory here!

Mr Killesteyn —This is about maximising the ability to get 40 million ballot papers printed.

CHAIR —So, in effect, if it is 10 days now for nominations you want nine.

Mr Dacey —That is correct.

CHAIR —So you want to take a day off. The next bit was on, I think, the timing of prepolling. Was there a change of a day on that as well?

Mr Dacey —There was the timing of the receipt—

CHAIR —Bear with me while I find it in the submission. I think it is important. There was the postal voting deadline, and we got that. There is a suggestion that prepolling does not occur till the Monday after the close of nominations.

Mr Killesteyn —That is a consequential impact.

CHAIR —Is it consequential? What does the section currently say?

Mr Killesteyn —My understanding is that it is currently 48 hours after the close of nominations and we are asking that that be held until the Monday.

CHAIR —So it is 72 hours, in effect.

Mr Dacey —Because it is very difficult—

CHAIR —I understand. I am not opposed; I just think it is important that these things be aired publicly rather than just in submissions. So the basis upon which you are recommending that an application for a prepoll cannot be made before the Monday 19 days before polling day is that otherwise, if the act as it currently stands is not changed, you would be prepolling on the Sunday.

Mr Killesteyn —That is right.

Mr Dacey —And we will not have ballot papers on the Sunday. The access cannot be made earlier than the second day after the day on which nominations for the Senate election are declared.

CHAIR —That is right. So Friday and Saturday means you have to give overtime on Sunday, haven’t you? Do you pay overtime?

Mr Dacey —We would probably be working overtime on Sunday anyway.

CHAIR —That is fine. It is just that there is no conspiracy in any of this. What you are saying, I suggest, is that it will allow a smoother process for you. There is a sufficient time for people to prepoll and for nominations to be lodged in the time frames you are suggesting.

Mr Dacey —It is also to reflect the reality of people being able to prepoll.

CHAIR —Yes The fact of the matter is you would need a decent ballot paper, not a hand written ballot paper, when it comes to prepolls.

Senator RYAN —You may not wish to comment on this or may wait until a supplementary submission, but we heard from Antony Green. He was particularly concerned about the size of the Senate ballot paper as it is starting to get in New South Wales and Victoria. He had some ideas, like increasing deposits and restrictions on names above the line. Does the commission have a view? The printing of the ballot papers reminded me. I would imagine it is getting beyond the limit of any quick printer. Do you have or are you going to consider some recommendations around what might need to be dealt with? I know in Victoria it is up both sides of the booth.

Mr Killesteyn —Yes, it is of concern to us. When you take evidence from our New South Wales manager, I have asked him to explicitly explore the size of the ballot paper and the sorts of issues that we might need to confront. If you want to wait until that time, he has a show and tell as well, so you can see what is happening.

Senator RYAN —Sure. I am very happy to do that.

CHAIR —All right.

Proceedings suspended from 11.23 am to 11.43 am