The ABS released its council and suburb level approvals data on Tuesday. It answers the question the monthly headline never does, which is where the work is actually forming.
Australia approved 18,901 new dwellings in July 2026. Fifty two council areas accounted for almost two thirds of them.
That is the picture in the small area data the Australian Bureau of Statistics released on Tuesday, covering local government areas and SA2 regions for the July reference period. There are 556 council areas in the collection. In July, 180 of them approved no new dwellings at all.
Brisbane approved more new homes in July than any other council
Brisbane City Council approved 1,182 new dwellings in July, made up of 278 houses and 904 in the other residential category, which covers townhouses, semi detached homes and apartments. In July 2025 the same council approved 451. That is an increase of 162 per cent, and it is the largest single council figure in the country. It also sits oddly against the national release published on 1 September, which recorded a 3.6 per cent fall for the month in seasonally adjusted terms.
Both things are true. The national series smooths and adjusts. The small area series does not, and it is published in original terms only, because the ABS does not produce seasonally adjusted or trend estimates at council level.
Melbourne was second with 673 new dwellings. Every one of them was other residential. Not a single new house was approved in the City of Melbourne in July, which is less strange than it sounds. Across the whole 2025 to 2026 financial year that council approved seven.
The Gold Coast was third with 494, down from 786 a year earlier. Logan approved 439, Wyndham 426, and Greater Geelong and Moreton Bay both approved 425.
The apartment recovery is concentrated in about eighteen councils
Nationally, 7,489 new other residential dwellings were approved in July, and the trend estimate for that category is running at its highest level since 2018. The small area data shows how narrow the base underneath that number is.
Of the 556 council areas, 394 approved no other residential dwellings whatsoever in July. Eighteen approved 100 or more. The ten busiest accounted for 43.6 per cent of the national total, and the five busiest for 32.1 per cent.
The list is short and it is urban. Brisbane 904, Melbourne 673, the Gold Coast 399, Parramatta 224, Sutherland Shire 207, unincorporated ACT 197, Ryde 192, Port Phillip 167, Greater Geelong 160 and the City of Sydney 140.
Several of those are close to standing starts. Parramatta approved 30 in July last year and 224 this July. Sutherland Shire went from 58 to 207, Ryde from 59 to 192, Port Phillip from 2 to 167 and the City of Sydney from 2 to 140. That is the shape of a recovery arriving in a small number of places at once rather than spreading evenly, which is the same pattern that showed up when TGB ran the same exercise on the April data.
Detached building is a national activity. Multi unit building is a big city activity, and in July it happened in about eighteen places.
Detached housing was spread across ten times as many councils
The contrast is the most useful thing in this release. Against the 162 councils that approved any multi unit housing, 371 approved at least one new house. The thirty busiest accounted for just over half the national house total, compared with the top ten taking nearly 44 per cent of multi unit approvals.
The leaders are the familiar growth corridors. Moreton Bay approved 353 new houses, Wyndham 347, Logan 341, Wanneroo 327, Melton 316, Hume 290, Brisbane 278 and Greater Geelong 265.
Most of them were flat or slightly lower than a year ago. Wyndham came off 372, Logan off 370, Wanneroo off 362, Melton off 359 and Casey off 312 to 258. Moreton Bay was the standout exception, up from 237, alongside Brisbane, up from 216.
That is the same story the national release told, where house approvals fell in every state the ABS publishes, viewed one council at a time. The corridors did not collapse. They eased, broadly and modestly, from a four year high.
What does the small area data add?
Building Approvals is published in two parts. The main monthly release, issued on 1 September for the July reference period, gives national and state figures in original, seasonally adjusted and trend terms. The small area release follows five business days later and breaks the same month down to local government areas and SA2 regions. Small area figures are original terms only. They are not seasonally adjusted and they carry no trend estimate, so they answer where approvals landed rather than which direction the market is moving.
The councils that went backwards
Falls at council level are as lumpy as rises. Cairns approved 655 new dwellings in July 2025 and 77 this July. Penrith went from 542 to 72, Lane Cove from 288 to 2 and Glen Eira from 225 to 25.
Others moved less dramatically but in ways that matter more to volume builders. Ipswich fell from 395 to 205, Whittlesea from 333 to 177 and Camden from 282 to 153, with the detached component doing most of the moving in each case.
One large approval can rewrite a council’s month
The clearest warning in this release comes from the New South Wales mid north coast. The Mid-Coast council area approved 280 new dwellings in July, up from 49 a year earlier. Of those, 260 were houses, and 227 of them sat in a single SA2, Forster.
For scale, Mid-Coast approved 431 houses across the entire 2025 to 2026 financial year. That is an average of 36 a month, and its strongest month in that year was 155. A result of 260 in July is not a regional building boom. It is almost certainly one large approval landing in one month.
The ABS says as much in the documentation for this collection, noting that estimates for small areas vary from month to month and that caution should be used in drawing conclusions from monthly movements. The same caveat applies to Melbourne’s 673, of which 673 came from the Southbank East region alone.
What the map does not tell you
Approvals describe intent. They describe it early, and at council level they describe it unevenly.
A permit issued in July becomes a slab months later and a completion a year or more after that, which is why the gap between an approval and a start remains the more useful number to watch than any single month’s ranking.
What the small area data does add is a sense of which labour pools are about to be asked for more. Read against the direction of the wider construction cycle, the split in this release is the part worth sitting with. The councils driving the multi unit recovery and the councils driving detached volumes are almost entirely different councils, drawing on different trade bases, different finance structures and different project timelines.
The read
July was a soft month nationally and a very uneven one locally. Both descriptions are accurate, and the second one is more useful if you are pricing work.
The multi unit segment is genuinely recovering, but it is recovering in roughly eighteen council areas, and in several of them it is arriving from close to nothing. Detached approvals eased across most of the corridors that carry national volume, from a level they had not held since 2021.
For anyone operating in one of the eighteen, the tightening will not feel gradual, because a base of near zero to 200 approvals in twelve months is not a gradual change in demand for concreters, formworkers and crane crews. That is the window in which how trade capacity is contracted and held tends to move up the priority list, usually about a year before anyone starts talking about it.
The pipeline is not evenly distributed. It never was. This is the release that shows by how much.
Frequently asked questions
Brisbane City Council, with 1,182 new dwellings approved in original terms. That was made up of 278 houses and 904 townhouses, semi detached homes and apartments, and compares with 451 in July 2025. Melbourne was second with 673 and the Gold Coast third with 494.
Highly. Of the 556 local government areas in the ABS collection, 52 approved 100 or more new dwellings in July 2026, and those 52 accounted for 65.9 per cent of the national total of 18,901. A further 180 councils approved none at all.
In a small number of urban councils. Nationally 7,489 new other residential dwellings were approved in July 2026, but 394 of 556 councils recorded none, and only 18 recorded 100 or more. The ten busiest councils accounted for 43.6 per cent of the national total.
The main monthly release covers national and state figures in original, seasonally adjusted and trend terms, and was published on 1 September for the July reference period. The small area release follows five business days later, on 8 September, and breaks the same month down to council and SA2 level in original terms only.
Yes. Approvals are revised in later releases, and this release added 320 dwellings to the 2025 to 2026 financial year. Separately, the ABS has confirmed that 2026 local government area boundaries are not expected until October 2026, so July to December 2026 data continues on 2025 boundaries and will be restated in the January 2027 publication, which also adopts a new statistical geography standard and republishes SA2 data back to July 2021.
RELATED ARTICLES
- House Approvals Fell in Every State in July. The National Total Still Held Above 10,000 for a Seventh Month.
- Where Australia’s Next Wave of New Homes Is Actually Being Approved
- The Approvals Are There. The Starts Are Not. What the Latest ABS Data Really Shows.
Last updated 10 September 2026. Building approvals figures are revised by the ABS in subsequent releases, and council level figures in this article will be restated onto 2026 boundaries in the January 2027 publication.
General information only
This article is general industry commentary and does not constitute financial, commercial, legal or professional guidance. All figures are drawn from the Australian Bureau of Statistics release Building Approvals, Australia, July 2026, published 1 September 2026, and its additional information release of small area data published 8 September 2026. Small area figures are original terms and are not seasonally adjusted. National figures described as seasonally adjusted or trend are identified as such. Counts, rankings, shares and year on year comparisons at council and SA2 level are TGB calculations derived from the published ABS small area data cubes and are not official ABS aggregates. Readers should seek independent guidance relevant to their own circumstances.
Author: TGB Editorial
0 Comments