One of the most muddled aspects of the recent South Australian election was reference to lower house ticket votes.
I’d started to write a post on the subject referring to an Electoral Backgrounder published by the Electoral Commission of South Australia (ECSA) in 2020. The backgrounder explains how the system works, provides background statistics on its use, and summarises the arguments for and against the provision. It’s difficult to link to the backgrounder so I’ve made it available on my site here.
The backgrounder references an article I published on the subject as part of the ABC’s 2006 South Australian election site. I barely remember it and the old election sites are no longer available. But digging around on the ABC’s old content server, I found the original article.
I intend to cover the subject in more detail tomorrow with statistics on how many votes are saved by the ticket voting system. But until then, I’m re-publishing my article from 2006. I’ve added a few notes in the text where the rules have been changed since 2006.
Unique Features of the South Australian Electoral System (Published 2006)
There are several features of South Australian elections that differ from procedures used at elections for the Commonwealth and in other states. These relate to instructions on the ballot paper, the way Independents appear on the ballot paper, and the use of preference tickets to assist voters and to ‘save’ votes that would otherwise be declared informal.
Despite attempts by the previous Liberal government to introduce voluntary voting, compulsory enrolment and voting still applies at South Australian elections. Initial enrolment on the state roll is voluntary, but as it is compulsory to enrol for Commonwealth elections, most people enrol for both. Once a voter is on the state roll, enrolment becomes compulsory. (2026 update – the rules are now consistent for both elections.)
While voting is compulsory, South Australian ballot papers are unique in including an instruction that informs the voter that they do not have to mark the ballot paper. This results in South Australian elections having the country’s highest proportion of blank ballot papers. At the 2002 state election, 42.3% of all informal ballot papers were left entirely blank, compared to only 21.4 per cent at the 2001 Federal election.
Like other states and the Commonwealth, South Australia maintains a register of political parties, and candidates of registered parties have the party name appear on the ballot paper. Party names can consist of up to six words. Uniquely in South Australia, Independents may also take advantage of this provision, allowed to appear on the ballot paper with the word ‘Independent’ followed by up to five words. (2026 update – this provision was shortened to three words in 2014 and abolished ahead of the 2026 election. Only the word “Independent” was used in 2026.)
At the 2002 state election, an array of Independents nominated, including ‘Independent Labour’, ‘Independent Liberal’, ‘Independent for Voluntary Euthanasia’, ‘Independent Youth Empowerment Save Live Music’, ‘Independent No Nuke Dumps No Paedophiles’ and ‘Independent Nuclear Free Green Granny’.
The final unique feature of South Australian elections concerns the use of ticket votes in the House of Assembly to assist voters in numbering preferences, and also in providing a safety net for informal votes.
Like all Australian lower house elections, elections for the South Australia House of Assembly are conducted using preferential voting, with electors numbering candidates on the ballot paper in the order they would wish to see the candidates elected. Electors must indicate preferences for all candidates on the ballot paper if they wish their vote to be counted.
If on the count of primary or first preference votes (votes marked with the number ‘1’), no candidate achieves more than 50% of the vote, the candidate with the fewest votes is excluded and their votes distributed according to the next available preferences, their 2nd or third choice candidate. This process of exclusion continues until one candidate achieves 50% of the vote. It is now standard practice to carry out a full distribution of preferences for information purposes, whether or not the count is actually required.
Some Australian jurisdictions allow optional preferential voting (New South Wales, Queensland, Tasmania and the ACT), but South Australian voters must number all squares on the ballot paper. One of the consequences of this compulsion is an increase in the incidence of spoiled or informal ballot papers.
However, South Australia uses a unique system of ticket voting in the House of Assembly which results in the state having the lowest informal vote of any jurisdiction using compulsory preferential voting.
All candidates contesting the House of Assembly are entitled to lodge either one or two ‘ticket’ votes. These ticket votes serve two purposes.
First, all ticket votes for an electoral district are displayed on the voting screens in each polling place. Many voters will have received how-to-vote material outside of the polling place. However, for those that did not, or who want to vote for a candidate not distributing how-to-vote material, the displayed ticket votes are provided to assist voters in casting a formal ballot.
(2026 update: reading this 20 years later, I realise I’m not describing the tickets correctly. The above paragraph should refer to registered how-to-votes. The registered tickets do not always correspond to the how-to-vote, especially on how-to-votes where the candidate recommends a 1-only vote with text saying to number all the squares. Tomorrow I’ll return to this subject and how tickets worked in 2026.)
Second, the ticket votes are also used as a ‘savings’ provision for votes that would otherwise be informal. Evidence at past Federal elections has shown that between two-thirds and three-quarters of the informal vote is caused by numbering errors. In some electorates, around half of the informal vote consists entirely of ballot papers that have been marked with a single ‘1’. Comparing informal voting votes in different states, it seems clear that the use of group ticket or ‘above the line’ voting in the Senate and state Legislative Councils induces voters to use a single ‘1’ on their lower house ballot paper.
To explain how the ticket votes are used to ‘save’ informal votes, the following table illustrates some sample ticket votes and ballot papers.
| Registered tickets | Example Ballot papers | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ballot Paper | Ticket 1 | Ticket 2 | 1 | 2 | 3 | 4 | |||||||
| Smith | 4 | 3 | (blank) | (blank) | (blank) | 3 | |||||||
| Jones | 3 | 4 | (blank) | (blank) | (blank) | (blank) | |||||||
| Nguyen | 5 | 5 | (blank) | (blank) | 3 | (blank) | |||||||
| Sgroi | 1 | 1 | X | 1 | 1 | 1 | |||||||
| Anderson | 2 | 2 | (blank) | 2 | 2 | 2 | |||||||
| Status of vote | Formal | Formal | Informal | Formal | |||||||||
In example ballot paper 1, the ballot has been marked with a single ‘X’, a vote which is taken to imply a single first preference. This vote would be informal at Commonwealth elections, but in South Australia is ‘saved’ by the ticket vote and deemed to have the ticket vote’s list of preferences. In this case, because candidate Sgroi has lodged two ticket votes, the total of all ballot papers saved by this provision is split evenly between the two tickets.
Example 2 is also saved by the ticket vote as the first two preferences match the lodged ticket. Example 3 is informal as the third preference has deviated from the registered ticket and therefore cannot be implied to have the ticket vote. Example 4 is more complex, matching one of the registered tickets and therefore implied to have one of the lodged tickets.
It should be stressed that the vast majority of votes saved by this provision are votes on which a single preference has been marked. Ballot papers with duplicate or missed preferences are far less likely to be saved.
The results of this provision is a halving of the informal vote. The following table shows the incidence of informal and ticket voting at South Australian elections since the provision was introduced in 1985.
| 1985 | 1989 | 1993 | 1997 | 2002 | |
|---|---|---|---|---|---|
| % Informal Vote | 3.5 | 2.8 | 3.1 | 4.0 | 3.1 |
| % Ticket Votes | 4.1 | 6.0 | 5.9 | 4.9 | 4.0 |
Ticket vote percentages calculated as a proportion of formal votes. These figures are slightly higher than similar tables calculated as a percentage of the total vote.
The informal percentage is calculated as a percentage of the Total vote, ticket vote as a percentage of the Formal vote. All of these ticket votes would have been informal at Commonwealth elections. There were a total of 30,537 informal votes at the 2002 South Australian election. Another 37,897 votes were ‘saved’ by the ticket voting provisions.
It needs to be made clear that these lower house ticket votes are very different to upper house ticket or ‘above the line’ votes. In the upper house, more than 95% of votes are ticket votes because parties actively encourage the use of the just vote ‘1’ above-the-line option. In the lower house, there is no ticket voting box, voters are encouraged with how to vote material to direct full preferences, and the ticket vote ‘saving’ provision only comes into play when a ballot paper has insufficient valid preferences.
I am an ACT voter and it took a bit of time to get my head around the complexities of the SA voting system. Or do I? It seems such a complex system with ballot votes needing to match registered tickets of candidates.
I think I’ll stick with the ACT’s voting system where voter’s can decide who to vote for with the option of leaving boxes blank and not being obliged to follow party lines.
Or do I have the wrong end of the stick?
COMMENT: Voting in SA is the same as other states – simply number all the squares in a correct ordered sequence. The ticket are a savings provision, designed to allow certain otherwise informal votes to count. This provision is not something for voters to worry about, just counting officials and scrutineers.
I’m absolutely in favour of saving votes that aren’t formal if there is a clear preference expressed amongst the relevant candidates, but the savings provision distributing in line with the party’s preferences is somewhat undemocratic (albeit unlikely to make much of a difference the vast majority of the time, and not nearly as much of an issue as the old Senate ticket voting). According to the ECSA the total number of votes saved in 2022 was 31,786 (pretty similar to 2002, about 800 per seat).
Unless there is a strong trend towards saving one candidate’s votes over another’s (e.g. if votes for one candidate were more likely to be saved because of how the ballot paper was shown on the HTV card) it would be unlikely to end up with much of a difference on first preferences (maybe 20-30 votes between the last two candidates), and genuinely reflecting the will of the voters.
For working out how it affects distributing preferences there are so many assumptions about how preferences would have flowed without the savings ticket (and if they would be different between different party’s) that it’s basically a guess, but I can’t imagine it would be more than another 20-30 votes difference in the final count between how the votes are actually distributed and how voters are likely to have distributed the preferences themselves (if they had filled in the ballot paper properly).
Overall probably only making any difference for seats with a final margin under 50 votes.
Antony: Do you know if there is any data released on saved votes and who the first preference is for?
COMMENT: It is absolutely made available. In the past there has been little difference by party in the defaut use of tickets. Data is normally included in SA election statistical returns though for technical reasons the data is not available for 2022.
I’ll have a post up on the 2018 data tomorrow. 70% of the saved votes were for a candidate in the top two so their preferences were never considered. Another 6% were in seats where candidates won with over 50% so preferences were not relevant. Only 24% defaulted to preferences that were required to be counted, representing 0.9% of the formal vote.