Like what you've read?

On Line Opinion is the only Australian site where you get all sides of the story. We don't
charge, but we need your support. Here�s how you can help.

  • Advertise

    We have a monthly audience of 70,000 and advertising packages from $200 a month.

  • Volunteer

    We always need commissioning editors and sub-editors.

  • Contribute

    Got something to say? Submit an essay.

 The National Forum   Donate   Your Account   On Line Opinion   Forum   Blogs   Polling   About   
On Line Opinion logo ON LINE OPINION - Australia's e-journal of social and political debate


On Line Opinion is a not-for-profit publication and relies on the generosity of its sponsors, editors and contributors. If you would like to help, contact us.


RSS 2.0

Above or below the line? Managing preference votes

By Antony Green - posted Wednesday, 20 April 2005

Before new electoral rules were introduced in 1984, elections for the Australian Senate were marred by rates of informal voting that sometimes rose above 10 per cent. Voters were forced to indicate preferences for every candidate on gigantic ballot papers with no party names, the process a recipe for error and confusion.

The 1984 legislative changes simplified voting by adding party names to the ballot paper and also introduced the current ballot paper divided by a thick horizontal line. Electors can now vote for a single group ticket voting square “above the line”, or vote by marking preferences for every candidate “below the line”. At the 2004 Senate election, 96 per cent of mainland electors voted above the line, a lower 81.2 per cent in Tasmania where state elections have given voters more experience with proportional representation.

Parties with a group voting square have lodged one, two or three group ticket votes showing preferences for all candidates on the ballot paper. Electors voting for a party using the group ticket voting option effectively cede their right to distribute preferences, ballot papers defaulting to the preference ticket of the selected party.


The good that has come from ticket voting is the dramatic fall in Senate informal voting. But the acceptance of group ticket voting has involved a trade-off. With the vast majority of ballot papers now tied to party tickets, increasingly Byzantine preference deals are being engaged in by political parties in an attempt to engineer election results. A democratic deficit has developed, with serious questions as to whether the results engineered by group ticket voting truly represent the will of the electorate.

The democratic deficit is clear when you look at the choices faced by voters. Group ticket voting produces the ridiculous situation where voters are forced to choose between voting above the line for a party ticket they don’t know, can’t find out about and probably wouldn’t understand if they could, or to vote below the line giving preferences to a vast array of candidates they don’t know and don’t care about just to have their vote count for the smaller number of candidates they do know.

Consider the choices faced by New South Wales voters at the 2004 Senate election. A well informed voter could have examined the preference deals on the Electoral Commission’s website, except the booklet on NSW preferences was a pdf file of 1.6 megabytes, unusable without broadband internet connection. Alternatively, electors could try examining the Group Ticket Voting booklet available in polling places. Assuming the staff knew what the voter was talking about, one of these 80 page booklets would be available for every thousand voters expected at a polling place. Even then, an elector would have needed an intimate knowledge of the electoral system to understand this booklet. Even for those with inside knowledge, some of the NSW preference deals were unfathomable.

If our informed voter passed these hurdles and decided they didn’t like the preference deal of their party, what could they do? In NSW, the only choice was to number preferences for all 78 candidates listed below the line. Party “how to vote” material is of no assistance, only recommending an above the line vote. Even the best informed voter would be lucky to know half the candidates and parties on the ballot paper. To cast a formal vote, the only option to complete a ballot paper would be some form of random process to allocate further preferences.

Yet the democratic deficit goes beyond the choices faced by voters. The tight exchange of preferences made possible by group ticket voting is distorting the preference choices of the political parties. The power over preferences delivered by group ticket voting is leading to strategic preference deals that owe more to the academic field of game theory than political ideology.

The Senate’s electoral system is one type of “Proportional Representation by Single Transferable Vote”, or PR-STV for short. It is not proportional representation based on the primary vote, as seen every where else in the world, but operates on a complex interplay of primary votes and preferences. Primary votes determine the number of Senators each party elects with full quotas in the initial stages of the count. The final vacancies are filled by the distribution of preferences between candidates remaining with partial quotas.


In all other systems of proportional representation, a party’s sole interest is to maximise its own vote. Under PR-STV, a party has two interests, maximising its own vote, but if one of its own candidates cannot be elected, there is a second interest in influencing which other parties win the final vacancies on preferences.

PR-STV counting is mind numbingly complex, but the simple strategy for a voter to follow is to simply number all the candidates in the order they would like to see them elected. Any attempt to vote tactically could fail, because it requires detailed knowledge of how many quotas each party will fill, and also in what order each party will finish after all candidates with full quotas have been excluded from the count.

Political parties have more knowledge of the order each ballot paper group is likely to finish, and thanks to ticket voting, they also have a mechanism to trade preferences with other parties. If it were it not for ticket voting, major parties would not do deals with smaller parties, as the smaller parties would not be able to deliver on their end of any deal.

Ticket voting has introduced two tactics that are distorting the proportionality of the Senate’s electoral system. The first I will call preference “harvesting”, and is a tactic employed by minor and “micro” parties to keep preferences away from major parties. The second I will call preference “corralling”, deals done between minor and major parties to engineer results.

Preference harvesting has produced unseemly “show and tell” meetings in foyers of Electoral Commissions across the country, as parties and candidates with no ideological affinity engage in the game of keeping preferences away from the bigger parties. Its first successful use was at the 1995 NSW Legislative Council election, when Alan Corbett was elected on behalf of a party called “A Better Future for Our Children”. The tactic was also successful at the 1997 South Australia election, when anti-poker machine campaigner, Nick Xenophon, was elected after harvesting the preferences of every other minor party on the ballot paper.

The game reached new heights at the 1999 NSW election. A plethora of so-called “micro” parties created a ballot paper the size of a tablecloth, with 264 candidates and 81 groups across 3 rows. Despite finishing 29th on the primary vote, Malcolm Jones from the Outdoor Recreation Party stormed to victory with just 0.2 per cent of the vote, or 0.04 of a quota. Jones harvested preferences from 21 other parties, including 8 that had achieved a higher primary count.

This success by micro-parties has produced the most disquiet concerning ticket voting. It has produced suggestions that a new “threshold” be introduced, a minimal level of primary vote that a party must achieve before it is allowed to stay in the count and collect votes as preferences. The problem with this idea is there would be no need for a threshold if it were not for ticket voting. It would also encourage larger parties to do even more preference swaps, secure that preference deals with small parties could never be reversed.

The second strategy of preference “corralling” has been less controversial to date. It was first seen at the 1984 Senate election when Labor and the Coalition did a preference swap in NSW to prevent the election of the Nuclear Disarmament Party’s Peter Garrett, despite him winning 9.1 per cent of the vote. Three years later, the preference deals were different, Labor helping to elect the far less threatening Robert Wood on behalf of Nuclear Disarmament, despite polling just 1.2 per cent of the vote.

The 1998 election saw the most extensive use of preference corralling as swaps between the major parties, Greens and Australian Democrats worked to prevent victories by candidates from Pauline Hanson’s One Nation. Despite One Nation easily outpolling the Australian Democrats in five states, it was the Democrats that elected five Senators and One Nation a single Senator. Perversely, the Democrats recorded by far their best vote in Victoria, the one state where it failed to elect a Senator, and where One Nation’s preferences instead elected Australia’s first Asian born Senator.

Now the 2004 Senate election has seen these strategies reach new heights. All sorts of bizarre deals were done in a bid to engineer the Senate result. In NSW, these complex deals elected a third Labor Senator at the expense of the Greens. In Victoria it resulted in the election of Stephen Fielding from Family First, despite receiving only 1.9 per cent of the vote. In Tasmania, it almost led to another Family First victory, the deal undermined only by the high incidence of below the line voting in Tasmania.

Parties have two major strategic objectives under PR-STV:

  • A party is more interested in electing one of its own candidates than candidates of any other party.
  • If a party cannot elect one of its own candidates, it has an interest in controlling its preferences to elect candidates and parties it prefers to be elected.

If you have no knowledge of the order candidates will finish, and cannot guarantee preferences, then the only strategy to meet these twin goals is to list candidates and parties in the preferred order you would like to see them elected.

But group ticket voting introduces an element of game theory. It allows preferences to be controlled, so parties can trade off the two objectives listed above. A party can gamble its preference list in an attempt to improve its chances of electing more of its own candidates. This is exactly what occurred at the 2004 election. Parties traded off chances to elect like-minded candidates and parties for improved chances of electing one of your their own.

Consider the objectives of parties contesting the 2004 Senate election in Victoria:

  • The Labor Party was more interested in re-electing its third candidate, Senator Jacinta Collins, than conceding a Senate seat to the Greens.
  • The Australian Democrats were more interested in staying in the count long enough to collect Coalition preferences and elect themselves than they were in helping to elect a Greens Senator.
  • Family First, the Christian Democrats and the DLP were always attracted to helping to elect Labor’s Jacinta Collins to stop the Greens winning a seat. All three groups chose to direct preferences to Collins ahead of the Coalition, improving her chances of holding off a Green challenge.
  • The Coalition was happy to help all the small Christian parties and the Australian Democrats at the expense of Labor and the Greens.

This confluence of interests explains how Family First came to win the final Senate place on Labor preferences. Labor did the deal in an attempt to elect one of its own Senators at the expense of the Greens. Had Labor not done deals and simply directed preferences to the Greens, it might have kissed goodbye to any chance of electing Jacinta Collins. The strategic preference swap with Family First was a gamble to elect a third Labor Senator. It was a gamble that failed, electing a Family First Senator when the preferred choice of the majority of Labor voters would have been a Greens Senator. It is fair to say that the will of the electorate was subverted by the wheeling and dealing created by ticket voting.

The deals that resulted in Labor winning a third seat in both South Australia and New South Wales at the expense of the Greens were essentially the same as the one that saw Family First elected in Victoria. At the final count in all three states, a Christian Party led the race for the final vacancy, Family First in South Australia and Victoria the Christian Democrats in NSW. In NSW and South Australia, the Greens failed to pass Labor, so Green preferences elected Labor senators. In Victoria, it was Labor that failed to catch the Greens, at which point Labor’s preference deal backfired and elected Family First.

Failed strategic deals have had a critical impact on the last two Legislative Council elections in Western Australia. In 2001, One Nation deprived itself of the balance of power in the Council by helping elect two Greens MLCs and giving the Greens the balance of power. In 2005, a decision by the National Party to swap preferences with the Greens in an attempt to gain an extra seat backfired and again delivered the balance of power to the Greens.

There are several different approaches that can be taken to overcome the democratic deficit created by group ticket voting. The first is to give voters more options to direct their own preferences, which will weaken the control parties have over preferences, making elections more reflective of the will of the electorate. The second is to change the way parties lodge ticket votes to discouraging micro-parties engaging in preference harvesting, and also to discourage larger parties from gambling with their preferences.

The easiest solution is optional preferential voting below the line, voters only having to fill in as many preferences as there are vacancies to fill. A second is to adopt the new NSW Legislative Council system, where voters are allowed to fill in their own preferences for parties above the line, again ideally using optional preferences. Both of these options give voters a much more manageable way of voting against the pre-determined preference tickets of parties.

The second approach would be to put an upper limit on the number of parties that could be included on a group ticket preference list. If a party could only give preferences to five other parties on the ballot paper, it would have two consequences. First, preference harvesting by micro parties would be made much more difficult. Second, with a limit on preferences, parties would be encouraged to list like minded parties on their preference tickets rather than gamble one of their precious preferences on a strategic deal.

The solutions above would all increase the number of exhausted preferences, which will weight the system against parties that rely on preference deals to get elected. This is not a controversial issue. An electoral system should encourage parties and candidates to campaign for votes rather than arrange deals with preferences.

Finally, the romantic would call for ticket voting to be abolished and Tasmania’s Hare-Clark system adopted. The first problem is that the quota at Tasmanian state elections is 10,000 voters compared to 550,000 in the NSW Senate. A very personal electoral system like Hare-Clark seems inappropriate for electing second chambers at conjoint elections. Second, any change must not result in a rise in informal voting, which despite its other faults, is one of the successes of group ticket voting.

  1. Pages:
  2. 1
  3. 2
  4. 3
  5. All

This article has been based on a submission to the current enquiry of the Joint Standing Committee on Electoral Matters into the conduct of the 2004 Commonwealth election. Submissions to the enquiry will be published at the end of April.

Discuss in our Forums

See what other readers are saying about this article!

Click here to read & post comments.

17 posts so far.

Share this:
reddit this reddit thisbookmark with Del.icio.usdigg thisseed newsvineSeed NewsvineStumbleUpon StumbleUponsubmit to propellerkwoff it

About the Author

Antony Green is an Election Analyst and produces regular publications on elections for parliamentary libraries, but he is best known for producing the reams of content at the ABC's election websites, and also as the face of ABC-TV election broadcasts. He is currently on extended leave from the ABC, and the opinions expressed in this article are his own and do not represent any view of the ABC.

Other articles by this Author

All articles by Antony Green

Creative Commons LicenseThis work is licensed under a Creative Commons License.

Photo of Antony Green
Article Tools
Comment 17 comments
Print Printable version
Subscribe Subscribe
Email Email a friend

About Us Search Discuss Feedback Legals Privacy