| Parliamentary Joint Committee on Electoral Matters

Navigation: Previous Page | Contents | Next Page

Chapter 7 Formality issues

7.1                   The Constitution[1] provides for Australia’s representatives to be directly chosen by the Australian people. Consequently, ensuring that the valid votes cast by eligible electors count towards the election of their representatives is fundamental to Australia’s democratic electoral system. Opposition members of the Committee note that the Constitution provides for Australia‘s representatives to be directly chosen by the Australian people, and, consequently, votes of individual electors should never be decided by a third party nor directed to a party the elector never had any intention of voting for. Opposition members believe the South Australian ticket system is not constitutionally sound.

7.2                   At every election, it has been the case that some ballot papers do not meet formality requirements and so cannot be included in the vote count. This may be the result of a deliberate choice or a genuine mistake by the elector. The effect is that these electors are not having a say in who will be their representatives in Parliament.

7.3                   For this reason, it is important after each election to tackle vote formality issues. It is vital to closely examine informal ballot papers and attempt to understand the intention of the voter, to explore the underlying contributing factors to informality, and to act to ameliorate the problem.

7.4                   The 2010 federal election saw informal votes for the House of Representatives at 5.55 per cent (729 304 votes), an increase of 1.6 per cent on the 2007 federal election.[2] Senate informality was 3.75 per cent (495 160 votes), an increase of 1.2 per cent on the previous election.[3]

7.5                   In its examination of the level of informal voting at the 2010 federal election, the Australian Electoral Commission (AEC) observed that:

In every election, it is likely that a small proportion of the votes cast will not meet the specified voting requirements and will therefore be deemed informal. Levels of informal voting can provide an indication of people’s engagement with (and understanding of) the electoral process and, together with enrolment participation rates and measures of turnout, are therefore a key indicator of democratic health.[4]

7.6                   In recognition of the importance of the informal voting issue, the AEC now routinely undertakes an analysis of informal voting in the House of Representatives following federal elections.

7.7                   In practical terms, the rising level of informality means that more and more people (hundreds of thousands) are turning up at designated polling places (or voting by post) and lodging ballot papers that then do not count towards the election of their representatives.

7.8                   Not all informality is unintentional. The AEC analysis of House of Representatives informality revealed that 51.4 per cent of informal votes were assumed to be unintentional, with the remaining 48.6 per cent assumed to be deliberately informal.[5] Opposition members contend the actual intentional informal figure is likely to be far higher than 48.6 per cent, with many electors who vote only ‘1’ or place a tick or a cross deliberately choosing not to number every box. It is important that those who intend to vote informally do not have their vote counted against their wishes for a candidate they did not wish to vote for.

7.9                   The level of informality for voters genuinely trying to cast a formal vote is of concern. This means that upwards of 370 000 voters are attempting to vote, but for various reasons are failing.

Committee view

7.10               The Committee believes that as part of a system that seeks to maximise participation in the democratic process to elect Australia’s representatives, reasonable measures should be taken to help ensure that votes are not wasted. In particular, in cases where it is clear that an eligible voter has attempted to cast a formal vote, but it is informal perhaps due to confusion over what is required to make their vote count.

Requirements for a formal vote

7.11               Ballot papers must satisfy certain requirements before being accepted into the vote count. Each ballot paper must first undergo authenticity checks to ascertain that:

• it is a genuine ballot paper—carrying the official mark and initials of the issuing officer; and
• it does not identify the voter.[6]

7.12               Each ballot paper is then checked to ensure that the vote cast is ‘formal’ (in keeping with requirements set out in the Commonwealth Electoral Act 1918).

7.13               A House of Representatives vote will be formal if:

• one first preference is indicated and all boxes are numbered consecutively; or
• one first preference is indicated and all boxes (except one) are numbered consecutively.

7.14               The Senate voting system provides voters with two options for casting a vote: above-the-line (group ticket voting) and below-the-line (indicating all preferences). A voter can vote above-the-line by indicating one, and only one, first preference against one of the group voting squares, and their preferences for all the other candidates will be taken to be in accordance with the group voting ticket (or tickets) lodged with the AEC by that political party or Senate group. Voters may number more than one preference above-the-line, but preferences will still be determined as per the voting ticket of their first preference candidate.

7.15               When voting below-the-line for the Senate the vote will be formal if:

• a first preference is shown by the number ‘1’ marked in the square opposite the name of one, and only one, candidate; and
• where there are 10 or more candidates, not less than 90 per cent of the squares opposite the names of candidates on the ballot paper are numbered as required, or would be if no more than three numbers were changed; or
• where there are nine or fewer candidates, all squares opposite the names of candidates on the ballot paper (or all but one of these squares with only one square left blank) are numbered as required, or would be if not more than two numbers were changed.[7]

7.16               In keeping with the principle of erring in favour of enfranchisement, some deviations in numbering are acceptable, and may allow a ballot paper to remain in the count. However, a repetition of a first preference is not acceptable in Senate or House of Representatives voting. Table 7.1 outlines some of the main deviations that may occur and compares how they will affect ballot papers in the House of Representatives and the Senate.

Table 7.1 Comparison of acceptable numbering on ballots papers

Ballot paper marking(a)	Acceptable on Senate ballot paper(b)	Acceptable on House of Representatives ballot paper
Single first preference (figure ‘1’)	ATL: Yes BTL: No	No
Single first preference (figure ‘1’, a tick or a cross)	ATL: Yes BTL: No	No
Incomplete numbering (consecutive preferences beginning with 1 are shown, but the last square is left blank)	ATL::Yes BTL: Yes, the blank square is deemed to express the voter’s last preference	Yes, the blank square is deemed to express the voter’s last preference
Incomplete numbering (consecutive preferences beginning with 1 are shown, but more than one square is left blank)	ATL: Yes BTL: Yes, if certain requirements are met (CEA s. 270(1)(b))	No
Number sequence errors (missed numbers)	ATL: Yes BTL: Yes, if certain requirements are met (CEA s. 270(1)(b))	No
Repeated numbers	ATL: Yes BTL: Yes, if certain requirements are met (CEA s. 270(1)(b))	No

(a)   The Commonwealth Electoral Act (CEA) prescribes the ballot paper formality requirements for federal elections.

(b)   The Senate’s two forms of voting are above-the-line (ATL) and below-the-line (BTL).

Source        Australian Electoral Commission, Scrutineers Handbook: Federal election 2010, Version 3.

7.17               On Senate ballot papers provision is also made for when a voter has attempted to vote both above and below-the-line. If both votes would have been formal if recorded on their own then the below-the-line is given precedence and used for the count, with the ticket vote treated as if it had not been attempted, as it is assumed that marking all preferences is a better indication of the voter’s intent.

7.18               If the votes attempted both above and below-the-line would have been informal if recorded on their own then the ballot is rejected. However, if the voter made an error when marking their ballot paper that made either, but not both, their ticket or full preferential vote informal, then the vote may still be formal in the following cases:

• where the ticket vote would have been formal if recorded on its own, but the preferential vote would have been informal if recorded on its own, the ballot paper is formal and  the preferential vote below-the-line is treated as if it had not been attempted; or
• where the preferential vote below-the-line would have been formal if recorded on its own, but the ticket vote would have been informal if recorded on its own, the ballot paper is formal and is treated as if the ticket vote had not been attempted.

7.19               The Senate voting system currently allows more opportunities for saving informal votes than the House of Representatives arrangements.

Informal voting in the 2010 federal election

7.20               In the 2010 federal election, the rate of informal votes for the House of Representatives was 5.5 per cent. Other than the 1984 election, this was the highest informal vote since the introduction of compulsory voting for federal elections in 1924. In 1984, informality in the House of Representatives ballooned as a result of confusion from the introduction of above-the-line Senate voting. However, the Committee accepts that optional preferential voting in New South Wales and Queensland, and the prevalence of ‘just vote 1’ campaigns in these jurisdictions, has, along with other factors, also contributed to increased levels of informality in the last few years.

7.21               The total House of Representatives informal vote in 2010 (729 304 informal votes) was equivalent to 7.8 average electoral divisions at the 2010 federal election. The rise in informality since the 2007 federal election is equivalent to 2.3 electoral divisions. These equivalents were calculated based on the national average for enrolment at 31 July 2010, which was 93 804 electors.

7.22               The 2010 federal election saw a substantial increase in assumed intentional informal voting in the House of Representatives, most readily identified by ballot papers that are left blank or have slogans or other messages written on them. This means that some electors see it as their right to submit an informal vote and Opposition members believe it is thus important that their votes are not given to bureaucrats to fill out the rest of the form for them. For the House of Representatives, 48.6 per cent of informal votes were assumed to be deliberately informal.[8] Notably, the AEC found that 2010 was the first election since the informal ballot paper survey began that the proportion of blank ballot papers (210 587)—a key indicator of a deliberate informal vote—was higher than the proportion of number ‘1’ only ballots (202 432).[9]

7.23               The AEC analysis of House of Representatives informality revealed that 51.4 per cent of informal votes were assumed to be unintentional, with the remaining 48.6 per cent assumed to be deliberately informal.[10] Opposition members believe the actual intentional informal figure is likely to be far higher than 48.6 per cent, with many electors who use only a ‘1’, a tick or a cross deliberately choosing to not number every box. It is important that those who intend to vote informally do not have their vote counted against their wishes for a candidate they did not wish to vote for.

7.24               The AEC noted the media coverage of the call by a former Member of Parliament, prior to the 2010 federal election, for voters to submit blank ballot papers.[11] The AEC observed that:

It is not possible to determine whether the increase in blank ballots is related to Mr Latham’s comments or indeed to any other public commentary. It is possible that the level of blank ballots was simply a reflection of the mood of the electorate.[12]

7.25               In its analysis of House of Representatives informal voting, the AEC provided a breakdown, by state and territory, of the categories of informal votes (see Table 7.2).

Table 7.2 2010 House of Representatives informality, by State and Territory

State/Territory	Number ‘1’ only	Ticks and crosses	Sum of  Number '1' only and Ticks and crosses	All other informal categories	Total Informal Votes
New South Wales	93 466	40 405	133 871	159 892	293 763
Victoria	31 005	13 606	44 611	105 088	149 699
Queensland	44 247	13 626	57 873	79 522	137 395
Western Australia	13 786	7 061	20 847	40 120	60 967
South Australia	13 124	7 258	20 382	36 183	56 565
Tasmania	2 595	1 440	4 035	9 756	13 791
Australian Capital Territory	2 969	1 535	4 504	6 422	10 926
Northern Territory	1 219	793	2 012	4 186	6 198
National	202 411	85 724	288 135	441 169	729 304

Source Provided by the Australian Electoral Commission, based on Analysis of Informal Voting: House of Representatives, 2010 Federal Election, Research Report Number 12, 29 March 2011.

7.26               It is evident that the highest numbers of informal votes were in New South Wales, Victoria and Queensland. Accordingly, most of the higher rates of informal voting by division are also predominantly in these states, in particular New South Wales (see Table 7.3). Opposition members of the Committee observe that this shows that many voters are confused by the difference between the state and federal systems, in particular in NSW and Queensland where optional preferential voting is used, this is confirmed by noting that the highest proportion of informal votes where the elector has just put a ‘1’, a tick or a cross come from these two states.

7.27               When examined by division, the ten divisions with the highest informality rates are all located in western Sydney: Blaxland, Fowler, Watson, Chifley, McMahon, Werriwa, Greenway, Barton, Reid and Parramatta.[13]

Table 7.3 2010 Informal vote—Divisions with highest informality rates

Rank	Division	State	Formal votes	Informal votes	Total votes		Informal %	Informal swing %
1	Blaxland	NSW	73 830	12 081	85 911	14.06		5.17
2	Fowler	NSW	76 882	11 314	88 196	12.83		4.35
3	Watson	NSW	76 757	11 265	88 022	12.80		3.71
4	Chifley	NSW	80 371	10 097	90 468	11.16		3.25
5	McMahon	NSW	79 860	9 710	89 570	10.84		3.24
6	Werriwa	NSW	75 314	8 692	84 006	10.35		3.77
7	Greenway	NSW	79 308	9 075	88 383	10.27		4.09
8	Barton	NSW	78 683	8 572	87 255	9.82		3.25
9	Reid	NSW	79 628	7 680	87 308	8.80		3.22
10	Parramatta	NSW	78 317	7 418	85 735	8.65		2.03
11	Banks	NSW	83 869	7 665	91 534	8.37		2.61
12	Lindsay	NSW	83 227	7 402	90 629	8.17		2.65
13	Kingsford Smith	NSW	82 029	7 280	89 309	8.15		2.84
14	Macarthur	NSW	78 203	6 899	85 102	8.11		2.54
15	Lingiari	NT	42 927	3 482	46 409	7.50		2.65

Source Parliamentary Library, Exhibit 3, p. 1.[14]

7.28               In the Senate, the informality rate at the 2010 federal election was 3.75 per cent (495 160 votes). There is clearly a gap between that and the higher House of Representatives informality of 5.5 per cent.

7.29               However, this was not always the case. Prior to the introduction of group ticket voting in the Senate for the 1984 federal election, Senate informality was approaching ten per cent (see Table 7.5 later in the chapter).

7.30               Since its first use at the 1984 election, the more user friendly ticket (above-the-line) voting is now firmly established as the voting option used by the vast majority of voters for Senate elections, especially in states with high candidate numbers. This is clearly illustrated in Table 7.4.

Table 7.4 Method of voting in the 2010 Senate federal election, by State and Territory

State/ Territory	Number of Senate Candidates	Number of ticket votes cast	Percentage of total Senate vote cast using Ticket	Number of Below the line Senate Votes cast	Percentage of total Senate vote cast using full preferential
NSW	84	4 059 558	97.76	92 966	2.24
Vic	60	3 122 603	97.01	96 148	2.99
Qld	60	2 374 789	96.91	75 722	3.09
WA	55	1 196 446	96.94	37 773	3.06
SA	42	950 000	94.1	59 578	5.9
Tas	24	263 944	79.82	66 747	20.18
ACT	9	174 086	75.93	55 186	24.07
NT	15	87 665	90.67	9 022	9.33

Source Australian Electoral Commission, Virtual Tally Room, Election 2010.[15]

7.31               As is evident in Table 7.1 earlier in the chapter, comparing the acceptable numbering on Senate and House of Representatives ballot papers, the voting safety net for saving potentially informal votes is much wider for the Senate. However, the above-the-line ticket voting option is what has had the biggest impact on reducing Senate informality.

Historical context to Senate voting changes

7.32               In 1983, the Joint Select Committee on Electoral Reform (JSCER) was formed to inquire into and report upon all aspects of the conduct of elections for the Parliament of the Commonwealth and matters related thereto. Its inquiry included an examination of federal voting systems.

7.33               The JSCER observed that there had been various experiments with aspects of voting systems since federation, and outlined key developments in its report.[16] The AEC also keeps timelines of Australia’s major electoral developments.[17]

7.34               One of the issues of concern at the time of the JSCER’s inquiry was the high level of Senate informal votes at the 1983 federal election. For example, in a parliamentary debate that year, it was observed that:

When we look at the election results around Australia for the 1983 Senate election, we find that the informal vote for the Senate was, in fact, the third largest bloc of votes nationally for the Senate. The informal vote for the Senate actually exceeded the vote for the Australian Democrats around Australia. That is the dimension of the problem. It is even more starkly highlighted by the fact that the number of informal votes in New South Wales and Victoria this year exceeded one electoral quota for the Senate. So, the informal vote could have elected one senator in both of those States.[18]

7.35               In the First Report, the JSCER concluded that the introduction of ‘proportional representation’ in 1948 had resulted in a more evenly balanced composition of the Senate. However, the troubling trend of rising levels of informal voting emerged in the Senate. The Australian Electoral Office Survey of informal voting at the 1977 federal election revealed nine per cent (731 555 ballot papers) informality for the Senate, as compared to 2.52 per cent (204 912 ballot papers) for the House of Representatives.[19]

7.36               Various stakeholders made submissions to the JSCER, advocating changes to address the informality issue. The Australian Labor Party advocated for optional preferential voting, as it argued that the full preferential system led to increased informal votes and forced voters to cast a preference for all candidates. It recommended that the voter only need express the number of preferences equal to the number of vacancies in the House of Representatives or the Senate. The Liberal Party and the National Party opposed optional preferential voting, as it was similar to, and came with, the disadvantages of a first-past-the-post voting system.[20]

7.37               The Australian Electoral Office proposed the introduction of a voting system in which a vote could be cast by ticking a box indicating a registered ‘list’ of party preferences. The vote would then be counted as if it has been fully completed. The Liberal Party expressed a preference for the ‘list’ system over optional preferential voting.[21]

7.38               The JSCER decided that the introduction of the ‘list’ system whilst retaining the existing system—having the option to allocate all preferences—was the most feasible solution. It recommended that:

15. the current system of voting for each house should be modified as follows—
    
    
    1. for the Senate, a ‘list’ system should be introduced together with the retention of the existing system as an option open to those who wish to exercise their allocation of preferences, provided that a vote is not considered invalid if a mistake in sequence is made, but the voter intention is clear, i.e. a Senate vote should be considered formal as far as its intention is ascertainable provided that numbers are placed in at least 90% of squares;
    2. a House of Representatives vote should be considered formal as far as its intention is ascertainable provided that all except one of the squares is numbered.[22]

7.39               The then Government sought to give effect to this recommendation with the introduction of the Commonwealth Electoral Legislation Amendment Bill 1983. This Bill was part of major electoral reform, and included: the establishment of an independent Australian Electoral Commission to administer the federal electoral system; changing the franchise qualification to Australian citizenship; registration of political parties; and public funding and disclosure arrangements.

7.40               In his second reading speech on the Bill, the then Special Minister of State expressed concern about the high number of informal votes in the Senate elections, stating:

On Senate voting, the Government has accepted the recommendation of the Committee to provide; firstly, that a voter may mark one square indicating the adoption of his preferred party's how to vote ticket and, secondly, for the validation of any person's vote up to the point where the voter's intention remains clear. It has been a matter of notoriety that the complexity of the ticket has contributed to a substantial informal vote and also just as significantly to the throw-away vote, known as a donkey vote. The Joint Committee had before it material from the Australian Electoral Office which showed how the most trivial mistakes were the major factors in the Senate informal vote. The figures were taken from an analysis of the 1977 Senate and House of Representatives vote. The Senate informal vote was 9 per cent nationwide compared to 2.5 per cent for the House of Representatives. Of the Senate informal votes almost eight out of 10, 78 per cent, had been disqualified on one or two grounds, that is incorrect numeric sequence, or some squares left blank. This meant the disfranchisement in that election for the Senate of almost 600,000 Australians, more than the population of Tasmania. The situation is palpably absurd and this legislation will go a long way towards correcting it.[23]

7.41               In the course of debate on the 1983 Bill various concerns were raised in relation to aspects of the change to the Senate voting system:

Senator Peter Baume argued that the attempt to address Senate informality would have other undesirable consequences:

It is true that the voting process in the Senate can be simplified, but the use of the list system will reduce the effectiveness of the whole concept which lies behind a preferential system of voting...

While this might minimise informal votes in one respect, it could well encourage them in another. The fact is, of course, that at present 90 per cent of all electors are able to cast formal votes. That is a credit to the voters. It just puts out of court the claim by the Government that the community cannot handle a full preferential voting system. Clearly, the community can handle it. The fact that 90 per cent of votes are formal indicates to us that the task should be rather to increase that number of formal votes than to try to change it and move to a list system which carries at least as many problems as it does advantages.[24]

• The then Leader of the National Party of Australia argued that the ‘list’ voting was an oversimplification and ‘an insult to the intelligence of the average Australian voter to assume that he cannot fill in a significant number of squares’ to indicate preferences.[25]
• Senator Sir John Carrick argued that the ‘pre-occupation with eliminating informal votes ends up in a first-past-the-post system...[that] will certainly weaken the integrity of the preferential and proportional systems’.[26]
• Mr Steel Hall MP, objected to the further simplification of the Senate ballot paper by also allowing a tick and a cross to be deemed a figure ‘1’.[27]

7.42               Many of the JSCER’s recommendations were given effect by the Commonwealth Electoral Legislation Amendment Act 1983. The Act introduced above-the-line voting and Group Voting Tickets (GVTs). Senate candidates could choose to form groups and submit a GVT to the AEC setting out the order in which preferences should be distributed if a voter chose to vote above-the-line by just marking one box with a number ‘1’, a tick or a cross.

7.43               These changes had the desired effect of reducing Senate informality. However, in most elections since 1984 the rate of informal votes in the House of Representatives has exceeded Senate informality.

7.44               Table 7.5 shows the percentage of informal votes in House of Representatives and Senate elections since 1977. The marked difference between levels of informal votes in the Senate and House in the late 1970s and early 1980s has again emerged, but the case has reversed with the House of Representatives now recording a higher level of informal votes than the Senate.

Table 7.5 Percentage of informal votes for the House of Representatives and the Senate, 1977 to 2010 elections

Election	1977	1980	1983«	1984	1987	1990	1993	1996	1998	2001	2004	2007	2010
House	2.5	2.5	2.1	6.3	4.9	3.2	3.0	3.2	3.78	4.8	5.2	3.95	5.55
Senate	9.00	9.6	9.9	4.3	4.1	3.4	2.6	3.5	3.24	3.9	3.8	2.55	3.75

Source Australian Electoral Commission, for 1977 to 2007: Informality percentage House of Representatives and Senate, http://www.aec.gov.au/Voting/Informal_Voting/summary.htm, viewed 22 June 2011.
«1983 figures include missing and discarded ballots

7.45               The examination of the historical context of the changes to the Senate voting system to address high levels of informal votes and the higher level of informal votes in the House of Representatives currently being experienced, reveal compelling similarities between the type of mistakes made on ballot papers and the groups whose votes are affected.

7.46               When discussing the issue in 1983, South Australian Senator Graham Maguire stated:

The Australian Electoral Office conducted a study of all informal ballot papers at the 1977 federal election and the report showed uniformly in all States that the errors which caused informal votes were basically of two types: Incorrect numbering sequence and cases where people had left some squares blank. It was quite striking that in virtually every State about 77 per cent or 78 per cent of informal votes could be put down to those two types of errors. So clearly people were trying to fill in their ballot papers. People were trying to cast formal votes, but they were just frustrated by this system which required 30 or 40 preferences to be placed on a ballot paper...

...

Who votes informally is a very important question that has to be faced up to. We carried out an investigation in my State some years ago. We found that, in many suburbs where there are large numbers of senior citizens, where migrants have arrived recently from certain countries or where there are citizens with low levels of schooling, we got high informal votes.[28]

7.47               Today, the Joint Standing Committee on Electoral Matters faces many of the same policy and practical challenges in reducing informality in the House of Representatives as faced by the Joint Select Committee on Electoral Reform when it looked at the high level of Senate informality in 1983.

Factors affecting informal voting

7.48               In its analysis of informal voting  for the House of Representatives in the 2010 federal election, the AEC observed that:

There are many factors that could influence a voter to intentionally or unintentionally cast an informal vote and it is not possible, in many cases, to accurately quantify or even separately identify the impact these factors might have. Of those factors identified as significant influences on (unintentional) informal voting at previous HoR elections, English language proficiency and the number of candidates appear to be the strongest predictors of informality rates (or changes in informality rates) in 2010.[29]

7.49               The key factors contributing to informal voting brought to the attention of the Committee included: lack of English language proficiency, high number of candidates, socioeconomic considerations, and the differences between voting systems at both the federal and state and territory levels. However, Opposition members believe none of these factors justify authorising a bureaucrat to deem the rest of the elector’s ballot paper being cast in accordance with a registered ticket.

English language proficiency

7.50               In its analysis of informality in the 2010 federal election, the AEC found English language proficiency to be one of the strongest predictors of unintentional informality.[30] It stated that:

Five out of the 10 divisions with the highest informality rates at the 2010 House of Representatives election also had the five highest proportions of persons who, at the 2006 Census of Population and Housing, indicated that they did not speak English well, or did not speak English at all.[31]

7.51               Elections analyst, Mr Antony Green, also acknowledged the links between high migrant populations and higher informal voting rates. He advised the Committee that:

…certainly there are many people who came to Australia from overseas who have voted in other systems, and we are one of the only countries in the world that numbers a ballot paper. We are certainly the only country in the world that insists you number every box on the ballot paper.[32]

7.52               The AEC strategies to assist people from culturally and linguistically diverse (CALD) backgrounds and minimise informality in the 2010 federal election included:

• translating a range of information (including the Official Guide to the 2010 Federal Election - a leaflet that was distributed to all households) into 22 different languages for use in the polling place. Additionally, all translated election communication materials were available on the AEC website in an “information in your language” section. This information was also accessible through an AEC telephone translation service which provided assistance in 16 languages...[;and]
• employing multi-lingual staff in divisions with a high percentage of people with culturally and linguistically diverse (CALD) backgrounds; an interactive “How to vote practice tool” was available on the AEC website to enable electors to practise filling in their ballot papers; this was promoted through the advertising and public relations materials.[33]

7.53               The AEC also undertook a pilot project in western Sydney in the lead-up to the 2010 federal election. Areas that had high informality rates in previous federal elections were targeted: Blaxland, Watson, Chifley, McMahon, Fowler, Reid, Parramatta, Werriwa, Banks and Bennelong. The AEC noted that:

The primary objective of this project was to increase voter knowledge in relation to casting a formal vote. The secondary objectives were to engage and build community connections and to evaluate the content methodology used in delivering community education.[34]

7.54               The pilot project was managed by consultants working closely with the AEC. Bilingual educators, in many cases with extensive links in the relevant community, were engaged and trained by AEC staff. In conducting the workshops, the AEC noted that:

The educators were provided with appropriate tool kits and information and presentation material. Each workshop was attended by an AEC staff member who provided technical expertise and support for the relevant bilingual educator.[35]

7.55               The pilot project was cut short by the announcement of the 2010 federal election. In total, 90 workshops were conducted, with 1772 participants across 13 language groups in the three week period leading up to the 2010 federal election, at a total cost of just under $210 000.[36]

7.56               While formal written feedback on the courses was limited, educators confirmed an increase in the knowledge and understanding exhibited during the course of the workshop.[37]

7.57               In the 2010 federal election, eight of these western Sydney divisions in which the workshops were conducted were ranked in the top ten highest percentages of divisions with informal votes.[38]

7.58               The AEC conducted a voter survey at seven locations in western Sydney on polling day. Forty-five of those interviewed had attended a workshop and found it useful. However, the AEC found that those interviewed were ‘...no less likely than other western Sydney respondents who speak languages other than English to vote informally, nor were they more confident about voting than those who had not attended a workshop’.[39]

7.59               In its evaluation of the project targeting voters from non-English speaking backgrounds (NESBs) in western Sydney the AEC concluded that the project had little impact on informality rates in those divisions, but was optimistic about the potential impact of a longer term project of this kind. The AEC found that:

The workshops did not directly result in a reduction in informality in the targeted divisions during the 2010 federal election. However, they were perceived by attendees as useful. Given an average of 170-180 workshop participants in each of the target divisions consisting of around 100,000 voters, an immediate impact in informality statistics could not be expected. However, the adoption of the program on a continual basis with opportunity for repeat visits and broader topic coverage could be expected to have an impact on informality figures over time.[40]

7.60               SydWest Multicultural Services provides support services in the south western Sydney area to refugees and humanitarian entrants, including many from NESBs. It recognises that a lack of functional English is a significant challenge for people from NESBs carrying out a range of necessary activities, including casting a valid vote. Mr Agwa stated:

We do admit that language is the big issue, therefore it is important to engage with these people and give them resources so they can be able to cast their vote and exercise their democratic rights effectively and efficiently.[41]

7.61               The problem is not only restricted to new arrivals, as some Indigenous voters face the challenge of low functional English language skills. The Hon Warren Snowdon MP, the Member for Lingiari in the Northern Territory, raised this as a matter of concern in relation to Indigenous voters. He commented that:

...we had the highest enrolment figures but lowest turnout by percentage and highest informal vote over the 2001-2010 period. That is an issue. It reflects one issue that we need to be confronting that prior to 1996, there was a very active Aboriginal education voter enrolment division within the Electoral Commission. That was subsequently removed and therefore the capacity for people to get educated about their voting obligations was not what it ought to be. I think that is reflected in these voting figures, in the turnout and in the high informality, and is a question which I think we need to be confronting.[42]

Number of candidates

7.62               Generally, there has been a steady rise in the number of candidates contesting House of Representatives and Senate elections over the last few elections. In addition to cost implications, increasing numbers of candidates have resulted in a more complicated—and at times unwieldy—ballot paper that represents a logistical challenge to the AEC and a practical challenge to voters trying to cast a formal vote.

7.63               The AEC expanded on how this has evolved since federation, and observed that:

…the task faced by the average voter is now clearly more complex than when full preferential voting was introduced in 1918. Up until then, there had been seven general elections, at which a total of 525 vacancies were filled, and for which a total of 1060 nominations were received. Over those seven elections, the overall average number of candidates per vacancy was 2.02. Over the last eight general elections, from and including that of 1990, 1,191 vacancies have been filled, for which a total of 7,775 nominations have been received, at an average of 6.53 candidates per vacancy. This trend came to a climax at the 2009 Bradfield by-election, contested by 22 candidates, at which the informal vote reached 9%, by a substantial margin the highest ever recorded in the division, and more than double the rate for that division at the 2007 election.[43]

7.64               The impact of the increasing number of candidates in Senate and House of Representatives election is discussed in more detail in Chapter 9.

Differences between voting systems

7.65               Having differences between the voting systems for the Senate and the House of Representatives at the federal level itself, and with states and territories, can be confusing and pose a problem for voters trying to cast a formal vote.

7.66               Mr Antony Green observed in his analysis of previous elections that occurrences of a first preference only being marked on House of Representatives papers was higher in both New South Wales and Queensland, where optional preferential voting is used for state elections.[44]

7.67               This is supported by AEC statistics that show a substantially higher informal vote (with only number ‘1’, ticks or crosses) for New South Wales (133 871 votes) and Queensland (57 873 votes).[45]

7.68               When considering the differences at the federal level itself, Mr Green commented that:

While the interaction with state laws on optional preferential voting plays a part in varying the incidence of '1' only voting by state, what we also know from research is that the use of the Senate ballot paper, and its instruction that voters place a '1' above the line on the ballot paper plays a part in inducing '1' only voting in the House.

Research at by-elections has shown that without the distraction of the senate ballot paper, the incidence of '1' only voting declines...[46]

7.69               When the Commonwealth Electoral Legislation Amendment Bill 1983 (which introduced group voting tickets in the Senate) was considered, the potential for confusion was foreshadowed in debate, with Senator Baume stating:

We think it is far simpler for people to know that when they vote, the process they use will be to follow a full preferential listing of numbers and that that will occur whether they are voting for the House of Representatives or for the Senate. It is very difficult to ask people to use ticks or crosses on one of their voting papers and to write numbers on the other voting paper and not to expect that some inefficiency or unavoidable error will occur as a result.[47]

Committee view

7.70               The Committee agrees that greater harmonisation of voting systems is desirable. It would be much easier for a voter to be able to look at a ballot paper for the election of representatives to the Senate, House of Representatives or their state or territory Parliament, and have confidence that the same thing is expected of them to make their vote count.

7.71               While the Committee acknowledges that at present national harmonisation of voting systems is unlikely, it believes that the Government must take the necessary steps to mitigate the impact that confusion over—or indeed ignorance of—different voting systems may have on levels of informal voting.

Socioeconomic considerations

7.72               In evidence to the Committee, the Victorian Electoral Commissioner, Mr Steve Tully, noted the correlation between socioeconomic disadvantage and levels of informal voting. He observed that:

The data is also there to support the view that informality is directly related to the electorate in which people live. In the eastern suburbs of Melbourne, informality is a lot lower than it is in the western suburbs. The population make-up is also different, and I do not think it is drawing too long a bow to say that there are socioeconomic and other factors involved in informality—and literacy, I suspect, is also a big issue in informality, particularly with blank ballot papers.[48]

7.73               A comparison of socioeconomic indexes for electoral divisions[49] with the electoral divisions ranked by levels of high informal voting[50], reveal some correlations. For example, of the top 15 electoral divisions ranked by index of relative socioeconomic disadvantage and ranked by highest levels of informal voting in 2010, there were four divisions present on both lists (Lingiari, Fowler, Watson and Chifley).[51]

Addressing informality through education

7.74               The AEC delivers various education services, including school and community visits, sessions at their electoral education centre, and professional development workshops for educators.

7.75               Information is available on the AEC’s website regarding the conduct of elections. In the lead up to the 2010 federal election the Official Guide to the 2010 federal election went directly to households, and was available in a range of languages, Braille and a sound recording to make it accessible to a wide range of electors.

7.76               The AEC also conducts targeted programs to help specific groups more effectively exercise their voting franchise. Over many years these activities have included:

• providing printed How-to-Vote information in up to 21 different languages;
• including formality information in the Official Guide to the Election at federal elections;
• providing telephone translation services through specialist providers; and, more recently, providing special programs in targeted divisions in the lead up to an election;
• adapting press advertising for placement in Indigenous media, and placing translated press and television advertising in media for persons from non-English speaking backgrounds;
• directly mailing community organisations and groups, and migrant resource centres, with translated how to vote correctly fact sheets, posters and DVDs featuring translated television advertisements;
• employing staff with particular language skills in polling places in targeted divisions;
• providing extra training to polling staff in divisions with a high level of persons from non-English speaking backgrounds;
• translating the three questions issuing officers are required to ask of electors into 21 languages for use in targeted polling places;
• playing looped video information in targeted polling places in divisions with persons from non-English speaking backgrounds;
• providing information to Candidates through the Candidate’s Handbooks, advising candidates about minimising unintentional informality through the design of their How-to-Vote cards;
• providing an interactive How-to-Vote practice tool on the AEC website; and
• conducting school and community education sessions in areas with high levels of persons from non-English speaking backgrounds (with a view to providing information to students and other persons who speak English from such families in order that they may assist in passing information to parents).[52]

7.77               As discussed earlier in the chapter, the AEC pilot project in western Sydney sought to address the high levels of informal voting in those electoral divisions.

7.78               The Indigenous Electoral Participation Program (IEPP) provides targeted assistance to Indigenous electors and potential electors. The IEPP commenced in 2010 and is funded for four years as part of the Government’s Closing the Gap initiatives. It is a national program that aims to:

• increase levels of knowledge of democratic and electoral processes;
• increase levels of enrolment;
• increase levels of participation in democratic and electoral processes; and
• decrease levels of informal voting.

7.79               Ms Bright, the AEC State Manager for Queensland, outlined the work of the IEPP in Queensland, stating that:

Education sessions for Indigenous Queenslanders were conducted by staff of the AEC’s Indigenous Electoral Participation Program at some 18 very diverse locations around the state, from the Gold Coast right up to the Torres Strait, as far west as Cunnamulla and St George and right out west to Mount Isa. Staff also used these events to promote employment opportunities for Indigenous people at the upcoming election. In addition, an electoral awareness officer program was trialled throughout Queensland for the election. One hundred and fifty Indigenous officers were engaged to assist with educating electors and to provide information in communities about enrolling, the voting process and how to cast a formal ballot. In the Cape, the Torres Strait and areas of metropolitan Brisbane where these officers worked, the rate of informality decreased.[53]

7.80               The AEC Manager for the Northern Territory, Mr Pugsley, advised the Committee that IEPP staff were also in the field in the Northern Territory in the weeks leading up to the 2010 federal election.[54]

7.81               At the 2010 federal election, there was a decrease in voter turnout in the Northern Territory to 82.7 per cent.[55] Statistics for various mobile polling stations in the Northern Territory also revealed high levels of vote informality in some remote areas.[56]

7.82               However, Mr Pugsley was optimistic about the future impact of the IEPP program. He noted that the IEPP program had only been operating for a few weeks prior to the 2010 federal election being called.[57]

Committee conclusion

7.83               The Committee acknowledges the role that education must play in helping to address informality. Targeted programs conducted by the AEC such as the Indigenous Electoral Participation Program assisting Indigenous people, and the pilot project in western Sydney to assist voters from non-English speaking backgrounds, are valuable and should continue.

7.84               The Committee notes the range of education-based and other activities that the AEC has undertaken in recent years to try to decrease the levels of unintentional informality, especially in areas where there are high levels of persons from non-English speaking backgrounds. It is clear, however, that despite these efforts informality continues to rise.

7.85               The Committee believes that unintentional informal voting is a growing and complex problem that requires action beyond the improvements that education programs have been able to deliver. Opposition members of the Committee point out that there has been no extensive research as to what proportion of informal voting is deliberate and what is accidental. At the moment, the AEC merely assumes that all ballots that are numbered ‘1’ only are accidentally informal and those that are blank are deliberately informal. Clearly there needs to be more extensive research done. The Opposition members believe education programs do deliver better outcomes, particularly in states with optional preferential voting and believes that people are entitled to vote informally if they wish.

7.86               The Committee notes the 1983 amendments for reducing informality in the Senate. However, the increasing amounts of informal votes in House of Representatives elections means action is required to reduce these numbers before the House of Representatives is faced with the alarmingly high levels of the Senate prior to 1983. The Opposition members note that this is a false comparison as no House of Representatives ballot paper would ever resemble a Senate paper in terms of numbers of candidates nominating. It is also necessary to point out that Senate electors are still able to fill out every square if they wish.

Saving informal votes in the House of Representatives

7.87               The Committee notes the Australian Labor Party’s observation about addressing informality:

The AEC has made significant efforts to try to address it, but those do not seem to be really improving the situation. I think we do need to look at something a bit more fundamental.[58]

7.88               There are various options that could be explored to reduce the amount of informal votes. Mr Antony Green proposed that the Committee consider optional preferential voting, a system of progressive informality and the South Australian ticket voting system, as options for lowering the informal vote in the House of Representatives.[59]

Optional preferential voting

7.89               Some have suggested optional preferential voting as a way to reduce informality in the House of Representatives, as it provides voters with the flexibility to indicate as few or as many preferences as suits.[60] Groups such as the Electoral Reform Society of South Australia support optional preferential voting on the basis that it would make it easier for voters and fairer for ungrouped candidates, in the case of the Senate.[61]

7.90               Others object to an optional preferential voting system.[62] Not only opposing its introduction at the federal level, The Nationals advocated for greater harmonisation of voting systems, calling for full preferential voting to be standard at the federal and state and territory levels.[63]

7.91               In terms of addressing informality, Mr Andy Becker, a former South Australian Electoral Commissioner, noted that:

Optional preferential voting (OPV) does go some small way toward saving such votes but the rationale is not directed solely to that end. The main purpose of OPV is to enable a voter not to have to indicate a preference if he or she does not have one. The consequence is that a great many ballot papers exhaust in the process of preference distribution and take no part in the final distribution.[64]

Committee conclusion

7.92               The Committee has repeatedly considered the optional preferential voting system when proposed in the course of its various post-election reviews and other inquiries, but continues to support a system of full preferential voting at the federal level.

Progressive informality voting system

7.93               Elections analyst, Mr Antony Green, proposed a new voting system of ‘progressive informality’ that retains compulsory preferential voting, but relaxes the formality criteria.Under his proposed system ballot papers with valid first preferences would be admitted to the count and the ballot papers with incomplete preferences would only be excluded at the point that preferences were required to be counted.[65]

7.94               Mr Green outlined his counting procedure as follows:

1. Initial count admits any vote with a valid first preference. 
   
   
2. If one candidate has a majority of first preference votes, no further checks for formality are required on ballot papers admitted to the count under Step (1).
   
   
3. If preferences require to be counted to determine a winning candidate, the ballot papers of a candidate are re-examined for formality before they are distributed. Any ballot papers that do not have a valid next preference are excluded from the count. 
   
   
4. Having excluded some first preferences as informal, a check is made to determine that the leading candidate has not now reached 50% of the new formal total. If preferences are still required to determine the winner, proceed to step (5). 
   
   
5. Distribute preferences. Return to step (3) and determine if further distributions need to be undertaken.[66]

7.95               Under this model an elector’s vote is counted if their first preference candidate secures a majority of first preference votes. However, if further preferences are required, and the ballot paper has no other preferences marked, the vote will be deemed informal at that point.

7.96               In a situation where a candidate secures an absolute majority on first preferences, those incomplete ballot papers will count toward the result. The candidate will win the election; however, during the full distribution of preferences, those ballot papers (if any) that were deemed formal at the first preference count for the remainder of the candidates would be informal at the point where the candidate was excluded, if no further preferences were indicated.

7.97               Under that situation, ballot papers for the winning candidate may be treated differently to ballot papers for the other candidates.

7.98               Another relevant factor is that the number of ballot papers deemed formal at the first count, may be different to the number of ballot papers deemed formal at the end of the full distribution. It is likely, therefore, that the number of ballot papers counted toward the final result would be different to the number of ballot papers that counted toward the candidate being elected. Opposition members are in strong disagreement with this assessment, the idea that a vote which exhausts under an optional preferential system because the elector deliberately chooses not to preference certain candidates does not make their vote informal. Opposition members feel that pointing out that some votes may exhaust is a very poor argument against ‘progressive informality’ or optional preferential voting, and that saying that these votes are informal is very deceptive.

7.99               Similarly, the result achieved in a division in which a candidate achieves an absolute majority of first preference votes, would treat incomplete ballot papers in a different way than in a division where a full distribution of preferences is required to determine the result.

7.100           Mr Green acknowledged that while progressive informality gives effect to the intent of more voters than the current rules, there are disadvantages in that: it does disadvantage minor parties and independents as the ballot papers given their first preference are more likely to have to be excluded and thus become informal if full preferences are not indicated; and election night counts would be less reliable, as ballot papers with first preferences counted on election night may later need to be excluded.[67]

7.101           In evidence to the Committee, Mr Green argued that his progressive informality model struck the right balance between saving votes and respecting compulsory voting. He asserted that:

Under progressive informality, you cannot vote 1 and hope your vote will exhaust before it reaches somebody and have an effect on the count. If you voted 1 and your preferences need to be counted, your vote will still end up informal. But what I am arguing, which the AEC has argued since the 1987 informal vote report and which is evident if you look at the South Australian research in detail, is that for every vote we reject from the count to protect compulsory preferential voting—in other words, if somebody has not filled in all their preferences and those preferences are required—there are nine votes with a valid first preference that could have counted and did not need to have their preferences counted. So we have a very high test of formality to protect compulsory preferential voting when nine in 10 of those votes are not damaging compulsory preferential voting anyway.[68]

7.102           At the Committee’s request, the AEC considered Mr Green’s progressive informality model and what effect it could have had on House of Representatives informality in the 2010 federal election. The AEC concluded that of the 729 304 informal votes, 273 035 could possibly have been saved at the first preference count, with a further 85 724 potentially saved if ticks and crosses were also accepted, and potentially another 4 816 if other symbols such as alphabetic characters were accepted. However, a number of ballot papers may then be rendered informal if a distribution of preferences was required.[69]

7.103           The Committee also notes the following AEC observations about the progressive informality model:

• the progressive informality system and optional preferential voting systems are practically equivalent, sharing many of the same merits in minimising unintentionally formality;
• the system would be more effective than the current system in giving effect to first preferences expressed by voters;
• broadly, progressive informality would be simpler to implement than the South Australian ticket system; and
• the impact of the system on the clarity of election night results is likely to only be marginal and have less impact than uncounted declaration votes.[70]

7.104           Dr Brent of the Democratic Audit of Australia also commented on the similarity between optional preferential voting and progressive informality, suggesting that progressive informality was ‘really OPV with a bit of a semantic change’.[71]

Committee conclusion

7.105           The Committee supports a system of full preferential voting. It notes the similarities between progressive informality and optional preferential voting.

7.106           The Committee does not consider the progressive informality voting system proposed by Antony Green to be a viable option. In particular, the Committee is concerned about ballot papers being treated differently from one electoral division to another. It is not equitable to say that in the event an elector has voted with a single ‘1’, that if the elector’s first preference happens to be for a candidate that received an absolute majority then it can be treated as formal, but if their first preferences is for another candidate and additional preferences are required then the ballot paper will be deemed informal.

7.107           The Committee believes that the progressive informality system could potentially be more confusing, for voters and election officials administering the count, than the current system that the Committee is seeking to improve.

7.108           The Committee also believes that, wherever possible, ballot papers deemed to be formal should be treated consistently, both within and across electoral divisions. Progressive informality, unlike the South Australian savings provision dealt with below, does not treat ballot papers deemed formal in an equal and consistent manner.

South Australian ticket voting

7.109           The South Australian voting system is full preferential, but has a savings provision (SA ticket voting) for the House of Assembly which permits candidates to lodge one or two preferences ticket, and provides that some ballot papers may be rendered formal if a voter has only indicated some preferences on their ballot paper where the preferences indicated are consistent with the ticket or tickets lodged.

7.110           In the 2010 South Australian state election, 32 638 House of Assembly votes were saved by South Australia’s ticket voting system. In the two previous elections in 2006 and 2002, 43 553 and 37 897 votes were saved.[72]

7.111            The Committee received evidence that at the 2010 state election the informal vote in South Australia, which was 3.3 per cent under ticket voting provisions, would have been 6.5 per cent had federal rules applied.[73]

7.112           In the South Australian House of Assembly, provided a ballot paper is not marked in a way that identifies the voter, it is formal if:

• consecutive preferences are indicated against the names of all candidates, commencing with a ‘1’, a tick or a cross;
• consecutive numerical preferences commencing with a ‘1’ (or a tick or a cross) are indicated against the names of all candidates except one; or
• the previous criteria are met, and the elector has corrected an error (for instance by crossing out and renumbering), but the correction still leaves the elector’s intention clear.[74]

7.113           South Australian House of Assembly ballot papers are informal if:

• it is marked in a way that positively identifies the voter;
• it is blank or if no first preference is indicated by either a ‘1’, a tick or a cross;
• more than one first preference is indicated (i.e. if a ‘1’, tick or cross appears in or against two or more squares);
• there is a break in the consecutive numerical preferences, a duplication, or two or more preferences are omitted; or
• the ballot paper was not placed in a ballot box, even if it meets other formality requirements.[75]

7.114           Uniquely, a savings provision in section 93 of the Electoral Act 1985 (SA), allows some votes to be saved in certain circumstances. Under the South Australian ticket voting system, candidates are entitled to lodge one or two voting tickets.[76] An incomplete ballot paper may then be saved when:

• a first preference has been marked (with a ‘1’, tick or a cross) for a candidate who has lodged one or more tickets; or
• a first preference and some—but not full—preferences have been marked that are consistent with the ticket(s) lodged.

7.115           If these requirements are met, then the vote can be saved and preferences will be allocated according to the voting ticket(s) with which it is consistent. In cases where there is only a first preference indicated and two tickets are lodged, subsection 93(3) provides that:

...then the ballot paper is to be grouped with other ballot papers marked in the same manner and—

3. if the number of those ballot papers is an even number—half of them will be taken to have been marked in accordance with one ticket and half in accordance with the other; or
4. if the number of those ballot papers is not an even number—
   
   
   1. one of the ballot papers will be taken to have been marked in accordance with whichever of the 2 tickets is determined by lot by the returning officer; and
   2. half the remainder (if any) will be taken to have been marked in accordance with one ticket and half in accordance with the other.[77]

7.116           In cases where a first preference and some further preferences are indicated, subsection 93(5) provides that:

5. Where—
   
   
   1. a voter marks a ballot paper by placing the number 1 in the square opposite the name of a particular candidate and proceeds to indicate further preferences by consecutive numbers; and
   2. there are 2 voting tickets registered for the purposes of the election in relation to the candidate; and
   3. the preferences indicated by the voter are consistent with one or both of those voting tickets; and
   4. the ballot paper would, apart from this subsection, be informal,

   the ballot paper, if consistent with both voting tickets, will be treated as if it had been marked only with the number 1 and dealt with in accordance with subsection (3), but if it is consistent with one only of the voting tickets, it will be taken to have been marked in accordance with that voting ticket.[78]

7.117           This means that if the preferences indicated on an otherwise informal ticket deviates from the preference on the ticket(s) lodged by the candidate for whom first preference is marked it will not be saved by the ticket voting provision.

7.118           On polling night, the votes that may potentially be saved are still counted as informal at the polling place, with the savings provision subsequently applied by Returning Officers once they are satisfied that the necessary requirements have been met, specifically, that the candidate has lodged one or two tickets and that the preference(s) indicated are consistent with one or both of those tickets.[79]

7.119           The South Australian Electoral Commissioner acknowledged that the ticket voting requirements do represent an additional administrative challenge for election workers. Ms Mousley noted that:

You will find that the voting ticket provisions are somewhat difficult for polling place staff to understand. Primarily, they are employed once every four years and, whilst some staff have problems with preferences, particularly with the number of candidates, the higher the number, just following the preferences and sorting to informal or formal, is problematic in itself. With the voting ticket provisions we advise our polling place managers on the night of election itself that a fully preferential vote is sorted to formal and anything that is not fully preferentially marked is put out to informal.[80]

7.120           The likelihood of ballot papers with some preferences being saved by tickets reduces as more preferences are indicated. The Electoral Commission of South Australia (ECSA) confirmed that:

If the partial numbering does not match a voting ticket it will fail in formality. We do some analysis on informality. Unacceptable preferencing is around 21.8 per cent. The great majority of informals are blanks or messages.[81]

7.121           Mr David Gulley of the ECSA, advised the Committee that the majority of South Australian House of Assembly votes saved by their ticket voting system are first preferences only (marked with a ‘1’, tick or a cross). He observed that:

In our election reports we have reported a figure of less than 0.1 per cent of total formal ballot papers are partially preferenced where they fail the ticket, so the great majority of ticket votes are single 1s, ticks or crosses. The highest figure that I could find was in the 2002 report where it referred to the previous election. In the 2002 election it was less than 0.1 per cent of the formal vote, but 0.6 per cent in 1997.[82]

7.122           A vote cannot be saved unless a candidate has lodged at least one preference ticket with the ECSA. Candidates can choose not to lodge tickets.

7.123           The AEC advised that in the 2010 South Australian election, there were 40 candidates who did not lodge a voting ticket (28 Greens, 9 Fair Land Tax – Tax Party, and three independent candidates).[83] As Table 7.6 illustrates, the number of candidates in 2010 choosing not to lodge tickets was unusually high compared with previous years.

Table 7.6 Number of candidates who lodged no, one or two tickets for the 2010 South Australian election

Year	Number who lodged no voting ticket	Number who lodged a single voting ticket	Number who lodged two voting tickets
1993	6	148	73
1997	2	94	101
2002	2	244	56
2006	1	157	111
2010	40	206	7

Source Australian Electoral Commission, Submission 87.4, Attachment B, p. 12.

7.124           Despite this unprecedented rise in candidates not lodging tickets in 2010, discussions during the inquiry revealed that, generally, having first preference votes for them saved is a good incentive for parties and candidates to lodge tickets.[84]

7.125           Concerns about this model of ticket voting being seen as a move away from full preferential voting are addressed in section 126 of the South Australian Electoral Act 1985, which prohibits advocating forms of voting other than full preferential voting.

7.126           The provision prohibiting advocating methods of voting other than full preferential helps protect against problems encountered in the past.[85]

7.127           It is pertinent for the Committee to consider the issues associated with Langer-style voting, as they highlight how individuals or groups, for a variety of reasons, may seek to subvert a savings provision aimed at reducing informality.

7.128           The background to this is that as part of the major electoral reforms of 1983, a safety net was introduced for ballot papers where full preferences had been indicated, but a sequencing error had occurred. It provided that on ballot papers on which all squares were numbered, if there was a mistake in the sequence and numbers were repeated, for example 1,2,3,4,5,5,6,7,8, then the ballot paper would be formal, and the preferences would remain valid until the point at which the error occurred.

7.129           In the lead up to the 1996 federal election a political activist, Mr Albert Langer, encouraged electors to deliberately make sequencing errors on their House of Representatives ballot papers, so as to deny major parties preferences, but still have the elector’s desired preferences counted. Mr Langer’s activities included an advertisement in The Australian with instructions on how to cast their vote in this way. This style of voting became known as Langer voting.

7.130           The AEC took legal actions against Mr Langer, as his activities were in breach of the Commonwealth Electoral Act. At the time, section 240 of the Act provided that a voter must indicate their first preference with a ‘1’ and then number all the remaining squares, and section 329A made it an offence (with a maximum penalty of six months imprisonment) to encourage voters to vote other than in accordance with section 240.

7.131           This matter and associated issues were heard in the Victorian, Federal and High Courts. The outcome was that Mr Langer served three weeks of a ten week sentence, and the constitutional validity of section 329A was upheld.[86]

7.132           In 1998, following consideration of this issue in the inquiry into the 1998 federal election, section 240 of the Commonwealth Electoral Act was subsequently amended to explicitly provide that numbers could not be repeated, and the savings provision allowing the repetition of numbers was removed.

7.133           The Langer experience is a useful cautionary tale that savings provisions must be carefully designed to strike a balance between seeking to better give effect to voters’ intentions and upholding the system of full preferential voting.

7.134           During its hearings, the Committee considered the possible effect of Langer voting on a system based on SA ticket voting. The ECSA confirmed that under the SA savings provision a Langer voter would be informal.[87]

7.135           While the South Australian Electoral Act 1985, section 126, prohibits the advertising of the ticket voting provision, the reality is that a campaigner would not benefit from conducting a Langer-style campaign.

7.136           Unlike the sequencing savings provision that was in place at the 1998 federal election, which Langer sought to exploit, the ticket voting system is still a full-preferential voting system. If an elector only indicates one or two preferences and their ballot paper is saved, their vote will still be full-preferential in keeping with the ticket lodged by their first preference candidate. It cannot be manipulated to become a de-facto optional preferential system.

7.137           Former South Australian Electoral Commissioner, Mr Tully, acknowledged that while administering a provision that is not widely known was ‘tricky’, the amount of votes saved made it worthwhile. He observed that:

The complexity of having what some people regarded as a secret provision was often a bit challenging to explain but, nonetheless, its impact was significant. My recollection in two elections that I conducted in South Australia—both were extraordinarily close—was that ticket votes accounted across the state for about four per cent. In other words, if there were not ticket votes informality would have increased by a further four per cent.[88]

7.138           However, the fact that the operation of the SA ticket system in the House of Assembly is not widely publicised, and by extension, not widely known, could give rise to concerns about the extent to which voters are aware that their ballot papers are being deemed formal and preferences distributed as per their first preference candidate’s voting ticket.

7.139           While having acknowledged that the SA ticket system does have some positive input into the democratic process by saving unintentional informal votes of voters confused by the different voting systems,[89] Emeritus Professor Dean Jaensch argues that ticket voting makes too many assumptions and turns the informal ballot into a formal one, without asking the permission of the voter concerned.[90]

7.140           In its analysis, the AEC similarly made the distinction between the Senate group ticket system—which allows the voter to make the choice to vote above-the-line and have their preferences distributed as per the group ticket or to vote below-the-line and indicate their own preferences—and the South Australian system that does not explicitly provide voters with a choice.[91]

7.141           The AEC observed that the effect of potentially allocating preferences that were not intended by the voter is to some degree mitigated by the fact that further preferences will not necessarily need to be drawn on. The AEC commented that:

...[in most cases] the second and later preferences on many ballot papers never need to be revisited after the formality check. In particular, many voters in seats where a preference distribution is not needed to determine the result, and many voters in other seats who cast first preference votes for candidates not excluded during the count (e.g., most major party candidates), would be significantly advantaged by the adoption at the federal level of the South Australian system, since in many cases their first preference votes would be counted, and the later preferences attributed to them due to the operation of the voting ticket system either would not be counted, or would not change the result in the seat.[92]

7.142           Elections analyst, Mr Antony Green, advised that:

The South Australian provision almost overwhelmingly captures one-only votes. There are a small number of cases where someone has gone 1 and 2 and it gets saved, but the vast majority are people who have just voted 1, and we know from past research that between a third and a half of the ballot papers are from people who just voted 1. Of course, it is higher in New South Wales and Queensland because of the experience of optional preferential voting at the state level.[93]

7.143           In the context of the 2010 South Australian election, Mr Green found that:

Of the 32,638 ticket votes admitted to the South Australian count, only 2,020 or 6.2% would have been required to have their preferences examined to determine the winning candidate in a contest. The other 93.8% could have been formal based on their first preferences because further preferences did not need to be examined to determine the winner.[94]

7.144           Mr Andy Becker, a former South Australian Electoral Commissioner, was involved in the development and implementation of the SA ticket voting system. Mr Becker has a long professional history in the administration of elections in South Australia and nationally.

7.145           He informed the Committee that the rationale for the savings provision was that:

...with a compulsory system every effort should be made to make it as easy as possible for an elector to comply with the legislation and in doing so be as effective as possible in casting a meaningful vote.[95]

7.146           The SA ticket voting savings provisions that emerged were administrator-driven rather than government or politically driven. Mr Becker recalled that in 1985 the drafting instructions provided by the SA Government at the time were ‘fairly loose’ and he, as Electoral Commissioner, working in conjunction with the Parliamentary Counsel, had considerable latitude in developing South Australia’s new Electoral Act.[96]

7.147           Mr Becker outlined for the Committee the emergence of ticket voting as the preferred savings option. He recollected that:

We then went and reviewed a lot of the legislation interstate and some of the provisions that they had like the safety net provisions in New South Wales for optional preferential voting. That in itself did not appeal to me greatly because it seemed to me that, in many cases in New South Wales in the lower house, they were electing people with fewer than 50 per cent of the vote. So that is when the alternative of putting in voting tickets came up for consideration. What that really enabled us to do was to separate out the instructions that we give to the elector from the instructions that we give to the scrutineer people performing scrutiny. Everything up to and including polling day is saying you shall vote—number every square and do not leave any blank. There are safety provisions saying if you leave the last square blank that that is still considered a preference. However, we then said when it comes to the scrutiny we put in the safety provision in exactly the same way as we have for the Senate that, in the event of a voting ticket having been lodged, we should give effect to that ballot paper that might have a tick, a cross or a 1 and a voting ticket has been lodged for that tick, cross or 1, and we will give effect to that ballot paper in accordance with the voting ticket.

Generally speaking that was considered fairly well by all sides. There was a heck of a lot of debate about it but it passed in the end.[97]

7.148           In evidence to the Committee, Mr Becker advised that twenty-six years later the provisions had served South Australia well for seven elections, and that he was not aware of any objection to, nor any attempt to amend, the legislation covering voting tickets.[98]

7.149           That has also been the experience of South Australian Electoral Commissioner, Ms Mousley, who advised that:

I have only been commissioner in South Australia for the 2006 and 2010 elections and I cannot see any evidence from the election reports where there has been any record of concerns that were raised throughout the election because of that provision.[99]

7.150           The South Australian Deputy Electoral Commissioner, Mr David Gulley, also shared that view, telling the Committee that:

I have been with the commission since 1995 and deputy commissioner since 1998. I am not aware of any great concerns with the process, other than some punters out in the community think it is undemocratic, as they do with other things.[100]

7.151           Senator Nick Xenophon, a Senator for South Australia, similarly found that:

From my observation, the South Australian system—and I may be wrong in terms of the comments of the Liberal Party and the Labor Party; I do not know what they commented—has not been controversial. In my 10 years in the legislative council, I am not aware of a push to change that system.[101]

7.152           Table 7.7 is a comparison of acceptable number on ballot papers in the Senate, House of Representatives and the South Australian House of Assembly.

Table 7.7 Comparison of acceptable numbering on ballots papers in the Senate, House of Representatives and the South Australian House of Assembly

	Acceptable on ballot paper for the:
Ballot paper marking(a)	Senate(b)	House of Representatives	South Australian House of Assembly (Ticket voting)
Single first preference (figure ‘1’)	ATL: Yes BTL: No	No	Yes If candidate lodged ticket(s)
Single first preference (figure ‘1’, a tick or a cross)	ATL: Yes BTL: No	No	Yes If candidate lodged ticket(s)
Incomplete numbering (consecutive preferences beginning with 1 are shown, but the last square is left blank)	ATL::Yes BTL: Yes, the blank square is deemed to express the voter’s last preference.	Yes, the blank square is deemed to express the voter’s last preference.	Yes If candidate lodged ticket(s) and is consistent with one or both of the tickets
Incomplete numbering (consecutive preferences beginning with 1 are shown, but more than one square is left blank)	ATL: Yes BTL: Yes, if certain requirements are met (CEA s. 270(1)(b))	No	Yes If candidate lodged ticket(s) and is consistent with one or both of the tickets
Number sequence errors (missed numbers)	ATL: Yes BTL: Yes, if certain requirements are met (CEA s. 270(1)(b))	No	No
Repeated numbers	ATL: Yes BTL: Yes, if certain requirements are met (CEA s. 270(1)(b))	No	No

(a)   The Commonwealth Electoral Act (CEA) prescribes the ballot paper formality requirements for federal elections.

(b)   The Senate’s two forms of voting are above-the-line (ATL) and below-the-line (BTL).

Source Australian Electoral Commission, Scrutineers Handbook: Federal election 2010, Version 3; and the Electoral Commission of South Australia, Scrutineers Handbook, R018, 2010.

7.153           At the Committee’s request, the AEC undertook an analysis of the House of Representatives informality under the current system, and the potential impact on informality under the proposed options of progressive informality and the SA ticket voting provision.[102] Accordingly, the AEC made certain assumptions, including that: all candidates lodge tickets; ballot papers do not identify voters; ballot papers match tickets; and it relates to the first count only.[103]

7.154           From this analysis, it appears that the South Australian ticket voting system, if applied to House of Representatives ballot papers, could save a significant portion of informal votes. For the 2010 federal election, this could have been as much as 42.12 per cent (307 156 votes), assuming that all the relevant candidates had lodged tickets.

Navigation: Previous Page | Contents | Next Page

Back to top