Repayment Strain Grew 7.3% in Half a Year While New Borrowers Came In 12.9% Safer. Your Back Book and Your Pipeline Are Now Moving in Opposite Directions

Two numbers out of the same report this week point in opposite directions, and both of them describe clients you already have.

Experian’s Housing Pressure Report, covered by Australian Broker, Savings.com.au and The Adviser on 22 and 23 October, has the number of home loan accounts showing signs of repayment strain growing 7.3 per cent across the first half of 2026. In the same six months of 2025 that figure grew 2.3 per cent. The pace has roughly tripled.

The same report has new borrowers arriving with a predicted default risk 12.9 per cent lower than the cohort that entered the market a year earlier, on average new loan sizes that fell 1.7 per cent between the March and June quarters.

So the book is getting worse and the pipeline is getting better, at the same time. For a broker that is not a contradiction — it is a description of two different populations sitting in one CRM. The clients you settled in 2021 and 2022 are the ones carrying the strain. The clients you are writing now are being filtered harder on the way in. Most of the commentary around this report has been written for lenders and for borrowers. This piece is about what the split asks of a broker’s existing database in the next fortnight.

What the report actually measures

The strain figure is not a stress-test estimate or a survey of feelings. Experian builds it from de-identified credit bureau and spending data, and defines strain through observable indicators on the account: arrears, hardship arrangements and loan variations. That matters, because it means every account counted has already had something happen to it — a missed payment, a formal hardship arrangement, or a change to the contract.

The level is still low. In June 2026, 1.8 per cent of home loan accounts showed strain on that definition. The headline is the growth rate, not the level.

Alongside the bureau data, Experian ran a national survey of 1,010 Australians in September, after three cash rate increases earlier this year. In that survey, 55 per cent said rent or mortgage repayments placed a fair amount or a great deal of pressure on their finances. Interest rate changes were the main reason repayments had changed for 34.5 per cent of respondents nationally.

Two survey findings are worth writing down because they are the ones that predict your next six months of phone calls. 20.9 per cent said they regularly or sometimes rely on savings or credit to get through a typical month. And 42.4 per cent said they were not confident they could absorb a further 10 per cent increase in repayments without cutting essential spending or taking on more debt.

That second number is the one to sit with. It is not a forecast of default. It is a statement about how much headroom is left in the household budget before something has to give — and on a loan at current rates, a 10 per cent lift in a repayment is not a large move.

The growth is concentrated, and that is the useful part

Barrett Hasseldine, Head of Data Science at Experian Australia and New Zealand, made the point directly in the report’s coverage: “Housing pressure isn’t showing up evenly across Australia; national averages can mask important differences.” He also said that “understanding housing pressure requires more than a single metric.”

The distribution is where the broker value sits. On Experian’s figures:

• Borrowers aged 35 to 44 recorded the fastest growth in repayment strain of any age group, up 11.8 per cent across the first half of 2026.
• Borrowers aged 45 to 54 sat at a strain level about 13.6 per cent above the national average in June.
• Regional home loan accounts were around 10.5 per cent more likely than the national average to show strain. Capital-city accounts sat 3.5 per cent below it.
• Greater Melbourne and regional Victoria recorded the highest strain levels among the locations Experian analysed.

The age pattern lines up with the cause. Of borrowers aged 35 to 44 whose repayments had changed, 47.5 per cent named interest rate changes as the main reason, against 45.5 per cent of 45 to 54 year olds and 34.5 per cent overall. These are not households that took on the wrong loan. They are households whose loan did what a variable loan does, on a balance large enough that the move mattered.

If you want one practical read of the above: the strain is growing fastest in the cohort most likely to be mid-life, mid-mortgage and carrying dependants, and it is doing it faster outside the capitals than inside them. That is a filter you can run on your own database today. Age bracket, postcode, settlement year, and loan size at settlement are four fields most CRMs already hold.

Why the RBA and Experian can both be right

Brokers reading the Experian coverage against the Reserve Bank’s October Financial Stability Review will notice a gap in tone. The RBA’s assessment of household resilience runs under the heading “Arrears rates remain low despite a recent pick-up.” It says the share of housing loans more than three months behind “has increased a little over the year to date but remains around pre-pandemic levels,” and that arrears among lower-income borrowers and those with high LVR or loan-to-income ratios “remain higher than for other borrowers but have not picked up significantly since the start of the year.”

On equity, the Review puts “less than 1 per cent of borrowers” in the position of owing more than their property is worth, and estimates that even a uniform 20 per cent fall in housing prices would push “only around 5 per cent of mortgages” into negative equity. On hardship, it says “the share of loans in formal hardship arrangements increased but remains low.”

Both readings hold. The RBA is describing the stability of the system; Experian is describing the rate of change inside a cohort. A 7.3 per cent increase on a 1.8 per cent base is not a solvency event for any lender. It is, however, a meaningful increase in the number of households who have already asked for a variation or missed a payment — and every one of those is a client conversation that someone is going to have. The question for a broker is whether it is you, or the retention team at their current lender, or nobody.

The constraint nobody can refinance their way around

The obvious broker response to a client under repayment pressure is a refinance. For a growing share of these households, that door is narrower than it was.

The MFAA’s August 2026 Market Sentiment Survey, taken across 588 brokers and released on 23 September, found 40.7 per cent of brokers expected more of their clients to have difficulty meeting home loan repayments over the following six months. The same survey recorded the share of brokers reporting clients blocked from refinancing on serviceability grounds at 49.2 per cent, more than double the 24.4 per cent recorded six months earlier — a figure The Broker Times covered when the survey landed.

What is less discussed is what those same brokers actually did instead. In the six months to August, 96 per cent reported helping a client secure a discount from their existing lender. 91 per cent helped a client restructure a home loan. 84 per cent helped clients with budgeting strategies.

Read against the Experian data, that is the shape of the work for the rest of this year. For a household in the 35 to 44 bracket, in a regional postcode, with a balance written in 2021 and a repayment that has moved three times this year, the answer is frequently not a switch. It is a repricing request, a structural change, or a conversation about where the budget is actually going — and sometimes it is a referral into the lender’s hardship team, which is the credit provider’s function, not yours.

When the recommendation is “stay where you are”, RG 273 still applies

This is the part that gets missed, and it is worth being precise about because the guidance is explicit.

ASIC’s Regulatory Guide 273 does not treat a recommendation to stay put as a non-event. At RG 273.117, the guide says that “if you provide credit assistance suggesting that a consumer remain in a home loan you previously helped them obtain, you will need to act in the consumer’s best interests when making that suggestion.” At RG 273.119 it goes further: “You must not suggest that a consumer remain in a credit contract … without considering whether this would be in the consumer’s best interests.” And at RG 273.120, for consumers thinking about refinancing, “you should consider the costs associated with refinancing.”

The guide also speaks directly to the constrained client. RG 273.59 acknowledges that “access to certain products may be limited based on consumers’ specific circumstances … including their credit risk and desired loan amount.” Where that is the case, RG 273.61 asks you to “explain why the products are not available and the limitations involved with the available option(s)” and to “still consider the cost of the available options.”

On records, RG 273.162 says ASIC expects you “to keep records of how you have acted when providing credit assistance,” and RG 273.163 notes that “good records will help you to demonstrate that you have complied with these obligations.” RG 273.104(b) asks that “you clearly record the recommendations you presented, including any alternatives that are not selected by the consumer.”

Put together, the compliance implication of a strained back book is not complicated, but it is specific. A reactive call where you tell a client there is nothing better available right now is credit assistance. It needs the same reasoning and the same file note as a switch: what you looked at, why the alternatives were not available or not better once costs were counted, and what the client decided. A file that records a refinance you placed, and says nothing about the twelve clients you advised to stay, is a file that records only half of your year.

This is general information rather than advice on your own obligations. Your licensee or aggregator compliance team sets the standard your files are actually audited against, and that is the version to work to.

A back-book triage you can run this week

The Experian segmentation gives you a credible basis for prioritising calls rather than working alphabetically. A workable sequence:

1. Build the list. Filter your database for clients who settled in 2021 or 2022 on a variable rate, are in the 35 to 54 age bracket, and sit in a regional postcode. Those three filters sit on top of each other in Experian’s data, so the overlap is where the strain is growing fastest. Add Greater Melbourne and regional Victoria as a separate slice.
2. Sort by headroom, not by balance. The 42.4 per cent figure is about the ability to absorb another 10 per cent on the repayment. Run that as a rough calculation on each file using the repayment you last recorded. The clients whose repayment has already moved most since settlement go to the top.
3. Lead with the repricing ask. 96 per cent of brokers secured a discount from an existing lender in the six months to August. That is the highest-probability result on the list, it does not depend on a fresh serviceability assessment, and it is the step a client is least likely to take alone.
4. Separate the three outcomes before you call. A reprice, a restructure, and a hardship referral are three different conversations with three different file notes. Deciding which one a file is likely to need before you dial keeps the call short and keeps the note accurate.
5. Write the note for the clients who stay. Per RG 273.104(b) and RG 273.162, the alternatives you considered and did not place belong on the file. Build the template once rather than writing it eighteen times.

A figure to treat carefully

One number in the coverage of this report does not reconcile cleanly, and it is worth flagging rather than repeating. Australian Broker and Savings.com.au both report that new borrowers had a predicted default risk 12.9 per cent lower than those entering the market a year earlier. The Adviser‘s account of the same report describes new borrowers’ predicted default risk as 35 per cent above the average existing home loan borrower’s, drawn from Experian’s June-quarter bureau data.

Those are different comparisons — one is a change against last year’s intake, the other a level against the whole book — and they are not necessarily in conflict. But if you intend to use either figure in a client conversation or a lender submission, confirm it against Experian’s own report first rather than against the trade coverage. The 12.9 per cent figure is the one two outlets agree on, and it is the one used above.

What to watch next

The September CPI prints on 28 October — a monthly release, since the ABS’s complete Monthly CPI replaced the quarterly headline measure — and it is the input the market will read ahead of the next rate decision. If it lands on the firmer side, the 42.4 per cent of households with no capacity to absorb another 10 per cent on the repayment becomes a larger operational issue for the broker channel, not a smaller one. Watch also for the next Experian update and the next MFAA sentiment survey — the useful signal in both is not the national average but whether the regional-versus-capital gap widens.

The practical significance

The strain in Australian home loans is not spread evenly, and the national figure is the least useful number in the report. A 1.8 per cent strain level across the market tells a broker nothing. A 10.5 per cent regional skew, an 11.8 per cent increase in the 35 to 44 cohort and a 13.6 per cent above-average reading for 45 to 54 year olds tell you which forty files to open first.

The front book is being written more conservatively and the back book is where the pressure is. That is an argument for treating database review as revenue work rather than housekeeping — and for building the file note that covers the clients who, this year, are better off staying exactly where they are.