An Australian postmortem

With all the seats now declared from last month’s federal election in Australia, there have been a number of reviews of how the new voting system in the Senate worked. Kevin Bonham’s, in two parts, is particularly thorough, but you can get similar perspectives from Tim Colebatch and Stephen Morey.

The consensus verdict is that the combined effect of the voting reforms and holding a double dissolution has been to produce a Senate that’s unusually faithful to the voters’ wishes – although, democracy being what it is, one can say that that’s not an unmixed blessing. It’s especially interesting that the Senate’s extreme malapportionment (votes in Tasmania being worth 13 times those in NSW) makes so little difference to the result. The state-by-state differences pretty much cancel out.

I might have more to say about the Senate myself at another time. What’s attracted less attention, however, is the effectiveness of our voting system for the House of Representatives, where governments are actually made and unmade. I think it’s worth a look. I’ll present the result in three tables: first, from the electoral commission’s website, are the straight totals of votes and seats by party:

Party Votes Seats % seats Proportional
Australian Labor Party 34.7% 69 46.0% 53
Liberal 28.7% 45 30.0% 44
The Greens 10.2% 1 0.7% 16
Liberal National Party (Qld.) 8.5% 21 14.0% 13
The Nationals 4.6% 10 6.7% 7
Independent 2.8% 2 1.3% 4
Nick Xenophon Team 1.8% 1 0.7% 3
Family First 1.5%     2
Christian Democratic Party 1.3%     2
One Nation 1.3%     2
Animal Justice Party 0.7%     1
Katter’s Australian Party 0.5% 1 0.7% 1
Rise Up Australia Party 0.5%     1
Liberal Democrats (LDP) 0.5%     1
Others 2.2%      
TOTAL 100.0% 150   150

 

The final column, “proportional”, is a Sainte-Laguë calculation of how seats would be allocated if they were strictly in proportion to votes. You can already see that it’s very different from the actual result, but to make things clearer, here’s a version that consolidates the various components of the Coalition into one line:

Party Votes Seats % seats Proportional
Coalition 42.0% 76 50.7% 64
Australian Labor Party 34.7% 69 46.0% 53
The Greens 10.2% 1 0.7% 16
Independent 2.8% 2 1.3% 4
Nick Xenophon Team 1.8% 1 0.7% 3
Family First 1.5%     2
Christian Democratic Party 1.3%     2
One Nation 1.3%     2
Animal Justice Party 0.7%     1
Katter’s Australian Party 0.5% 1 0.7% 1
Rise Up Australia Party 0.5%     1
Liberal Democrats (LDP) 0.5%     1
Others 2.0%      
TOTAL 100.0% 150   150

 

So the system didn’t do too badly as far as the relativity between the Coalition and Labor goes; in fact, it advantaged Labor slightly. But both of them received a huge boost as compared to everyone else. The minor parties are all under-represented – enormously so in the case of the Greens.

With a proportional system, neither major party (even counting the Coalition as a single party) would have been anywhere close to a majority. Labor plus the Greens would have had 69 seats, the same as the Coalition with its reasonably natural allies (Family First, Christian Democrats and LDP). The balance would have been held mostly by independents and Xenophones, with a few extremist wild cards thrown in.

On one view, that’s an argument against proportional representation. But since on anyone’s story this was a very close election, it’s not obvious why the electoral system should be tasked with creating an artificial majority that the voters did not choose. Particularly since we know from past experience (see 1990 and 1998) that our system is quite capable of getting close results wrong, constructing a majority for the minority party. (And if you do think majority government is especially important, there are ways of arranging that without sacrificing proportionality and introducing arbitrary unfairness in the way that our system does.)

That brings us to the final table, which reduces everything to the two-party-preferred vote:*

Party Votes Seats % seats Proportional
Coalition 50.4% 79 52.7% 76
Australian Labor Party 49.6% 71 47.3% 74
TOTAL 100.0% 150 100.0% 150

 

On that basis, the system got the result right, although counting the cross-bench MPs who lean towards the Coalition (I put Katter, McGowan and Sharkie in that category) it gave the government a bit of a bonus. But there’s no systematic reason to expect that. In a close election, getting the right result is a matter of luck rather than design.

I find it interesting that we’re so willing to talk about the representativeness or otherwise of the Senate – whether with pride, anger or wonder – but never seem to have a discussion about the obvious defects of the lower house.

 

* Note: For some reason the AEC’s count of the two-party-preferred vote still lags slightly behind the main count, to the tune of about 10,000 votes. Almost all of them are from the inner-Melbourne seat of Wills, so the final total will be slightly better for Labor than appears at present; the difference will be only a couple of hundredths of a percentage point, but it might just tick the rounded Coalition total back from 50.4% to 50.3%.

One thought on “An Australian postmortem

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.