New Victorian State Electoral Boundaries Finalised

Last week the Victorian Boundaries Commission released its final determination of the state’s new electoral boundaries. The new boundaries will apply for the next Victorian election in November 2022.

The draft boundaries were released at the end of June and I analysed their political impact in a previous post. There were major changes to the boundaries used at the 2018 election.

Of the 88 districts proposed at the draft stage, 56 remain unchanged in the final version.

I’ve prepared a listing showing the composition of all new electorates based on movements between old and new electorate. You can find it at this link.

Maps of all new districts, details of changes, and the Commissioner’s reasoning for the changes can be found on the Electoral Boundaries Commission website.

Political summary in a paragraph – a permanent shift of two seats from Liberal parts of Melbourne to Labor parts as a result of differential population growth rates. But it is not electoral boundaries but the scale of the Liberal Party’s 2018 defeat that is the bigger problem for the Coalition at the 2022 state election.

In this post I’ll analyse the political impact of the final boundaries.

Changes Between the Draft and Final Boundaries

The following are the more significant changes to boundaries between the draft and final boundaries.

  • Some of the changes to Bass were reversed. Inverloch was brought back into the electorate and the areas originally transferred from Narracan were handed back. Labor won Bass in 2018 with a margin of 2.4%, the draft boundaries created a Liberal 2.4% electorate, and the final version is Liberal 0.8%.
  • Reversing the Bass-Narracan boundary change required Narracan to lose Moe to Morwell, which in turn lost areas north and south of Traralgon to Gippsland South. These changes substantially strengthened Narracan for the Liberal Party and at 11.1% it is the Liberal Party’s safest seat. The addition of Moe considerably strengthens Labor’s position in Morwell and probably weakens Independent Russell Northe’s hold on the seat.
  • The incursion of Ripon into Ballarat’s western suburbs on the draft boundaries has been reversed. Ripon now gains western parts of Golden Plains Shire from Buninyong. The change swaps a Labor voting part of Wendouree for a Labor voting part of Buninyong which means Ripon still becomes a notional Labor seat. At least the new areas of Ripon are a better fit with the rural nature of Ripon than Ballarat suburbs.
  • The proposal to re-name Wendouree as Eureka has been abandoned, Eureka instead adopted as the new name for Buninyong.
  • The above changes to Buninyong/Eureka have resulted in changes to boundaries for Polwarth, Geelong and South Barwon.
  • The major boundary change between Albert Park and Prahran was abandoned in favour of a much smaller alteration. This change restored Albert Park as a safe Labor seat and restored the Green position in Prahran.
  • There were also a re-arrangement of boundaries to electorates around Eltham and Mill Park, and an adjustment to the boundary between Bayswater and Rowville. The proposed Morang has reverted to its original name Mill Park.

Comparing the Final Boundaries with 2018 Election Results

In broad summary, the major changes produced by the redistribution are –

  • The Electoral Boundaries Commission lists 9 seats as abolished and 9 created, but some of these are merely name changes to represent shifts in boundaries.
  • In terms of geography, the redistribution has abolished two seats in Melbourne’s eastern suburbs and one middle distance south-east seat. They are replaced by three new outer suburban seats, one each in Melbourne’s outer west, outer north-west and south-east fringe.
  • In terms of politics, two seats are abolished in eastern Melbourne, traditional Liberal territory where Labor currently holds several marginal seats due to the scale of its 2018 landslide victory. In south-east Melbourne, the safe Labor seat of Keysborough is abolished, replaced by more marginal Labor seats further to the south-east. There are two new and very safe Labor seats in Melbourne’s north and west. Looking past estimated margins based on the lopsided 2018 result, the impact of the new boundaries is there are two fewer seats where the Liberal Party wins most of its seats in eastern Melbourne, and two extra seats in western and northern Melbourne where Labor wins most of its seats. The tale in south-east Melbourne is more mixed and there is little change in regional Victoria.
  • Boundary changes are extensive in the eastern suburbs, but the impact can be summarised as follows. Burwood (ALP 3.3%) has been replaced by Ashwood (ALP 2.1%) and Forest Hill (LIB 1.2%) replaced by Glen Waverley (LIB 0.8%). Ferntree Gully (LIB 1.6%) and Mount Waverley (ALP 1.8%) have been abolished. Bayswater flips from being a Labor seat (ALP 0.4%) to a notional Liberal seat (LIB 0.7%).
  • South-eastern Keysborough (ALP 14.9%) has been abolished and the new seat of Pakenham (ALP 3.0%) created on Melbourne’s south-east fringe. In associated changes to create Pakenham, Gembrook (LIB 0.8%) is in part replaced by Berwick (LIB 1.0%), Bass flips from (ALP 2.4%) to (LIB 0.8%), and Hastings goes the opposite way, switching from (LIB 1.1%) to (ALP 0.4%).
  • The major changes that threatened the Greens’ hold on Prahran (GRN 7.5% v LIB) were reversed by the final boundaries. It is still a close race for second on first preferences between Labor and the Greens, but the main impact of new boundaries is to weaken the Liberal vote. The new Prahran is now less of a three-way contest and more a race between Labor and the Greens.
  • West of the Yarra, Altona (ALP 14.6%) is replaced by Point Cook (ALP 12.4%) and there is a new seat called Laverton (ALP 24.0%).
  • In Melbourne’s north-west, well over quota Yuroke (ALP 20.3%) has been split into two news seats, Greenvale (ALP 22.3%) and Kalkallo (ALP 20.4%).
  • On the final boundaries, Buninyong (ALP 12.2%) has been re-named Eureka (ALP 9.4%). Wendouree re-gains its original name and most of its former boundaries. The final boundaries for Ripon (LIB 0.02%) removed the Ballarat suburbs added on the draft boundaries, but now include smaller Golden Plains Shire communities west of Napoleans previously in Buninyong. The estimated new margin for Ripon is (ALP 2.8%).
  • The Liberal margin in Polwarth slips from 5.4% to 2.2% as the seat moves deeper into the Surf Coast.
  • The Green margin in Brunswick increases from 0.6% to 2.3%.
  • Labor won 55 seats at the 2018 election. They lose Keysborough and Mount Waverley (abolished) and Bass and Bayswater (become notional Liberal). They gain Hastings, Morwell and Ripon on the new boundaries, and the new seats of Greenvale and Kalkallo (in place of Yuroke), Laverton and Pakenham. Labor now has a notional 57 seats.
  • The Liberal Party have slipped from 21 to 20 seats. The party has lost Ferntree Gully (abolished), Hastings and Ripon are notional losses to Labor, but notional gains are Bass and Bayswater.
  • A new Independent margin for Mildura is impossible to calculate. The same comment applies to Morwell but Labor’s position is significantly strengthened. If Russell Northe does not contest the next election then Morwell will be a notional Labor seat.
  • The uniform swing needed for a majority Coalition government falls from around 12% to around 11%, but the Coalition need two extra seats in that swing.

A Note on Calculations

Calculating estimated margins after a redistribution is always difficult, but particularly difficult given the huge pre-poll vote in 2018. You can’t isolate pre-poll votes to a single part of the electorate unlike polling place results. The way the Victorian Electoral Commission reports pre-poll makes the problem worse. Only a single total is provided for pre-poll votes, even in districts with more than one pre-poll centre.

For instance, Bass had three pre-poll centres in different parts of the electorate, while South Barwon had one at the Surf Coast end of the electorate, and a second large centre in the suburbs of Geelong. It is almost certain that pre-poll results varied in different parts of the electorate, but the impact cannot be untangled from the data.

Comments I’ve received from parties on my previously published margins suggest my estimates for Bass, Berwick, Pakenham and South Barwon are particularly rubbery.

New Electoral Pendulum

Labor (57 seats) Liberal/National (20/6)
Margin Electorate Margin Electorate
ALP 0.4 Hastings (LIB MP) LIB 0.1 Caulfield
ALP 0.4 Hawthorn LIB 0.4 Sandringham
ALP 0.6 Nepean LIB 0.6 Brighton
ALP 1.7 Northcote (v GRN) LIB 0.6 Croydon
ALP 2.1 Ashwood LIB 0.7 Bayswater (ALP MP)
ALP 2.8 Ripon (LIB MP) LIB 0.7 Glen Waverley
ALP 3.0 Box Hill LIB 0.8 Bass (ALP MP)
ALP 3.0 Pakenham LIB 0.9 Eildon
ALP 3.5 Ringwood LIB 1.0 Berwick
ALP 3.6 South Barwon LIB 1.9 Evelyn
ALP 4.6 Morwell (IND MP) LIB 2.2 Polwarth
ALP 5.8 Melton LIB 3.3 South-West Coast
ALP 5.9 Richmond (v GRN) LIB 4.2 Warrandyte
ALP 8.4 Monbulk LIB 4.5 Kew
ALP 9.0 Cranbourne LIB 5.0 Mornington
ALP 9.4 Eureka LIB 5.4 Bulleen
ALP 10.2 Frankston LIB 5.4 Rowville
ALP 10.3 Geelong LIB 6.2 Malvern
ALP 10.5 Eltham LIB 9.4 Benambra
ALP 10.5 Narre Warren North LIB 11.1 Narracan
ALP 10.7 Narre Warren South NAT 12.0 Ovens Valley
ALP 11.2 Wendouree NAT 14.0 Gippsland South
ALP 11.4 Bellarine NAT 15.3 Euroa
ALP 11.5 Bentleigh NAT 17.6 Gippsland East
ALP 12.0 Carrum NAT 21.6 Lowan
ALP 12.1 Bendigo East NAT 24.0 Murray Plains
ALP 12.4 Ivanhoe
ALP 12.4 Point Cook Independent (2) / Green (3)
ALP 12.5 Niddrie IND 0.3 Mildura (v NAT)
ALP 12.9 Albert Park GRN 1.6 Melbourne (v ALP)
ALP 13.4 Macedon GRN 2.3 Brunswick (v ALP)
ALP 13.4 Mordialloc IND 5.3 Shepparton (v LIB)
ALP 13.6 Werribee (9.2 v IND) GRN 9.0 Prahran (v LIB)
ALP 14.6 Sunbury
ALP 15.0 Clarinda
ALP 15.8 Essendon
ALP 16.1 Oakleigh
ALP 16.4 Mulgrave
ALP 16.6 Bundoora
ALP 16.8 Yan Yean
ALP 17.7 Sydenham
ALP 17.7 Tarneit
ALP 18.6 Bendigo West
ALP 19.0 Williamstown
ALP 19.1 Lara
ALP 20.6 Kalkallo
ALP 22.1 Pascoe Vale
ALP 22.1 St Albans
ALP 22.4 Greenvale
ALP 23.4 Dandenong
ALP 24.0 Laverton
ALP 24.9 Mill Park
ALP 25.2 Broadmeadows
ALP 25.2 Kororoit
ALP 27.4 Thomastown
ALP 28.3 Preston (21.1 v GRN)
ALP 28.7 Footscray

Old and Estimated New Margins for all Districts

Note – all margins in the table below are estimated two party preferred margins. Seats actually held by Greens and Independents are indicated. Estimated two-candidate preferred margins for selected seats are shown below the table.

Division Old Margin New Margin Change
Albert Park ALP 13.1% ALP 12.9% 0.2 to LIB
Ashwood (ex-Burwood) ALP 3.3% ALP 2.1% 1.2 to LIB
Bass (Changes Party) ALP 2.4% LIB 0.8% 3.2 to LIB
Bayswater (Changes Party) ALP 0.4% LIB 0.7% 1.1 to LIB
Bellarine ALP 11.5% ALP 11.4% 0.1 to LIB
Benambra LIB 8.9% LIB 9.4% 0.5 to LIB
Bendigo East ALP 12.1% ALP 12.1% no change
Bendigo West ALP 18.6% ALP 18.6% no change
Bentleigh ALP 11.9% ALP 11.5% 0.4 to LIB
Berwick (ex-Gembrook) LIB 0.8% LIB 1.0% 0.2 to LIB
Box Hill ALP 2.1% ALP 3.0% 0.9 to ALP
Brighton LIB 1.1% LIB 0.6% 0.5 to ALP
Broadmeadows ALP 30.3% ALP 25.2% 5.1 to LIB
Brunswick (GRN held) ALP 34.4% ALP 34.6% 0.2 to ALP
Bulleen LIB 5.8% LIB 5.4% 0.4 to ALP
Bundoora ALP 17.4% ALP 16.6% 0.8 to LIB
Carrum ALP 11.9% ALP 12.0% 0.1 to ALP
Caulfield LIB 0.3% LIB 0.1% 0.2 to ALP
Clarinda ALP 17.4% ALP 15.0% 2.4 to LIB
Cranbourne ALP 11.0% ALP 9.0% 2.0 to LIB
Croydon LIB 2.1% LIB 0.6% 1.5 to ALP
Dandenong ALP 23.9% ALP 23.4% 0.5 to LIB
Eildon LIB 2.4% LIB 0.9% 1.5 to ALP
Eltham ALP 9.1% ALP 10.5% 1.4 to ALP
Essendon ALP 15.9% ALP 15.8% 0.1 to LIB
Eureka (ex-Buninyong) ALP 12.2% ALP 9.4% 2.8 to LIB
Euroa NAT 15.4% NAT 15.3% 0.1 to ALP
Evelyn LIB 2.6% LIB 1.9% 0.7 to ALP
Ferntree Gully (abolished) LIB 1.6% .. ..
Footscray ALP 28.1% ALP 28.7% 0.6 to ALP
Frankston ALP 9.7% ALP 10.2% 0.5 to ALP
Geelong ALP 10.1% ALP 10.3% 0.2 to ALP
Gippsland East NAT 17.6% NAT 17.6% no change
Gippsland South NAT 15.3% NAT 14.0% 1.3 to ALP
Glen Waverley (ex-Forest Hill) LIB 1.2% LIB 0.7% 0.5 to ALP
Greenvale (new seat) .. ALP 22.4% ..
Hastings (Changes Party) LIB 1.1% ALP 0.4% 1.5 to ALP
Hawthorn ALP 0.4% ALP 0.4% no change
Ivanhoe ALP 12.4% ALP 12.4% no change
Kalkallo (new seat) .. ALP 20.6% ..
Kew LIB 4.8% LIB 4.5% 0.3 to ALP
Keysborough (abolished) ALP 14.9% .. ..
Kororoit ALP 25.6% ALP 25.2% 0.4 to LIB
Lara ALP 19.1% ALP 19.1% no change
Laverton (new seat) .. ALP 24.0% ..
Lowan NAT 23.5% NAT 21.6% 1.9 to ALP
Macedon ALP 13.2% ALP 13.4% 0.2 to ALP
Malvern LIB 6.1% LIB 6.2% 0.1 to LIB
Melbourne (GRN held) ALP 25.3% ALP 25.0% 0.3 to LIB
Melton ALP 4.3% ALP 5.8% 1.5 to ALP
Mildura (IND held) NAT 5.6% NAT 6.2% 0.6 to NAT
Mill Park ALP 24.9% ALP 24.9% no change
Monbulk ALP 8.6% ALP 8.4% 0.2 to LIB
Mordialloc ALP 12.9% ALP 13.4% 0.5 to ALP
Mornington LIB 5.0% LIB 5.0% no change
Morwell (IND held) ALP 2.4% ALP 4.6% 2.2 to ALP
Mount Waverley (abolished) ALP 1.8% .. ..
Mulgrave ALP 12.7% ALP 16.4% 3.7 to ALP
Murray Plains NAT 24.0% NAT 24.0% no change
Narracan LIB 7.3% LIB 11.1% 3.8 to LIB
Narre Warren North ALP 9.8% ALP 10.5% 0.7 to ALP
Narre Warren South ALP 6.9% ALP 10.7% 3.8 to ALP
Nepean ALP 0.9% ALP 0.6% 0.3 to LIB
Niddrie ALP 12.6% ALP 12.5% 0.1 to LIB
Northcote ALP 33.2% ALP 33.2% no change
Oakleigh ALP 15.8% ALP 16.1% 0.3 to ALP
Ovens Valley NAT 12.6% NAT 12.0% 0.6 to ALP
Pakenham (new seat) .. ALP 3.0% ..
Pascoe Vale ALP 18.3% ALP 22.1% 3.8 to ALP
Point Cook (ex-Altona) ALP 14.6% ALP 12.4% 2.2 to LIB
Polwarth LIB 5.4% LIB 2.2% 3.2 to ALP
Prahran (GRN held) ALP 7.6% ALP 9.5% 1.9 to ALP
Preston ALP 28.5% ALP 28.3% 0.2 to LIB
Richmond ALP 31.7% ALP 31.3% 0.4 to LIB
Ringwood ALP 2.8% ALP 3.5% 0.7 to ALP
Ripon (Changes Party) LIB 0.0% ALP 2.8% 2.8 to ALP
Rowville LIB 5.7% LIB 5.4% 0.3 to ALP
Sandringham LIB 0.6% LIB 0.4% 0.2 to ALP
Sheppartonm (IND held) LIB 11.7% LIB 11.7% no change
South Barwon ALP 4.6% ALP 3.6% 1.0 to LIB
South-West Coast LIB 2.3% LIB 3.3% 1.0 to LIB
St Albans ALP 21.5% ALP 22.1% 0.6 to ALP
Sunbury ALP 14.3% ALP 14.6% 0.3 to ALP
Sydenham ALP 17.9% ALP 17.7% 0.2 to LIB
Tarneit ALP 18.0% ALP 17.7% 0.3 to LIB
Thomastown ALP 27.2% ALP 27.4% 0.2 to ALP
Warrandyte LIB 3.9% LIB 4.2% 0.3 to LIB
Wendouree ALP 10.3% ALP 11.2% 0.9 to ALP
Werribee ALP 12.6% ALP 13.6% 1.0 to ALP
Williamstown ALP 22.1% ALP 19.0% 3.1 to LIB
Yan Yean ALP 17.0% ALP 16.8% 0.2 to LIB
Yuroke (abolished) ALP 20.3% .. ..

Estimated Labor/Green Margins

Division Old Margin New Margin Change
Brunswick (GRN v ALP) GRN 0.6% GRN 2.3% 1.7 to GRN
Melbourne (GRN v ALP) GRN 1.3% GRN 1.6% 0.3 to GRN
Northcote (ALP v GRN) ALP 1.7% ALP 1.7% no change
Prahran (GRN v LIB) GRN 7.5% GRN 9.0% 1.5 to GRN
Preston (ALP v GRN) ALP 20.7% ALP 21.1% 0.4 to ALP
Richmond (ALP v GRN) ALP 5.5% ALP 5.9% 0.4 to ALP

Further notes on Non-2PP Contests

Benambra – Liberal margin versus Independent nudges upwards from 2.4% to 2.6%.

Geelong – Labor margin versus Independent 6.2% in 2018, estimated new margin 7.1%.

Mildura – Independent Margin verses National 0.3%, gains 2,252 electors from Ripon, new margin not calculated

Morwell – A new margin is difficult to calculate given the transfer of Moe from Narracan. The Labor vote increases significantly on the new boundaries and almost certainly overturns the majority of Independent Russell Northe.

Pascoe Vale – Labor margin 8.6% versus Independent, major changes to boundaries make new margin impossible to calculate.

Shepparton – Unchanged, Independent margin v Liberal 5.3%.

Werribee – Labor margin versus Independent nudges up from 8.8% to 9.2%.

11 thoughts on “New Victorian State Electoral Boundaries Finalised”

  1. Not sure if this would be likely, but if Libs end up shrinking to 3rd in Prahran, could that end up giving Labor the seat? Before, Greens would accumulate preferences before outlasting Labor and getting preferences from them; but now maybe Libs would be eliminated before Greens get 50%, with Liberals preferences going to Labor instead.

    COMMENT: If the Liberals slip to third place, then if they continue with their current policy of putting Labor ahead of the Greens on how-to-votes, then Labor would probably win the seat.

    1. … would this scenario then count as a spoiler or something else? For some reason, this possibility (libs slipping to 3rd leading to a labor win) feels counter-intuitive, but it seems to be different to what spoilers usually are (similar ideology parties splitting their vote and harming each other)

    2. I think if the proposed boundaries went ahead (with the remainder of St Kilda transferring into Prahran in exchange for the “Postcode 3004” corridor), the Liberals slipping to third place would have been a very likely scenario. I don’t see it being possible on these boundaries though, so that 3PP result between ALP & GRN will be the real race in this seat, which will make it an interesting count on election night.

  2. On these boundaries, would Footscray also now be a Labor v Greens seat? It’s lost all of its Sunshine component and gained some of the strongest Green parts of the old Williamstown seat.

    COMMENT: On first preferences the numbers are Labor 54.8%, Greens 20.2% and Liberal 17.1%.

    1. Tom the first and best

      Given that the Animal Justice primary was 7.2% in the part of Footscray that has not been redistributed out (The Greens had a controversial candidate in Footscray last time, causing their vote to leak, including to Animal Justice), I think the Greens (with a different candidate to last time) have a decent chance of getting second place in the 3CP vote, even without the Liberal primary being as low as last time.

  3. Do you See Labor winning seats off the back of Liberal preferences in contests with the Greens?
    And what is your prediction on Glen Waverley? Would it be the 2 incumbents from Forrest hill and Mount Waverley running against each other?

    COMMENT: It depends what the Liberal Party recommends with its how-to-votes. As a rough guide, 60-70% of Liberal preferences will flow to whichever party the Liberals choose to recommend. I have no idea who the candidates for Glen Waverley will be.

  4. Would the Liberals have won in 2010 on these boundaries. Back then the Coalition won Swan Hill, Doncaster, Mount Waverley and Ferntree Gully all of which are now abolished. The problem for the Liberals is that the Population growth is in North and Western Melbourne and SE Melbourne. It is only SE Melbourne which they are competitive in so they would have likely won Pakenham then (only new seat which they are competitive in). What TPP does the Liberals need to win Government.

    COMMENT: The 2013 redistribution abolished Swan Hill and Doncaster and created Werribee and Sunbury in Melbourne’s west. This redistribution repeats the process.

  5. Bob, I think you are right that Glen Waverley will likely have two incumbents in Matt Fregon and Neil Angus facing off, due to Mount Waverley being abolished, which is also likely the case in Bayswater with Jackson Taylor and Nick Wakeling, due to Ferntree Gully being abolished.

  6. Antony, a genuine question if you don’t mind: the scale of the Coalition defeat in 2018 is undeniable but you’re concluding this redistribution indicates a net gain in ALP seats of 2.

    How does the Commission justify this in view of the ALP 2PP share was 57.3%, yet its share of seats won was, by my calculations, 62.5%. Conversely, the Coalition’s 42.7% share of the 2PP won only 30.7% of seats. A net gain of two seats to the ALP notionally increases its proportion of seats won to 64.8%.

    I fully understand the Legislative Assembly isn’t determined by proportional representation but doesn’t it seem anachronistic for the party over-represented at the last election to be favoured by the subsequent redistribution?

    What am I missing here?

    COMMENT: Several seats won by Labor in 2018, including Bass, Cranbourne, Yuroke and Yan Yean, were more than 50% above the enrolment quota. Every seat in eastern Melbourne was under quota. The redistribution redrew boundaries to ensure all electorates were within 10% of the state average. It was always the case that seats in eastern Melbourne were going to be abolished and new seats created in the north, north-west and south-east of Melbourne. I wrote three years ago that Labor would gain 2-3 new safe seats from the redistribution. All you needed to do was look at the enrolment numbers. The Boundaries Commission draws boundaries based on enrolment, not voting patterns, and drawing on enrolment was always going to favour Labor.

  7. To highlight my point even further, you say that “The uniform swing needed for a majority Coalition government falls from around 12% to around 11%, but the Coalition need two extra seats in that swing…”.

    Are you expressing that required swing in 2PP terms? If so, are you suggesting the Coalition needs approximately 53.7% of the 2PP to win government?

    COMMENT: The swing to Labor in 2018 was much larger in marginal seats and many marginal Labor seats ended up with double digit margins. That’s why the swing back for the Coalition to win enough marginal seats for government is greater than what is needed for a majority of the 2PP vote. The Coalition could win with 49.%, or 51% or 54%, depending entirely how uniform the swing is.

Leave a Reply