Members of the Assembly are now elected by the Hare-Clark proportional representation system. Canberra is divided into five electorates—Brindabella, Ginninderra, Kurrajong, Murrumbidgee and Yerrabi. Five members are elected in each electorate, making a 25 member Assembly. Members are elected for a fixed four-year term and elections are held every four years on the third Saturday in October.
Votes are counted according to the preferences indicated on the ballot paper. To be certain of election under the Hare-Clark system, a candidate must obtain a set proportion (or quota) of the votes for his or her electorate. Any candidate whose votes equal or exceed a quota is elected. A quota is calculated using the following formula:
The quota for each of the ACT’s five-member electorates is one-sixth, plus one, or roughly 16.67% of all votes cast.
The modern Hare-Clark system is based on methods developed by Englishman Thomas Hare (1806-1891) and Tasmanian Andrew Inglis Clark (1848-1907). Introduced in ACT Assembly elections in 1995, it was first used in Tasmania in 1897 and is still used for Tasmanian House of Assembly elections today.
The Hare-Clark system is a kind of proportional representation known as a ‘single transferable vote’ system. Each voter is given one vote, which can be transferred from candidate to candidate according to preferences shown by the voter. To vote, voters simply number preferences for candidates by writing 1, 2, 3, 4, 5, 6, 7 and so on against the names of candidates on the ballot paper.
After polls close, every number ‘1’ vote is counted. The quota is calculated after all the valid first preference votes have been counted. Candidates who record enough number ‘1’ votes to gain a quota are elected. Votes received by elected candidates, which are surplus to the quota, are allocated to other candidates according to voters’ preferences. After that, the candidate with the fewest votes is excluded from the count and his or her votes are allocated to other candidates according to voters’ preferences. This process of first transferring surplus votes and then excluding the candidate with the fewest votes continues until all vacancies are filled.
Another feature of the ACT electoral system is the Robson Rotation method of rotating the positions of candidates within each column on the ballot paper. This minimises the impact of ‘Donkey votes’ where all boxes are numbered in consecutive order down the ballot paper (although donkey votes are a valid way to vote, they advantage candidates at the top of the ballot paper and disadvantaging those at the bottom).
The ACT also has a ban on canvassing and the distribution of how-to-vote cards within 100 metres of a polling place. This gives voters the opportunity to consider each candidate on their own merits and to allocate their preferences for individual candidates free of guidance or influence from third parties.
Under the Hare-Clark system, vacancies arising from the resignation, retirement or death of a member during the term of an Assembly are filled (if possible) by recounting the ballot papers that were received by the vacating member, to determine the candidate favoured next by voters. Only candidates who contested the original election, and who indicate that they wish to contest the casual vacancy, may participate in this process.
If it is not possible to fill a vacancy using this method (for example, if no candidates from the election wish to contest the vacancy), the Legislative Assembly chooses a person to fill the vacancy. If the vacating member was a member of a registered political party, the new member must be a member of the same party.
If there is no available member of the relevant party, or if the vacating member was elected as an independent, the person chosen to fill the vacancy cannot have been a member of a registered political party within the 12 months preceding the filling of the vacancy.
Members of the first two Assemblies were elected in 1989 and 1992 under a modified d’Hondt system of voting before changing to the Hare-Clark system mentioned above.
Modified d’Hondt is a combination of a European party list system of proportional representation known as the d’Hondt system, the Australian Senate system of proportional representation, and various preferential voting methods that enable voters to place votes for candidates and parties, both within and across party lines. The electoral system, part of the federal self-government legislation passed in 1988, proved to be complex and time consuming.
A referendum held in conjunction with the 1992 election, gave voters the opportunity to choose a new electoral system for the ACT. The choice was between a single-member electorate system and the proportional representation (Hare-Clark) system.
A clear majority, 65.3%, favoured the Hare-Clark system. The Electoral Act 1992 established the new Hare-Clark electoral system for the ACT. A second referendum was held simultaneously with the 1995 election to formalise certain principles of Hare-Clark.
The referendum was successful with a majority (65.0%) of voters agreeing with the proposal to entrench Hare-Clark. As a result, any law that is inconsistent with the entrenched principles can only be passed by a two-thirds majority of the Legislative Assembly (17 Members) or by a majority of electors at a referendum.
The ACT’s electronic voting system is the first of its kind to be used for parliamentary elections in Australia and was introduced at the October 2001 election. The electronic voting system is used in selected pre-poll voting centres, which are open for three weeks before polling day. Electronic voting is also available at selected polling places on election day, where voters are given a choice of voting either electronically or on paper.
For more information visit the Elections ACT website.
The Electoral map used in the 2016 Election is also available from the Elections ACT website.