Prospects for the Federal Redistribution in Victoria

In brief – Victoria will lose a seat at the next Federal election. After gaining a 39th seat for the 2022 election, a relative decline in Victoria’s population sees the state revert to 38 seats. The two year shut down of immigration combined with on-going internal migration of Victorians to other states is behind the state losing a seat.

But with only three years since the last redistribution, there are no hot spots of enrolment growth that make it obvious which seat will disappear. It seems most likely that a Melbourne seat will be abolished, maybe east of the Yarra given population growth is higher to the west. But as is always the case, abolishing a metropolitan seat will have major consequences for seats across large parts of Melbourne.

Why will there be a Redistribution?

One year after every Federal election, the Australian Electoral Commissioner is required to make a determination on how many House of Representatives members each state will elect at the next election.

Commissioner Tom Rogers will make that determination in the last week of July. The determination will be made based on the most recent Australian Bureau of Statistics (ABS) quarterly population statistics. The latest population figures by state and territory, for the fourth quarter of 2022, were published on Thursday 15 June.

The Commissioner has no personal choice in making the determination. The method is strictly defined in law. For states the Commissioner will apply the formula set out in Section 24 of the Constitution. For the Territories the Commissioner will use the formula set out in the Electoral Act. The Constitution also states that the Commissioner will use the “latest statistics of the Commonwealth”, a phrase the High Court and Parliament has determined will be the quarterly population statistics, that is Thursday’s ABS release.

(I’ve published a post explaining how seats are allocated to states and why they are changing at the next election.)

The published figures show Victoria will lose a seat, as will New South Wales. Western Australia will gain a seat. (see my related post on the Western Australian federal redistribution). The House of Representatives will be reduced from 151 to 150 seats at the next election.

Change in a state’ entitlement triggers a redistribution with new boundaries drawn to match the new allocation of members. In this post I will run through where the redistribution could have greatest impact on Victorian seats.

(Note: happy to add comments with people’s views on how the new boundaries might be drawn.)

Redistribution Rules

While seats are allocated to states based on population, redistributions within each state are conducted based on enrolment. All new seats must have an enrolment within 10% of the state average. A second quota also applies based on projected enrolments. All newly drawn districts should have an average projected enrolment within 3.5% of the state average.

The narrower projected quota has more impact on the redistribution than current enrolment and forces the drawing of boundaries to take account of slow and fast growing areas.

Within the quota restrictions, the Redistribution Commissioners can take account of –

  • community of interests within proposed districts, including economic, social and regional interests
  • means of communication and travel within the proposed district
  • the physical features and area of the proposed district
  • the boundaries of existing districts

It is hard to make comment on the projected enrolment numbers as they will not be published until after the redistribution process begins.

But we do have the most recent enrolment figures as at 30 April 2023. These will be updated to the start date of the redistribution process, but this post will comment on how the redistribution might unfold based on current enrolments.

Current Enrolment Numbers

As at 30 April there were 4,416,875 voters on the electoral roll in Victoria.

With 38 seats to be drawn, a provisional quota based on the 30 April enrolments is 116,233. This is around 3,000 electors more than required for a 39-district quota.

All new districts must have an enrolment within 10% of the state average, that is between 104,609 and 127,856. Currently all 39 divisions lie within these bounds, but the new apportionment means the Redistribution Commissioners must divide the state into 38 districts within these bounds.

The last redistribution in 2020/21 was based on projected enrolments that were upended as Covid scrambled growth trends. Rapidly growing seats were set below the state quota, and stalled enrolment growth has left some of those seats below state average. The best example is the new seat of Hawke, created in 2021 in rapidly growing areas north-west of Melbourne, but now significantly under-quota. That doesn’t mean seats like Hawke won’t resume their former fast rate of population growth.

Regional Victoria

The 2020/21 redistribution set many regional seats above quota based on projected slow growth. Some remain above quota in 2023 even with the increased quota created by abolishing a seat.

The map below colour codes all 39 divisions based on percentage above or below a 38 seat quota. Colours range from red for most under quota through orange, light blue and on to dark blue for most over-quota. You can use this map to drill down and look at Melbourne seats, but a larger Melbourne-only map is included and discussed further down the page.

When Victorian electoral boundaries are redistributed, the first seats drawn tend to be those in the state's corners, Gippsland, Indi, Mallee and Wannon. As the map shows, these seats are all at a variation from quota that requires at most minimal change. The scale of change will be determined by projected enrolments numbers.

Between the corner seats, Nicholls will probably need to grow, dependent on the projected enrolment numbers. In the west of the state, where Mallee and Wannon shift interacts with the regional seats of Ballarat, Bendigo and Geelong-based Corangamite and Corio. The size of the regional cities in these districts mean they won't be abolished. These seats plus Nicholls may grow by eating into the fringe Melbourne seats of Hawke and McEwen.

East of Melbourne, Monash covering the La Trobe Valley and west Gippsland needs to grow, which means eating into neighbouring La Trobe on the south-east fringe of Melbourne. Another seat that gets drawn early is Flinders on the Mornington Peninsula, most of the seat's boundaries defined by two bays and Bass Strait.

All up, the fate of seats in Rural and Regional Victory depends on how much they must cannibalise seats on Melbourne's periphery to stay within 3.5% of the projected quota. Projected enrolments won't be released until August at the earliest.

In my view, it is unlikely that a rural seat will be abolished. Keeping the current number of regional seats can be achieved by creeping into Melbourne's periphery if required. The alternative of abolishing a rural/regional seat would pull one of Melbourne's seats deep into regional Victoria, which is effectively the same as abolishing a metropolitan seat.

Abolish a Melbourne seat - east or west of the Yarra?

The map below makes clear that the new quota leaves most Melbourne's seats under quota. In the inner-east both Higgins and Chisholm are more than 5% under quota. Four seats west of the Yarra are more than 5% under, inner-city Cooper Wills and Maribyrnong, and the new peripheral Melbourne seat of Hawke. Created ahead of the 2022 election, Hawke was set below quota, but the expected population growth evaporated as Covid hit. Hawke may resume rapid population growth, but the five inner-city seats do not have enrolment growth prospects.

If regional seats need to grow, then Hawke and McEwen could lose their less urban edges. If this occurs, it becomes more likely that a seat will be abolished west of the Yarra.

If regional seats don't eat into Hawke and McEwen, the decision on where to abolish a Melbourne seat becomes more complex. In total, seats west and north of the Yarra add to half a seat more than the area in entitled to, as in total are the seats east of the Yarra. If this balancing of east and west over-representation exists in the projected enrolments, then we may see a seat drawn astride the Yarra, the most likely culprits either Jagajaga or Menzies.

Port Phillip Bay and the course of the lower Yarra tend to be starting points for drawing boundaries. East of the Yarra, this geography means seats such as Macnamara, Higgins, Kooyong and Goldstein are drawn first. Low current enrolment and and slow enrolment growth mean all these seats must push east. Enrolment growth is concentrated in Holt and La Trobe on Melbourne's outer south-east fringe. As re-drawn inner-city seats stretch east to increase enrolment, there is a chance one of the middle distance seats will disappear.

If a seat is abolished in Melbourne's east, it could be Chisholm or Hotham. Both seats lack a clear geographic grounding, Chisholm having been detached from Koonung Creek by the 2021 redistribution. In the 2021 WA redistribution, the Commissioners had to choose between abolishing a seat named after a woman (Cowan) or a 19th century Governor (Stirling), which suggests that Chisholm could be retained over a seat named after Governor Hotham.

As I keep saying, whether a seat in Melbourne's east is abolished will be determined by the yet to be released projected enrolments. Plus, abolishing an eastern seat could pull Jagajaga south of the Yarra,

If a seat is to be abolished west of the Yarra, then the seats to watch are Cooper, Wills and Maribyrnong. Hawke may also be under quota, but a newly created electorate named after Australia's third longest serving Prime Minister is not going to be abolished, even if its boundaries must undergo radical surgery. Burke has been abolished so why not Wills as well, and Maribyrnong is one of the few remaining Melbourne seats with a name based on geography.

What may save western seats is the resumption of enrolment growth in Scullin, Calwell, McEwen, Hawke, Gorton and Lalor. Under-quota inner-city seats in Melbourne's west are adjacent to outer suburban growing seats. In Melbourne's east, there are an array of under-quota middle distance seats between inner-Melbourne and the growth hotspots on the road to Cranbourne and Pakenham.

Current 39 Victorian Divisions - Variation from 38-seat Quota

Enrolment Variation
Division Margin Enrolment Votes Quotas
Aston LIB 2.8 110,434 -5,799 -5.0
Ballarat ALP 13.0 112,359 -3,874 -3.3
Bendigo ALP 12.1 113,192 -3,041 -2.6
Bruce ALP 6.6 113,572 -2,661 -2.3
Calwell ALP 12.4 112,830 -3,403 -2.9
Casey LIB 1.5 115,302 -931 -0.8
Chisholm ALP 6.4 110,268 -5,965 -5.1
Cooper ALP 8.7 110,270 -5,963 -5.1
Corangamite ALP 7.6 115,653 -580 -0.5
Corio ALP 12.8 113,554 -2,679 -2.3
Deakin LIB 0.2 113,327 -2,906 -2.5
Dunkley ALP 6.3 112,534 -3,699 -3.2
Flinders LIB 6.7 114,251 -1,982 -1.7
Fraser ALP 16.5 112,423 -3,810 -3.3
Gellibrand ALP 11.5 111,508 -4,725 -4.1
Gippsland NAT 20.6 116,149 -84 -0.1
Goldstein IND 2.9 110,721 -5,512 -4.7
Gorton ALP 10.0 117,200 +967 +0.8
Hawke ALP 7.6 110,249 -5,984 -5.1
Higgins ALP 2.1 108,914 -7,319 -6.3
Holt ALP 7.1 111,793 -4,440 -3.8
Hotham ALP 14.3 117,330 +1,097 +0.9
Indi IND 8.9 118,640 +2,407 +2.1
Isaacs ALP 6.9 112,502 -3,731 -3.2
Jagajaga ALP 12.3 114,198 -2,035 -1.8
Kooyong IND 2.9 113,256 -2,977 -2.6
La Trobe LIB 8.7 111,875 -4,358 -3.7
Lalor ALP 12.8 114,296 -1,937 -1.7
Macnamara ALP 12.2 112,756 -3,477 -3.0
Mallee NAT 19.0 121,230 +4,997 +4.3
Maribyrnong ALP 12.4 109,907 -6,326 -5.4
McEwen ALP 3.3 112,734 -3,499 -3.0
Melbourne GRN 10.2 114,924 -1,309 -1.1
Menzies LIB 0.7 113,079 -3,154 -2.7
Monash LIB 2.9 112,971 -3,262 -2.8
Nicholls NAT 3.8 114,283 -1,950 -1.7
Scullin ALP 15.6 110,651 -5,582 -4.8
Wannon LIB 3.9 116,211 -22 0.0
Wills ALP 8.6 109,529 -6,704 -5.8

15 thoughts on “Prospects for the Federal Redistribution in Victoria”

  1. Does Hotham being above quota make it less likely to be abolished? Surely a smaller seat is easier to parcel out to its neighbours.

    While Covid disrupted Melbourne’s growth, it seems unlikely to change the medium term trend which involves faster growth in the West. If so that would mean a South-east seat is for the chop?

    COMMENT: Hotham may be currently over-quota, but the bay and Yarra mean Macnamara, Higgins, Kooyong and Goldstein will be drawn first, eating into Hotham’s western end and moving the seat around. Population growth is much higher in the south-east mean abolition there is less likely.

  2. The major reason Vic and NSW lose a seat and WA gain is the ABS rebasing of population estimates from the 2021 census. Vic/NSW revised roughly 100k downwards and WA 60k upwards. If one believes the post census estimates are correct, then Vic should never have been allocated another seat last time around.

  3. Great read! I have a few thoughts:
    ● Is the distribution pre 2021 viable to be reused?

    (A) for both WA and Victoria no because the boundaries drawn in 2021 corrected large inequalities in enrolment. You couldn’t revert back to those boundaries as they breach the quota rules.

    ● Are the commissioners really that attached to names that it’d affect which and how districts are changed?

    (B) To some extent they draw the boundaries before sorting out names. Geography, enrolment and order drawn drive the redistribution, not names. In the 2021 WA redistribution, Cowan and Stirling were effectively merged and one name abolished. In Victoria, it is possible Hotham and Chisholm are effectively merged and re-drawn and a decision is made on which name to drop.

    ● Seeing the regional west is above quota and the metro West below, the lines taking a jump to the west seems logical to me…
    ● I just double checked the maps, Gorton will need to significantly shrink! There’s thousands of homes going in there every year! (Also means my first point is a definite no 😅)
    ● Also, brimbank shouldn’t be a part of Gorton, the difference in demographics is chalk and cheese

  4. I’m not sure Jagajaga is a good candidate if the Yarra needs to be crossed. There was a lot of objection to Menzies having basically the same arrangement last time.

    I think if the Yarra needs to be crossed, better alternatives could be:

    1) Melbourne pushing into the northern part of Macnamara, since Southbank is essentially an extension of the city anyway.

    2) McEwen regaining the Upper Yarra Valley, as the river is much less of a divide in this area (arguably it actually unites communities grouped around it, rather than divides them)

    COMMENT: Which would continue the tradition of McEwen being the state’s bits and pieces electorate, vacuuming up all the left over bits where northern Melbourne meets rural Victoria. The only problem is whether McEwen will still extend far enough east to include the Yarra Valley.

    Your other suggestion would guarantee the seat abolished is in eastern Melbourne.

    1. I guess in my scenario, seats like Hawke, Bendigo, Nicholls, Calwell etc would take most of McEwen’s western half, so the seat would become a bit more of a coherent Yarra Valley/Diamond Valley/Whittelsea based Division rather than ‘bits and pieces’

  5. Antony, brilliant post as always. I’m curious as to how likely you think it is that Higgins would be the seat to be abolished if there were to be one south of the Yarra?

    I notice that enrolment is lower in Higgins than either Hotham or Chisholm. It also is easy to see how Macnamara would be able to shed enrolment on the St Kilda end to be a seat anchored moreso to the Yarra than to the bay. I’d imagine that it also wouldn’t be the hardest job in the world to put to sleep a division named after Henry Higgins, who was very briefly AG shortly after federation, in a similar strain to the argument for getting rid of the names Hotham or Gellibrand.

    COMMENT: I think the issue is whether the Caufield extension to Macnamara finally gets straightened, swapped out for the western end of Higgins. As someone else suggests, Melbourne could be drawn across the Yarra creating an entirely new configuration.

    I suspect we will see three electorates in eastern Melbourne come down to two, or four re-combined into three. Sorting out which seat name remains will come after the boundaries are drawn.

  6. Great read however you have used old figures showing Aston as Lib seat with a 2.8% margin which is now a Labor held seat margin over 3%

    COMMENT: Done deliberately. When you come to do the calculations for a redistribution you use the general election results, not by-elections. You might make a separate calculation afterwards to assess the by-election.

  7. I find it interesting, given most of these suggestions seems to be suggesting Mcnamara migrating east slightly, the notional calculations in such a situation. Given that the three candidates in the three candidate preferred count were within one percent, the areas of Windsor and Prahran, both with Greens ahead in polling place results, might tip it over to the Greens. Yet it might also slacken the Liberal vote enough to see Labor win in a TCP contest against the Greens. How will you be calculating such a situation as this?

    COMMENT: Calculate the first preferences, see what they look like and then apply appropriate preference counts if required.

  8. This is more of a general question regarding redistributions moving forward, but what are your thoughts on the impact of the larger numbers of people casting pre-poll votes? Given there are generally fewer pre-poll voting centres than ordinary poll voting centres, along with the fact that it is generally easier to cast a pre-poll vote at a centre outside of your electorate, I assume this would greatly impact Electoral Commissions’ abilities to consider the impact of boundary adjustments on margins since there is less location specific data to make the decisions on.

    COMMENT: It has no impact on how Redistribution Commissioners draw boundaries. Boundaries are drawn based on where people live, not where they vote. The impact on margins is not a criteria for Commissioners to draw boundaries. The issue you raise has an impact on people like me who estimate post-redistribution boundaries, not on the Redistribution Commissioners.

  9. Hi Antony. During the previous Victorian distribution the Liberals submission included a proposal to merge the southern ends of Wills and Coopers to effectively create a notional Greens seat to the north of Melbourne. Given the number of seats in this area that are now below quota do you think this is a viable option this time around?

    COMMENT: It can be put forward. The current north-south boundaries for both seats align strongly with local government boundaries and this has been used in the past as a strong argument for community of interest.

    1. The current North-South running boundaries also align with natural features, notably Darebin and Merri Creeks and, to a lesser extent, Moonee Ponds Creek.

      COMMENT: The fact those natural features are also the council boundaries means that north-south alignment can be argued on the basis of both community of interest and geography.

  10. Hi Antony. I know a lot of it is pure guesswork, but in the scenario that Chisholm is abolished, how far would you see Menzies being pulled south? Would this also result in Menzies perhaps losing suburbs like Warrandyte, Wonga Park or even Templestowe?

  11. Hi Antony. I know a lot if it is guesswork, but in the scenario that Chisholm was abolished, how far would you see Menzies being pulled south? Would this result in Menzies potentially losing suburbs like Warrandyte, Wonga Park, or even Templestowe?

    COMMENT: I couldn’t comment. There are many ways to draw the boundaries.

  12. Antony, big fan btw, can you elaborate on the Macedon Ranges reposed seat of Hawkes boundaries

    COMMENT: I’ve moved on to other work at the moment. I suggest you use the links in the post to check details in the submissions themselves. There are aspects of the geography of Melbourne’s rural fringe that I’m not familiar with.

Leave a Reply