The most useful number in this week’s default forecast is not the record 24,546 households. It is the 3%. That is how much the modelled count grew in the September quarter, after an 18% jump in June. Growth has cooled. The level has not.
The figures come from a September quarter 2026 default report covering 70 ranked postcodes. It estimates how many borrowers are likely to default within 12 months. Sixty-six of those 70 postcodes saw more at-risk borrowers than a year earlier. The national count is up 24% on the prior year.
One caution first. These are modelled projections. They are not actual defaults, and they are not lender arrears data. A forecast can overshoot or undershoot, so treat the direction as the signal and the exact figure as a rough guide.
Why a slower rise is not good news
A 3% quarterly gain sounds calm next to 18%. But growth on top of a record is still growth. The report’s own research head warned that slower growth should not be read as improvement, and that is the right reading.
The other complication is timing. The estimates predate the latest rate rises. So the model is looking at borrowers before the newest repayment increases have landed. That is a reason to think the next reading carries more risk than this one, not less.
Other data points the same way:
- A credit bureau found strained loan accounts up 7.3% in early 2026.
- A survey found one in ten borrowers cannot cope with higher repayments.
- Banks report rising hardship requests.
None of these proves a wave of defaults. Together they show a borrower base with thin buffers.
Where the stress has moved
The older story was outer Sydney and Melbourne growth corridors. That picture is now incomplete.
| State | At-risk borrowers | Notable detail |
|---|---|---|
| Victoria | 6,026 | Largest state count; Cranbourne is the top hotspot at 1,364 |
| Western Australia | 5,895 | Second-highest state total |
| New South Wales | 3,782 | Only state to fall in the quarter; Blacktown still up 76% on a year ago |
| South Australia | 1,874 | Fastest annual rise, up 31%, across ten postcodes |
In South Australia, Encounter Bay jumped 78% in the quarter to 166 at-risk borrowers. Regional Queensland pockets such as Manunda and Craignish also climbed sharply, even though the Queensland total was flat. A flat state number can hide sharp local moves. That is the point of reading by postcode.
The catch
- A state total that falls can sit beside a suburb that surges. New South Wales fell in the quarter, yet Blacktown rose 76% year on year.
- Small bases exaggerate percentage jumps. Encounter Bay’s 78% means 166 borrowers, not thousands.
- Every figure here is a model estimate, not reported arrears.
A practical way to read the spread
Why would stress migrate to cheaper markets? The report does not say, so here is a reasoned view, not a finding. Cheaper capitals saw strong price growth over the past cycle. Buyers who entered late may have borrowed heavily against their income. When repayments rise, those borrowers have the least room to move, whatever the headline price of their suburb.
So the useful question is not “is this suburb expensive?” but “how much of household income does the loan consume, and how big is the cash buffer?” A modest-priced home with a high loan-to-income ratio can be more fragile than a dearer one with a bigger deposit.
For readers, a rule of thumb: stress-test your own loan at a repayment well above today’s, and check how many months you could cover from savings. If the answer is under three, that is the number to fix first. Our mortgage stress and arrears guide walks through the steps, and the pressure on Sydney and Melbourne households is covered in our look at who is squeezed by rate rises.
What a forecast cannot tell you
Default risk is not default. Most at-risk borrowers will find a way through: cutting spending, extending hours, using offset balances, or refinancing. Lenders also have hardship tools, and a forced sale is usually the last step, not the first.
Still, a bigger at-risk pool raises the odds of more arrears and, later, more distressed listings. Whether that happens depends on rates, wages and employment, none of which this model controls for. That is the honest limit of the numbers.
If you want the weekly read on rates and credit without the spin, subscribe to the Australian Property Review newsletter.
General info, not financial advice.
