Zoning mix drives house prices, rents, growth and yield
2,856 NSW and Victorian suburb-localities, house repeat sales 1992–2026.
Headline for the retail investor buying existing housing stock. A house sitting inside a district with heavy industrial zoning trades at a persistent capital-growth headwind of roughly 0.2 percentage points per year for every ten-percentage-point uplift in that surrounding industrial share. A house inside the established middle-ring of medium-density residential zoning trades at a mild growth tailwind of similar size and runs expensive. A house with heavy rural surrounds trades at 40–60 per cent of the typical NSW or Victorian house price and earns a gross rental yield two to three percentage points above the benchmark. Heavy apartment and mixed-use surrounds in the inner ring compress yield by around one percentage point. These patterns show up in the established housing stock over the full 1992–2026 window, with repeat-sale medians across thousands of suburbs.
Scope note: this whitepaper focuses on established housing stock in long-settled suburbs. We exclude fringe urban-growth areas (Victoria’s UGZ precincts on the outer Melbourne ring) from the worked examples because their headline repeat-sale growth is inflated by developer-to-market handoffs and new-build repricing rather than the investment thesis a retail buyer of existing stock is running.
Contents
1. What we measured
For each of 2,856 suburb-localities in NSW and Victoria with thirty or more house repeat sales, we computed four outcomes: median annualised capital growth rate per property (buy price to sell price of the same house, held at least two years), median house sell price 2020–2026, median advertised weekly rent, and gross rental yield. Every growth figure is a per-property repeat sale, not a change in suburb median. The features are area shares of sixteen zoning categories in six distance rings around the suburb centre.
2. Capital growth: how much each zone shifts it
Effects below are from the full multi-target fit, restricted to established-stock signals relevant to a retail investor. Each row reads as “every ten-percentage-point uplift in the share of land within the given ring that falls in the named meta category shifts annualised capital growth by this many percentage points per year, other things held equal.”
| Zoning meta | Ring | Effect on capital growth per +10pp share |
|---|---|---|
| Industrial | 1,600 m | -0.21 pts/yr slower |
| Commercial | 1,600 m | -0.17 pts/yr slower |
| Medium density | 3,200 m | +0.16 pts/yr faster |
| Mixed use | 1,600 m | +0.15 pts/yr faster |
| Capital-city centre | inside 400 m | -0.14 pts/yr slower |
| Park | 800 m | -0.14 pts/yr slower |
| Medium density | 1,600 m | -0.13 pts/yr slower |
| High density | 3,200 m | +0.11 pts/yr faster |
Read two ways. A suburb with 30 per cent industrial zoning inside the 1,600 m ring carries a growth headwind of 0.6 percentage points per year from that alone. A suburb set inside a 3,200 m district where medium-density residential zoning covers 60 per cent of the land picks up roughly 1 percentage point per year of growth tailwind, a reliable pattern in the established middle ring of Melbourne and the inner-west of Sydney. The park and capital-city coefficients turn negative because they co-locate with inner-city apartment stock that grows slower than detached-house stock further out.
2.1 Industrial-surround illustration
Chester Hill (Sydney, NSW, south-west) sits in a district where 27 per cent of the land within 1.6 km is industrial and a further 16 per cent of the 3.2 km outer ring is industrial. Across 197 house repeat sales 2020–2026, median sell price is $950,000, close to the NSW state median. The full-period median annualised growth runs 0.5 percentage points per year below the NSW and Victoria house benchmark. The entry price is accessible, but the industrial surround sits as a persistent drag on long-run capital growth.
2.2 Industrial right next door is the sharpest single signal
The 3.2 km coefficient summarises the broad-district industrial effect on capital growth, but it understates how much the immediate neighbourhood matters. Restricting the same analysis to the 200 m near-neighbour ring shows a much sharper pattern. Of 2,795 NSW and Victorian suburbs with thirty or more house repeat sales, only about 80 carry an industrial share above 10 per cent inside the immediate 200 m ring. Among those, the binned median capital growth runs 0.5 to 0.75 percentage points per year below the NSW and Victoria house benchmark across the 20–50 per cent share bands.
| Industrial share within 200 m | Suburbs in band | Median CG vs benchmark, 1992–2026 |
|---|---|---|
| 0–10% | 2,732 | −0.04 pts/yr (essentially benchmark) |
| 10–20% | 17 | +0.25 pts/yr |
| 20–30% | 15 | −0.75 pts/yr |
| 30–40% | 9 | −0.58 pts/yr |
| 40–50% | 7 | −0.54 pts/yr |
| 50–60% | 5 | +1.41 pts/yr (5-suburb outlier band) |
The retail-investor read: a check of the 200 m walking ring around any candidate property catches the immediate-frontage industrial signal in a way the broad-district figure cannot. The 100 m ring is even sharper but the sample is too thin to read with confidence outside the very heavy industrial bands. For most established residential suburbs the question is binary — there is no industrial inside the 200 m ring at all — and the negative signal only kicks in for properties with depots, warehouse strips, or workshop clusters in the immediate next-door land.
2.3 Medium-density middle-ring illustration
Mount Waverley (Melbourne, VIC, south-east) sits at 63 per cent medium-density residential zoning inside its 3.2 km ring. Across 762 house repeat sales 2020–2026, median sell price is $1,550,000, around 1.8 times the NSW and Victoria house median. The full-period median annualised growth runs roughly 2.4 percentage points per year above the benchmark. This is the pattern retail investors recognise as the established middle-ring premium.
3. Price levels: where rural surrounds cut prices in half
The zoning mix explains nearly half of the suburb-level variation in median house prices, and the picture is led by two categories at the broad 3.2-kilometre scale.
Suburbs whose outer ring is heavily rural price well below the NSW and Victoria house median. At one extreme, Castlecrag (Sydney, NSW, Lower North Shore) has zero rural share inside its 3.2-kilometre ring and a median house price of about $3.68 million, more than four times the NSW and Victoria house median. At the other, 2,044 suburb-localities in our sample sit above median rural shares and trade for between forty and seventy per cent of the NSW and Victoria house benchmark. Industrial surrounds depress prices nearly as strongly as rural, and the effect is roughly linear in the share of surrounding land used that way.
High-density residential surrounds (apartment zones, typically within two kilometres of a commuter rail station) push prices up for the detached houses inside them. This reads as the urban premium: proximity to density is proximity to amenity. Mixed-use surrounds produce a similar, smaller positive effect.
4. Rent levels: the same rural discount, softer
The zoning-mix picture explains a third of the variation in median weekly rent across suburbs. The direction mirrors the price picture but at about half the strength. Rural surrounds cut rent roughly 40 per cent relative to the NSW and Victoria house rent benchmark. Industrial surrounds cut it by a similar margin. Commercial surrounds, unlike their effect on capital growth, lift rent, because jobs and amenity are close by even when long-run capital growth is slower. Mixed-use surrounds push rent up on the same logic.
The two effects that price and rent share most consistently are direction and rank order. If zoning mix makes a suburb expensive to buy, it usually also makes it expensive to rent. The offset between the two determines yield.
5. Yield: the cleanest zoning story of the four
Because rural surrounds cut prices more than they cut rent, suburbs with rural surrounds run higher gross yields. The zoning mix predicts gross yield more accurately than growth and rent, and close to how well it predicts price. A suburb with a fully rural 3.2-kilometre ring earns a yield more than two and a half percentage points above the NSW and Victoria house yield benchmark. A suburb with a heavily high-density surround, inside the classical inner-city ring, earns a yield more than a percentage point below the benchmark. In gross-yield terms, that is the difference between inner Sydney and the New England Tablelands.
For illustration, Brighton (Melbourne, VIC), a low-density beachside suburb with some commercial amenity but no rural surrounds, runs a gross yield around 0.7 percentage points below the benchmark on a median price near 3.7× the benchmark. Cessnock (NSW, Hunter Valley), a long-established regional centre with 40 per cent rural within 3.2 kilometres, runs a gross yield about 0.8 percentage points above the benchmark on a median price around 0.65× the benchmark. Both outcomes are consistent with the zoning-mix picture.
6. Worked examples
The eight suburbs below illustrate how the zoning mix at two different distances tracks each of the four outcomes. All figures are relative to the NSW and Victoria house benchmark (capital growth, gross yield) or expressed as a multiple of the NSW and Victoria house median (price, rent) and all growth figures span 1992–2026.
| Suburb | City | Zoning mix (3.2 km) | Capital growth vs benchmark | Price vs benchmark | Rent vs benchmark | Gross yield vs benchmark |
|---|---|---|---|---|---|---|
| Cessnock | Hunter Valley, NSW | 40% rural, no industrial nearby | -0.7 pts/yr slower | 0.65× | 0.80× | +0.8 pts higher |
| Sale | Gippsland, VIC | 49% rural, limited commercial | -0.9 pts/yr slower | 0.55× | 0.81× | +1.4 pts higher |
| Chester Hill | Sydney SW, NSW | 27% industrial in 1.6 km ring | -0.5 pts/yr slower | 1.11× | 1.11× | +0.1 pts higher |
| Mount Waverley | Melbourne SE, VIC | 63% medium density in 3.2 km ring | +2.4 pts/yr faster | 1.81× | 1.16× | -0.9 pts lower |
| Sunshine North | Melbourne, VIC | 32% medium density, no rural | +1.6 pts/yr faster | 0.91× | 0.79× | -0.3 pts lower |
| Brighton | Melbourne, VIC | Low density bayside, no rural | +3.6 pts/yr faster | 3.66× | 2.63× | -0.7 pts lower |
| Carlton | Melbourne, VIC | Capital city + mixed use | +2.0 pts/yr faster | 1.69× | 1.49× | -0.2 pts lower |
| Castlecrag | Sydney, NSW | Low density + park, no rural | +0.3 pts/yr faster | 4.29× | n/a | n/a |
| Pyrmont | Sydney, NSW | Capital city + high density | -0.9 pts/yr slower | 1.78× | 1.69× | -0.0 pts |
The nine suburbs span the retail-investor decision set: regional rural surrounds that compress price and lift yield (Cessnock, Sale), industrial surrounds that drag growth (Chester Hill), established medium-density middle ring that runs above benchmark on growth (Mount Waverley, Sunshine North), blue-chip low-density surrounds (Brighton, Castlecrag), inner-ring capital-city and mixed-use surrounds (Carlton), and the inner-Sydney compressed-yield apartment belt (Pyrmont). Each example is a real suburb and the numbers are the actual medians for its houses.
7. One crosswalk, two states
An incidental contribution of the study is the zoning crosswalk itself. Until now, NSW and Victoria have been effectively impossible to compare like-for-like, because NSW uses land-use labels (“Low Density Residential”, “Environmental Living”) and Victoria uses schedule codes (“GRZ3”, “NRZ1”, “UGZ4”). We built an explicit mapping table that unifies 262 source labels into the same sixteen meta categories with 99.9 per cent coverage. The crosswalk is the layer that makes a national, multi-state zoning study possible.
Queensland is included descriptively only. The statutory zoning polygons we can load for Queensland cover the South East Queensland regional plan and distinguish only three very broad categories (Urban Footprint, Rural Living Area, Regional Landscape and Rural Production Area). That is too coarse to enter the model. A QLD extension of this work would require council planning-scheme polygons from Brisbane, Gold Coast, Sunshine Coast, Moreton Bay, Logan, Ipswich and Redland.
8. Defence against criticism
Is this a zoning effect or a location effect? The honest answer is that it is both, and we do not attempt to identify one from the other. Zoning correlates with distance from the central business district, with school catchments, with coastline, and with historic subdivision patterns. We claim only that a buyer who knows the zoning mix within a few kilometres of a candidate property knows something real and portable about how it will price, rent, grow and yield. An identification strategy that isolates the marginal effect of zoning itself would require an exogenous rezoning event and is a separate question.
Why house-only? Unit stock is concentrated in the high-density and mixed-use categories by construction. If we measured growth on mixed dwelling types we would see what looks like a penalty on those zones, but most of that penalty would just be the unit-to-house composition effect. The study restricts to detached houses throughout so that the zoning signal is not contaminated by dwelling-type mix.
Is the signal real or an artefact of sample size? We scrambled every suburb’s zoning mix inside its own state and re-ran the same comparison. Every explanatory figure fell back to the state-only comparison. The observed signal does not survive random rearrangement, which rules out sample-size inflation and state-level level effects as the explanation.
9. Limitations
The zoning polygons are current, not historical. A suburb whose zoning was different a decade ago is measured as if the current configuration held all along. The unit of analysis is the suburb-locality, not the individual parcel. A parcel-level version of this study would need property-specific surroundings rather than suburb-level surroundings. Queensland, South Australia, Western Australia, Tasmania, ACT and Northern Territory are out of scope for the predictive portion. The crosswalk is extensible but is not yet extended.
10. Conclusion
Before an offer on any NSW or Victorian house, check the zoning mix inside the 1.6-kilometre ring around the candidate. A high share of industrial zoning is the clearest repeat-sale drag on long-run capital growth and is the single most useful filter for a retail investor. A high share of medium-density residential in the 3.2-kilometre district tracks with a mild tailwind. A heavy rural surround means cheaper entry and a yield premium, at the cost of slower long-run price growth. Apartment and mixed-use surrounds near the CBD deliver the trophy-stock premium and the compressed yield that comes with it.