Wednesday, 26 June 2013

The Australian electoral system: Preferential voting vs Proportional voting

You were warned...


Australia has long been regarded as a nation at the forefront of the innovation and design of electoral systems. So much so that it has a reputation as a country more willing than any other to experiment with electoral systems in pursuit of improving parliamentary democracy (Farrell & McAllister 2006 p. 1). It can come as no surprise then that the voting systems for electing members according to population (House of Representatives) and equally among states (the Senate) are different (Uhr 1997, p.68-9). This article will look at how and why these systems are different, how well they reflect the people who vote under them and how these systems could potentially be improved to better reflect the will of the people and confirm Australia’s reputation as an electoral trailblazer.



Australia is somewhat of an oddity when it comes to electoral voting systems. Not only is it one of a handful of countries that has legislated the compulsion to vote in federal elections, it is part of an even smaller group of nations that also enforces the compulsion to vote.  On top of this, the method of electing members to the House of Representatives is so rare at a national level that it is known as being quintessentially Australian (Farrell & McAllister 2006, p. 52). This idiosyncrasy has even led to a tendency for the voting system, known elsewhere as alternative voting, to be referred to in Australia as preferential voting.

Alternative or preferential voting is a non-proportional system that requires voters to rank candidates in the order that they would like to see them elected. In this system the candidate, in order to be elected, must achieve a majority of support from the constituents (Farrell & McAllister 2006, p. 53). This can be done it two ways, the simplest of which is that a candidate wins fifty per cent of the popular vote and is automatically elected. The second and increasingly more common method in Australia is that candidates rely on preferences from other candidates in order to receive majority support (Hughes & Costar 2006, p. 68).  This is done by sequentially re-distributing the votes of the least popular candidates up until the point where one candidate earns a majority of votes. Another feature of preferential voting is that it operates on the single-member constituencies’ model, in that each division or electorate votes in one member to represent them in parliament (Heywood 2007, p. 259). During a federal election this creates, in essence, one hundred and fifty separate elections across Australia (one per division) with the winners becoming the members of the House of Representatives.

How well the system of preferential voting represents the will of the people is perhaps better seen through the Rousseauian idea of the ‘general will’ of the people, because by and large the results of preferential voting aligns with the mood of the electorate.  Even though candidates increasingly have to rely on the re-distribution of preferences in order to attain fifty percent of the vote, very rarely is the winning candidate unpopular with the electorate. Since the introduction of preferential voting for the House of Representatives 1919, there have been less than a dozen instances where the winning candidate was not one of the top two candidates following the first count (Hughes & Costar 2006, p. 67). So despite the theoretical possibility of a candidate with a low primary vote winning overall, in practice this rarely happens. Furthermore, research suggests there is a high level of confidence and satisfaction in the preferential voting system in its representation the community’s wishes and in the maintenance of democracy (Farrell & McAllister 2006, p. 165).

Preferential voting also ensures that parties with similar voter demographics aren’t unfairly disadvantaged. Whereas under other voting systems (such as first-past-the-post) these parties would cannibalise each other’s vote and thus be far less likely to win the election, preferential voting limits the effect of that vote-splitting (Farrell & McAllister 2006, p. 55). It is even suggested that the system of preferential voting is the reason behind the formalising of the coalition of the Liberal and National party as the best means to combat the Australian Labor Party, as the system benefits the formation of alliances in the face of strong opposition (Lucy 1993, p. 92). It can come as no surprise then that the conservative Hughes government first introduced the idea of preferential voting, to prevent the split in the centre-right vote from benefitting the Labor Party (Jaensch 1997, p. 265).

That’s not to say that preferential voting is a fool-proof way to ensure the will of the people is conveyed. Many researchers contend that the system of preferential voting, in conjunction with the enforced compulsory voting has contributed to the entrenchment of the major parties in the House of Representatives to the exclusion of independents and minor parties (Bean 1997, p. 102, Lucy 1993, p. 92, Jaensch 1997, p. 368, Hughes & Costar 2006, p. 68). A good example of this is that in the 2010 Federal Election only five of the 150 members elected to the House of Representatives were either independents or from a minor party (Australian Electoral Commission 2013). Preferential voting requires voters of minor parties to balance their explicit will to vote for a particular party with particular ideals with their general will to direct their preferences to a major party which is somewhat in line with their own beliefs.

One final quirk of preferential voting is that it doesn’t necessarily equate with equality, in terms of either overall representation or the general consensus of the nation.  There is no guarantee that winning the majority of support nationwide will result in the ability to form government, nor is there a guarantee the make up of Parliament is proportional to support amongst the electorate (Jaensch 1997, p. 374). Such a result occurred three times during the 1950s and 1960s, where Labor won the first preference vote as well as the two party preferred, yet were unable to claim enough seats to form government (Bean 1997, p. 106), and even as recently as 1998 where the Kim Beazley-led Labor party won only 67 of 148 seats despite winning 51 percent of the two-party vote. Of course this works the other way too with parties with strong support in concentrated areas representation winning disproportionate numbers of seats. In Australia the National Party has been the greatest beneficiary of this, and as a result has been over-represented in the House of Representatives since 1946 (Jaensch 1997, p. 405).

Although preferential voting is targeted at maximising voter choice (Farrell & McAllister 2006, p. 3), and does succeed in both giving the voter choice and general expression of their political will, there are some areas that can be explored to create greater choice and political expression for the voter. One idea is to change the method of preferential voting. At a national level Australia employs the compulsory preferential voting method, where for a vote to be considered valid a number must be placed next to all the candidates on the ballot paper. An argument put forward is that by changing that system to an optional preferential system where the voter can choose the number of candidates to give preferences to. Lucy (1993) argues that doing this would reduce ‘voter fatigue’ and lower informal voting (p. 93). This would also give voters the surety that there is no possibility that their vote would count toward electing a candidate they do not want. Small parties would also be major beneficiaries of optional preferential voting, as it would give them a greater chance at securing seats in the House of Representatives (Farrell & McAllister 2006, p. 56). The main obstacle to any change of this nature occurring would be the opposition of the major parties, who would be incredibly unlikely to legislate anything that would endanger their primacy.

Another potential change could be in the way electoral districts are arranged. In discussion of the Hawke Government’s 1983 electoral reforms, Hughes and Costar (2006) explored the inclusion of qualitative criteria in the determination of electoral boundaries in addition the quantitative already in place (p.19).  This would be difficult to implement across the board, but the idea of creating electorates based on regional and communal similarities instead of just equal numbers may have some merit. For example instead of the twin cities of Albury-Wodonga being represented by two vast electoral districts on either side of the border, they would be better served by a smaller district encompassing both sides on the Victoria/New South Wales border. Although there is legislation which ensures the minimum numbers of seats and the prevention of electoral boundaries crossing state lines (Hughes and Costar 2006, p. 20), under a federal system with one house (the Senate) already divided along state lines perhaps those laws can be reformed to create electoral divisions that have more in common than just statistical equality.

In contrast to the House of Representatives, the voting system for the Senate uses a proportional voting model to elect candidates. Again the use of this system makes Australia somewhat of an oddity on an international level, with Ireland and Malta being the only other countries to use a similar system (Farrell & McAllister 2006, p. 2). The system essentially relies on voters to rank candidates preferentially, similar to preferential voting. But whereas preferential voting returns one member per electorate, proportional voting returns multiple members, with candidates being elected if they reach the required quota of votes. The quota is calculated using the Droop formula that divides the total number of votes by the number of seats to be filled plus one. One is then added to that total to determine the number of votes a candidate needs to win election (Heywood 2007, p. 262). The fundamental idea behind this system is that a candidate or parties representation in parliament is in direct proportion to their electoral strength.

Despite support from a number of areas for the introduction of proportional voting for the Senate, support that goes as far back as Federation (Farrell & McAllister 2006, p. 32), it wasn’t until 1948 that the single transferable vote (a form of proportional voting) was introduced for the Senate. This was due to a number of factors including Australia’s understanding and enthusiasm for preferential voting (Hughes & Costar 2006, p. 73) and the fear of major parties that proportional voting would adversely affect their representation in the Senate (Lucy 1993, p. 95). However the underlying factor for the continued push for an introduction of a separate system of voting for the Senate was to ensure that the Houses had different voting systems, neither of which utilised a plurality, or block, system (Farrell & McAllister 2006, p. 26, and p. 40).

Theoretically, proportional voting should result in the most accurate reflection of the will of the electorate. Voters can be much more explicit in their choice of candidate and have more control over the way they direct their preferences (Jaensch 1997, p. 375). Furthermore the quota system of a multi-member electorate enables minor parties and independents a to have a greater opportunity to win representation as they do not have to acquire a majority of support, as with preferential voting. The Senate also enables those minor parties to have a real voice in the legislative process, evidenced by the fact that until 2005 a collection of these minor parties (including the Democrats, Greens and Independents) had held the balance of power since the fourth term of the Fraser Government (Hughes & Costar 2006, p. 76). The system of proportional voting allows for a greater representation of the people’s will by acknowledging and representing the support for smaller political entities while not disadvantaging the larger political parties.

But the problem with proportional voting is that it may in fact give the voter too much choice. One of the greatest criticisms of proportional voting is that voters fail to grasp the complexities of the system and as a result the system can throw up unexpected results (Farrell & McAllister 2006, p. 43). For instance, in the 2010 Federal Election Victorians elected Democratic Labor Party (DLP) member John Madigan to the Senate despite Madigan receiving just over one thousand first preference votes and the DLP as a whole receiving just 2.5% of first preferences (Australian Electoral Commission 2013). The failure to understand or fully appreciate the weight their preferences can have affect how well proportional voting reflects the people’s will. It also doesn’t help that some see their vote in the Senate as a way to express their discontent towards the government without affecting their majority in the House of Representatives (Bean 1997, p. 104).

The Hawke Government’s reforms of the electoral system in 1983 had a mixed effect on how well the system reflects the people’s will. The simplified voting method, enabling voters to vote ‘above the line’ for a party instead of individual candidates helps reduce informal voting and combat what Lucy’s describes as ‘voter fatigue’. Following the introduction of the Electoral Act 1983 informal voting for the Senate dropped from nearly 10% to just over 4% in the next year (Lucy 1993, p. 96). This had the duel benefit of making it easier for voters to express their general will, and made sure more voters could validly express it. But while the reform made voting simpler it also benefitted the major parties. The Act transformed the voting system, allowing parties to choose the rank order of candidates on the ticket, effectively determining whom the electors voted for (Farrell & McAllister 2006, p. 44). Furthermore, the act contained a provision whereby a party may distribute preferences equally between any numbers of parties (Jaensch 1997, p. 373). The will of the people can not possible be reflected if they do not know which way their preferences are going.

For proportional voting to function as well and as fairly as it should theoretically, an emphasis should be placed on education. Despite potential for the more explicit expression of the voters will through proportional voting, less than 5% of voters take the time to list individual preferences when casting a Senate ballot paper (Hughes & Costar 2006, p. 74). Although this increases to around 20% in Tasmania and the Australian Capital Territory (where voters have more experience with the system), the fact that such a large majority of voters allow major parties to direct their preferences means that the people’s will cannot be accurately reflected. Even more disturbingly there is the suggestion voters have difficulty understanding even basic information about the proportional method of voting, and that without how-to-vote cards could fail to vote at all (Lucy 1993, p. 90-97). Without a greater understanding of what their vote actually means, proportional voting will not represent the will of the people.

The voting systems for the House of Representatives and the Senate are designed with similar goals in mind. Both are aimed at giving the voters choice and flexibility (Farrell & McAllister 2006 p. 1). They also reflect the people’s will, although in slightly different ways. Preferential voting in the House of Representatives supports the general will of the people, while proportional voting in the Senate gives voters the chance to express their views more explicitly. But as Hughes and Costar (2006) point out “even the most democratic of electoral systems need fine-tuning” (p. 89). Australia may have one of the fairest and most egalitarian electoral systems, but that doesn’t mean that it cannot be improved to better reflect the will of the people.

No comments:

Post a Comment