It was lodged in April
We commissioned research to show that a group structure had hidden the exposure. It showed the opposite, which is the more useful answer
The intended article was about invisibility. A property group running several hundred project vehicles, forty-odd lenders each facing a different one, and an aggregate that nobody could see from inside any single file. It is a good hypothesis, it fits what has been written about the Bathla collapse, and we asked for it to be tested before we wrote it. Frankly it was what we expected after MFS.
But I was wrong, the hypothesis did not survive. What follows is what the research showed, it was all there, this could be just another story of headwinds and aggressive leverage.
The dates
Universal Property Group, the parent of the Bathla group, lodged consolidated financial statements for the year ended 30 June 2025. They record total liabilities of $3.19bn and total borrowings of $2.85bn, of which $1.99bn was short-term debt falling due within twelve months.
Voluntary administrators were appointed on the 25th of August 2026, across 542 companies. At the first creditors meeting on the 4th of September the administrators put known creditors at approximately $3.4bn, of which around $3.08bn is secured. Reporting puts the number of private credit funds with exposure at around forty, with individual positions ranging from A$1.5m to A$340m.
A lender approached in June 2026 to write a facility against a Bathla project vehicle could have bought a company extract on that vehicle, read the ultimate holding company field, bought the parent’s consolidated accounts, and seen $2.85bn of borrowings with two thirds of it falling due inside a year. The accounts had been available for six weeks. The total cost is under a hundred dollars and the work is an afternoon of reading…assuming you don’t outsource it to a bot!
And before the registers, there was a newspaper
This is the part that changes the argument rather than merely correcting it.
On the 19th of June 2025, fourteen months before the administration, the Australian Financial Review published an investigation into the group. It reported borrowings of more than $2.7bn against a cash balance of $6.5m, nearly $200m paid in interest and other financing costs over the year, and that a number of larger private credit firms had declined to take on the company’s debt.
The registers were the second-best source available. The first was a broadsheet, and it was a year old. By the time the papers pick up a story, in particular these days…it’s normally “known” within the lender market.
So why does the second search not get made
Because diligence is scoped to the counterparty, and the counterparty was a project vehicle. The accounts were lodged by the parent. The newspaper was writing about the parent. Nothing in a conventional underwriting scope requires you to climb the ownership chain and start again at the top. But equally…most should.
That is a convention rather than a failing, and it is close to universal. We share it, which is the honest reason this article exists at all: we assumed the structure was doing some of the lift in concealing borrowings through layered or partner vehicles, and the structure was doing nothing of the kind.
The field on the extract that most people do not read
One mechanical detail is worth more than the rest of this article, and it costs money so you can’t just lean on an LLM to find it.
Under section 319 of the Corporations Act, a large proprietary company must lodge audited financial statements within four months of its financial year end. For a 30 June balance date that is the 31st of October. A company extract carries a financial reports table, and the entries in it are not all the same. A report marked as lodged, even lodged late, has been filed. A report marked outstanding is due and has not been filed. The first is a compliance lapse. The second is an unmet statutory obligation sitting on a public register.
Bathla’s lodgement history is visible on that table. The FY2022 accounts were lodged in January 2024. The FY2023 and FY2024 accounts were both lodged in June 2025, and the Financial Review reported that the FY2024 accounts were filed days after it had asked questions about their late disclosure. For context on how seriously the regulator now takes this, ASIC issued infringement notices to twelve large proprietary companies in late 2025 for failing to lodge FY2024 reports on time. I don’t have the time, or the inclination, to chase down the data on companies house arears.
A pattern of chronic late lodgement is not proof of anything. It is a question, and it is one document answers before you have read a single covenant.
What genuinely was not discoverable
Half the original hypothesis does survive, and it is the half worth keeping.
You could see how much the group owed. You could not see who ranked where. Australian security over a development book splits across two registers that do not reconcile: real property mortgages sit on state land titles registers, while general security agreements sit on the Personal Property Securities Register, which does not record mortgages over land at all. Assembling the security position across $3.08bn, several hundred titles and more than one state is an exercise in aggregation that no single lender is equipped to perform. The administrators did it because it is their job and because they had the group’s own records.
Nor does the market fill the gap. Australian commercial credit bureaux aggregate defaults, judgments, insolvency notices and credit enquiries, and they would have shown a searcher that a great many other lenders had recently been looking at this borrower. None of them aggregates outstanding balances owed by a commercial borrower across its lenders, because comprehensive credit reporting in Australia is a consumer regime. A bureau would have told you that others had looked. Only the lodged accounts would have told you how much they had lent.
The argument against all of this
La Trobe Financial has publicly disclosed exposure of approximately $38.1m across four residential loans and one development loan, a weighted-average loan-to-value of 62.5%, and presales covering 89% of the current loan balance. On those numbers, what the sponsor owes elsewhere is arguably second-order. A first-mortgage construction lender recovers from completion cost, presale contracts and site value. Group leverage is somebody else’s problem until it is not.
The counter is that it becomes your problem at the point the sponsor can no longer complete. The administrators told the court the group had 219 live projects, of which 45 were in construction, roughly 349 employees, and a projected cash burn of about $40m between September and December. Reporting puts presales at 1,198 against 14,873 planned homes. Whether a particular project finishes is a function of group solvency long before it is a function of that project’s loan-to-value.
What we do, and what it does not do
D3T searches the corporate family at underwriting and again through the life of a facility, on the borrowing entity and on the group above it. That is a description of scope, not a claim about outcomes. What it means is that our residual exposure is to active concealment rather than to information we did not go looking for. It does not make a facility safe. It changes what would have to happen for one to fail.
And it has a limit worth naming, because the same research exposes it. A search returns what has been lodged, and lodged accounts are stale by construction. The figures a lender could have read in June 2026 were as at the 30th of June 2025, twelve months old, and the group’s position had moved since. Nothing about searching harder closes that gap. It is closed, if it is closed at all, by what the borrower is contractually obliged to tell you between filings, how often you ask, and what happens when the answer changes.
Diligence has a boundary and the boundary is a convention. All of it looks like hard work, in particular when you end up finding what you expect.
D3T Capital LTD is an Appointed Representative of Capricorn Fund Managers Limited, which is authorised and regulated by the Financial Conduct Authority. D3T Capital LTD is incorporated in England and the registered office is at Unit 17 Orbital 25 Business Park, Dwight Road, Watford, WD18 9DA, United Kingdom.
The investment products and services of D3T Capital LTD are only available to professional clients and eligible counterparties. They are not available to retail clients.
This article is general market commentary. It does not constitute an offer, invitation or inducement to invest in any product or service, and it is not directed at retail clients.
The administration described is ongoing and no finding of wrongdoing has been made against any person or entity referred to. Figures are as reported on the public record at the date of publication and administrator estimates are preliminary.
