The numbers at a glance
Inside the RBA’s October 2026 downside scenario
The Reserve Bank published a fully specified severe downturn on 1 October and told the market how many mortgagors fail it. Here is the scenario, the result, and the three characteristics that separate pressure from loss.
The scenario parameters
What breaks, and what does not
| Measure | Where it sits now | In the severe scenario |
|---|---|---|
| Mortgagors in cash flow shortfall | Around 2% of variable-rate owner-occupiers | About 5% of mortgagors, just surpassing the 2023 peak |
| Borrowers in negative equity | Less than 1% | Around 5% of mortgages on a uniform 20% price fall (modelled separately) |
| Cash flow shortfall + thin buffers + negative equity | Not separately published | Less than 1% of mortgagors |
| Housing loans 90+ days in arrears | Up a little year to date, around pre-pandemic levels | Not separately published |
| Median liquidity buffer | Over a year of scheduled repayments in offset and redraw | Two-thirds of those in shortfall hold at least six months |
The three triggers the RBA names
Cash flow shortfall
Income does not cover scheduled debt payments plus essential expenses. On its own, this is a hardship conversation.
Low savings buffers
Not enough in offset or redraw to ride out the shortfall. The RBA’s working line is six months of payments and essentials.
Negative equity
The loan exceeds the property value, removing the option to sell and repay in full. Only all three together mark the highest-risk group.
Read the bases before you quote the numbers. The two “about 5 per cent” figures in the Review are different measures of different populations, and the 2 per cent figure is a third series again. The RBA also states that its scenario analysis does not routinely incorporate behavioural adjustments or feedback loops, and that results “are not precise, due to data and model limitation”.
Market insight · 4 October 2026
The RBA modelled 6.3% unemployment and a 5.6% cash rate. About 5% of mortgagors fail it — and under 1% carry all three triggers
The October Financial Stability Review landed on 1 October with a headline about resilience. Underneath it is something more useful to anyone managing a loan book: a published severe-downturn scenario, the share of borrowers who fail it, and a named list of what separates pressure from loss.
Key takeaways
- The RBA’s severe scenario assumes unemployment at 6.3%, inflation at 7%, a cash rate of 5.6%, a 20% fall in housing prices and a 1.4% fall in GDP.
- In that scenario the share of mortgagors in a cash flow shortfall rises to about 5%, just surpassing the 2023 peak. Around two-thirds of them hold at least six months of buffers.
- The group the RBA identifies as most likely to enter foreclosure carries three characteristics at once — shortfall, thin buffers and negative equity — and is under 1% of mortgagors even in that scenario.
- High-DTI lending sits well below APRA’s 20% limits, which the Review says are unlikely to be binding at system level — which points to the 3 percentage point serviceability buffer at a 4.60% cash rate, rather than the DTI cap, as what is squeezing capacity.
- Serviceability exemptions — the mechanism that absorbed the 2023 refinancing squeeze — rose only slightly in the first half of 2026.
On 1 October the Reserve Bank published its Financial Stability Review for October 2026. The line that travelled was that Australian households and businesses remain resilient. That is what the document concludes, and the evidence in it supports the conclusion. It is also the least useful sentence in the Review for a broker.
Underneath the conclusion sits something the channel has not had in this cycle: a fully specified downturn, modelled on loan-level data, with the share of mortgagors who fail it published alongside the characteristics that distinguish a borrower under pressure from one at genuine risk of losing the property. The cash rate moved to 4.60% on 30 September, the fourth increase this year and a cumulative 100 basis points over 2026. The Review effectively hands brokers a stress test, the parameters, and the answer.
What follows is what the RBA actually modelled, what it found, where its numbers do and do not line up with one another, and how the same three-part test can be run across a database without inventing figures the RBA did not publish.
What the Reserve Bank modelled
The scenario sits in Focus Topic 4.2 of the Review, which the RBA describes as “deliberately much more pronounced than the adverse scenarios considered in the May 2026 Statement on Monetary Policy, in order to stress test borrowers and lenders from a financial stability perspective”. It is a stress test, not a forecast, and the Review is explicit about that.
In the scenario the unemployment rate increases to 6.3%, GDP falls by 1.4%, housing prices fall by 20% from current levels, inflation increases to 7% and the cash rate increases to 5.6%. The RBA states the scenario “assumes current geopolitical tensions worsen and risks materialise, leading to further negative supply shocks”, and notes the labour market deterioration is similar in magnitude to the 2008–09 downturn.
For brokers, the cash rate assumption is the one to hold onto. At 5.6% the scenario sits exactly 100 basis points above today’s 4.60% — inside the 3 percentage point serviceability buffer APRA requires lenders to apply. The Review makes the related point itself: borrowers who took out loans in recent years “have faced loan servicing capacity requirements greater than those faced by pandemic-period borrowers and well above current interest rates”. That is one reason the RBA expects stress to peak below its 2024 level even with rates above their 2023–24 peak. It does not extend to older loans assessed against a much lower starting rate.
What fails, and by how much
The RBA’s key household vulnerability indicator is the share of borrowers in a cash flow shortfall, which it defines as income being insufficient to cover debt payments and essential expenses. Using loan-level mortgage data from its Securitisation System, the Review finds that in the severe scenario “the share of mortgagors in a cash flow shortfall increases to about 5 per cent, but remains contained, just surpassing the peak observed in 2023”.
That is the number most of the coverage led with, and on its own it overstates the position. The Review immediately qualifies it: “About two-thirds of borrowers in cash flow shortfall have sufficient savings buffers to cover at least six months of mortgage payments and essential expenses.” A shortfall is a cash flow event, not an outcome.
The number that matters most
The passage brokers should mark is the one that follows, and it is worth quoting in full.
“The borrowers most likely to enter foreclosure, where the lender takes possession of the property, are those in a cash flow shortfall, with low savings buffers and in negative equity. The share of borrowers with all three of these characteristics is less than 1 per cent of mortgagors, even in this very adverse scenario.”
RBA, Financial Stability Review, October 2026, Focus Topic 4.2
Three characteristics, not one. The Review is applying the double-trigger framing that has long underpinned Australian default research — it cites Bergmann’s 2020 RBA discussion paper on the point — and states plainly in Chapter 2 that “negative equity is insufficient to trigger default if borrowers remain able to service their loans”.
A book segmented on one variable will be ranked wrongly. High LVR alone is not a risk list. Nor is a tight servicing position on its own.
This is the practical content of the Review for a broker. If your approach to reviewing a back book in a softening market is to sort by LVR, you are sorting on the characteristic the RBA says is insufficient by itself. The borrowers who need the call first are the ones where the three overlap.
Where the book sits today
The current-state figures in Chapter 2 give the starting point:
- The estimated share of variable-rate owner-occupier borrowers in a cash flow shortfall “increased a little over the first half of 2026, but remains relatively low at around 2 per cent”, and is projected to remain a little under 2% for some time.
- For those in deficit, the RBA estimates the median shortfall at around 6% of income — a gap that overtime, a lodger or a repayment restructure can close for many households.
- The median mortgagor can cover more than a year of scheduled mortgage payments at current interest rates from offset and redraw balances, which the RBA describes as a stronger position than before the pandemic.
- Fewer than 1% of borrowers are currently estimated to be in negative equity. On a uniform 20% fall in values from current levels, the Securitisation data suggest around 5% of mortgages would fall into negative equity.
- Housing loans more than three months behind on repayments have “increased a little over the year to date but remain around pre-pandemic levels”.
- Arrears are higher among lower-income borrowers and those with high LVRs or high loan-to-income ratios, but the Review says these have “not picked up significantly since the start of the year” and that these cohorts are a very small share of borrowers.
Before you put any of this in a client email
The Review contains two different figures of roughly 5 per cent and they are not the same measure. One is the share of mortgages in negative equity after a uniform 20% price fall with all else held constant. The other is the share of mortgagors in a cash flow shortfall under the full macroeconomic scenario. The 2 per cent figure is a third series again — variable-rate owner-occupiers only, measured today.
The RBA also sets out its own limits. Its scenario analysis “does not routinely incorporate behavioural adjustments or feedback loops between the financial system and real economy”, and it states that “scenario analysis results are not precise, due to data and model limitation, but they provide insights as to where vulnerabilities may be present”. The negative equity scenario is explicitly a uniform shock that “does not consider the broader macroeconomic implications of declines in housing prices”. Drawing a single arrow from 2% to 5% in client correspondence would misrepresent all three series.
What the Review says about credit supply
There is a second story in the Review that bears directly on what brokers can place. APRA activated debt-to-income limits on 27 November 2025 under Attachment C of APS 220, effective 1 February 2026, allowing authorised deposit-taking institutions to fund up to 20% of new investment loans and up to 20% of new owner-occupied loans at a DTI of six times or greater. On 28 May 2026 APRA confirmed it would hold its settings: the serviceability buffer stays at 3 percentage points, the countercyclical capital buffer at 1% of risk-weighted assets, and the DTI limits unchanged.
The October Review adds the assessment that matters. The aggregate share of new high-DTI lending “remains well below APRA’s recently implemented limits of 20 per cent”, and the limits are “unlikely to be binding at the system level or constrain credit over the period ahead given the outlook for interest rates”. The RBA also records that it supported APRA’s position to keep macroprudential settings unchanged — advice it gave through the Council of Financial Regulators, whose 3 September 2026 statement notes the RBA was “supportive of APRA’s recent decision to keep macroprudential policy settings unchanged given the current environment and outlook”.
The practical read is unglamorous but clarifying. If a file is failing on capacity, the DTI speed limit is almost certainly not why. The constraint that bites looks arithmetic rather than macroprudential: a 3 percentage point assessment buffer applied on top of a 4.60% cash rate. Brokers hoping a macroprudential easing will rescue a pipeline are waiting on a lever the RBA has just said is not the one pressing down, and both regulators have signalled no appetite to move it.
The footnote worth reading twice
Endnote 12 of Chapter 2 is the most immediately operational paragraph in the document. It records that lenders are permitted in certain circumstances to apply an exemption to serviceability requirements, “including, for example, in cases of like-for-like refinancing”. It then notes that in late 2023 the share of loans granted exemptions increased sharply to about 5% of new lending, following an APRA open letter clarifying how it expects banks to manage exemptions, and that liaison suggested many of those exemptions were for borrowers refinancing who no longer passed the stricter serviceability assessment after successive rate increases.
Against that history, the Review’s current observation is striking: the share of new lending issued with exemptions to serviceability requirements “increased only slightly over the first half of 2026”.
The mechanism that absorbed the last refinancing squeeze exists, has an established precedent, and on the RBA’s numbers is barely being used this time. Two qualifications matter. The exemption is prudential — it sits between APRA and the lender, not with the broker — and policy on it varies between lenders. The Review does not say which lenders are applying exemptions or on what terms. What it does justify is asking the question explicitly of a panel rather than assuming the answer, particularly for a client who is comfortably servicing an existing loan but cannot pass a fresh assessment.
Where this meets the best interests duty
A published downside scenario is not a forecast and does not create an obligation. It is worth being precise about what the regulatory framework does say.
The best interests duty applies to mortgage brokers under sections 158LA and 158LE of the National Credit Act, and has done since 1 January 2021. In RG 273, ASIC sets out a non-exhaustive list of factors a broker may need to consider in assessing a consumer’s individual circumstances. That list includes, at RG 273.48(d), “reasonably foreseeable changes to the consumer’s personal circumstances and financial situation”. ASIC also states that there is “no prescribed process or set of factors” for the assessment.
Two points follow for brokers reading the Review. First, RG 273 describes the assessment of what products would be in a consumer’s best interests as a point-in-time assessment. A Review published on 1 October does not reach back into recommendations already made. Second, ASIC says it expects “evidence of compliance with the best interests obligations will come predominantly from the broker’s records”, and that those records should cover the inquiries made into the consumer’s circumstances and the consideration, investigation and assessment of the products recommended.
The connection is therefore modest and worth stating carefully: the Review is a public, regulator-sourced description of what an adverse environment could look like, which may be useful material when a broker is documenting how a recommendation was considered. It does not impose a new test. ASIC has separately warned that the risk of non-compliance is substantially increased where a broker’s processes typically produce a one-size-fits-all outcome — which is an argument for segmenting a book rather than sending it one email. Brokers should confirm their own obligations with their licensee or aggregator compliance team.
Run the three-trigger test across your book
- Narrow the cohort first. The whole database is not the exercise. Start with loans written in the last three years at higher LVRs, construction files, and anyone who has already contacted you about repayments.
- Trigger one — cash flow headroom. Re-run servicing at the current rate, not the rate at origination. The question is whether income covers scheduled repayments plus essentials, which is the RBA’s own definition.
- Trigger two — buffer depth. Six months of payments and essential expenses is the line the RBA uses. Offset and redraw balances are visible on most lender portals; this is the fastest of the three to check.
- Trigger three — equity position. Compare the current balance against a current valuation estimate, not the purchase price. Apply a haircut if the security is in a market that has moved.
- Rank by trigger count, not by any single variable. Three triggers is the priority call. Two warrants a conversation. One, on the RBA’s own framing, is usually manageable.
- Decide what the call is actually offering before you make it — a restructure, a rate review, a hardship referral to the lender, or simply a documented check-in. A call with no option attached erodes trust.
- Record the review. File notes showing which clients were assessed, on what basis, and what was recommended are the records ASIC says it expects to see.
What to watch next
- The serviceability exemption share. This is the single series in the Review that would move first if refinancing pressure becomes acute. It rose to about 5% of new lending in late 2023 and has moved only slightly so far in 2026.
- The next Monetary Policy Board meeting in November. The scenario assumes a cash rate of 5.6%, 100 basis points above where the rate sits now.
- APRA’s next macroprudential update. Settings were last confirmed on 28 May 2026 and endorsed through the CFR on 3 September.
- The next Financial Stability Review, scheduled for 25 March 2027.
The strategic read
The temptation with a document like this is to take the reassuring conclusion and move on, or to take the scariest number in it and build a marketing campaign around fear. Neither serves a client.
What the October Review actually offers the broker channel is a method. The RBA segmented the mortgage population on three variables, published the result, and showed that the aggregate picture masks a small group with overlapping vulnerabilities — which is precisely the argument for why a broker relationship is worth more in a softening market than in a rising one. The regulator’s point, in its own words, is that “aggregate outcomes can mask significant vulnerabilities within specific sectors or groups”.
A broker who can tell a client where they sit against that framework — and can show the work — is doing something a rate comparison cannot. The data to do it is now public. The question is whether the review of your book happens before the client calls you, or after.
Frequently asked questions
No. The Review describes the scenario as deliberately more pronounced than the adverse scenarios in the May 2026 Statement on Monetary Policy, constructed to stress test borrowers and lenders. It is a stress test used to assess resilience, not a forecast of where rates are going. The cash rate was set at 4.60% on 30 September 2026.
No. Around 5% of mortgages would fall into negative equity on a uniform 20% fall, but the RBA states that negative equity alone is insufficient to trigger default where the borrower can still service the loan. The group it identifies as most likely to enter foreclosure — cash flow shortfall, low buffers and negative equity together — is under 1% of mortgagors even in the severe scenario.
On the RBA’s assessment, unlikely. The Review states that the aggregate share of new high-DTI lending remains well below the 20% limits and that those limits are unlikely to be binding at the system level over the period ahead. The larger constraint on capacity is the 3 percentage point serviceability buffer applied on top of a higher cash rate. Individual lender policy may of course differ from the system picture.
Exemptions are permitted in certain circumstances, including like-for-like refinancing, but they sit between APRA and the lender rather than with the broker, and policy varies by lender. The Review does not identify which lenders apply them or on what terms. It is a question worth putting to a lender directly rather than assuming the answer, and any client conversation about it should be framed as a possibility rather than an outcome.
No. The best interests duty sits in sections 158LA and 158LE of the National Credit Act and is unchanged by an RBA publication. ASIC’s RG 273 includes reasonably foreseeable changes to a consumer’s circumstances among the factors that may need to be considered, and describes the assessment as point-in-time. Brokers should confirm how any of this applies to their own process with their licensee or aggregator compliance team.
- Reserve Bank of Australia, Financial Stability Review — October 2026, published 1 October 2026: Chapter 2 (Resilience of Australian Households and Businesses) and Focus Topic 4.2 (Informing Our Financial Stability Assessment Using Scenarios).
- Reserve Bank of Australia, Cash Rate Target statistics (cash rate 4.60%, effective 30 September 2026).
- APRA, ‘Activation of debt-to-income limits as a macroprudential policy tool’, 27 November 2025.
- APRA, ‘APRA maintains current macroprudential policy settings in highly uncertain environment’, 28 May 2026.
- Council of Financial Regulators, Quarterly Statement, Media Release 2026-05, 3 September 2026.
- ASIC, Regulatory Guide 273: Mortgage brokers — Best interests duty, issued 24 June 2020.
Get the broker read, not the press release
The Broker Times unpacks the regulator releases, lender policy changes and market data that actually land on your files — twice a day, written for brokers.
Interactive · Broker tool
The three-trigger file check
Answer three questions about a single client file to see where it sits against the framework the RBA used in its October 2026 Review. Nothing is recorded or sent anywhere — this runs entirely in your browser.
Step 1 — Assess the file
Trigger 1: Cash flow headroom
Re-run at today’s actual rate, not the rate at origination. Does income cover scheduled repayments plus essential expenses?
Trigger 2: Buffer depth
Offset and redraw balances. The RBA’s working line is six months of mortgage payments and essential expenses.
Trigger 3: Equity position
Current balance against a current valuation estimate, not the purchase price.
Loading…
Step 2 — What the Review does and does not say
On borrower resilience
On negative equity
On APRA’s DTI limits
On refinancing and serviceability exemptions
This tool is a reading aid for the RBA’s Financial Stability Review, October 2026. It is general information, not credit assistance, a credit assessment or advice about any particular borrower. It does not replace your own servicing calculations, lender policy checks, or your obligations under the National Consumer Credit Protection Act 2009. Confirm your process with your licensee or aggregator compliance team.
Disclaimer: This article is for general information and professional development purposes only. It does not constitute legal, compliance, or financial advice. Brokers should consult their aggregator’s compliance team and, where required, seek independent legal advice regarding their obligations under the National Consumer Credit Protection Act 2009 and ASIC’s responsible lending guidelines.
