Specialist Lending Index · Issue #1
The Specialist Lending Index — July 2026
What Australia's specialist and non-bank lenders will actually consider, counted across one broker's research library of 60 lender policy sets. Issue #1.
By John Carson-Zangor — mortgage broker, Bethania QLD. Credit Representative 537545 of QED Credit Services Pty Ltd, Australian Credit Licence 387856.
General information only — this is not credit advice and doesn't consider your situation. Lender policies change without notice. All figures are as at July 2026, counted from one broker's research library, not from any official industry dataset.
· Figures as at July 2026
I maintain a research library of lender credit policies — 60 lender brands as of July 2026, from major banks through mutuals to private lenders — and I built an AI system that reads it, so I can search the whole market's appetite in seconds when a client's bank says no.
The individual policy documents are commercial-in-confidence, so I will never publish what a named lender's policy says. But nobody, as far as I can find, has ever published what the market says in aggregate. That's this index: counts, shares and ranges across the library, updated monthly. No lender names, no rates, no promises — just the shape of the specialist end of the Australian mortgage market, measured.
- The five numbers that matter this month
- What do I mean by "specialist lending"?
- How many lenders actually have appetite for impaired credit?
- Can you get a home loan with unpaid defaults?
- How soon after bankruptcy can someone borrow?
- What about ATO and tax debt?
- How big a deposit does specialist lending really need?
- And self-employed income?
- What this means if your bank said no
- Key numbers — July 2026
- FAQ
- Methodology and limitations
The five numbers that matter this month
What do I mean by "specialist lending"?
Specialist lending (sometimes "non-conforming" lending) is the part of the market built for borrowers who don't fit a mainstream bank's credit box: defaults or arrears on the credit file, a discharged bankruptcy, income that's real but hard to evidence the standard way, or debts — like a tax bill — that banks won't touch. In Australia it lives almost entirely with non-bank lenders. It generally prices higher than bank lending, usually swaps lenders mortgage insurance (LMI) for a lender "risk fee", and most of it is structured in tiers — commonly named something like prime → near prime → specialist — where the pricing and maximum loan-to-value ratio (LVR) step down as the credit history gets heavier.
Terms used below: a default is an overdue debt listed on your credit report — paid if since settled, unpaid if not. Alt-doc means proving self-employed income without full tax returns. LVR is the loan as a percentage of the property value.
How many lenders actually have appetite for impaired credit?
Of the 52 policy sets in my library with enough substance to count (see methodology), 12 — about one in four — document a genuine impaired-credit program: 10 are residential tiered specialist models, 2 are commercial/private lenders that assess credit events case-by-case. A further handful are well-known specialist brands whose deeper adverse-credit matrices simply aren't captured in my sources yet — so if anything, this undercounts the specialist market.
The other three-quarters of the market documents, at most, tolerance for small paid defaults. Across the 17 policy sets that put an actual dollar figure on default tolerance, the same two numbers keep appearing: $500 (in 11 of the 17) and $1,000. That's the practical boundary of the mainstream market: a paid default over about $1,000 and you're usually shopping in specialist territory whether you know it or not.
Can you get a home loan with unpaid defaults?
Mainstream policy is blunt here: at least 10 of the 52 policy sets state that any unpaid default is an outright decline, and most of the rest are silent or case-by-case. But at least 9 policy sets — all non-banks — document appetite to consider unpaid defaults at some tier. The documented tolerances range from capped amounts (around the $1,000–$3,000 mark, sometimes only if the listing is older than 12–24 months) up to one policy set whose specialist tier documents defaults with no stated restrictions, priced accordingly.
What decides placement isn't a credit score. Across the specialist policy sets, the same four questions do the work: how much, how old, paid or unpaid, and was it one event or a pattern. One specialist policy set documents using no credit score at all. That's the biggest practical difference from bank lending — specialists read the story, not the number.
How soon after bankruptcy can someone borrow?
This is the widest spread in the entire library. Of the 52 counted policy sets, 27 state a position on bankruptcy. The answers run from "never — even discharged bankrupts are unacceptable" through the mainstream norm (unacceptable while undischarged, then a 3–5 year wait or case-by-case referral) all the way to the specialist end:
- 7 policy sets document a pathway within roughly two years of discharge (one more starts at exactly the two-year mark). All are non-banks.
- 6 of those document appetite from as little as one day after discharge.
- 1 documents considering borrowers still in an undischarged bankruptcy, where it was entered into more than two years ago.
Even inside the specialist set, appetite isn't uniform — at least one otherwise-flexible specialist lane documents that bankruptcy, discharged or not, is never acceptable. Which is exactly why this gets checked lender by lender, on the day, against current policy.
What about ATO and tax debt?
Tax debt is quietly one of the sharpest dividing lines in the market. 9 of the 52 policy sets document appetite to refinance or consolidate ATO/tax debt — every one of them a non-bank — at least one of which documents accepting a tax debt that stays on a payment plan after settlement. On the other side, at least 8 policy sets explicitly list paying out tax debt as an unacceptable loan purpose. The rest are silent. If a business owner with a tax bill walks into the wrong lender, the decline isn't about their income — it's a purpose rule they were never told about.
How big a deposit does specialist lending really need?
Here's the aggregate that matters for planning. Among the residential specialist policy sets with clearly documented tier caps, the maximum LVR at the deepest impaired-credit tier sits between 75% and 85% in every case — and two-thirds of them cap at exactly 80%. Lighter "near prime" tiers typically run 85–90%, with a small number of full-doc options documented into the 90s.
Translation: heavy credit history usually means bringing a 15–25% deposit (or equivalent equity) to the table. And the support structures mainstream borrowers lean on mostly aren't there: 9 of the 10 residential specialist policy sets document no family-guarantee product, while about two-thirds of the bank and mutual policy sets I reviewed do. The flip side: 4 of the 10 document that genuine savings aren't required — the deposit can come from a gift, sale, or other lump sum. Specialists trade guarantee support for flexibility about where the money comes from.
And self-employed income?
19 of 52 policy sets (37%) document an alt-doc pathway — and the split is absolute: 19 of the 27 non-banks (70%), zero of the 25 banks and mutuals. All 19 use some combination of the same three verification tools: an accountant's declaration, BAS, or business bank statements. 13 policy sets — a quarter of the counted library, again all non-bank — flag appetite for young ABNs, some from as little as roughly six months' trading.
What this means if your bank said no
Three practical takeaways from this month's counts:
- A bank decline is one lender's credit box, not a market verdict. Roughly a quarter of the counted market documents appetite for the exact things banks decline on — but that quarter is nearly invisible unless you (or your broker) can read it.
- The variables that move a file are knowable: the size, age, paid status and count of credit events; how income can be evidenced; the purpose of the money; and the deposit. A default paid last month versus 13 months ago can move a file whole tiers.
- The price of appetite is structure: lower maximum LVRs, risk fees instead of LMI, and higher pricing than bank lending — usually as a rehabilitation step, refinanced back to mainstream once the file is clean. Nothing here is a promise that any particular loan will be approved.
Bank said no and you want to know where your file actually sits? That is a conversation, not a guess.
Book a call with Johnor start with the declined home loans pageKey numbers — July 2026
| Measure | Count | Denominator |
|---|---|---|
| Lender brands in the library | 60 | 62 notes minus 2 duplicates |
| Policy sets counted for this index | 52 | 8 low-confidence stubs excluded |
| Market mix of counted sets | 27 non-bank / 13 bank / 12 mutual | 52 |
| Documented impaired-credit program | 12 (23%) | 52 |
| Alt-doc pathway documented | 19 (37%) — all non-bank | 52 |
| Unpaid defaults considered at some tier | ≥9 (17%) — all non-bank | 52 |
| Any unpaid default = documented decline | ≥10 | 52 |
| State a bankruptcy position | 27 (52%) | 52 |
| Pathway within ~2 yrs of discharge | 7 (6 from day one) | 52 |
| ATO/tax-debt appetite documented | 9 — all non-bank | 52 |
| Explicitly exclude tax-debt payout | ≥8 | 52 |
| Deepest-tier max LVR range | 75–85% (mode 80%) | 9 sets with clear caps |
| Short-ABN appetite | 13 (25%) — all non-bank | 52 |
| Checked during 2026 | 42 (37 in Jun–Jul 2026) | 52 |
How this was counted (short version)
This index is counted from my research library: one structured note per lender brand, built from product guides, credit policies, broker portals and BDM communications, each tagged with source, date checked and confidence. For Issue #1 I counted 62 notes, removed 2 duplicates (60 brands), excluded 8 low-confidence stubs, and tallied the remaining 52 by hand. "Documented" means the policy set states the position — silence is never counted as yes or no, which is why many denominators are smaller than 52. Median "last checked" date: 20 June 2026. Full methodology and limitations are published alongside this index.
FAQ
Is the list of lenders available?
No. The underlying policy documents are commercial-in-confidence, so this index only publishes aggregates — counts, shares and ranges. Matching a scenario to specific lenders is broker work, done against current policy on the day.
Does "documented appetite" mean I'd be approved?
No. It means the lender's policy, as at the date checked, states it can consider that scenario. Every application is assessed individually against full policy, serviceability and security rules.
Are non-bank lenders regulated?
Yes — they're licensed credit providers under the same National Consumer Credit Protection Act as banks. They fund loans differently and don't hold deposits, which is why their credit appetite can differ so much.
Will these numbers change?
Yes — that's the point. The index is republished monthly from the same library at the same URL, so the shares can be tracked over time.
Can I use these figures?
Yes, with attribution and a link: Specialist Lending Index, July 2026, counted from a 60-lender research library. They describe one broker's library at one date, not the whole market.
How this index is built
Methodology and limitations
What the library is
A private research library of structured lender-policy notes maintained by John Carson-Zangor for broker triage, one note per lender brand. Sources per note are a mix of: full broker credit policies and product guides (PDFs), broker-portal extracts, public broker-site pages, BDM emails, and archived niche sheets. Every note carries frontmatter: source description, date_checked, confidence (high/medium/low), tagged strengths/avoids, a written credit-appetite summary, max-LVR notes and open BDM questions. The library is triage-grade, not advice-grade: its own standing rule is "recheck live lender/aggregator policy before relying on any note for a client."
How the counting was done
- Universe: 62 lender-note files in the library on 16 July 2026 (the marketing plan's "64" counted the folder's README and template files; the honest note count is 62).
- Deduplication: 2 notes are pure alias pointers to other notes (same policy, different brand entry). Removed → 60 lender brands.
- Exclusions: 8 notes rated
confidence: low(public source-maps, stubs, or stale single-product flyers — including one dated 2021) are counted in library size but excluded from every policy-parameter tally → 52 counted policy sets (33 high confidence, 19 medium). - Shared procedure: two bank brands in one merged group operate one shared lending procedure; both are retained as brands because brand-level settings differ (e.g. government-scheme availability). Where this double-counts an appetite position, it affects tallies by at most 2.
- "Documented" standard: a policy set only enters a tally if the note states the position (for or against). Silence, "unknown", or "confirm with BDM" is excluded from that tally's numerator AND denominator. This is why sub-denominators (e.g. 27 of 52 stating a bankruptcy position; 17 stating dollar default caps; 9 with clear deepest-tier LVR caps) are smaller than 52.
- Segment classification (non-bank / bank / mutual) is John's judgement call by brand ownership and funding model: 27 / 13 / 12 within the 52.
- Repayment-features dataset: the library's companion
repayment-policy.json(extra repayments, offset, redraw, splits) was checked and excluded entirely — 61 of its 62 entries carryconfidence: "unfilled", leaving a denominator of 1, which cannot support any aggregate claim.
Freshness
date_checkedspread of the 52 counted sets: oldest 13 Feb 2024, newest 15 Jul 2026, median 20 Jun 2026.- 42 of 52 checked during 2026; 37 of 52 checked in June–July 2026; 10 last checked in 2024–2025.
- Freshness describes when JOHN's note was last verified against a source — not when the lender last changed policy. A lender may have changed settings after the check date.
Limitations (state these plainly wherever the index is cited)
- This is one broker's library, not a census. 60 brands is a large working set but not the whole Australian market; lender selection reflects John's accreditations and research priorities, which oversample the specialist/non-bank end (27 of 52 counted sets are non-banks — a higher share than the market at large).
- Sources vary in depth. High-confidence notes come from full credit policies/product guides; medium-confidence notes may rest on product sheets or portal extracts that omit the adverse-credit matrix. Several well-known specialist brands are undercounted because their deep-adverse matrices aren't in the captured sources — stated in the article.
- Counts are of documented policy positions, not approval behaviour. A documented appetite is not an approval rate; lenders apply full policy, scoring, servicing and security rules per file.
- Aggregation choices matter. "Impaired-credit program", "alt-doc pathway" and segment labels required judgement; the classification list is retained internally so Issue #2 uses identical rules and the month-on-month deltas are real.
- Two data points within tallies rest on medium-confidence archive documents (both commercial/private-lender appetite claims). Removing them would change the impaired-credit program count from 12 to 10 and no other headline number.
- Dates: all figures speak as at July 2026 only. Policies change without notice.
John Carson-Zangor is a mortgage broker in Bethania, Queensland, specialising in complex and declined scenarios. Credit Representative 537545 of QED Credit Services Pty Ltd, Australian Credit Licence 387856. This article is general information only and is not credit advice; it does not consider your objectives, financial situation or needs. Lending policies change without notice. Figures are as at July 2026 and are counted from one broker's research library.
Start with the full picture
A review should leave you clearer, not more pressured
Repayments, costs, equity, credit history, timing and lender criteria all belong in the same conversation.
John Carson-ZangorDirect help from a residential mortgage broker based in Bethania, Logan.