It cost $890,000 to deliver a detached house on Sydney greenfield land in 2025, against $1,049,000 for a single apartment in a seven storey infill building. The house then sold for $1,184,000 against the apartment at $1,103,000. Cheaper to build, dearer to sell, by $159,000 and $81,000 respectively.

Those numbers come from the same piece of modelling, prepared for NSW Treasury by the Centre for International Economics and released with the NSW Productivity and Equality Commission housing paper on 29 July. They go a long way to explaining why Australia keeps building detached houses on the fringe while approved apartments sit unbuilt.

They also carry a catch that is easy to miss, and it sits in the land line.

Where the money goes on a greenfield house

This is the full delivery cost for one detached house on Sydney greenfield land in 2025.

Construction  $418,000

Required developer margin  $148,000

Site preparation and demolition  $80,000

Land acquisition  $76,000

Financing  $43,000

Local government infrastructure contributions  $41,000

Sales and marketing  $36,000

Professional fees  $26,000

State and other contributions  $16,000

Land tax and stamp duty  $7,000

Total  $890,000

Construction is 47 per cent of the total. Site preparation and demolition is nine per cent, which at $80,000 a lot is a larger number than most people outside civil work would guess. It is more than local contributions, state contributions, land tax and stamp duty combined.

Construction costs moved faster here than on apartments

Between 2023 and 2025 the total delivery cost of a Sydney greenfield house rose $136,000, or 18 per cent. Construction alone rose $90,000, or 28 per cent.

That is a steeper construction increase than apartments recorded over the same two years, where construction rose 18 per cent. Detached housing has absorbed more cost escalation, not less.

Sale prices rose 11 per cent over the same period. So the pattern is the same as apartments in direction, costs outrunning prices, but it started from a far more comfortable position. The feasibility gap fell from 42 per cent in 2023 to 34 per cent in 2024 and 33 per cent in 2025. On apartments it fell from 13 per cent to five.

Greenfield housing has taken a bigger construction cost hit than apartments and is still comfortably feasible. That is not resilience in the builder. It is headroom in the land.

The land line is not what you pay

Here is the catch. The $76,000 land acquisition figure is not the price of a residential lot. CIE is modelling the opportunity cost of the land in its rural use, meaning what the ground is worth as a paddock, because that is the economically correct input for asking whether development should proceed at all.

It is not what a developer or builder hands over for a rezoned block. CIE says so directly and provides the cross check. In greenfield suburbs such as Schofields, Marsden Park and Austral, the median value of large residentially zoned land runs to $373 per square metre. Rural land in the same areas runs to $26 per square metre. Applied to a typical 600 square metre lot, including the land allocation for local roads, that is $223,871 against $15,668.

What is a rezoning premium?

The rezoning premium is the jump in land value that happens the moment a site is zoned for housing rather than rural use. Nothing about the physical site changes. The legal permission to build on it does, and in a supply constrained market that permission is worth a great deal.

It matters commercially because the premium is captured by whoever owns the land at the point of rezoning. A builder or developer buying in afterwards pays it as a cost. The modelled feasibility gap is therefore an estimate of the surplus available from development, not a description of who ends up with it.

The margin is real, but it is mostly already spent

CIE runs that comparison deliberately, as a sanity check on its own result. If greenfield development carries a large surplus, that surplus should show up in the value gap between rezoned land and rural land, because landowners capture some or all of it.

It does. The gap between the two land values, per 600 square metre lot, is around $200,000. The modelled feasibility gap per dwelling is $293,000. In other words, the great majority of the apparent greenfield margin has already been paid across to the landowner before a builder prices a slab. That is the same mechanism as the value a rezoning adds to land, showing up in the cost stack rather than the sale price.

CIE is careful about the limits of the comparison. Some large residentially zoned lots carry low values because they are flood prone or otherwise constrained, and some rural lots carry high values because a future rezoning is anticipated. But the direction is not in doubt.

For a builder, the practical reading is straightforward. A 33 per cent feasibility gap on paper does not mean 33 per cent is available to whoever builds the house. Most of it was transacted upstream.

The council line that swings $25,000

Local government infrastructure contributions are the single most variable cost in the greenfield stack, and the variation is not driven by anything a builder controls.

CIE estimated the average charge per dwelling across greenfield local government areas.

Blacktown  $40,657

Camden  $40,625

Liverpool  $34,673

The Hills  $30,478

Lake Macquarie  $30,000

Campbelltown  $25,851

Wollondilly  $24,272

Newcastle  $22,397

Penrith  $18,016

Wingecarribee  $16,980

Maitland  $16,854

Central Coast  $15,971

The spread between Blacktown and the Central Coast is $24,686 per dwelling. On the modelled greenfield cost stack that is nearly three per cent of total delivery cost, decided entirely by which side of a council boundary a lot sits on.

IPART sets maximum contribution rates of $30,000 per dwelling for greenfield development and $20,000 for infill, and councils can apply to charge above the cap. Four of the areas above sit over the greenfield maximum. CIE also identifies Blacktown, Baulkham Hills and Hawkesbury, and South West Sydney as the areas where local infrastructure charges are highest, alongside higher construction costs.

These are CIE estimates drawn from council contribution plans and averaged where boundaries cross multiple plans, so they are a comparison tool rather than a rate any single project will pay. Used that way they are genuinely useful when pricing work in a growth corridor, because the number is knowable before a lot is bought and it moves the feasibility of an entire estate.

Where it still stacks up

Every greenfield market CIE modelled returned a positive feasibility gap in 2025. That is a genuinely different picture to apartments, where development is generally only feasible east of Parramatta.

Baulkham Hills and Hawkesbury leads at 43 per cent. Only two markets fall below 30 per cent, being Sydney South West at 26 per cent and the Hunter Valley excluding Newcastle at 28 per cent. The strongest feasibility by area sits in Sydney’s north west.

That is the backdrop for anyone weighing detached work against infill over the next few years. Fringe housing is under the same cost escalation as everything else in the conditions across the Australian housing market, but it has a buffer that apartment projects have already spent.

Why the pipeline keeps producing detached houses

Put the two typologies side by side on the same modelling and the market behaviour stops being mysterious. A detached house on the fringe costs less to deliver, sells for more, and clears its required margin by 33 per cent. A mid rise apartment costs more, sells for less, and clears by five per cent, or not at all west of Parramatta.

None of which settles the policy question. The Commission’s argument is that housing should be concentrated where infrastructure already exists and demand is highest, which points inward rather than outward, and it costs more per home to service growth on the fringe. The counter argument, that greenfield can deliver at scale, is not weakened by this modelling. If anything it is supported by it.

What the modelling does settle is the commercial question. On current numbers, the fringe is where the feasibility is.

The Good Builder Take

Three things worth taking from this modelling.

One. Greenfield detached housing has absorbed a 28 per cent construction cost increase in two years, steeper than apartments, and is still comfortably feasible. The buffer is real.

Two. The buffer is in the land, and most of it is captured at rezoning. A 33 per cent feasibility gap is a measure of the surplus that exists, not the surplus available to whoever builds. Anyone reading these numbers as builder margin is reading them wrong.

Three. Council contributions swing nearly $25,000 a dwelling between the highest and lowest greenfield LGAs. That is one of the few large cost variables you can establish before committing to a site.

Frequently asked questions

Sources: Centre for International Economics, Cost and feasibility estimates for supplying new dwellings in New South Wales, 2025 Edition, prepared for NSW Treasury, 27 July 2026, chapter 4 and appendix A. NSW Productivity and Equality Commission, Building more homes where the numbers stack up, July 2026, published 29 July 2026. Local infrastructure contribution caps as set by IPART.

Disclaimer: This article is general information only. It is not legal, financial, planning or investment advice. All figures are modelled estimates published by the sources cited and are averages across defined greenfield areas, so they will not reflect the economics of any individual site or lot. The land acquisition figure is an opportunity cost of rural land, not a market price for residential land. Developer contribution figures are estimates derived from council plans and rates change. Recommendations by the NSW Productivity and Equality Commission do not become NSW Government policy unless explicitly adopted. Builders, developers and trades professionals should seek independent professional input relevant to their own circumstances before making commercial decisions.

Last updated: 4 August 2026.

Author: TGB Editorial