I didn’t get around to blogging on the outcome of the 2010 Australian federal election at the time because I was busy revising for an exam — but better late than never, I suppose. (You can read my pre-election wrapup here.)
The House of Representatives
The electoral results were pretty much what the pre-election polls had suggested: the Liberal-National Coalition received 43.3% of the vote, Labor 38.0%, and the Greens 11.7%. This amounts to a 5.4% swing away from Labor, a 1.3% swing to the Coalition, and a 4.0% swing to the Greens. Of the remaining 7.0% of the vote, 2.5% went to independents, 2.3% to Family First, 0.7% to the Christian Democratic Party, 0.3% to the National Party of Western Australia, and various other minor parties received less than 0.25% of the vote each. (A full list can be found here.)
Also as the polls had predicted, the national two-party-preferred (2PP) vote was very close to 50-50 (with Labor leading the Coalition at just 50.12%), but the results varied wildly from state to state. Labor received more than 55% of the 2PP vote in the states of Tasmania, Victoria, and South Australia (all in Australia’s southeast), but less than 45% in Western Australia and Queensland (west and northeast, respectively). The 2PP vote was closer in New South Wales but with the Coalition slightly ahead. Of Australia’s two main territories, Labor was only slightly ahead in the Northern Territory but way ahead in the Australian Capital Territory at 62%.
The 2PP swings were similarly diverse across the nation:
In hindsight, the most reliable polls were those which broke down the results state by state. When the state-by-state results were plugged into ABC election analyst Antony Green’s election calculator, the calculator kept predicting a hung parliament. The state-by-state polls were right.
The outcome of the election was that Labor won 72 of the 150 seats; the Coalition 72 (Liberals 44, Liberal Nationals 21, Nationals 6, Country Liberals 1); the WA Nationals 1; the Greens 1; and independents 4 (Bob Katter, Rob Oakeshott, Andrew Wilkie, and Tony Windsor). For Labor this meant a net loss of 11 seats. In the weeks that followed the election, Labor negotiated with the 1 Green and 3 of the independents (Oakeshott, Wilkie, and Windsor) to form the barest of minority governments.
Minority governments are very unusual in Australia (this is our first in nearly seventy years), for the same reason that the Greens won 12% of the vote but only 1 out of 150 seats: because Australia’s electoral system strongly favors major parties. The members of the House of Representatives are elected from single-member electorates using a preferential system (more details in a previous post). This makes it very difficult for minor parties to win seats (at least unless their vote is highly concentrated in particular regions, as is the case for the Nationals). Personally, I don’t think this is fair, but it’s the way Australia’s electoral system works.
This also explains how Labor won the same number of seats as the Coalition despite winning fewer votes. Because 79% of Greens voters preferenced Labor ahead of the Coalition, Greens preferences helped Labor win several key seats. Meanwhile, the Greens could not have won any seats without Liberal preferences. In the one seat they did win, Melbourne, 80% of Liberal voters preferenced the Greens ahead of Labor. You can read more about preference flows at this election here.
(On a side note, this is an interesting situation because the flow of preferences is largely determined by the recommendations of the parties themselves. Party members stand outside voting booths on election day handing out How-to-Vote Cards listing their preferences, and those tend to be decided by deals between political parties. Evidently most voters, when presented with a ready-made list of preferences, will take it. If any party decides to change its recommended preferences, particularly in certain key seats, it could change the result of an election.)
Anyway, back to my original point: the numbers of seats won do not accurately reflect what the people are actually voting for. So let’s look again at first-preference votes. Despite a lot of crowing from the Liberals, three-quarters of the swing against Labor went to the Greens, so three times as many voters have swung to Labor’s left than to its right. Even more strikingly, there were swings to the Greens in virtually every electorate (with just four exceptions).
On the same day as the lower house election there was also a half-Senate election. Of the national Senate vote, the Coalition received 38.3%, Labor 35.1%, and the Greens 13.1%. Interestingly, the combined vote for the two major parties is only 73%, compared to 81% in the House of Representatives. Minor party members play a much more important role in the Senate, since it has a proportional representation system (again, see my previous post for more details). Most of the extra votes were distributed among various minor parties who each received 1-2% of the vote (2.1% Family First Party, 2.0% Australian Sex Party, 1.8% Liberal Democratic Party, 1.7% Shooters and Fishers Party, 1.1% Democratic Labor Party, 1.0% Christian Democratic Party, and the remaining 3.8% went to others).
In the Senate as in the lower house, the largest swing was against Labor, but unlike in the lower house there was also a swing away from the Coalition. However, we did see the same 4% swing to the Greens, only in the proportional voting system of the Senate it actually had an effect.
The Coalition won 18 seats, Labor 15, the Greens 6, and the DLP 1. The winning candidates will join 36 continuing Senators, taking the Coalition down 3 seats to 34, Labor down 1 to 31, the Greens up 4 to 9, the DLP up 1 to 1, and Family First down 1 to 0 (the one other Senator is continuing independent Nick Xenophon). These numbers mean that the Greens will hold the balance of power in the new Senate, hence for Labor to get legislation through the Senate it will have to negotiate with either the Coalition or the Greens. However, the newly-elected Senators will not take their seats until mid-2011.
Although I prefer the Senate’s proportional representation to the lower house’s single-member electorates, I have some other issues with the Senate system. I think they are illustrated by the fact that the DLP won a seat despite only getting 1% of the vote.
Before a Senate election, each party, group, or independent registers a full list of preferences, called a group voting ticket, with the Australian Electoral Commission, which makes the tickets publically available. At the ballot box, voters have the choice of either numbering all the candidates, or simply writing the number 1 in the group voting ticket box of their choice, giving that group the right to distribute their preferences according to the pre-existing group voting ticket. Because there can be 60 candidates on a Senate ballot paper, it is much easier to just tick the group voting ticket box, and that is what around 95% of voters do. So the flow of preferences is, in the Senate even more so than in the lower house, engineered by the parties.
At a half-Senate election, each state elects 6 Senators. To win a seat, a Senator Senate candidate must win a certain quota of votes (14.3%, to be precise). In 2010 in the state of Victoria, Labor won 37.8% of the first preference vote, the Liberal/National Coalition 34.4%, the Greens 14.7%, Family First 2.6%, and the Democratic Labor Party came fifth at just 2.3%. The first five seats were won by Labor (Kim Carr), Liberal (Michael Ronaldson), Greens (Richard Di Natale), Labor (Stephen Conroy), and Nationals (Bridget Mackenzie). The election of these candidates used up the quotas won by those parties, leaving Labor with 9.2%, the Coalition 5.9%, and the Greens 0.4%. Because none of the remaining candidates had 14.3% of the votes, the sixth seat went to preferences.
To fill the sixth quota, the remaining candidates were eliminated one by one, starting with the candidate with the least votes and working up, and their preferences were distributed. During this process, the DLP’s John Madigan edged ahead of Family First’s Steve Fielding, and at the key count was ahead 3.4% to 3.1%, so Fielding was excluded. Fielding’s preferences boosted Madigan to 6.3%, ahead of the Sex Party’s Fiona Patten on 5.4%. Although Patten’s first preference votes went to Labor, many of the other votes she had gathered went to Madigan. At this point there were three candidates still in the race: Labor’s Antony Thow on 12.8%, Madigan on 8.3%, and Liberal Julian McGauran on 7.5%. McGauran was excluded and his preferences flowed to Madigan, giving him 15.6% of the vote, more than a quota. Madigan was elected, and the rest will be history. But is this really what the voters intended?
The problem is that a result like this depends on the majority of voters using group voting tickets. I do not think it is clear that all those minor party and Liberal voters intended to preference the DLP. Though I don’t have any statistics to back it up, I believe that most of the people who group ticket vote have no idea where their preferred party is directing their preferences; they are content to just tick the box and leave it to the party machine. If minor party voters chose their own preferences, they would probably be a lot more random (as they are in the lower house). And how many Liberal voters realised that by ticking the group ticket voting box they were preferencing the DLP?
For that matter, how many DLP voters themselves intended to vote for the DLP? According to Antony Green, a vocal critic of group ticket voting, when a party has such a small amount of the vote there is a large element of randomness in their electoral result. For example, it is possible that some voters confused the Democratic Labor Party with the Australian Labor Party and accidentally gave their vote to the DLP. Any such “accidental” votes are amplified by the careful engineering of preferences that is group ticket voting.
With these problems in mind, I think it is difficult to say whether the Democratic Labor Party was democratically elected. Green says that no other electoral system in the world would have elected a two-percenter like John Madigan (not even a Hare-Clark system, which is similar to the Australian Senate but without group ticket voting):
Every other electoral system in the world would have favoured McGauran, but the Senate’s ticket voting system and its party controlled preferences allows the will of the electorate to be distorted by pre-arranged deals.
In conclusion, with Australia’s first hung parliament since 1940 and the Greens now holding the balance of power in the Senate, the next government promises to be an interesting one.