Reading the ABS CPI housing component: rents, new dwellings, and what the RBA actually watches

When the RBA says housing inflation is contributing a given number of percentage points to headline CPI, it is adding two distinct subindexes that move on completely different drivers: market rents and the cost of new dwellings. They lead and lag each other by quarters, they respond to different policy levers, and the headline that ends up in the press conference hides the tension between them. A reader who understands how the housing group is built can pull a far richer story out of each CPI release than the topline number provides on its own.

The ABS publishes the Consumer Price Index quarterly under Catalogue 6401.0, with a monthly indicator running alongside it since 2022. The housing group is one of eleven groups in the basket, and within housing the two subindexes that dominate are rents and new dwelling purchase by owner-occupiers. The other components matter for completeness, but the rate-setting conversation almost always comes back to those two.

What sits inside the housing group

The housing group weight in the CPI basket is roughly 22% of total household spending, by far the largest single group. The composition matters because each subindex moves on its own clock.

• Rents, weight around 7% of the total CPI basket. The ABS measures stock rents, meaning what existing tenants are actually paying, not what landlords are advertising. Re-leases cycle the asking-rent shock into the index slowly, because only roughly 30 to 50 per cent of tenancies turn over in a given year. In normal markets the lag from market rents to CPI rents runs 12 to 18 months.
• New dwelling purchase by owner-occupier, weight around 9% of the total basket. This captures the cost of a new house build: materials, labour, builder margin, and project-management overhead. Land value is explicitly excluded. The subindex responds to construction-cost inflation, immigration-driven labour tightness, and government stimulus. The HomeBuilder spike of 2021 and 2022 remains the canonical recent example of a policy-driven surge through this line.
• Maintenance and repair of the dwelling: small weight, included for accounting completeness. Tracks tradie costs more than anything else.
• Property rates and charges: council rates, water service charges, building insurance. Sticky and slow-moving, usually re-set annually on local government cycles.

One thing is explicitly not in the basket. Interest paid on mortgages was removed from CPI in the late-1990s reforms, and it has stayed out since. This is the most common confusion when readers try to reconcile the cash rate with the housing print. The RBA cash rate transmits through to property prices via borrowing capacity and wealth effects, not via direct mortgage-cost capture in the index.

The lead-lag dynamic between rents and new dwellings

The two large subindexes do not move together. They respond to different signals, and they peak at different points in the cycle.

Advertised asking rents, the sort tracked by SQM Research weekly and CoreLogic monthly, are the leading indicator for the rents subindex. When a tight rental market pushes asking rents up sharply, that pressure shows up in CPI rents 6 to 12 months later, because only the fraction of tenancies that re-lease in a given quarter feed the new contract rent into the stock rent measure. The result is a CPI rents series that is smoother and slower than what rental-listing data is showing in real time.

New dwelling inflation tracks the construction Producer Price Index with a shorter lag, typically one to two quarters. Builder quotes pass through to the CPI new dwelling line more quickly than asking rents pass through to CPI rents, because each new contract feeds the index directly without waiting for a re-lease cycle. That gives the new dwelling subindex more in common with timber and labour costs than with the rental market.

Through 2023 and 2024, rents inflation peaked well after new dwelling inflation began moderating. The two diverged, and that divergence complicated the RBA's read of underlying housing pressure. A reader looking only at the housing group total would have missed the change in composition.

Why the trimmed mean keeps the housing story

The RBA's preferred underlying inflation measure is the trimmed mean, published by the ABS in the same release. The trimmed mean removes the most extreme 15 per cent of CPI components at each end of the distribution each quarter, and averages what remains. The point is to strip out volatile or one-off price moves so the underlying signal becomes visible.

Housing components rarely get trimmed. Rents and new dwellings are both large and steady, and they almost always sit in the middle of the distribution rather than at the extremes. The consequence is that when housing inflation is hot, the trimmed mean tends to run hot alongside it, because the largest persistent contributors survive every trim. When housing inflation cools, the trimmed mean cools with it. This is why housing prints carry so much weight in the RBA Statement on Monetary Policy: they show up in both the headline and the underlying measure at the same time.

How to read each quarterly release

Four habits separate a useful read from a headline skim.

• Look at the quarterly contribution. Which subindex added the most percentage points to the housing group this quarter. Rents and new dwellings move on different drivers; knowing which is doing the work tells you whether tightness is in the rental market or the construction pipeline.
• Compare annualised quarterly with year-on-year. The annualised quarterly rate is noisier but timelier. A YoY of 5% with an annualised quarterly of 7% says momentum is building; the same YoY with an annualised quarterly of 3% says the heat is coming off.
• Cross-check rents against asking-rent data. If SQM weekly or CoreLogic monthly show asking rents decelerating while CPI rents are still hot, the deceleration is in the pipeline and a turn in the CPI line is likely in the next two to four quarters.
• Check trimmed mean against headline. When trimmed mean runs above headline, housing is usually a big part of the explanation. When trimmed mean runs below headline, one-off shocks such as fuel or fresh food are driving the topline and the underlying picture is calmer.

What each subindex means for property thinking

The two large subindexes carry different signals for anyone making a property decision.

• Hot CPI rents tell you vacancy is low and demand is outpacing supply at the household level. That is bullish for rental yields on completed stock, and bearish for the renter cohort and rental affordability. Investor-focused readers can compare the CPI rents trajectory against current asking rents to decide whether the heat is fading or still building.
• Hot new dwelling CPI tells you the construction supply chain is constrained. That is bullish for the replacement value of completed homes, and bearish for off-the-plan buyers who locked in a price before the build cost moved. It also tends to slow the supply response that would otherwise relieve rental pressure, which is how the two subindexes eventually feed each other.

The combination is the actual housing inflation story. Reading the headline alone, without unpacking which subindex is driving it, is the single most common error in property commentary built around CPI prints.

A worked numeric example

Numbers help. Suppose a quarterly release shows CPI rents running 8% year-on-year and the new dwelling subindex running 5% year-on-year. With the weights set out above, the contribution arithmetic looks like this.

• Rents contribution: 7% basket weight multiplied by 8% YoY inflation equals 0.56 percentage points of headline CPI.
• New dwellings contribution: 9% basket weight multiplied by 5% YoY inflation equals 0.45 percentage points of headline CPI.
• Combined housing contribution: 0.56 plus 0.45 equals 1.01 percentage points, before counting the smaller maintenance and rates lines.

If the headline CPI in that release prints around 3.2% year-on-year, then housing is delivering roughly a third of the entire top line on its own. A trimmed mean print of 3.5% against that headline would tell you the underlying measure is running hotter than the topline, and housing is the most likely reason. That is the kind of composition read the RBA Statement on Monetary Policy typically dwells on for several paragraphs. Investors running their own scenarios can check current yields on the rental yield calculator before deciding how much of that CPI rent print is already in their numbers.

Sources to cross-reference

No single number tells the housing inflation story on its own. Four published series make up most of what a careful reader needs.

• ABS Consumer Price Index, Catalogue 6401.0. The headline release, quarterly, with a monthly indicator running alongside it since 2022. The housing group breakdown is in the detailed tables, not the press release.
• ABS Selected Living Cost Indexes, Catalogue 6467.0. A complementary quarterly release that re-weights the basket for different household types, including employee, pensioner, and self-funded retiree. Useful for reading how housing inflation lands differently across cohorts.
• RBA Statement on Monetary Policy. The quarterly document gives the RBA's own read on housing inflation in plain English, and is the single best place to see how the central bank is weighting rents against new dwellings in any given cycle.
• RBA Bulletin housing-themed articles. Periodic deeper dives into the construction pipeline, rental dynamics, or the transmission of cash-rate moves into housing markets. Less timely, more analytical.

Reading the housing print alongside the rest

CPI housing does not stand alone. It sits in a small constellation of releases that together describe the actual state of housing inflation in Australia. The RBA Statement on Monetary Policy is the document the rest of the calendar feeds into. For the supply side of the new dwelling story, ABS Building Approvals is the leading indicator of completions two to three years out. For the leading indicator on the rental side, SQM Research vacancy data moves several quarters ahead of CPI rents, and the broader rental vacancy picture sets the backdrop. For the demand-side driver of both rents and new dwellings, ABS net overseas migration is the population pulse the construction pipeline cannot keep up with in any short window.

On Burbfinder, suburb and region pages surface rents, vacancy, and median prices alongside the macro context. The national CPI housing print is the backdrop; the local rental market and construction pipeline are where the decision actually lives. Reading the two layers together, rather than treating the headline as the whole story, is the habit worth building.