The Albanese Government has framed the 2026-27 Federal Budget’s negative gearing and CGT reforms as a deliberate funnel — investor demand will be steered out of established stock and into newly built housing supply. From 1 July 2027, negative gearing against salary income will only be available on eligible new builds, and the 50 per cent CGT discount will be replaced for everyone else with cost-base indexation plus a 30 per cent minimum tax on net capital gains.
On paper, the policy is a present to developers. In practice, the underlying numbers — pre-sale ratios, completion pipelines, household income, and post-February APRA lending limits — point in the opposite direction. The carve-out arrives just as the structural conditions that historically rescued distressed projects, principally cheap and abundant investor credit, are being tightened by the same government and regulator that designed the carve-out. For many apartment projects already in the pre-sale phase, the gap between policy intent and balance-sheet reality is the cliff edge.
The “new build” status has a short shelf life
Treasury’s definition of a qualifying new build is narrower than the headlines suggest. A newly constructed apartment qualifies as a “new build” when sold by the developer to the first investor purchaser, provided the unit has not been occupied — by a tenant or owner-occupier — for more than 12 months before that sale. The 12 months is a ceiling, not a precondition: an off-the-plan sale with zero prior occupancy qualifies, a unit rented for eight months and then sold qualifies, but a unit rented for 18 months before sale does not. Once the 12-month threshold is crossed, new-build status is permanently extinguished. The status also does not transfer to subsequent investor-to-investor sales in any case — the benefit is one-off, attached to the first sale out of the builder’s hands.
For a developer who finishes a tower and cannot clear stock, this is the architecture of the trap. Renting the units to cover holding costs is the obvious commercial response, and the first twelve months of rental income are safe. But the developer is then managing a rolling 12-month deadline across each individual unsold unit, not a single building-wide one. Cross the threshold on any given unit and that unit’s marketability to the next investor drops sharply — because the next buyer can no longer negative gear it against salary income.
The supply target the developers are supposed to deliver
Under the National Housing Accord, all three levels of government committed to delivering 1.2 million new homes over the five years to June 2029. Progress has been well below the trajectory required. Australia built around 219,000 homes in the first five quarters of the Accord — but the 1.2 million target requires roughly 280,000 per year, and no state is on track. Master Builders Australia’s modelling projects national completions of about 1.03 million by June 2029, leaving a shortfall of approximately 166,000 dwellings. NSW alone is forecast to finish 303,280 homes against a 377,000 target.
The National Housing Supply and Affordability Council estimates the 1.2 million target will not be met until September 2030 — over a year past the Accord period — and that was before the Middle East conflict pushed input costs higher. Sydney is the city under the most pressure: 82 per cent of new dwellings in the pipeline are townhouses or apartments, and the number of NSW construction company insolvencies tripled between 2022 and 2024.
The political need for delivery is therefore at its highest exactly when the financial conditions to deliver are worst.
The buyer pool the policy is rerouting
Investors are not a marginal participant in this market. According to Cotality, the value of new mortgage lending going to investors hit 40.6 per cent in September 2025 — the highest concentration since December 2016 and well above the decade average of 33.4 per cent. Investors account for roughly 41 per cent of all home lending nationally.
The Budget reforms do not destroy that demand, but they do redirect it. From 1 July 2027, an investor who buys an established apartment can only deduct rental losses against other rental income or future capital gains — not against salary. The arithmetic for an investor on the top marginal rate shifts materially: the after-tax cost of holding a negatively geared established unit rises, and the tax-effective premium that an investor can pay narrows.
The policy assumes that pool of demand will simply migrate to the new-build market. The unanswered question is whether the new-build market is priced to absorb it.
The affordability wall
Even before the reforms, the affordability picture for the buyer base that developers actually need was severe. PropTrack’s Housing Affordability Report found that only 14 per cent of median-income households nationally can afford the median-priced home in 2026 — down from 43 per cent just three years earlier. In Sydney that figure is 10 per cent. The national median multiple sits at 8.2 times income; Sydney’s is 10.1, placing it among the most unaffordable cities in the OECD. Sydney’s median dwelling now sits at about 13.8 times median household income.
An average-income Australian household would need to spend about 32.7 per cent of gross income to service the mortgage on a median-priced home — above the conventional 30 per cent housing stress threshold. In NSW, a median-income household can afford just 11 per cent of homes sold in the past year. To service a Sydney median unit at $850,000 comfortably, household income of roughly $160,000 is required.
This is the demand-side constraint that no tax reform addresses. A buyer who cannot service the loan does not become a buyer because negative gearing is available; the negative gearing benefit only crystallises after acquisition.
Bondi Junction as a microcosm
The user-supplied figures for Bondi Junction make the structural problem visible. A new one-bedroom apartment in the area is being marketed at roughly $35,000 to $40,000 per internal square metre. On a 55-square-metre internal floor plate, that puts the build at $1.925 million to $2.2 million. A second-hand one-bedroom in a modern building, with parking and views, transacts at around $1.1 million — approximately $18,000 per internal square metre. Current market rent for a comparable unit is around $900 per week, or $46,800 a year.
The gross yields tell the story:
- New build at $2 million → 2.34 per cent gross yield
- Established unit at $1.1 million → 4.25 per cent gross yield
At 80 per cent LVR, the $2 million new build carries a $1.6 million loan. At a mortgage rate around 6 per cent, interest alone runs to roughly $96,000 a year — and that’s only one part of the burden. APRA has been tightening interest-only lending standards since 2017, and most investment loans now convert to principal-and-interest within five years. On a 30-year P&I term at the same rate, total annual repayments rise to about $115,000, split roughly $96,000 interest and $19,000 principal in year one. The principal portion is not tax-deductible — it comes straight out of after-tax income — so the extra $19,000 is the most expensive money on the balance sheet. Strata, council and water rates, insurance, and a property manager’s fee add another $13,000 or so.
Pre-tax, the cash loss is well north of $60,000 a year on interest-only and closer to $82,000 on P&I. The tax position softens it but does not rescue it: deductible expenses (interest, holding costs, depreciation on new fixtures) net against rental income to produce a tax loss of around $70,000, which at a 47 per cent marginal rate returns roughly $33,000 in tax saving. That leaves an after-tax cash outflow of between $30,000 (interest-only) and $48,000 (P&I) each year. The investor then needs capital growth of between 1.5 and 2.4 per cent on the purchase price annually just to break even after tax, before transaction costs.
The same investor, buying the second-hand unit at $1.1 million, has a substantially smaller loan, a much smaller negative carry on both interest and principal, and starts at a yield more than 180 basis points higher. The Budget reforms are designed to remove the negative gearing benefit from that second option for new purchases — but they do nothing about the fact that the second-hand unit was always the better risk-adjusted bet. The premium the new-build buyer is paying is, in large part, a premium for the developer’s land cost, build cost, marketing, profit margin, and GST — none of which are economic value to the investor.
This is the gap that policy alone cannot close. The new build will only sell at the marketed price if either rents rise sharply, or interest rates fall sharply, or the buyer pool accepts a much lower yield in expectation of capital growth. Each of those depends on conditions outside the developer’s control.
The developer’s bind: pre-sales, holding costs, and the bank’s clock
Australian apartment development funding is structured around pre-sales. The typical residential construction loan requires roughly 50 per cent of apartments in a project to be pre-sold — under unconditional contracts — before construction can commence, with the residual settlement risk transferring to completion. This means a developer who cannot hit the pre-sale hurdle does not get to begin; one who hits the hurdle but watches the market deteriorate during the 18-to-30-month build cycle inherits the gap between contract price and completion value.
When that gap turns adverse, three things happen simultaneously. Off-the-plan buyers struggle to settle — valuations on completion come in below contract price, lenders reduce loan amounts, and individual purchasers walk away or default. Holding costs accumulate against unsold stock, with interest, rates, strata, marketing, and insurance running every day the building sits. And the financier, exposed across multiple completion-phase projects, pushes for sales — through price reductions, incentives, agent rebates, or, ultimately, mortgagee-in-possession sales.
Brisbane in 2017 and 2018 is the most recent template. Lendlease publicly acknowledged inner-Brisbane apartment oversupply in 2018; RiskWise documented 14,813 units in the inner-Brisbane pipeline in July of that year — equivalent to a 20 per cent addition to existing stock in 24 months. Landlords in the worst-hit corridors were offering free rent periods, gift vouchers and iPads to secure tenants. Apartment values fell 0.7 per cent in 2018 against a 0.4 per cent rise in house values, and Mirvac confirmed rising default rates on off-the-plan settlements as foreign and local buyers struggled to complete. Melbourne’s CBD, Docklands and Southbank traversed the same path on a longer timeline.
Sydney’s current situation is not an exact analogue — vacancy rates are tight at around 1.3 per cent according to SQM Research and unit rents are rising — but the pipeline composition, construction insolvency rate, and gap between new-build and second-hand pricing all point to a similar mechanism, particularly in transit-oriented high-density corridors where multiple projects compete for the same investor and downsizer demand.
The rental guarantee playbook — and why it’s harder to run in 2026
When direct sales fail, the historical fallback for developers has been to bridge the gap with rental guarantees. A typical structure: the developer (or a related marketing vehicle) guarantees a fixed rent for one to three years, often pitched above genuine market rent, to make the investment maths work on paper for an off-the-plan investor. The cost of the guarantee is built into the purchase price, so the investor is in effect paying upfront for their own rent.
The history of these arrangements in Australia is poor. The Charterhill Group collapse in 2014 — founder George Nowak’s company sold properties with long-term rental guarantees through related entity Australian Leasecorp — left investors with no recourse when the guarantor became insolvent. Rental guarantees are typically offered most aggressively in suburbs in construction boom, precisely when underlying rental conditions are most vulnerable to the very supply the guarantees are funding. When the guarantee period expires, the unit reverts to genuine market rent — usually well below the guaranteed figure — and the valuation correction follows.
In 2026 the playbook is structurally harder to run for two reasons.
First, the pool of investors with capacity to absorb stock at these prices is constrained by APRA’s debt-to-income cap, effective 1 February 2026. Authorised deposit-taking institutions can write no more than 20 per cent of new mortgage lending at a DTI of six times income or above, and the cap applies separately to owner-occupier and investor portfolios. APRA has explicitly identified high-DTI investor lending as the driver behind activating the tool. Loans for the purchase or construction of new dwellings are exempt — a deliberate alignment with the Budget’s new-build incentive — but that exemption sits with the lender’s quota, not the borrower’s serviceability. An investor whose DTI is over six because of an existing portfolio still faces a tighter lender shortlist. APRA has also signalled that the limit could become more influential if the share of high-DTI investor borrowing keeps rising.
Second, the negative gearing reform itself means a rental guarantee is no longer a stand-alone marketing tool. To be commercially attractive to the next investor, the guarantee period now has to end before the 12-month occupancy clock disqualifies the unit’s new-build status — or the unit has to be sold and settled to an investor before that clock starts. A developer who rents the units to bridge holding costs, even with a guarantee structure, has to manage the 12-month threshold across the entire building, not unit by unit. That converts a relatively simple sales-and-leasing operation into a tax-status race against time.
What this means
The 2026 Budget’s negative gearing reform is being marketed as supply-side policy. Read as a financial flow, it is a redirection of the same tax subsidy from established stock into the developer’s product — but it works only if the new-build market clears at the price the developer needs. The Budget does not deliver that condition; it depends on it.
For Sydney apartment developers carrying inventory at $35,000-$40,000 per internal square metre, the conditions to clear that inventory are deteriorating on three sides at once. The buyer pool can borrow less under APRA’s new framework. The household-income base in Sydney can afford roughly 10 per cent of the existing stock. And the alternative tax-effective vehicle — established stock — was always priced more attractively than the new product, even before the new product became the only path to salary-offset negative gearing.
What’s left is the rental-conversion option, hemmed in by a 12-month tax-status cliff, and the rental-guarantee option, hemmed in by APRA, by investor scepticism shaped by Charterhill and similar collapses, and by the underlying yield gap that no marketing structure removes.
The cliff edge for developers is therefore not the Budget. The Budget is the policy assumption that the cliff isn’t there. The cliff is in the spread between what the new-build market is priced at and what the buyer pool can actually pay.
Note: The negative gearing and CGT measures take effect from 1 July 2027 and are subject to the passage of legislation. Treasury fact sheet detail on the 12-month occupancy threshold for “new build” status is from the 12 May 2026 Budget papers; final scope will be confirmed in draft legislation. APRA’s DTI limit took effect 1 February 2026.
