The first step in the redistribution of NSW federal electoral boundaries began yesterday with a call for submissions and the release of base enrolment data.
The major scale of boundary changes required has been revealed by the released projected enrolment figures.
NSW is losing a seat at the next Federal election, the state’s representation reduced from 47 to 46 seats.
In addition, with seven years having passed since the last redistribution, enrolments by electoral division have diverged widely from the state average.
Abolishing a division while bringing all divisions back within the permitted variation from quota will require major surgery to some electorates.
And boundary changes will almost certainly have big political consequences.
Several electorates in the state’s west are well below quota and require major changes. Seats in Sydney’s west and south-west are well over quota, in contrast to under quota Sydney seats closer to the coast.
Evening out the enrolment numbers across the Sydney basin will not be easy. Sydney’s many bays and inlets give the city a distinctive political geography. Wholesale boundary changes are going to jumble the electoral margins of many seats.
In the immediate firing line are the four ‘teal’ Independent seats in Sydney’s east.
The seats of Wentworth (Allegra Spender), Mackellar (Sophie Scamps), Warringah (Zali Steggall) and North Sydney (Kylea Tink) are all well under quota. All these seats must increase their enrolment, eating into the territory of seats to their west.
Sydney’s Liberal heartland north of the Harbour looks certain to lose a seat, possible forcing a Liberal MP to nominate against one of the ‘teals’.
There will be a new seat created in Sydney’s outer south-west and the possible abolition of a seat further east. This creates a complex electoral jigsaw that the redistribution commissioners will first have to unpick and then re-assemble.
Inside the post I have maps highlighting the enrolment variations and provide analysis of how new boundaries might be drawn.
In a previous post on NSW redistribution prospects, I looked only at what current enrolment numbers could tell us about the redistribution. In this post I am using the more important projected enrolment numbers.
(And I’m happy to receive and publish suggestions on how the new boundaries might be drawn.)
Setting the Quotas
The abolition of a seat means that all divisions need an average gain of 3,000 voters. That is in addition to changes made to iron out the over and under enrolments that have developed across the state in the seven years since boundaries were last re-drawn.
Two quotas apply for the redistribution, a current enrolment quota set for 9 August 2023, and a projected enrolment quota set for 10 April 2028.
On current enrolments, the 46 new districts must have enrolments within 10% of the current enrolment quota of 121,011 electors. All new divisions must have current enrolments between 108,910 and 133,112.
But the projected quota is much tighter, with all districts required to be within 3.5% of quota. The projected enrolment quota is 129,621. All new divisions must have a projected enrolment in the narrower band between 125,085 and 134,157.
Almost all districts that fit within the projected quota limits will also fit within the current enrolment quota limits. The projected quota is the more important value for drawing boundaries and can force redistribution commissioners to divide areas of high enrolment growth.
Country seats
The map below shows variation from projected quota for all seats. The two lightest shades show seats within the permitted 3.5% variation from quota. The darker the shade of brown, the further a division is below quota, and the darker the shade of green, the more a division is over the projected quota. If you are using a mouse you can hover to see names and variations from quota, or you can tap the map if you are on a mobile device.
This map can drill down into the Sydney metropolitan area, but there is a more detailed map for Sydney further down this post.
On the North Coast, only Cowper (NAT held, +4.6% quota) is outside the permitted range. On the edge of the Hunter, Paterson is over quota (ALP held, +11.7%), Hunter (ALP, +4.6%) with Shortland (ALP, -5.3%) and Robertson (ALP, -3.4%) under quota. It is possible that imbalances with these seats can be resolved without adjusting the boundary with New England. If required only a minor change is necessary.
Moving west, major changes will be required to bring rural divisions within the projected quota. New England (NAT held, -11.0% from quota, Barnaby Joyce MP), Parkes (NAT, -15.5%), Riverina (NAT, -9.4%, Michael McCormack) and Eden-Monaro (ALP, -4.4%) are all under the permitted variation. Hume (LIB, +6.5%, Angus Taylor) and Gilmore (ALP, +4.1%) are the only seats over quota. Macquarie (ALP, -9.1%) on the western edge of Sydney will likely be resolved by a boundary adjustment with the Sydney seats of Lindsay or Mitchell.
Alternatively, the Hawkesbury region of Macquarie could be added to a Sydney north shore seat and Macquarie stretch west to Lithgow and Bathurst. That is an alignment that has only been used once in the last forty years. It would have the consequence of making it more likely a country seat would be abolished.
Overall the area from New England anti-clockwise round to Hume and Cunningham has 0.4 seats too many seats for voters. If New England moves west, Parkes needs even more voters, but where from? You can see Riverina being scavenged to shore up enrolments in other seats, but Riverina would then need to take voters from Eden-Monaro and Hume. That either pushes Eden-Monaro north with the likely consequence of pushing the southern Sydney seat of Hughes west into Liverpool, or Hume must push further into outer Sydney.
An alternative is to abolish a country seat. Abolish Riverina and put Wagga Wagga in Farrer. Abolish Parkes and put Broken Hill into Farrer, and re-unite the Riverina districts in Riverina. Calare would face a major re-arrangement.
Abolishing a country seat would draw Hume out of the metropolitan area, and drag Macarthur along with it and leave room for a new seat in south-west Sydney. Currently an entirely urban seat, Macarthur would return to one of its previous alignments combining the edge of Sydney with the Southern Highlands.
Northern Sydney Changes
The projected enrolments point to Sydney electorates requiring major surgery. Central Sydney is filled with under quota divisions, the metropolitan area framed by a halo of over-quota divisions.
The north shore has seven seats but only 6.4 quotas of electors. Working in from the coast, everything is under quota until you reach Mitchell.
The three teal seats along the coast and harbour are all under quota, Mackellar (-9.0%, Sophie Scamps), Warringah (-18.3%, Zali Steggall) and North Sydney (-13.2%, Kylea Tink). If is an the past, Mackellar shifts south and Warringah west, then North Sydney is either squeezed off the map, or pushes significantly into Bradfield (LIB held) or Bennelong (ALP held).
With Liberal held Bradfield (-15.5%, Paul Fletcher) and Berowra (-15.8%, Julian Leeser) both well under quota, there is a high likelihood one of these two seats or North Sydney will be abolished. There is likely to be a seat centred on Hornsby on the upper north shore, and Mitchell (LIB held, +14.6% quota, Alex Hawke) will move east, which has consequence for over quota Labor held seats to the west.
The current Windsor Road boundary between Mitchell and Greenway (ALP held, +11.2%, Michelle Rowland) is one of the clearest in Sydney and stands out on electoral maps going back decades. It is also the boundary between Blacktown City Council and The Hills Shire. It is the boundary between Liberal heartland on the north shore, and Labor's territory in Sydney's western suburbs.
On the published enrolments, it seems hard to find a way not to draw a boundary that crosses Windsor Road. Perhaps the shortfall in the north can be solved by transferring voters to Macquarie (ALP held, -9.1%), but this may still not be enough to stop Windsor Road being crossed. As I mentioned in discussing country electorates, Macquarie could be split, Hawkesbury Council added to Mitchell or Greenway, keeping the Windsor Road boundary and pushing Macquarie west to Bathurst.
Another way to keep Windsor Road is adjustments in the lower Hills between Mitchell, Bennelong (ALP, -7.8%, Jerome Laxale) and Parramatta (ALP, -7.8%, Andrew Charlton). As with skinning the proverbial cat, there is more than one way to draw electoral boundaries. How the boundaries are drawn along the line from Parramatta to Windsor will be interesting to observe in the party submissions.
South of the Harbour
The area between Parramatta, Villawood and the coast in Sydney's east is 0.75 quotas short of the current seat numbers. Will this result in a seat being abolished, or will there just be a general shift west of all existing seats?
The problems begin at South Head with Wentworth (IND held, -21.0% quota, Allegra Spender), Kingsford Smith (ALP, -6.8%, Matt Thistlethwaite), Sydney (ALP, -6.1%, Tanya Plibersek), Grayndler (ALP, -14.0%, Anthony Albanese), then every seat to the west between 5% and 8% under quota. How will these shortfalls be dealt with?
Bringing Wentworth back within quota causes major changes to Sydney and Kingsford Smith, and that propagates through Grayndler, Barton, Reid and beyond. Do Watson or Blaxland further west disappear, or do all seats get pushed further west to absorb surplus to quota votes in western and south-western Sydney?
Does in-quota Cook, held by Scott Morrison, get pushed back south of the Georges River. That and other possible changes south to the Victorian border have implications for Hughes (LIB, -7.0%). State electorates in the area covered by Hughes have been pushing into south-west Sydney in recent state redistributions.
Another way to fix Hughes would be to abolish Banks (LIB, -7.8%, David Coleman) and push Hughes into southern parts of Canterbury-Bankstown Council. Depending on which name Commissioners want to retain, the name Banks could be retained, and Blaxland (ALP, -5.3%, Jason Clare) disappear, freeing up numbers for Reid (ALP, -3.0%) and Watson (ALP, -4.8%, Tony Burke).
Commissioners want to keep seats named after former Prime Ministers, which is why Banks and Blaxland are more likely to disappear than Hughes, Reid, Watson or Barton (ALP, -5.9%, Linda Burney). Of all the current electorate names, Grayndler has the least historical importance, but abolishing the Prime Minister's seat would be unusual. There are other options.
The New Seat
In Sydney's south-west, Macarthur (ALP held, +32.5% to quota) and Werriwa (ALP, +22.7%) between them have half a seat of excess electors. Further north Greenway (ALP, +11.2%), Chifley (ALP, +14.7%) and Lindsay (LIB, +11.4%) need to shed voters. In-between McMahon (ALP, -2.9%, Chris Bowen) and Fowler (IND, -1.5%, Dai Le) do not need to change, but will change to cope with the excess in surrounding seats.
The general rule is that the redistribution commissioners start drawing Sydney seats on the coast and work inland. As each redrawn division eats into the territory of its neighbours to the west, the changes to seats grow in magnitude.
By the time you reach western Sydney, the changes to electoral boundaries can be massive. Decisions made on abolishing seats further east will impact on what happens in western and south-western Sydney.
Whatever else happens elsewhere, there will be a new seat in Sydney's outer south-west. If the redistribution does not involve crossing Sydney's south-west boundary, then it is possible a seat will be abolished both north and south of the harbour and a new seat created in Sydney's south-west. This scenario might still involve Hume moving into Sydney's south-west.
If a country seat is abolished, then there will still be a Sydney seat abolished, a new seat created in Sydney's south-west, and a seat like Macarthur might end up extending into the Southern Highlands.
Whichever way the redistribution goes, the boundary changes for NSW are going to be major and with important political implications.
In Northern Sydney, North Sydney seems the logical candidate to be eliminated. Warringah needs to move west taking most of the North Sydney LGA (it can’t go north, south or east). Then you solve the Bradfield problem by moving it south to take in Willoughby LGA. And Bennelong goes east back its old boundaries with Hunter Hill and Lane Cove LGAs. Mackellar takes some of Warringah and the problem is solved.
this is exactly what il be proposing
I think a problem with this suggestion is that it would fail the ‘community of interest’ test. The residents of the northern beaches from Manly to Dee Why don’t have any community of interest with the residents on the lower north shore: completely different public transport links, different local newspapers, different business and shopping hubs, no common schools. The demographics are probably very different too. A better solution would be for North Sydney to lose Hunters Hill LGA, pick up the rest of both Willoughby & North Sydney LGAs, which it currently only has part of. That would see North Sydney hit 120,000. Warringah pushes north into Mackellar and Mackellar moves west, picking up St Ives.
I agree with Ford. Don’t forget that North Sydney is a federation electorate and should be retained whenever possible.
I also think the option to abolish North Sydney will likely be resisted by both major parties. Note that in the 2022 federal election, the Liberal Party won the 2PP vote decisively in the Hunters Hill Council area, but the Labor Party won the 2PP vote in the Lane Cove Council area.
For the Labor Party, Bennelong will take in Hunters Hill (which will halve Labor’s margin in Bennelong from 1.0% to 0.5%) regardless of whether or not North Sydney is abolished, but losing more Labor leaning territory in the west of Bennelong and adding less Labor leaning territory to the east of Bennelong could reduce Labor’s margin further, therefore not abolishing North Sydney is probably better for Labor.
For the Liberal Party, abolishing North Sydney will put the Northern part of North Sydney, where independent Kylea Tink polled well, into Bradfield, putting Liberal MP Paul Fletcher at greater risk of losing the seat to second-time independent challenger Nicolette Boele. Although abolishing North Sydney will likely reduce Labor’s margin in Bennelong, abolishing North Sydney will hurt the Liberal Party in Bradfield more than it will hurt the Labor Party in Bennelong, not to mention that Jerome Laxale should get a sophomore boost as a first term Labor MP.
Besides, abolishing North Sydney may also cause some problems with the community of interest test. Therefore I don’t think North Sydney will be abolished.
On another note, Ford’s suggestion will result in Bradfield rather than North Sydney being abolished, which I think, for the reasons I mentioned above, makes more sense than abolishing North Sydney.
COMMENT: The Commissioners will draw the boundaries on the north shore starting from the coast, probably with Mackeller whose western boundary has been unchanged for decades. They will probably draw the electorates without names initially and add the names later. North Sydney will move generally north or maybe west. The main issue that would count against the name North Sydney being retained is if the North Sydney CBD is removed or split.
Whether the name Bradfield or North Sydney disappears has little relevance to the boundaries drawn which may be exactly the same. The name is of most importance to MPs, permitting them to call themselves MP for (insert electorate name), and avoid the disappearance of an electorate name creating pre-selection problems.
Agreed Grahame. On north side, abolish North Sydney and expand the seats west. On the south side, abolish Wentworth, rename it East Sydney and expand it south, and once again start moving boundaries westward for the adjoining seats. I think this will all be a mute point anyway, as once again Sydney is being called upon to absorb the lions share of overseas migration which within two years will result in most Sydney based Federal electorates being well over quota.
COMMENT: Whether you call it Wentworth or not, there will be a seat based on the South Head peninsula and extending south and west. Given the name has historic connections and is a Federation seat, I expect it to be retained. The allocation of seats to each state is done based on population but the drawing of boundaries is based on enrolment, and enrolment depends on citizenship. So while migration can impact the allocation of seats to states, until migrants become citizens it does not impact enrolment or the drawing of boundaries.
Wentworth is going to be a tough one. I almost would advise the commissioners to start with Kingsford-Smith and take it as far into Wentworth as it needs to go to be in quota then deal with Wentworth. K-S boundaries with Sydney are pretty solid (LGA borders and Motorways) and it’s also bounded by water on 2 sides so the easiest way for it to grow is to grow into Wentworth.
Wentworth can then move west and south-west into Sydney. Such a Wentworth would eat up much of the northern end of Sydney. There is then the obvious knock on effect to the south-west. Somewhere in the middle ring a seat would need to be abolished and a new seat in the outer south-west created.
On the Northern side of the harbour, North Sydney seems to be the most logical abolition. Will be interesting to see what the impact that has on Teal MPs. The North Sydney Teal would likely be best placed in Bradfield.
The problem with retaining the Sydney-Botany boundaries for Kingsford Smith is that this would cause Wentworth into things like Surry Hills, fully crossing the eastern distributor.
If Wentworth takes Darlinghurst and Woolloomooloo (but no further), then I am Kingsford Smith would STILL have to go well into Zetland and Alexandria. Even parts of Waterloo.
I would think that after this redistribution, Kingsford Smith would become a “South Sydney” electorate instead of a Randwick + Botany one.
Agree Leon – Bondi, Randwick and Coogee although different politically share common interests much like the adjoining riverside suburbs of New Farm, Teneriffe and Hamilton in Brisbane.
Putting Randwick and possibly Coogee into Wentworth would be a better option than trying to stretch Wentworth all the way to the CBD. Kingsford Smith as mentioned would absorb the inner south suburbs close to Redfern, which are a good fit with Mascot and suburbs adjoining the airport (the airport line runs through Green Square which adjoins Alexandria and Zetland). Sydney, having lost its inner south suburbs can then gain more of the Inner West around Balmain from Grayndler.
Disagree, disagree, disagree.
Mackellar need not be limited to expending south – it can very easily expand West, perhaps by a long way. This allows North Sydney and Warringah to move anti-clockwise. North Sydney loses Hunters Hill to under quota Bennelong. Parramatta is also under quota and should move east and north. North Sydney pushes well up into Ku-ring-gai Council. A new seat named Bradfield takes in a large arc of northern Sydney from south of Hornsby, the current Berrowa and parts of over quota Mitchell.
In the east, K-S can straddle the airport – no problem. Gardners road is a functional road and adequately connects Mascot with Kingsford. This allows Wentworth to work in its current corner location, which really it must. This would mean Barton would have to retake Cook north of the Georges River, which surely any sensible Commissioner would be looking to do.
As Antony states, the name of Banks might go, but similar to Reid, a seat based on most of its current position must surely stay as it is fully bounded on one side by water. It does seem one of Watson or Blaxand must go, which means a search for new seat for a NSW ALP Right Cabinet Minister. Some say Burney will have had enough by 2025, or does pressure come upon someone else, further east.
I tried expanding Mackellar and Warringah west to include St Ives, North Turrumurra, Middle Cove etc. that is the only way North Sydney can retain North Sydney and it stretches all the way from North Sydney to Gordon and parts of Pymble, whist lacking most middle harbour suburbs from Middle Cove to North Turramurra. This is probably worse than abolishing North Sydney because this is no longer anything close to a Lower North Shore seat. The new Warringah will end up with 2 communities of interest though.
Where does the April 2028 time come from?
COMMENT: The boundaries are due to be gazetted October 2024. April 2028 is 3.5 years after that, half way through the seven year-cycle for re-drawing boundaries.
Hi Antony,
Can you please clarify this?
“…abolishing the Prime Minister’s seat would be unusual.”
Is this “unusualness” of abolishing a PM’s seat simply a reference to the very low probability that the PM’s seat is affected (1 in 151 / 1 in 47 in NSW)?
“There are other options.”
Indeed, but surely the AEC’s impeccable impartiality would mean that it should be blind to which representatives are more or less affected in the redistribution? Shouldn’t forming clear communities of interest be more important than the position of the MP who holds the seat?
I have no preference for whether Grayndler or any of its immediate neighbours should be abolished, but as a democrat who would like to see the house and sentate expanded to reduce the distance between the people and their elected representatives, I think it would make for an interesting public debate if the PM’s seat was to be one of the more affected seats in this redistribution. Even though the legal necessity of the redistribution can’t be disputed, it’s awful for NSW to be going backwards in our representation.
Looking for clear communities of interest, Sydney’s deep valleys and waterways are often the clearest dividers between different places. This is especially so in the north as well as around George’s River. Therefore Banks looks the most obvious seat to abolish, due to it being easily geographically split along Salt Pan Creek.
COMMENT: Drawing boundaries and naming electorates can be two separate processes. The last WA redistribution merged the inner-northern Perth seats of Cowan and Stirling. The boundaries were very logical and had strong communities of interest. There was very little challenge to the drawn boundaries, the complaints from some being about keeping the name Cowan rather than Stirling. As the sitting Labor MP for Cowan and sitting Liberal MP for Stirling contested the new Cowan, the concern over names came down to who can campaign as “Member for [insert electorate name]”, and that’s not an issue about community of interest.
From examining redistribution map data in the past, Commissioners often number in electorates in the order they are drawn rather than name them at the start.
Please, please let them re-arrange FARRER…. I’m so tired of living in a seat where Sussan Ley romps it in with no effort required
Just a thought bubble given how much certain seats are projected to decline/grow in population (relative to state)…
Would this point to any chance that NSW State Electoral Map have to be redrawn again before the 2027 election?
I saw a post elsewhere that certain state seats are already far over quota (*COUGH* RIVERSTONE *COUGH*)
I also had difficulty drawing certain federal seats to be within both 10% of current quota and 3.5% of projected quota (for example my big blue seat on my map called Ruse (that would majorly overlap with State Riverstone) that I created from excesses from northwest Sydney as well as Hawkesbury LGA being split from Macquarie – I had to remove Marsden Park from my first draft and instead have it take more of Schofields and even some Hornsby LGA)
https://www.google.com/maps/d/edit?mid=1zjp8bTN7_lxMspLIqQIucxa7OtVwvzU&usp=sharing
Just a thought bubble, but could this be a good reason to (somehow) trigger another redistribution in NSW State Electoral Districts in advance of the 2027 election?
I notice that some state seats are already far from quota (*COUGH* RIVERSTONE *COUGH* LONDONDERRY *COUGH*), and my rough impression the enrolment projections show that it may get worse before 2027. If someone has actually computed the numbers then please tell me exactly how bad this will be on these projected figures.
Also, my experience drawing that new seat based on Hawkesbury LGA (in blue on the map below – the one I called Ruse) is that it was initially quite difficult to draw it in a way that passes both the (within 10% of current quota) and (within 3.5% of projected quota) requirements. I eventually fixed this issue by returning Marsden Park back to Chifley and completing readjustments elsewhere, but this suggests that both State Londonderry and Riverstone could have crazy amounts of electors by 2027.
Reference:
https://www.google.com/maps/d/edit?mid=1zjp8bTN7_lxMspLIqQIucxa7OtVwvzU&usp=sharing
COMMENT: NSW conducts redistributions every two terms of parliament, that is eight years. Redistributions can also be triggered by a change in the number of Assembly members. A third trigger is the malapportionment provision in Section 28A of the Constitution. If more than one-quarter of seats, 24 in the current Assembly, are outside of the 5% range from the average enrolment for more than two months, then a redistribution must take place. This provision can’t be triggered within 12 months of the next election, so it applies until March 2026. There are current nine districts outside the 5% band and it is speculated that on current enrolment trends a redistribution will be triggered before the next election.
Leon, I really like your map. Where did you get the data to know the number of electors in the particular areas? The ABS Statistical areas don’t seem particularly helpful as they refer to suburbs, but it only uses numbers to precisely identify certain parts of a suburb are referenced
COMMENT: The data is somewhere inside the AEC’s NSW Redistribution page.
Hi Antony, I’m interested in your opinion/comments regarding Shortland. It’s obvious there is going to be boundary changes between Hunter and Paterson and Shortland could be the answer.
Word on the street is that Pat Conroy is desperate for his seat to move north-west into Argenton/Glendale and not sweep around the south to Wyee and the Morisset Peninsula which is fertile ground for the Liberals. One reason it was suggested was because the south-western corner of Lake Macquarie has little to nothing in common with the majority of Hunter. It’s basically a forgotten area with a mass of electors that can be moved easily. It then frees up the Commissioner to move Kurri Kurri freely out of Paterson anf into Hunter and only minor changes thereafter.
Would appreciate your logic on this proposition.
COMMENT: It will be driven partly by numbers and also by party and candidate interest. I really couldn’t comment on the specifics of an individual seat’s boundaries.
Has anyone considered Kingsford-Smith moving round Botany Bay to pick up Brighton-le-Sands and Ramsgate from Barton and Cook? This pushes Barton westward. Although this is different to Barton’s historical location, it avoids seriously impacting the seats of Sydney, Grayndler and Reid. This then has a flow-on effect through Watson, Blaxland, Fowler and through to the growth area of South-West Sydney.
In the north, Berowra, Bradfield or North Sydney needs to be abolished. I would recommend moving Mitchell north/east rather than pulling Bennelong across the Lane Cove River. Parramatta can then move north to the M2 rather than having Windsor Rd being crossed.
In the south, I would put Cook back in Sutherland Shire, and push Hughes into south-west Sydney.
Having a quick look over the proposed boundaries, it seems as though Kylea Tink is going to have some problems being re-elected, because I believe that in every major party’s proposal North Sydney gets consumed (most other proposals seem to go this way as well, except of course for Kylea Tink’s own proposition). The way it is being divided, you would think the options for a new seat are problematic for Tink. She would not really contest the seat of Warringah, as trying to defeat Zali Steggall is a trouble even for the Liberal Party, let alone by an independent with close to the same views. Bennelong is a two-party minefield, even though it tends to consume some of North Sydney, it would be hard for Tink to break ground. From these other two you would assume that the last option would be Bradfield, but it could be hard given that a re-run of Nicholette Boele is not out of the question.
Antony, what do you think will happen to the Member for North Sydney in her presumptive second run for Parliament?
COMMENT: Depending on where the boundaries are drawn, I expect her to contest Bradfield. That’s with no knowledge of Tink’s plans and accepting the decision is entirely for her to take.