Prime Is Now 79% of Pepper’s Mortgage Flow. The Non-Bank You Keep for Hard Files Is Bidding on Your Clean Ones

1. What Pepper actually reported

Pepper Money released its results for the six months to 30 June 2026 on 20 August. The headline was a record half, but the number brokers should sit with is the mix.

Total originations across mortgages and asset finance reached $6.3 billion, up 40%. Group assets under management hit a record $24 billion, up 20%, and pro-forma net profit after tax rose 15% to $53.9 million. Near-prime originations were around $900 million, up 14%, and specialist lending roughly $100 million, up 65%. Asset finance was comparatively flat at $1.7 billion, up 2%.

Two things follow from that mix. First, the growth is not coming from the borrowers non-banks are traditionally associated with. Specialist lending — the credit-impaired, the recently discharged, the genuinely difficult — is a rounding error next to the $3.6 billion of prime. Second, the book is actually growing. Mortgage assets under management were up 32% year-on-year. For context, total housing credit across the Australian system grew 7.5% in the year to June 2026, according to the Reserve Bank’s financial aggregates released on 31 July. Pepper’s mortgage book grew at roughly four times that pace.

Chief executive Mario Rehayem framed the half around reach rather than risk appetite, telling investors that “Pepper Money delivered very strong results, with 1H2026 setting a new record for total AUM,” as reported by The Adviser. Arrears moved the right way at the same time: 90-plus days past due improved to 1.66% from 1.89%.

2. Liberty: record originations, smaller residential book

Days later, Liberty Financial Group reported its FY26 full-year result — and it reads very differently once you separate group from residential.

Group loan originations hit a record $6.11 billion. Statutory net profit after tax was $144 million, up 8%, with underlying NPATA of $156 million, up 7%. Net interest margin was 2.50% and the total loan book closed at $15.24 billion.

Residential originations were $3.53 billion — strong, but weighted to the first half at $1.85 billion against $1.68 billion in the second. And Mortgage Professional Australia, reporting on the same result, put Liberty’s residential mortgage book at $7.67 billion, down from $7.75 billion a year earlier. The group grew. The residential book did not. The lift came from motor vehicle, commercial and SMSF lending, and from the financial services division housing nMB and Liberty Network Services, whose net annual margin rose from $52.7 million to $63.6 million.

Chief executive James Boyle put the positioning plainly, in remarks reported by The Adviser: “We’re not a deposit taker, and we get to set our own risk appetite, which means we’re able to help customers outside the parameters that APRA set.” That is the classic non-bank proposition — and on the residential side, in FY26, it produced record originations that still did not grow the book.

3. Why “non-bank” is now a funding signal, not a credit signal

For most of the last decade, brokers could use one shorthand: banks for clean files, non-banks for the ones that fall outside policy. That shorthand is now doing less work than it used to.

What the two results show is that the non-bank sector has split into at least three distinct strategies, and each one implies something different about where a file should go.

None of that is visible from a rate sheet. It is visible from results announcements, and it is the reason a panel review built on 2023 assumptions will steer files to the wrong places in 2026.

The aggregator data points the same way. AFG’s index for the June 2026 quarter, as reported by The Adviser, had non-major lenders at 42% of residential lodgements, up from 40.3% a year earlier, with the majors slipping to 58% from 59.7%. The non-major share was strongest in refinancing at 48%, up from 44.5%, and investor lending at 46%, up from 42.9%. First home buyers still leaned to the majors at 72%.

4. What this changes in your best interests reasoning

The best interests duty has always required a broker to act in the consumer’s best interests, and ASIC’s Regulatory Guide 273 sets out how the regulator expects that to work in practice. Two paragraphs are worth re-reading in light of a shifting panel.

RG 273.89 says brokers should present a consumer with more than one option, unless there is a good reason not to. RG 273.54 states that a failure to consider cost and investigate the lowest cost options available to the consumer may suggest non-compliance, with RG 273.51 identifying cost as a factor brokers should prioritise. RG 273.112–113 addresses maintaining a reasonably representative panel of credit providers.

The practical consequence of a non-bank moving into prime is that the set of options a reasonable broker would consider for an ordinary owner-occupier file has widened. Two failure modes now exist where previously there was one:

• Not considering a competitive lender. If a non-bank is pricing prime aggressively and it sits on your panel, a file note that never mentions it may be harder to defend than it was two years ago.
• Reaching for a non-bank without a cost rationale. The reverse is equally live. Placing a clean file with a non-bank because turnaround is quicker or the BDM is responsive is a legitimate consideration — but the reasoning needs to be documented alongside cost, not instead of it.

This is general information rather than compliance advice, and RG 273 is guidance about how ASIC administers obligations under the National Consumer Credit Protection Act 2009 — not a checklist that substitutes for your own licensee’s policy. If your process for documenting lender selection has not been revisited since your panel changed shape, that is a conversation for your aggregator or licensee compliance team, not for a blog.

5. The runoff problem sitting inside the same numbers

There is a second, quieter lesson in Liberty’s result, and it applies to your business as directly as it applies to theirs.

Record originations produced a smaller residential book. Whatever came in the front door left through repayments and discharges slightly faster. Every broker with a trail book faces the same arithmetic on a smaller scale: settlements are a flow number, and the book is a stock number, and it is entirely possible to have a record year for one while the other shrinks.

The AFG data suggests where that churn is concentrated. Non-majors took 48% of refinance lodgements in the June quarter. Refinance activity is, by definition, someone else’s book leaving. If a meaningful share of your settlements are refinances into the non-major end of the panel, then a meaningful share of somebody’s book — possibly, in time, your own — is the thing being refinanced.

6. The prime non-bank test: five questions before you place

When a non-bank is a genuine contender on a clean file, these five questions produce both a better recommendation and a more defensible file note.

• Is the comparison on total cost over the client’s realistic horizon, not the headline rate? Establishment, ongoing, valuation, discharge and any risk fee, modelled over the period the client actually expects to hold the loan.
• Is the pricing risk-graded, and does this client sit where you think they sit? Rate-for-risk means the advertised prime rate and the rate this file is offered can differ. Confirm before you present it as the cheaper option.
• What happens at the back end? Discharge fees, break costs, redraw and offset mechanics, and whether the loan can be varied without a full reassessment. Clients who refinance in three years feel these more than the front-end rate.
• Who will service this loan in five years? With servicing portfolios changing hands, the entity taking hardship and discharge calls may not be the one on the application. Say so up front rather than fielding the call later.
• If a major was cheaper and you still recommended the non-bank, is the reason written down? Turnaround, policy fit, servicing calculator outcome or product features are all real reasons. An undocumented one is the problem, not the choice.

7. What to review this week

• Re-sort your panel by current appetite, not by category. Split it by who is actively competing for prime, who is selective on residential, and who is strongest outside resi. The results announcements tell you this for free.
• Pull your last twenty clean owner-occupier files. Count how many considered a non-bank at all. If the answer is zero, that is worth understanding — it may be right, but it should be a decision rather than a habit.
• Calculate your runoff rate. Twelve-month book, plus settlements, minus current book. Then look at which lenders your discharges are going to.
• Check your file-note template. Does it capture why a lender was chosen over the cheaper alternative, or only that one was chosen?
• Brief your support team on servicing transfers. If a client’s loan is migrated to a new servicer, the first call comes to you.