Tags

,

As a follow-up to my earlier Shifting Boundaries post I am going to write some posts on how federal redistributions work in Australia.

My main reference for this post is, obviously, the Commonwealth Electoral Act 1918 (CEA). By the way if you know where to look, the CEA contains some fascinating glimpses into Australian electoral history[1]. I expect to be referring to both the Australian Electoral Commission‘s website and Antony Green’s election blog a fair bit, but that shouldn’t be a surprise either.

Note: any errors are mine, although I’ve done my best to make this as accurate as I can.

Determining Entitlement

The place to begin with federal boundaries is with the number of seats, and determining how many seats each state or territory is entitled to.

12 months[2] after a new parliament sits for the first time[3], the Electoral Commissioner (EC) will use Australian Bureau of Statistics population data to determine the population of each state, territory, and of the commonwealth[4]. Note that for the purposes of the quota calculation, the population of the commonwealth excludes the population of the territories[5].

The current parliament first sat on 12 November 2013, so the most recent calculation was conducted on 13 November 2014, and can be found here.

The next step is to divide this number by 144, which is twice the number of Senators from the states which get 12 each. This is the core formula:

Total population of the six states / (Number of Senators for the states x 2) = Population quota[6]

Then the population of each state or territory gets divided by that number with fractions of ½ or more rounded up, and less than that rounded down[7]. That’s the number of seats each state or territory is entitled to at the next election.

In other words:

Total population of individual state or territory / Population quota = Number of members[8]

Now, there’s a bunch of shenanigans in section 48 that I’ve skipped over, including the statistical gymnastics that let the Northern Territory keep two seats[9], but don’t let the Australian Capital Territory get a third[10].

Oh, and Tasmania gets 5 because it’s a state[11].

But for the purposes of a rough guide this will do for now.

Triggering a Redistribution

Redistributions are basically triggered automatically when certain conditions are met. The EC has some[12] discretion to delay redistributions in particular circumstances (such as an election being due), but for the most part if the trigger is met, the redistribution has to happen.

Once the number of seats per state and territory for the next parliament is known, any difference from the current parliament will trigger what I usually refer to as an entitlement redistribution[13].  In the current parliament this happened to NSW (entitlement down one) and WA (entitlement up one).

The other common trigger is what I refer to as the 7 year itch. If nothing else happens, a redistribution will be automatically triggered 7 years after the determination of the last redistribution for the state or territory[14]. In the current parliament this happened to the ACT and NT (still in progress).

The final trigger is a failsafe mechanism to protect one vote, one value: Malapportionment[15]. It’s important to note that a malapportionment redistribution has never been triggered since these rules were established in the early 80s, later posts in the series will explain why[16].

The CEA defines malapportionment as more than 1/3rd of the divisions in a state/territory varying from the average enrolment by more than 10% for two months running[17]. And, yes, the AEC reports on enrolment numbers by division monthly.

To use Victoria as an example, 13 of 37 seats would need to be out by more than 10% to trigger malapportionment. The last redistribution of Victoria was gazetted on 24 December 2010. So Victoria is approaching the next seven year itch and only McEwen[18] is malapportioned[19], although Aston is close. Even that much surprised me.

Conclusion

This post covers (mostly) the when and why of federal redistributions; in the next post I’ll start tackling the how. That may take a while to draft since it can get fairly complicated.

As always happy to take comments and questions. For a more detailed look at specific redistributions I recommend checking out the relevant tag on Antony Green’s election blog.

Footnotes (Some Snark, Mostly cross references)

[1] See for example sections 39 and 48A and leave a comment if you want to know the background.
[2] Technically it’s the day after the anniversary of the parliament sitting. See section 46 (1).
[3] Which means in the case of a really early election nothing will change.
[4] Section 46, see section 47 for some specific rules applicable to the supply of population data by the Australian Statistician.
[5] Consider the jokes about NT and ACT not counting to have already been made please. See also section 45 (1).
[6] Source: AEC Website.
[7] Section 48.
[8] Source: AEC Website.
[9] ***COUGH*** S48A ***COUGH***
[10] But I’m not bitter about that. Much.
[11] Section 24 of the Australian Constitution. By the calculation Tasmania would usually get about 3 – 4.
[12] Section 59 (3) – (5)
[13] Section 59 (2) (a)
[14] Section 59 (2) (c)
[15] Section 59 (2) (b)
[16] Because of the importance of electoral representation there’s one other failsafe in the CEA re redistributions. That one’s never been used either, and it will also come up in a later post.
[17] Section 59 (10).
[18] Oh yes, for the benefit of American readers, federal divisions are named not numbered like Congressional seats in the U.S.
[19] McEwen is currently 20% above average as of 31 March 2016