Australian House Price Forecast 2026–2027: How Far Will Prices Fall?

Published 3 October 2026 · Prepared by the Property Investment Professionals research desk · Data last checked 2 October 2026

Australian home values fell for a sixth straight month in September. Cotality's national Home Value Index, released on 1 October, dropped 1.1% and is now 5.2% below its March 2026 peak, and 97% of capital-city suburbs lost value over the quarter (Cotality, 1 October 2026). On 29 September the Reserve Bank lifted the cash rate to 4.60%, its fourth rise this year and the highest level since 2011.

This house price forecast for Australia in 2026 and 2027 starts by making the published numbers comparable, because as released they aren't. Some measure a fall from peak to trough, others the change over a calendar or financial year. Some cover the five largest capitals, some all eight, some the whole country. Most were finalised before the 29 September hike, on rate assumptions that are now out of date.

Three layers come out of that work. The main cluster of formal peak-to-trough forecasts sits at 9% to 13%. The full published range runs from 7.3% (Westpac, five capitals) to 15% (the top of AMP's range). Beyond that sit stress cases that aren't forecasts: AMP's worst case of about 20% and the RBA's hypothetical further 20% fall. On Cotality's index, about 40% to 60% of the fall in that cluster has already happened.

How far will house prices fall in Australia?

The forecaster scorecard

Source: as listed in the Source tier column; full links under Sources. Forecasts are the institutions' own. Primary means the institution's own page or report; secondary means a named news outlet reporting it. We could not reach a primary page for AMP's October range or ANZ's August update; each figure matches across at least two outlets. SQM's March forecast stays, although it predates the Budget, because its published "aggressive rate rises" case covers a cash rate of 4.5% or more.

Source: the forecasters in the scorecard (published 11 August to about 1 October 2026); Cotality Home Value Index, September 2026. Values in the table above.

Primary notes for the figures that carry most weight: CBA, 1 September, Westpac Housing Pulse, 25 September, NAB Forward View, 17 September and KPMG, August.

Rate assumptions and status at 2 October

Source: the forecasters' publications in the scorecard; CommBank newsroom (21 and 30 Sep 2026); Westpac IQ (30 Sep 2026); ABC News (30 Sep 2026) for ANZ's November call. The status column is our assessment.

What changed since the August forecasts

We sort the updates into three categories and use them the same way throughout.

• Formal forecast revisions (before the hike): Westpac cut its 2026 call from −3% (June) to −6% in September; NAB moved from −5% to −6%; Macquarie from −5% to about −10%; HSBC from −8% to −13%. CBA moved from a 3% rise for 2026 (May) to a 9% peak-to-trough fall on 1 September, and on 21 September shifted its rate path to cuts in August and November 2027.
• Since 29 September: no bank has made a formal forecast revision. AMP has published a new range of 10% to 15% (from about 10% on 1 September). Two analyst comments followed on 1 October, the day the September index data came out: UBS's Tharenou (Reuters) and Cotality's Lawless (ABC).
• Unchanged or overtaken: ANZ (11 August) and KPMG (4 August) still stand as published; Domain and PropTrack (June) and SQM (March) assumed a 4.35% peak.

Where we can trace an institution's 2026 revisions (ANZ from June to August, Westpac, NAB, Macquarie, HSBC and CBA), each has been downward.

The peak-to-trough calls on a common basis

Only the peak-to-trough calls answer the question "how far will prices fall". The formal ones are 7.3% (Westpac, five capitals), 9% (CBA, national), about 10% (Macquarie), 10.6% (ANZ, capitals), 10% to 15% (AMP, national) and 13% (HSBC, national). The middle of those figures is about 10%; that is our arithmetic, not an industry consensus.

Calendar-year numbers cut the fall at 31 December. Westpac's and NAB's −6% for 2026 sit alongside 2027 forecasts of +3% (Westpac) and "modest growth in the second half" (NAB), so both expect most of the fall this year. ANZ's 2027 figure of −3.4% shows it expected the decline to run longer.

When forecasters expect the trough

CBA expects the adjustment "to continue through the first half of next year". AMP sees the bottom "around the June quarter of next year". NAB expects the market "to stabilise in early 2027 before returning to modest growth in the second half of the year, as the RBA begins its easing cycle." Westpac's 7.3% peak-to-trough figure is deeper than its −6% for 2026, so part of its fall lands in 2027. A first-half-2027 trough is therefore what these named forecasts point to; the scenarios below test what would push it later.

A trough in the June quarter of 2027 would land on top of the 1 July 2027 start of the negative gearing and capital gains tax changes, covered below.

How much of the forecast fall has already happened?

Two index houses, two starting points

Cotality and PropTrack measure the same market but differ on how far it has fallen. Both adjust for the mix of homes sold; they differ in data and method.

Source: Cotality Home Value Index, September 2026 (released 1 Oct 2026); PropTrack Home Price Index, September 2026 (released 1 Oct 2026, via propertyupdate.com.au). Melbourne's Cotality figure is from its November 2025 cyclical high; values are 7.5% below the March 2022 record. PropTrack's combined-capitals from-peak figure wasn't in the release we used.

PropTrack's Eleanor Creagh said "the spring selling season has so far failed to deliver the usual lift in momentum." The two indexes agree on direction and differ on depth by about two points nationally and three in Sydney. Most forecasters don't name their index, so we show the arithmetic on both. Our September Home Value Index deep-dive covers the Cotality release suburb by suburb.

Our arithmetic: what's left of each forecast

The remaining fall isn't the forecast minus the fall to date, because each percentage has a different base. If values are 5.2% below the peak, a 9% peak-to-trough fall means values end at 91% of the peak, which is 4.0% below today (0.91 ÷ 0.948 − 1). Each forecast is matched to the index with the same geography: national forecasts to the national index, capital-city forecasts to Cotality's combined capitals.

Source: our calculation from the forecasts in the scorecard and the September 2026 indexes above. Remaining fall = (1 + forecast) ÷ (1 + fall to date) − 1. The result translates each forecast to today's level; it is not a predicted monthly path. PropTrack's combined-capitals from-peak figure wasn't in the release we used, so capital-city rows are blank on that index.

Source: our calculation from the scorecard forecasts; Cotality Home Value Index and PropTrack Home Price Index, September 2026 (released 1 Oct 2026). Percentage points of peak value; the AMP worst case (a stress case) is in the table only.

At September's pace of −1.1% a month, Cotality's national index would reach CBA's 9% trough in about four months, around the January 2027 reading (0.948 × 0.989⁴ ≈ 0.907). Monthly falls have usually narrowed before a trough, so our view is that the pace is more likely to slow than to hold, but the calculation shows how little room the shallower forecasts have left.

Calendar-year forecasts made before August have already been passed

Cotality's combined capitals index is down 4.6% for 2026 to the end of September. On that index, ANZ's −4.3% for the capitals in 2026 has already been exceeded with three months to go. PropTrack's June forecast of 0% would need a 4.8% rise in the December quarter, though PropTrack forecasts against its own index, which has fallen less. Westpac's and NAB's −6% for 2026 would need a further 1.5% fall over October to December on Cotality's combined capitals; Westpac's figure covers five capitals rather than all eight, so the match is approximate.

What is the house price forecast for Sydney, Melbourne, Brisbane, Perth and Adelaide in 2027?

City-by-city forecasts

Source: Cotality Home Value Index, September 2026; CommBank newsroom, 1 Sep 2026; ANZ Research via Capital Brief and The Nightly, 11 Aug 2026; HSBC via savings.com.au, 8 Sep 2026; Westpac Housing Pulse, 25 Sep 2026; NAB Australian Forward View, 17 Sep 2026. The Westpac and NAB columns are calendar-year changes; the CBA, ANZ and HSBC columns are falls from each city's peak.

Source: Cotality Home Value Index, September 2026 (released 1 Oct 2026); CommBank newsroom (1 Sep 2026); ANZ Research via Capital Brief and The Nightly (11 Aug 2026); HSBC via savings.com.au (8 Sep 2026). Peak-to-trough basis only.

A positive calendar-year forecast doesn't mean prices will rise from today's level, or that they will recover the previous peak. Westpac's +2% for Brisbane in 2026 is measured from December 2025, and Brisbane's index rose before it started falling in May.

Sydney house price forecast 2027

Sydney led the downturn. Values are 8.6% below February's peak, down 8.2% this year and 7.0% over 12 months, and Cotality notes the decline is "marginally deeper than the equivalent stage of the 2022-23 downturn." Sales over the three months to September were an estimated 26.5% lower than a year earlier. ANZ described its −14.5% Sydney call as the worst fall since 1983.

On Cotality's Sydney index, CBA's −13% leaves about 4.8% to fall, ANZ's −14.5% about 6.5% and HSBC's −16% about 8.1%. For calendar 2027 Westpac expects +2%, the weakest of its cities.

Melbourne house price forecast 2027

Melbourne fell 0.7% in September, a milder pace than each of the mid-sized capitals. Values are 7.2% below the November 2025 cyclical high and 7.5% below the March 2022 record. Measured from the cyclical high, CBA's −12% leaves about 5.2% to fall. Westpac's +5% for 2027 is its equal-strongest city call, alongside Perth. Our four signals for buying in Melbourne set out the measures we track, including final clearance rates of 55% or more, which is our editorial rule of thumb rather than a validated threshold.

Brisbane, Perth and Adelaide forecasts

Brisbane fell 1.5% in September, the sharpest monthly fall of any capital, and estimated sales were 27.2% lower than a year earlier. Perth fell 1.2% and is 6.0% below its April peak, already past ANZ's −5.2% peak-to-trough forecast. Adelaide fell 1.3% and is 2.9% below its May peak.

On Cotality's city indexes, CBA's −8% for each city implies a further fall of about 2.7% in Brisbane (0.92 ÷ 0.946 − 1), 2.1% in Perth (0.92 ÷ 0.940 − 1) and 5.3% in Adelaide (0.92 ÷ 0.971 − 1). Adelaide's smaller fall to date means more of CBA's forecast is still ahead, not that Adelaide will outperform.

Westpac's calendar forecasts for 2026 of +2% (Brisbane), +3% (Perth) and +3% (Adelaide) would need values to rise from here, since Brisbane is up only 0.1% for the year to date, Perth 1.5% and Adelaide 1.8%. NAB expects the mid-sized capitals to be "broadly flat to slightly positive" over 2026. Our reading is that the peak-to-trough calls fit the September data better than the calendar-year calls.

Hobart, Canberra and Darwin

Coverage is thin. HSBC is the only peak-to-trough forecaster here, at 12% for Canberra and 7% for Hobart. Westpac has calendar-year calls for Hobart only, KPMG's August houses forecast covers all three, and CBA and ANZ publish none. No September forecast covers Darwin. Darwin was the only capital to rise in September (+0.4%), with annual growth of 11.9% and a gross yield of 6.5%. Treat any view on these three cities as having little forecast support.

Why do house price forecasts disagree?

The rate path

The cash rate is the input that changes fastest. In June, Domain, PropTrack and SQM's base case assumed 4.35% would be the peak. In August, ANZ assumed the RBA had finished hiking. By August and September, CBA, NAB, Westpac and KPMG assumed 4.60%, which arrived on 29 September.

Since the hike the banks have split. Westpac expects another rise in November ("RBA hikes with one more expected", 30 September) and ANZ says a November hike to 4.85% "is more likely than not". CBA, NAB and AMP expect a hold, though CBA says risks are "tilted towards further rate rises".

Market pricing is a dated, market-implied estimate, not RBA guidance. After the August CPI on 30 September, LSEG pricing reported by ABC News implied about a 20% chance of a November rise; on 2 October, rbaratewatch.com's model of ASX 30-day futures put it at about 25%. Beyond November, that model's market-implied cash rate sits around 4.7% to 4.9% through the first half of 2027. Those are levels, not probabilities.

The RBA's August Statement on Monetary Policy used a market-based cash rate assumption of 4.4% at December 2026 and 4.5% through 2027, a technical conditioning assumption rather than a forecast. It said housing prices "are assumed to continue to decline gradually for a period." The November Statement, released with the decision at 2:30pm on 3 November, will update both.

Jobs, migration and supply

Cotality's outlook names the offsets: a labour market that "remains reasonably tight, although conditions are gradually loosening", and "persistently low levels of newly built housing supply". Unemployment was 4.6% in August (ABS Labour Force), and the RBA's August forecasts had it at 4.5% in December 2026 and 4.7% in December 2027. Population growth slowed to 1.4% over the year to March 2026 (ABS). Forecasters at the shallow end are mostly assuming the jobs market holds; AMP's worst case assumes the Middle East conflict drags on, oil reaches US$150 a barrel and job losses follow.

Cotality's September release gives four measures of demand and supply, each on its own basis:

Source: Cotality Home Value Index, September 2026 (released 1 Oct 2026); SQM Research national vacancy, August 2026 (released 15 Sep 2026). The Cotality release doesn't state the measurement window for the listings figures. The two vacancy series use different methods and months and aren't directly comparable.

Fewer owners are listing, yet unsold stock is accumulating. In Lawless's words, "the rate of sale has fallen even faster", so homes stay on the market longer and total inventory builds even as the flow of new listings shrinks.

The 2027 tax changes

The Treasury Laws Amendment (Tax Reform No. 1) Act 2026 (Act text, Federal Register of Legislation) received Royal Assent on 26 June 2026. Two property-relevant changes start on 1 July 2027.

• Negative gearing: net rental losses on established dwellings acquired after 7:30pm (ACT legal time) on 12 May 2026 are quarantined from 1 July 2027. They can offset residential property income, including residential capital gains, and unused losses carry forward. Holdings acquired before the cut-off are grandfathered while they are held (s 26-155), and eligible new dwellings are carved out.
• Capital gains tax: for most assets held by individuals and trusts, not only property, growth from 1 July 2027 moves from the 50% discount to CPI indexation of the cost base, with a Division 119 floor so that tax on post-2027 gains is at least 30% of the minimum tax capital gain. A deemed disposal (s 112-155) preserves the 50% discount on gains accrued to 30 June 2027. Eligible new residential dwellings keep a 50% discount on later growth (ss 115-100(a) and 115-102). Complying super funds keep their one-third discount and are excluded from the new method.

Cotality says the changes "have already seen a sharp reduction in investor demand". The RBA's October Financial Stability Review says "recent tax changes affecting housing, as well as the higher level of the cash rate, have contributed to lower demand for housing credit". The August Statement cited external modelling that the changes "could reduce housing prices by 0–5 per cent in the long run." Because the deemed disposal preserves the discount on gains to date, the law doesn't by itself create a reason to sell before 30 June 2027. Our CGT countdown guide and negative gearing transition plan cover the detail; get personal tax advice before acting on either change.

How accurate were forecasts in the last downturn?

In 2022 the major banks forecast falls of 15% to 18% and AMP 15% to 20%, assuming cash rates would peak between 2.6% and 3.35%. The cash rate went to 4.35%, yet Cotality's national index fell 9.1% as first published (April 2022 to February 2023) and 7.5% in the revised series (to January 2023). It took about 10 months from the early-2023 trough to a new record in November 2023.

Source: CBA via MacroBusiness (Nov 2022); Westpac via Broker Daily and Broker News (Aug 2022) and MPA (26 Apr 2023); ANZ via API Magazine (Aug 2022); NAB via MacroBusiness (1 Jul 2022); AMP via MacroBusiness citing the AFR (6 Jul 2022); Reuters poll via Investing.com (Nov 2022); CoreLogic/Cotality index releases, March and November 2023; RBA cash rate table. Forecasts are secondary reports of the banks' notes.

Forecasters got the direction right and the depth wrong, and they missed the speed of the recovery, with Cotality's national index rising 8.1% in calendar 2023. SQM Research's November 2022 Boom & Bust report, which forecast capital-city rises of 3% to 7% for 2023, came closer.

The measured conditions differ now. Population growth was 1.4% in the year to March 2026 (ABS), against a 2.5% peak in 2023. Unemployment was 4.6% in August 2026 (ABS), against a 3.4% low. Fewer than 5% of outstanding home loans were fixed in 2025, a record low (RBA Bulletin, May 2026), against almost 40% in early 2022 (RBA Bulletin, March 2023). Our assumption, not a measured fact, is that weaker supports make a fast rebound less likely this time, so 2023 is a less likely template. Our September rate-hike analysis compares every RBA hiking cycle since 1994.

When will house prices bottom in Australia?

Three scenarios

Source: our analysis. Ranges draw on the published forecasts in the scorecard (Westpac, CBA, ANZ, HSBC, AMP), the RBA's August 2026 forecasts and AMP's stated worst case. They aren't model outputs. "Still to fall" = (1 + scenario fall) ÷ 0.948 − 1.

Source: our analysis; Cotality Home Value Index, September 2026, for the fall to date. Scenario assumptions in the table above.

We set the middle range at 10% to 13% for two reasons. It brackets the formal forecasts made with a further hike in view (Macquarie about 10%, HSBC 13%) and the low end of AMP's post-hike range. It also assumes the November or February rise that Westpac and ANZ expect. CBA, NAB and AMP expect a hold, which points shallower, and the RBA's FSR finds large offset buffers and little sign of forced selling. Against that, the labour market is loosening, investor demand has fallen since the tax changes, and stock is building.

Each scenario can be tested against the monthly index. The shallow case needs national falls to average about 0.4% to 0.8% a month over the five readings to an early-2027 trough (0.93 ÷ 0.948 and 0.91 ÷ 0.948, spread over five months), against September's 1.1%. The scenarios assume the decline ends within 2027. If inflation, rates or unemployment surprise on the upside, the fall could run longer than any of the named forecasts expect, as the deep case allows.

The deep case and the RBA's stress tests

The RBA's Financial Stability Review, released on 1 October (data to 25 September), runs hypothetical stress scenarios that go well beyond any published forecast. They aren't forecasts, and each measures something different.

Source: RBA, Financial Stability Review, October 2026, chapters 2.1 and 3.1 and Focus Topic 4.2. The RBA presents these as stress scenarios, not projections. Negative equity (owing more than the home is worth) and cash-flow shortfall (income not covering debt payments and essential expenses) are separate measures.

A further 20% from today would put Cotality's national index about 24% below its March peak (1 − 0.948 × 0.80), far beyond the 7.5% to 9.1% fall of 2022–23. The RBA isn't predicting it. But it did write that its "housing price at risk modelling suggests ... downside risks to prices are currently elevated", and it flagged the investor behaviour that can deepen a fall: "In periods of declining housing prices, investors may be less willing to enter the market and more inclined to sell properties to limit losses."

The system-wide numbers hide the exposed borrowers. The FSR notes that "recent buyers and those who took out higher LVR loans are more likely to be in negative equity". Offset and redraw balances let the median borrower cover "over a year of scheduled mortgage payments".

Worked example (illustrative): an investor bought an $800,000 property at the March 2026 peak with a $640,000 interest-only loan (80% LVR), and the property tracks Cotality's national index.

Source: our calculation. Interest-only loan, so the balance doesn't fall; individual properties move more or less than the index.

Signals that a trough is forming

Source: RBA board meeting schedule and events calendar; ABS release calendar; SQM Research (15 Sep 2026); CommBank newsroom (21 Sep 2026); Cotality. The 55% finals threshold and the other trough signals are our editorial rules of thumb, not validated thresholds.

What happens to rents, yields and regional markets while prices fall?

Rent forecasts

Source: CommBank newsroom (1 Oct 2026); SQM Research national vacancy release, August 2026 (15 Sep 2026); KPMG Residential Property Market Outlook (4 Aug 2026).

Vacancy measures differ. SQM's national rate was 1.3% in August, while Cotality's measure rose to 2.0% in September from a record low of 1.5% in February, its highest since January 2025 but well below the pre-COVID decade average of 3.3%. Our vacancy rate tracker follows SQM's series monthly.

Gross yields by capital

Source: Cotality Home Value Index, September 2026 (yields from the index tables; rent changes from Cotality's rent charts).

A gross yield is annual rent divided by value. It isn't a net yield and it says nothing about cash flow: interest, property management, insurance, maintenance, council rates, land tax and strata all come out of the rent. Cotality expects yields to rise further "as rents continue to push higher and housing values trend lower", but adds that "opportunities for neutral to positive cash flow remain low" given those holding costs.

Our guide to rental yield expansion in a falling market covers how to use rising yields without overpaying for them.

Regional markets

Combined regional values fell 0.7% in September and are 2.2% below their May peak, but still 5.6% higher than a year ago. Some 71% of regional SA3 areas fell in the month. Regional South Australia, Western Australia and Tasmania are within about 1.2% of their peaks; regional NSW and Queensland are 2.8% below. No major forecaster publishes a regional price call. The direction of the nearest capital is a rough starting point only, because regional markets differ in their employment base, population flows, property mix, new supply and rents; local vacancy and sales volumes matter more.

Should you buy now or wait? How investors should use house price forecasts

Worked example: buy now or wait 12 months

Assumptions (illustrative): an $800,000 established property; 20% deposit plus 5% acquisition costs (stamp duty, legal and inspections; duty varies by state); a $640,000 interest-only loan at an assumed post-hike rate of 6.75% (the RBA's July 2026 average new investor interest-only rate of 6.5%, Table F6, plus the 25 basis point September rise, assuming full pass-through); gross rent at 4.0% of the price; $8,000 a year in rates, insurance, management and maintenance; cash earning 4.6% a year (the cash rate) while you wait.

Source: our calculation. Interest: $640,000 × 6.75%. Rent: $800,000 × 4.0%. Interest earned: ($160,000 + $40,000) × 4.6%. Gap: $19,200 + $9,200 = $28,400; $28,400 ÷ $800,000 = 3.55%.

How the pre-tax answer changes with prices over the next 12 months (acquisition costs scale with the purchase price):

Source: our calculation. Buy now = $800,000 × 1.05 + $19,200. Wait = new price × 1.05 − $9,200. Both investors own the same property at the end of the year.

The example leaves out principal repayments (the loan is interest-only), transaction costs beyond the 5% (such as lenders mortgage insurance), tax on the interest earned and tax deductions on the holding loss (both handled in the after-tax version below), and the return the deposit could earn in other investments or by reducing other debt. It compares two paths for one buyer; it doesn't establish a general advantage to waiting.

After tax (illustrative): the buyer is an individual on a 39% marginal rate (the 37% bracket plus the 2% Medicare levy, 2026–27), settles in October 2026 on an established dwelling acquired after the 12 May 2026 cut-off, and the loss accrues evenly through the year.

Source: our calculation under the stated assumptions. The quarantined $4,800 carries forward against future residential rental income or residential capital gains, so its eventual value depends on the investor's later position. A different marginal rate, settlement date or a grandfathered or new-build property changes the result.

What the forecasts imply for the next 12 months (our approximation): Westpac's −6% for 2026 needs about a further 1.5% fall in the December quarter, and spreading its +3% for 2027 evenly gives about +2.2% to September 2027, so values a year from now would be about 0.7% higher (0.985 × 1.03^0.75). CBA's trough is still about 4% below today, and with about 2% growth expected over the whole of 2027, values would still be below today's level a year out. AMP, HSBC and Macquarie expect deeper troughs. All of those sit below the 3.55% pre-tax and 2.4% after-tax gaps. PropTrack's +5.5% (capitals) for 2027 was built on a 4.35% peak. KPMG's +3.4% (houses) assumed 4.60% but predates the August–September falls, which have already taken national house values 3.2% lower this year (Cotality), against KPMG's −1.1% for all of 2026.

Run your own numbers in our cash flow calculator and check your capacity at a 4.85% cash rate with the borrowing capacity calculator.

How to read a forecast

1. Check the date and the rate assumption first. A forecast built on a 4.35% peak is out of date at 4.60%.
2. Check the basis. Peak to trough, calendar year and financial year numbers can't be compared directly.
3. Check the index and geography. Cotality and PropTrack differ by about two points on the fall so far, and a five-capital forecast isn't a national one.
4. Look at the range. In 2022 the forecasts were about twice the outcome; in 2026 every revision we can trace has been downward.
5. Apply it to your market. The scorecard covers national and capital-city indexes; a suburb with 1% vacancy and a 6% yield can behave differently.

Five investor profiles

These are illustrative investor profiles; the people and figures are hypothetical.

Profile 1: The buyer with finance ready, waiting for the bottom

Situation: pre-approved for $700,000, renting, with a 20% deposit in an offset account and a target of a Brisbane house.

Reasoning: Brisbane's peak-to-trough forecasts of 7.9% to 10% leave about 2.6% to 4.9% to fall on Cotality's index. A pre-approval expires and is reassessed at the rates applying then, so a November rise could reduce the limit.

Strategy: keep inspecting and track vendor discounts. Consider buying when a property is discounted to roughly where the forecasts put the index, rather than waiting for the index itself.

Profile 2: The recent buyer at a high LVR

Situation: bought a Sydney unit in late 2025 at 88% LVR with lenders mortgage insurance; the interest-only period ends in 2027.

Reasoning: on Cotality's Sydney unit index, down 5.0% this year, an 88% LVR from late 2025 is already about 93% (0.88 ÷ 0.95). Further falls of the size CBA to HSBC forecast for Sydney dwellings (another 4.8% to 8.1%) would take it to about 97% to 101%. Sydney units have fallen less than houses, so the dwelling forecasts may overstate the unit fall. At those levels a refinance may mean new lenders mortgage insurance or a declined application, depending on the lender.

Strategy: price the switch to principal and interest now, ask the current lender what an interest-only rollover requires, and build cash buffers before 2027 rather than relying on a refinance. Negative equity doesn't force a sale while the loan can be serviced.

Profile 3: The grandfathered holder thinking of selling

Situation: owns two Melbourne established properties bought before 12 May 2026 and is weighing a sale before 1 July 2027 to lock in the CGT discount.

Reasoning: the deemed disposal (s 112-155) preserves the 50% discount on gains to 30 June 2027 whether or not the properties are sold, and they stay grandfathered from loss quarantining while held (s 26-155). Selling into a falling market crystallises a lower price.

Strategy: model selling against holding with a registered tax agent, using the CGT sell-or-hold calculator, and let portfolio reasons such as cash flow or concentration set the timing.

Profile 4: The yield investor looking beyond Sydney and Melbourne

Situation: wants positive or near-neutral cash flow and is comparing Darwin, Hobart and regional Western Australia.

Reasoning: Darwin's 6.5% gross yield and regional WA's 5.2% are well above Sydney's 3.4%, and both markets are within about 1.2% of their peaks (Hobart is 2.0% below its 2022 peak). They are also small, thinly traded markets with little forecast coverage.

Strategy: underwrite on rent and vacancy, not growth. Use postcode-level vacancy, check insurance and strata costs, and plan to hold through a full cycle. Our Top 10 Suburbs for September shows the interest-cover test we use.

Profile 5: The SMSF trustee with cash

Situation: an SMSF with $900,000 in cash and a long-term residential property allocation in its investment strategy.

Reasoning: since 10 August 2026, a new limited recourse borrowing arrangement can only be used to acquire business real property (SIS Act s 67A(2) as amended, with business real property as defined in s 66), so a new residential purchase by the fund is unleveraged. Existing LRBAs are grandfathered and can be refinanced. Complying super funds sit outside the loss quarantine and keep their one-third CGT discount.

Strategy: a falling market suits a patient cash buyer, but the fund still needs liquidity for pensions and costs. Check the purchase against the investment strategy, see our SMSF LRBA ban checklist, and get advice from an SMSF specialist.

Decision framework

• Buying now is worth considering when the property's rent covers most of the interest, you can negotiate a price near or below where the forecasts put the trough, and your loan still services at a 4.85% cash rate with a buffer.
• Waiting may suit you when the property would be heavily negatively geared, your market is still falling more than 1% a month, or your finance only works at today's rate.
• Selling may be worth modelling when cash flow is unsustainable, your LVR is rising towards a refinance limit, or the equity has a better use, rather than because of a forecast alone.

These conditions rest on the assumptions above. An index forecast describes an average across a city or the country; it can't tell you the fair value of an individual property or whether it suits you.

Risks and how to manage them

Source: our analysis; cash rate scenarios from the bank calls above; tax treatment from the Treasury Laws Amendment (Tax Reform No. 1) Act 2026.

The Bottom Line

National values were 5.2% below their March 2026 peak on Cotality's index in September after six monthly falls. The main cluster of formal forecasts puts the trough 9% to 13% below the peak, with CBA, AMP and NAB pointing to the first half of 2027, and most of those numbers were finalised before the 29 September hike.

On Cotality's index the cluster implies roughly another 4% to 8% off national values. On our illustrative numbers, buying a negatively geared property now rather than in a year costs about 3.55% of the price before tax and 2.4% after tax in a flat market. That argues against rushing, though a property whose rent covers most of the interest, bought at a discount, can still make sense now.

The next tests are the September-quarter CPI on 28 October, the RBA's decision and Statement on Monetary Policy on 3 November, and October's Cotality index in early November. A hold and a smaller monthly fall would point to the shallow end; a November rise and unemployment heading towards 5% would point deeper. We'll update this scorecard as the banks revise.