Australia’s construction pipeline is usually described as one story: a record volume of new dwelling approvals, concentrated in a handful of high-growth LGAs. The tenure data underneath that number tells a different story. The highest-volume approval LGAs split into two tenure profiles that are almost opposites of each other — and the split matters for who actually carries the risk sitting inside a national headline figure.
Same volume, opposite household
Melbourne (LGA) approved 7,794 new dwellings over the fixed window July 2024 – June 2026 (24 months). Melton, on Melbourne’s growth fringe, approved 9,119 — more, in absolute terms. But the households living behind those two numbers look nothing alike. Melbourne LGA is 67.9% rented and 16.1% mortgaged. Melton is 53.5% mortgaged and 23.1% rented. Almost the same construction volume, opposite tenure composition.
Mortgage share, high-volume approval LGAs
Source: ABS Census 2021 tenure (APN Codex series 158/159/165), ABS 8731.0 LGA small-area approvals (July 2024 – June 2026, 24-month total). All six are among the fifteen highest-volume approval LGAs nationally.
The pattern holds across the top fifteen approval-volume LGAs, not just this pair. Brisbane (16,566 dwellings approved, 38.8% rented, 33.8% mortgaged) and Gold Coast (11,678 approved, 34.4% rented) sit in the CBD/inner-metro cluster. Wanneroo (7,442 approved, 53.9% mortgaged), Casey (7,189 approved, 51.8% mortgaged), Wyndham (8,598 approved, 48.8% mortgaged) and Hume (6,822 approved, 48.0% mortgaged) sit in the growth-corridor cluster. Two clearly separable geographies, contributing similar volumes to the same national approvals figure.
The corridor is building faster relative to its size, not just in absolute terms
Absolute volume favours big cities simply because they are big. Scaling approvals by each LGA’s occupied-dwelling base removes that effect — and the tenure pattern gets stronger, not weaker. Mortgage share correlates with approval intensity (dwellings approved per occupied dwelling, July 2024 – June 2026) at +0.421 across 349 LGAs. Outright-ownership share correlates with the same measure at −0.295.
Read together: it is not simply that a few large growth-corridor LGAs happen to approve a lot of dwellings. Construction activity relative to the existing housing stock is systematically higher where mortgage share is higher, and systematically lower where outright ownership is higher. The corridor is not just contributing volume — it is building proportionally faster than the rest of the country.
The relationship across the full range, not just the top and bottom deciles
Ranking all 349 matched LGAs by mortgage share and grouping them into ten equal deciles shows the same pattern holds broadly, with one genuine feature worth naming rather than smoothing over: the relationship is fairly flat through the middle of the distribution and then accelerates sharply in the highest-mortgage decile.
Average approval intensity by mortgage-share decile, decile 1 (lowest) to decile 10 (highest)
D1 averages 17.5% mortgage share; D10 averages 49.3%. Intensity drifts within a narrow 2.4%–4.4% band through deciles 2–9, then roughly doubles in D10 (6.87%). The effect is concentrated at the top of the mortgage-share range, not spread evenly across it.
Decile 1 (the lowest mortgage-share LGAs, averaging 17.5%) shows a build intensity of 3.21% — almost as high as the middle deciles. That is not an anomaly in the data; it reflects a genuine second driver of high build intensity that sits alongside the mortgage-belt story: a handful of low-mortgage, high-intensity LGAs that are inner-city apartment markets rebuilding on a small existing base, not growth-corridor house-and-land estates.
The exception worth naming: Melbourne builds fast too, for a different reason
Ranking all matched LGAs by build intensity rather than absolute volume produces a top-12 list dominated by growth-corridor and regional LGAs — Melton (16.83%), Murray (16.03%), Playford (13.03%), Kwinana (12.80%), Wollondilly (11.58%), Serpentine-Jarrahdale (11.25%) — but Melbourne LGA itself sits at position nine, with 10.93% intensity, despite being the most renter-heavy, least-mortgaged LGA in the country (67.9% rented, 16.1% mortgaged).
Top 12 LGAs by build intensity (approvals per occupied dwelling, 24 months)
| LGA | Intensity | Dwellings (2yr) | Mortgage % | Rented % |
|---|---|---|---|---|
| Melton | 16.83% | 9,119 | 53.5% | 23.1% |
| Murray | 16.03% | 1,064 | 39.6% | 19.2% |
| Augusta Margaret River | 13.16% | 800 | 36.5% | 24.9% |
| Playford | 13.03% | 4,540 | 41.1% | 36.9% |
| Kwinana | 12.80% | 1,984 | 53.4% | 28.1% |
| Wollondilly | 11.58% | 1,994 | 49.4% | 16.6% |
| Mitchell | 11.44% | 1,940 | 47.8% | 21.3% |
| Serpentine-Jarrahdale | 11.56% | 1,173 | 63.6% | 12.8% |
| Melbourne | 10.93% | 7,794 | 16.1% | 67.9% |
| Wanneroo | 10.52% | 7,442 | 53.9% | 22.3% |
| Light | 10.46% | 564 | 50.1% | 12.8% |
| Adelaide Plains | 10.00% | 336 | 56.6% | 11.0% |
Source: ABS Building Approvals (8731.0) LGA small-area supplement and ABS Census 2021 dwelling tenure by LGA. Intensity = 24-month approved dwellings ÷ occupied-private-dwelling base.
Ten of the twelve are mortgage-heavy growth-corridor or regional LGAs, exactly as the mortgage-belt story predicts. Melbourne is the genuine exception, and it is worth explaining rather than ignoring: a small existing dwelling base combined with high-density apartment redevelopment can push build intensity high even in a renter-dominated market. The corridor effect and the inner-city-redevelopment effect are two separate routes to a high intensity score — they happen to overlap in volume terms but not in tenure terms. The mortgage-belt pattern remains the dominant one across the dataset; Melbourne is the one large exception rather than evidence against it.
What this means for the record pipeline
APN Codex’s national Work Yet To Be Done reading for residential construction sits at a record $80.3bn (Q1 2026, +2.53σ), with 243,864 dwellings under construction nationally. That figure is a single national number. The tenure-and-approvals geography shows it is not evenly distributed risk. A meaningful share of the pipeline sits in growth-corridor LGAs where the household on the other end of the build is, on the tenure data, more likely to be mortgaged than owning outright or renting — a young or middle-income family carrying both the mortgage and the wait for the dwelling to be finished, rather than an investor, a renter, or an outright owner.
This does not change the national pipeline figure. It changes what the figure represents: not a uniform national backlog, but a backlog concentrated disproportionately on the geography and household type least able to absorb a delay — the outer-suburb mortgage belt, not the CBD rental market or the paid-off rural home — with a smaller, separate concentration in inner-city apartment redevelopment that carries a different risk profile again.
Findings are presented on the basis of data and evidence alone.
