This article is general information only and does not constitute financial or tax advice. Consult a qualified tax professional for advice specific to your situation.
Key takeaways
- KPMG's August 2026 Residential Property Market Outlook, released this week by chief economist Dr Brendan Rynne, forecasts national house values -1.1% in 2026 and +3.4% in 2027.
- National units +2.2% in 2026 and +3.7% in 2027. Units are the resilient side of the ledger, protected by a supply shortage and demand pushed down the price curve by higher borrowing costs and negative gearing changes.
- Sydney houses -5.0% in 2026 and +3.3% in 2027. Sydney units +0.4% in 2026 and +3.6% in 2027. Sydney is the softest of the majors on the house line.
- The forecast is an 8.8 percentage point downgrade on KPMG's January 2026 call of +7.7% national house price growth. One of the largest mid-year Big 4 forecast reversals of the cycle.
- Rents +3.5% per year through 2026 and 2027 on the KPMG numbers. On an $850-a-week Sydney investor house that is roughly $30 a week or $1,560 a year in gross rent.
- Rynne calls it a 'V-shaped trajectory' with a correction in 2026 and a gradual recovery in 2027. For a landlord that means the deepest discounts of this cycle are landing between now and December, not next spring.
- The forecast assumes the RBA holds at 4.35% at Monday's 11 August meeting and only cuts appear in 2027. Consistent with the CBA and NAB positions.
- The reversal is driven by the June 4.35% hike, the 12 May 2026 negative gearing and CGT overhaul, and the vendor overhang captured in SQM's July listings print of 278,984 dwellings.
This article is general information only. It does not consider your personal circumstances and is not financial, credit or tax advice. Speak to a licensed mortgage broker, buyer's agent or registered tax agent before acting on any of the figures below.
What KPMG actually released
The August 2026 edition of KPMG's Residential Property Market Outlook landed this week, fronted by chief economist Dr Brendan Rynne. It is the third KPMG outlook of the calendar year and the first to model the market on the other side of the 12 May federal budget and the 25 basis point RBA hike to 4.35% on 5 May.
The headline forecast:
- National house values: -1.1% in 2026, +3.4% in 2027.
- National unit values: +2.2% in 2026, +3.7% in 2027.
- Sydney houses: -5.0% in 2026, +3.3% in 2027.
- Sydney units: +0.4% in 2026, +3.6% in 2027.
- National rents: about +3.5% per year in 2026 and 2027.
Melbourne is flagged as the weakest of the majors on the house line for the remainder of 2026, though the public materials at the time of writing do not publish a discrete Melbourne house number.
Rynne's summary is direct. Population growth is still firm, rental vacancy rates are still exceptionally low, and housing supply is still well below demand. What is different from January is the layering of higher borrowing costs, persistent inflationary pressures and the changes to investor taxation settings that the Treasury Laws Amendment (Tax Reform No. 1) Act 2026 put into the pipeline in May. The three drags together are enough to hold prices down through 2026 even with the demographic and rental fundamentals still pushing the other way.
An 8.8 percentage point downgrade in eight months
In January 2026, KPMG had national house values rising 7.7% for the year. Perth was pencilled at 12.8%, Brisbane 10.9%, Darwin 10.5%, Melbourne 6.8%, Sydney 5.8%, Hobart 5.4% and Canberra 4.7%. That was a bullish call built on the pre-budget rate glide-path and the assumption that lending capacity would rise into 2026 as inflation eased and the RBA cut.
Eight months later the same forecaster is at -1.1% on the national house line for 2026. That is an 8.8 percentage point round trip from the January call, and one of the largest mid-year Big 4 forecast reversals of the current cycle. What broke between January and August was not the population number or the rental vacancy number. It was the price of credit and the after-tax return on a marginal investor purchase.
Two policy changes moved between the calls:
- The 12 May 2026 federal budget introduced grandfathering rules that keep full negative gearing for existing investors on established stock they already own but progressively tighten deductions for new purchases of established stock from 1 July 2027. Investor purchase intent for established stock softened almost immediately after the announcement, and the investor share of new lending has been running around 41% since the May print.
- The RBA lifted the cash rate 25 basis points to 4.35% on 5 May 2026 rather than cutting as the January consensus had assumed. That single move added around $100 a month to the mortgage on a $600,000 principal-and-interest investor loan at full pass-through, and the risk premium priced into fixed rates through May and June widened further.
Neither move was in KPMG's January model. Both are in the August one, and the -1.1% national house call is the arithmetic that falls out.
Sydney is doing most of the work in the national number
Sydney at -5.0% for 2026 is the single largest city drag inside the -1.1% national line. On a $1.2 million Sydney investor house, a 5.0% fall is $60,000 of unrealised equity out through the back half of the year. The Cotality July HVI already recorded a 1.4% Sydney monthly fall and Sydney listings sit 28.0% higher than a year ago on SQM's July stock on market print. KPMG's number is not a forecast that comes from nowhere. It is the shape the June and July prints have already been drawing.
The 2027 recovery is +3.3%. On the same $1.2 million property that is $39,600 back on the equity line inside twelve months. Peak-to-trough-to-recovery inside 18 months is a shallow correction by historical standards, but the wall-clock date of the trough matters. A landlord who lists a Sydney investor property in November 2026 is selling into the modelled trough. A landlord who lists in September 2027 is selling into the recovery leg.
Sydney units are the mirror image. KPMG has them at +0.4% in 2026 and +3.6% in 2027. That is the resilience story running inside the same city. The unit segment did not carry the same listings surge in July on the SQM print and it is not carrying the same serviceability drag, because the median Sydney unit contract is a smaller loan than the median Sydney house contract.
The V-shape and what it means for buy-vs-sell
Rynne's characterisation of the two-year path is that the market follows 'a V-shaped trajectory over the next two years, with a correction in 2026 followed by a gradual recovery in 2027'. For a landlord that framing is more useful than the point forecast, because it tells you where the deepest discounts of the cycle are landing.
A V does not have a flat bottom. The middle of the V is a single month or two of transactions. If KPMG is right, the deepest print of this correction is somewhere between October and December 2026, and the price line starts recovering into Q1 2027. That has three concrete implications:
- A buyer looking for a Sydney entry point is looking at a four-month window that lines up with the traditional spring listings peak. Vendors who list into a soft market and then need to sell before Christmas are the marginal price-setters through October and November.
- A vendor considering an early 2027 sale has two calls to make. One, whether to accept the modelled trough print in Q4 2026 to close a chapter. Two, whether to sit through the trough and list into the modelled recovery in Q2 or Q3 2027. The V-shape only recovers 3.3% in Sydney over the full year, so an early 2027 seller captures only a portion of that.
- A refinance decision is time-sensitive in the other direction. The mid-July refinance window we flagged priced against a 4.35% terminal. That terminal is now the modal call. Second-tier lenders have been sharpening variable investor rates below 5.90%. A landlord refinancing into the modelled trough is refinancing against a lower valuation than they would have got in June, which can push a borderline loan out of the 80% LVR bracket into LMI territory. The refinance is still worthwhile if the rate saving covers the LMI, but the sums have to be run against a fresh valuation, not the one from six months ago.
Units are the fundamentally different story
The +2.2% national unit forecast for 2026 sits alongside the -1.1% national house forecast. That is a 3.3 percentage point spread inside the same year, from the same forecaster, on the same market. The gap is not noise. It is the mathematical output of two structural forces working in the same direction on unit prices:
- Serviceability. A median unit purchase in every capital except Perth is a smaller loan than a median house purchase. At 4.35% cash rate and investor variable rates in the 5.9% to 6.5% band, more buyers clear the unit hurdle than the house hurdle. As the negative gearing reset from 1 July 2027 pushes marginal investor purchases toward smaller loans against smaller deductions, the unit segment is the more tax-efficient entry point for a new investor.
- Supply. Apartment approvals only bounced to 17.8% for the month in June 2026 off a low base. The unit pipeline is still tight against household formation and net migration. That is a longer-run price floor than the house pipeline can offer, because detached housing supply has recovered faster.
For a landlord thinking about the next purchase inside a self-managed super fund that is now locked out of new residential LRBAs from 10 August 2026, the unit segment is the more resilient side of the ledger for the cash purchases and existing-LRBA refinances that remain available.
Rents are still rising, just slower
KPMG has rents running at about 3.5% per year through 2026 and 2027. That is a step down from the 3.6% ABS CPI rent print for the year to June, and a step further down from the Cotality-captured 5.7% asking rent growth in the March print. The direction is the affordability ceiling doing its work: wages growth has been under 3.5% for most of 2026 and asking rents cannot rise faster than tenants can pay, indefinitely.
3.5% per year is still a meaningful number for a landlord modelling forward cash flow. On the $850-a-week Sydney investor house from the FAQ, another 3.5% takes the letting price to about $880 a week. Over a full year that is roughly $1,560 of additional gross rent. It does not offset a 5.0% modelled capital fall on a $1.2 million property in dollar terms. It does lift the running yield on the property from 3.7% to 3.9% before outgoings and interest, which nudges a marginal hold decision toward hold rather than sell.
For landlords who are thinking about a rent increase inside the next 12 months, the KPMG number is another data point to sanity-check the market rent against. Setting the right rent price is a comparable-based exercise, not a CPI-based one, but a 3.5% forecast is a defensible starting point in a rent review conversation with a tenant.
The 11 August RBA meeting is the last variable
The KPMG August outlook assumes the RBA holds at 4.35% on Monday 11 August 2026 and only cuts show up in 2027. That is now the modal call across the Big 4 economist desks after the June quarter CPI printed 3.6% trimmed mean on 29 July. Westpac has floated a live 25 basis point cut on Monday off the back of the CPI undershoot. CBA still expects a hold through 2026.
The forecast does not depend on further hikes. If the RBA surprises with a cut on Monday, the -1.1% call is on the conservative side and the 2027 recovery could pull forward. If the RBA holds, the call is unchanged. The only path that materially worsens the KPMG number from here is a rate hike combined with a fresh negative gearing tightening, and neither is on the table for August.
For a landlord, that means the mortgage line of the P&L is settled for the next quarter at worst-case flat and best-case slightly better. The capital line of the P&L is a modelled 5.0% down in Sydney and 1.1% down nationally between now and 31 December. The rent line is a modelled 3.5% up. The three together define what a hold looks like through the trough.
Practical calls for a landlord this week
Three things worth doing before Monday's RBA decision:
- Get a fresh valuation if you are refinancing. The mid-July variable rate window has widened, not closed, but the LVR calculation has to be run against a current market appraisal rather than a June one. A mortgage calculator can model the repayment saving but the LVR gate is set by the valuer.
- Review your rental income against the KPMG 3.5% forecast. If the tenancy is at or below market rent, the rent review provisions in your state allow a defensible increase inside the next twelve months. On an $850-a-week letting, 3.5% is $30 a week or $1,560 a year of gross rent.
- Log expenses to the ATO's rental data-matching pool ahead of the FY26 return. The ATO is receiving 2.3 million property records this tax time. An expense tracker that captures every deductible item across the year is the difference between a full deduction claim and a partial one at lodgement.
Bottom line
KPMG has walked back its 2026 call by 8.8 percentage points in eight months, and the shape of the walk-back is a V-shaped correction that troughs in Q4 2026 and recovers through 2027. Sydney is the weakest of the majors on the house line at -5.0%. Units are the resilient side of the ledger at +2.2% nationally and +0.4% in Sydney. Rents keep running at about 3.5% per year and blunt some of the capital drag on running yield.
For a landlord already in the market, the practical read is: hold through the trough, refinance against the current window, lift rents to market at the next review. For a landlord looking at a new purchase, the read is: the deepest discounts of this cycle are landing between October and December 2026, and units clear the serviceability bar more readily than houses at 4.35% cash rate.
The 11 August RBA decision is the last live variable inside this forecast for the next quarter. Everything else is set.