South Australia’s election produced a result that has sparked a lot of interest: despite the Liberal Party winning a majority of the two-party-preferred vote (and by even more than in 2010), the Liberal Party has won less seats than the ALP, and we appear to have narrowly avoided the Labor government holding an outright majority.
It’s not an uncommon outcome, in South Australia and elsewhere in the country. The ALP has formed government in South Australia despite losing the statewide vote three times in the last 25 years: in 1989, 2002 and 2010, and in two of those cases the ALP won an overall majority.
In federal politics, the 1990 and 1998 elections both saw the sitting government maintain power despite losing the vote (Labor in 1990, and the Coalition in 1998).
Following Saturday night’s result, multiple Liberal figures have come out to complain about the outcome and to vaguely criticize our existing electoral system which allows such an ‘unfair’ result.
Tony Abbott described South Australia’s election laws as ‘extraordinary’, ignoring the fact that Saturday’s outcome could just as easily happen under federal electoral law.
South Australia already attempted to remedy the ‘unfair result’ problem following the 1989 election result. After a history of Liberal governments holding on despite majority Labor votes thanks to severe rural malapportionments, the Bannon Labor government, with support from the Liberal Party, amended the Constitution to require that electoral boundaries be drawn as to be ‘fair’.
At each boundary redistribution after each election, the Electoral District Boundaries Commission is required to take into account the political make-up of each district, and draw boundaries that ensure that a majority of votes results in that party winning a majority of seats.
Unfortunately, that approach simply doesn’t work. Firstly, it is necessary to assign independent seats as either ‘Labor’ or ‘Liberal’ (an approach that wouldn’t have worked with Lyne and New England in 2010). Secondly, it assumes that swings are uniform, which they never are. The 2010 pre-election pendulum would have given the Liberal Party a majority of seats if the swing was uniform, but Labor suffered the biggest swings in safe seats while holding on in marginal seats.
Thirdly, the ‘fairness’ requirement had the potential to punish more effective MPs by making more severe changes to seats that bigger swings, while MPs who gained less of a swing get to hold on to more of their seat.
After the 2010 election, the Commissioners threw up their hands, and gave it up. They argued that the result was not due to unfair boundaries, but due to differences in campaigning, and made relatively mild changes to boundaries.
Attempting to create a ‘fair’ system out of the system of single-member electorates that we use to elect the House of Representatives and all mainland states’ lower houses is a fools’ errand. By its very nature, single-member electorates aren’t fair and don’t produce a fair result in terms of the number of seats won by each party.
The reason Labor has benefited in South Australia is not due to a ‘gerrymander’ or any kind of structural unfairness, despite Alexander Downer’s vaguely menacing implications. It’s South Australia’s people, and geography, that have the impact.
Labor won nine seats on Saturday by margins of more than 10%. The Liberal Party won thirteen. Labor didn’t win a single seat by a margin of more than 20%, while the Liberal Party won three. Meanwhile Labor won ten seats by margins of less than 6% while the Liberal Party won only five by such small margins (including Mitchell, which is still in doubt).
Liberal voters tend to be more tightly concentrated in affluent suburbs and rural parts of South Australia, while Labor’s voters are more evenly spread through more diverse suburbs that dominate most of South Australia’s marginal seats.
The same phenomenom has been experienced in New South Wales, which was largely considered to be a ‘natural Labor state’ because Labor voters tended to be spread more evenly through the Sydney suburbs while the Coalition racked up huge margins on the north shore and in rural areas.
We also don’t know how much campaigns, and voting patterns, are influenced by the electoral system we have. After the Coalition won the 1998 election despite losing the national vote, Liberal director Lynton Crosby pointed out that the Liberal Party may have run a different campaign if the goal was to win a majority of the vote.
If you want a fair system, there is only one answer: proportional representation. An electoral system like the Tasmanian-style Hare-Clark, New Zealand-style MMP or European-style party list would ensure that parties won the number of seats they are entitled to.
The reason that neither major party will consider these options is because, while it would ensure they would win their fair share, it would massively reduce the amount of time each party spends running a one-party majority government, and would bring in a lot more diverse voices into the Parliament.
Well you can’t have your cake and eat it too. You can either have a ‘fair’ system, or a system that ensures major party domination and majority government (most of the time).
It’s unclear what Liberals like Abbott, Downer and Hamilton-Smith are proposing as the solution to their problem. Downer mentions moving towards Tasmanian-style proportional representation, which South Australia and New South Wales used in the early part of the 20th Century.
Another commentator (whose identity appears to be lost thanks to the Advertiser’s shoddy editing) proposes moving towards optional preferential voting or first-past-the-post. Both systems would have favoured the Liberal Party, due to the high Greens vote and the lack of a significant right-wing force in the House of Assembly due to the weak state of the Nationals in the state.
Yet either of those options wouldn’t make the system fairer – they would tilt it towards the Liberal Party in the short term while allowing for even more distorted outcomes in the long-term. As the Labor governments who introduced OPV in New South Wales and Queensland in the early 1990s could tell the SA Liberals of today, it’s unwise to design an electoral system that is biased towards your side in the short term: because in the long term you don’t know who that bias may effect.