26 Non-Banks Published Home Loan Data With No Interest Rate. One Funder Sits Behind Half the White Labels

Non-bank lenders became part of Australia’s open banking regime on 13 July 2026. It was the third sector to join the Consumer Data Right, it happened with very little noise in the broker channel, and a month later the first read on what actually got published is not flattering.

According to analysis by fintech Stryd, shared with Broker Daily and published on 20 August, 67 non-bank lenders have now published residential mortgage Product Reference Data. Twenty-six of them published without an interest rate. Fifty-four of the 67 carry white-labelled products, funded by just 12 core lenders — and a single funder sits behind half of every white-label arrangement in the set.

For brokers, none of this is a technology story. It is a panel story, a file-quality story, and — because open banking is now the most-cited defence against AI-generated payslips — a fraud story. It is also a story about the difference between a lender panel that looks diverse and one that actually is.

What went live on 13 July — and what did not

This is the distinction most brokers will get wrong, so it is worth being precise about it.

What commenced on 13 July 2026 was product data sharing. The CDR website states plainly that “product data sharing obligations commenced for relevant non-bank lenders on 13 July 2026”. Product Reference Data is the lender’s published product menu in a machine-readable, standardised form: rates, fees, features, eligibility, repayment types, loan purposes. It is not customer information. Nobody consents to it. It is simply the shelf, described in a common language so that software can compare like with like.

What has not commenced is consumer data sharing — the part brokers actually care about day to day, where a client authorises their transaction and account data to flow. The CDR rollout timetable sets that out in two stages: 9 November 2026 for initial providers, and 10 May 2027 for large providers.

So when a BDM tells you their non-bank is “in open banking now”, the accurate reading is that their product menu is machine-readable. You still cannot pull that lender’s own account data into a servicing assessment, and depending on which tier they sit in, you may not be able to until well into 2027.

The ACCC put the expansion in context in its 13 July media release, with Commissioner Dr Ian Oppermann saying: “The expansion of the Consumer Data Right to non-bank lenders is a significant step in giving consumers access to information about the broadest possible range of financial products.” The same release noted more than 1.3 million Australians are now using the CDR, an increase of around 135 per cent over the past year.

The 135 products with no price on them

The headline problem in Stryd’s analysis is the simplest one. Of the 67 non-banks that have published, Broker Daily reports that 26 did so without an interest rate — affecting roughly 135 Product Reference Data entries, with rate types, repayment types and loan purposes also missing in places.

Ruth Hatherley, founder and chief executive of Moneycatcha, which operates Stryd, was blunt about what that does to the dataset’s purpose. “A CDR data set that excludes rate information isn’t just incomplete, it’s unusable for its core purpose — you can’t help a consumer find a more competitive home loan if the product data doesn’t say what the rate is,” she told the publication.

She added: “Roughly 130 non-bank mortgage products are currently live in the CDR without rate information. On the ACCC’s own terms, that’s not expanding consumer choice — it’s just adding noise to the dataset.”

Two further findings in the same analysis matter for anyone relying on a comparison tool. Nineteen Product Reference Data entries across five lenders initially displayed rates inconsistent with those on the lenders’ own public websites. Fifteen entries across two lenders contained incorrect repayment types or loan purposes.

These are moving numbers, not a fixed scorecard. Banking Day, reporting on 21 July, put the gap at 25 lenders and 126 products on an earlier snapshot. Lenders are still onboarding and still correcting. The direction of travel is the point: a month in, a meaningful slice of the non-bank shelf sits in the national comparison dataset without a price attached to it.

The finding that should change how you read your panel

The white-label numbers are the part of this that deserves fifteen minutes of your attention rather than fifteen seconds.

Stryd found that 54 of the 67 non-bank lenders — over 80 per cent — feature white-labelled Product Reference Data. Those products are funded through just 12 core lenders. And, per the analysis, a single funder accounts for 50 per cent of all white-labelled arrangements.

There is also a direct link between the two findings. Broker Daily reports that 24 of the brands affected by the missing-rate problem are white-labelled through the same core lender. One funder’s data practices are propagating across two dozen brand names.

Consider what that means at the file level. A broker who presents three non-bank options, each under a different brand with a different BDM, may in substance be presenting one credit policy, one funder and one set of underwriting appetites — dressed three ways. If that funder tightens policy, all three move together. If it has a servicing quirk that kills the deal, it kills it three times.

White labelling itself is not new. It is a legitimate, disclosed business model that has widened access to credit for borrowers the majors will not write. What is new is that the CDR has made the concentration visible and countable at a whole-of-sector level. Before 13 July, “how much of the non-bank shelf is really the same funder?” was a question you answered with instinct and BDM gossip. Now it is a number.

Read the exemptions register before you plan around November

Here is a detail almost nobody in the channel has looked at, and it is publicly available.

The published list of non-bank lender data holders names seven initial providers due to share consumer data from 9 November 2026: Columbus Capital, Firstmac, Liberty Financial, Metrics Credit Partners, Pepper Money, Resimac Group and Toyota Finance Australia. A further 31 large providers follow on 10 May 2027 — a list that includes La Trobe Financial, Bluestone, Athena, Mortgage House, Think Tank, Redzed, MoneyMe, Plenti and, notably for the broker channel, Australian Finance Group.

But the ACCC also maintains a public Consumer Data Right exemptions register, and several of those names appear on it. Columbus Capital holds an exemption, granted 16 June 2026, relating to product data sharing for certain white-label brands. Metrics Credit Partners holds exemptions covering product data sharing until 30 June 2027 and consumer data sharing until 30 June 2028. Pepper Money and Pepper Asset Finance were granted an exemption on 9 December 2025. Among the 2027 cohort, MA Money, Credabl, De Lage Landen, Scottish Pacific, Shift and Volkswagen Financial Services all hold exemptions of varying scope and duration.

To be clear about what an exemption is and is not: it is a lawful regulatory accommodation granted by the ACCC, usually on grounds of proportionality, technical readiness or product scope. It is not a compliance failure, and it says nothing about a lender’s conduct or its suitability for your clients. But it does mean the practical answer to “which non-banks can I pull consumer data from after 9 November?” is narrower than the seven-name list implies. If you are building a 2027 process on the assumption that the non-bank sector will be fully data-enabled by then, check the register rather than the timetable.

Why this gets more important every quarter

The non-bank channel is not a niche you can afford to model loosely any more.

Analysis of ABS Lending Indicators by Money.com.au, reported by Australian Broker on 19 August, put non-ADI home lending at $10.49 billion in the June 2026 quarter, up from $6.35 billion a year earlier — a 65.2 per cent year-on-year increase. ADI lending over the same period grew 2.6 per cent, from $85.41 billion to $87.61 billion. On that analysis non-banks now hold 10.7 per cent of the market, against 4.8 per cent in September 2019. Total housing lending fell 5.2 per cent quarter-on-quarter; non-ADI lending rose 3.2 per cent.

Money.com.au mortgage expert Nick Burgess pointed to the structural driver: “Non-bank lenders sit outside APRA’s prudential rules, including the 3% serviceability buffer banks have to apply.”

Put the two stories together. A rapidly growing share of Australian home lending is being written by lenders whose sector has just entered the national comparison dataset with a meaningful proportion of its shelf priced at “unknown”, and with four in five brands sitting on someone else’s funding line.

The fraud angle nobody has connected yet

This one is timing-sensitive and worth thinking through before your next self-employed non-bank deal.

Speaking to The Adviser on 20 August, Experian A/NZ head of fraud and identity Richard Atkinson warned that AI-generated financial documents have become extremely difficult to spot by hand — he described it as “nearing on impossible to detect the fraud” when reviewing such material manually, adding that “the tech is in everyone’s hands, and it’s very simple for the man in the street to create that fraudulent document.” His recommended defences included collecting income data directly from the customer’s bank account through open banking channels, rather than relying on documents the applicant hands over. Those comments followed AUSTRAC’s Operation Claw work on suspected coordinated mortgage fraud.

The advice is sound. The catch is sequencing. Open banking consumer data sharing is well established in the banking sector — but in the non-bank sector it does not begin until 9 November 2026 at the earliest, and 10 May 2027 for most of the names on the list, with exemptions pushing some later still.

That produces an uncomfortable asymmetry. The lenders most likely to be writing complex, self-employed, credit-impaired and non-conforming deals — precisely the files where document fraud risk concentrates — are the lenders whose own CDR consumer data channel is not open yet. For those deals, for now, you are still working from PDFs. That is not a reason to avoid non-banks. It is a reason to be deliberate about which verification steps you apply where, and to write down why.

The trusted adviser question to raise with your licensee

One more piece of the CDR framework is directly relevant to brokers and is frequently misunderstood.

Mortgage brokers are a named class of “trusted adviser” under the CDR rules, alongside qualified accountants, legal practitioners, tax agents, financial counselling agencies and financial advisers. That pathway lets a consumer consent to their CDR data being disclosed to you.

The OAIC’s guidance flags a consequence that is easy to miss. Its material states that “unless the CDR data is disclosed to a trusted adviser who is also an accredited person, the CDR data will no longer be subject to the protections and safeguards of the CDR system”. In other words, data that arrives via the trusted adviser pathway sits outside the CDR’s own privacy regime and falls back on general obligations and professional requirements. The OAIC’s stated best practice is that trusted advisers receiving CDR data “should ensure that they handle that data transparently and in a way that the consumer would expect”.

This is general information, not advice, and the detail of how it applies to your business depends on your licensing arrangement. If your brokerage is using or planning to use the trusted adviser pathway, that is a conversation to have with your licensee or aggregator compliance team — specifically about where that data lands, how long you keep it, and what your privacy documentation says about it.

What to review this week

Five things, none of which take long.

1. Map your non-bank panel to its funders. Take the non-bank lenders you actually used in the last 12 months and ask each BDM, in writing, who funds the product. If three of your five “options” trace back to one funding line, your panel is narrower than your submission history suggests. This is a diversification question before it is anything else.
2. Stop treating a comparison screen as a verified rate. With rate inconsistencies found across a number of lenders in the first month, confirm pricing against the lender’s own current rate sheet or portal before it goes in front of a client — particularly for white-labelled brands. Note in the file where the rate came from and when.
3. Check whether your tech vendor sources non-bank pricing from the CDR. If it does, ask what it does with a product that has no rate published: does it exclude it, show it blank, or fall back to a scraped figure? You want to know the answer before a client asks why a lender was not considered.
4. Pull the exemptions register before you plan your 2027 verification workflow. The ACCC’s Consumer Data Right exemptions register is public. If your process assumes a lender will be data-sharing from 9 November 2026, verify that assumption against the register rather than the headline date.
5. Write down your document-verification standard for non-bank files. Given where open banking coverage currently sits in that sector, be explicit about what you do to satisfy yourself on income and expenses when direct bank-sourced data is not available. A dated file note describing your process is worth considerably more at audit than a recollection.

The strategic read

Open banking was sold to the broker channel as a mechanism that would widen choice. On the evidence of its first month in the non-bank sector, what it has widened first is visibility — and the first thing it has made visible is that a large part of the sector’s apparent variety is a smaller number of funders wearing more names, and that a meaningful slice of the shelf is currently in the national dataset without a price on it.

That is not a reason for cynicism about the CDR. Data quality problems in a new sector’s first month are ordinary, and the banking sector went through the same corrections. The 9 November milestone is real and worth preparing for, and a working non-bank consumer data channel would be a genuine upgrade to how brokers verify income in exactly the segment where verification is hardest.

But between now and then, the practical takeaway is narrower and more useful than the promise. Know who funds the products you recommend. Verify the rate before you quote it. Do not assume a lender is data-enabled because its sector’s start date has passed. And keep the best interests duty in mind as a live question rather than a form: if you are comparing brands that share a funder, a credit policy and a servicing calculator, you should be able to explain to a reviewer what your comparison actually compared.

The dataset just made that question answerable. It is worth answering it before someone else asks.