How the numbers were produced

The Australian Zoning Atlas is the first comprehensive account of residential zoning nation-wide. The Atlas is a compilation of geospatial data from planning systems and schemes from across the country, categorised and transformed into comparable measures.

The primary question the Atlas seeks to answer is this: how much residential land across Australia’s eight capital cities is subject to highly restrictive planning controls?

In the context of the Atlas, “highly restricted” is defined as one or more of the following:

• Height is capped at two storeys or fewer,
• Only detached dwellings are permitted,
• The zoning carries an explicit “low-density” designation, or
• The site is subject to heritage controls.

The Atlas is purposefully conservative in answering this question. Where controls are ambiguous, the methodology resolves in the direction that makes land look less restricted. The headline numbers should be considered a relatively conservative floor, even given a roughly 2 percentage-point margin of error. All results have been subjected to rigorous sensitivity testing.

The analysis finds resolutely that the majority of Australia’s capital city land within 20 kilometres of city centres is subject to highly restrictive planning controls.

All analysis is at the level of the individual parcel

The unit is the cadastral lot (parcel)—the individual title boundary on the ground.

• Results are weighted by land area. Every measure is weighted by the lot’s land area in square metres. A big lot has more development potential than a small one, and so the figures describe land, which cannot be changed, rather than lot counts, which can.
• Each parcel is matched to one zone. A parcel is assigned the zone that its central point lands in. This avoids assigning planning controls to parcels that merely touch the edge of a control polygon.
• Stacked strata lots are collapsed. Apartment strata lots that share a single footprint are collapsed back to their underlying parcel so each piece of land is only counted once.
• Only non-rural, residential-permitting land is counted. We only assess zones that typically permit housing, and explicitly exclude rural land from our analysis. Industrial, commercial, and special-use land is also dropped.

The per-city CBD anchors, study-area boundaries, data vintages and per-state source datasets live on the data-inputs page.

The headline “highly restricted” metric has four triggers

A parcel is counted as highly restricted if any one of four controls applies:

1. it is limited to no more than 2 storeys, or
2. it permits detached housing only, or
3. it sits in a “low-density” zone, or
4. it is heritage-listed (state or local).

restricted = no more than 2 storeys ∪ detached-only ∪ low-density ∪ heritage

The union is computed at the parcel level, so a parcel touched by two, three or all four of these controls is only counted once. In practice, these measures overlap heavily—land that permits only detached dwellings is almost always also designated “low-density” or capped at 2 storeys.

It is worth noting the following about the Australian Zoning Atlas:

• It measures what the planning rules permit, not what is built,
• It is deliberately biased toward overstating permissiveness, meaning that the figures may underestimate restrictiveness, and
• It is not a feasibility model, a demand forecast, or a measure of the gap between what is permitted and what exists.
• It is a resource for policy and academic analysis at the suburb level and above,
• It is not a resource for commercial decisionmaking at the lot-level.

The Atlas’s restrictiveness estimates are conservative

The whole Australian Zoning Atlas pipeline is biased toward overstating permissiveness. Every ambiguous call is resolved so that land looks less restricted.

Three choices carry most of this:

• Unknown height limits never count as restricted. Land with no published height limit stays in the denominator (it is residential) but is never added to the 2-storey numerator. Zones with no fixed cap are treated as high-density. We never claim a limit we cannot see.
• The representative-point overlay test is strict. A parcel is flagged as being under a control such as a heritage overlay only when its interior point falls inside a heritage polygon, rather than when the lot merely touches an overlay edge. This means that we likely underestimate heritage on the margins.
• Rural and fringe land is excluded. Rural-zoned land, even where it permits residential use and is within a city’s Significant Urban Area, is excluded from the Atlas. This systematically reduces the proportion of highly restricted residential land reported in our final measures.

The reform-uplift model follows the same principle, as set out above. The full catalogue of modelling choices, each tagged with its direction of bias, can be found on the assumptions page.

The numbers are well-tested and robust

Generating a headline such as “88% of Sydney’s inner-20 km residential land is highly restricted” depends on a chain of interpretive choices.

This includes codifying a metres-to-storeys ratio, interpreting heritage controls, choosing a city centre, and deciding the threshold for including a parcel within the bounds of a control.

To ensure that the result is not an artefact of choices made, the whole pipeline has constantly been re-run against a set of counterfactual assumptions. These include:

• Progressively stricter “carve-outs” of restricted land,
• Treating already-developed heritage land as unrestricted,
• Varying the metres-to-storeys conversion,
• Randomly flipping 1–10% of parcel flags,
• Recomputing the headline at radii from 3 to 30 km, and
• Dropping the smallest parcels from the analysis altogether.

The Atlas’s headline findings hold under every one of these controls. The mechanics of the perturbation harness live on the code appendix.

The most methodology-dependent outputs are Melbourne and Sydney’s inner-5 km figures. Under the most permissive carve-outs their inner-5 km figures fall to roughly 47% and 53%—due largely to the high levels of heritage controls within the cores of these cities, as opposed to the other three “highly restricted” triggers. The six other capital cities have a wider application of all four controls across their cores, and so their results vary far less than our nation’s two largest cities.

The data has three top-level limitations

This data is intended for a high-level, cross-city overview and is not suitable for detailed, lot-by-lot analysis. Accuracy varies with the availability of cadastral data and the manual coding required for local planning variations.

• Restrictive covenants and lease conditions are not measured. The analysis is solely planning-based; non-planning restrictions (covenants, ACT lease variations) are out of scope.
• Environmental constraints are not measured. Unless it’s codified in the available height limit data or the zoning itself, environmental planning controls, whether it be airport flight paths to flooding risks, are not measured or factored into the analysis.
• Height coverage varies by state. Some jurisdictions have rich local height data; others rely on statewide provisions or zone defaults. Where a height is unknown it is recorded as missing and not restrictive—never as zero storeys.

The full per-state accuracy ratings and the complete assumption-by-assumption limitation register live on data-inputs and assumptions.

What the data does not seek to measure

This data should not seek to measure constraints beyond the planning system. This analysis has not factored in the following:

• Infrastructure constraints,
• Developability of sites, and
• Economic feasibility.

The output data should be taken as an upper limit of what is possible when the above constraints are not considered.

The reform-uplift calculations are fully documented

The reform page of the Atlas tests the potential of nation-wide zoning changes.

In the initial release, four reforms are tested:

• Townhouses everywhere,
• Mid-rise near transit,
• Mid-rise everywhere, and
• High-rise near the city centre.

For each reform, the model cycles through every residential parcel and asks: would this rule actually let you build more homes than the current rules allow?

The number is conservative by construction, for four distinct reasons:

• It counts only the extra homes. Not the total a lot could hold—only the additional dwellings the reform adds on top of what is already permitted. A lot that already allows an eight-storey building gets zero uplift from an “eight storeys near transit” reform, even if the lot is currently not fully built out. The model measures changes in permissibility, rather than the total.
• Each lot is stepped down to the form its size and frontage can actually hold. A reform might say “eight storeys allowed here,” but it is often not possible to fit an apartment block on a tiny lot. Before a lot is credited with uplift, it is tested against a size-and-frontage gate and stepped down to the tallest form it can genuinely take:

• Each lot’s uplift is equivalent to the highest form it qualifies for. Under an eight-storey reform, a lot that only clears the townhouse gate is credited with a three-storey townhouse, not eight storeys. The eligibility check can only ever remove or shrink a lot’s contribution, never inflate it.
• Each lot is only counted once. When reforms overlap—a lot near the CBD covered by both “high-rise near the centre” and “mid-rise everywhere”—the two are not added together. The model takes the highest density a given lot would be able to build and counts it once.
• Unconfirmable lots count for nothing. Lots that already allow tall buildings count for nothing; lots whose eligibility we cannot confirm count for nothing.

What this is and isn’t. It is a measure of what the planning rules would permit, made realistic by checking each lot’s size and only counting genuine increases. It is not a market forecast—it does not predict demand, prices, whether owners sell or build, or construction costs. It answers “what would become possible”, not “what will get built.”

The exact eligibility parameters and per-hectare yield calculation can be found on the assumptions and data outputs pages.

Where to go next

• Data inputs — where the numbers come from, state by state: study area, vintages, source datasets with accuracy ratings, and per-city special handling.
• Assumptions — every modelling choice, categorised, each with its direction of bias. No code.
• Glossary — plain-language definitions of every term, metric and acronym.
• Code & mechanics — the technical appendix: how the computation actually runs, in the abstract.
• Data outputs — what the project produces: the headline metrics, geographic granularity, reform outputs, and the data product.
• FAQ — quick answers for a general and skeptical reader.