Hospitality Insolvencies Fell 15% — and That's the Wrong Number to Be Watching
Hospitality Insolvencies Fell 15% — and That's the Wrong Number to Be Watching
An insolvency count tells you what already happened. The data that would have warned you is sitting in your own receivables ledger.
CreditorWatch's June 2026 Business Risk Index, published 22 July, carries a genuinely good headline for a sector that hasn't had many. Insolvencies in Accommodation and Food Services fell 15 per cent across FY26. Overall insolvencies were down 3.9 per cent, sitting at around 0.5 per cent of all operating businesses as at 30 June.
CreditorWatch's own chief executive, Patrick Coghlan, described the picture rather more carefully: "The insolvency picture is improving, but the credit data tells us risk is quietly rebuilding."
That sentence is the entire post. And the reason it deserves a Friday deep dive rather than a news item is that it exposes something structural about how most finance functions monitor customer risk — including plenty that have nothing to do with hospitality.
What the June data actually says
Underneath the improving headline, three things moved the wrong way.
Trade payment defaults rose sharply in May 2026 and stayed elevated through June. ATO tax debts above the $100,000 disclosure threshold — the point at which the ATO reports a debt to credit bureaus — reached some of their highest readings since collections activity resumed after the pandemic. And the relationship between those debts and failure remained brutal: businesses carrying more than $100,000 in ATO debt recorded an average insolvency rate of 21.9 per cent over the past twelve months, against a national average of 0.7 per cent. Thirty-one times.
|
21.9% vs 0.7%
Insolvency rate over 12 months for businesses with ATO debts above $100,000, against the national average — 31 times higher. 35,361 businesses sat above that threshold at 30 June 2026.
|
10x
Increase in insolvency likelihood over the following 12 months from a single registered trade payment default. Multiple defaults raise it further.
|
And within hospitality specifically, the improvement is relative, not absolute: insolvency rates in the sector remain around three times the national average. A 15 per cent fall from a very high base is still a very high base.
The indicator problem
Here is the structural issue, and it applies well beyond hospitality.
An insolvency statistic is a record of businesses that have already failed. By the time a customer appears in it, your exposure is settled — you are a creditor in a process, and the only remaining question is recovery percentage. Insolvency data is useful for sector strategy and pricing risk into terms. It is almost useless as an operational warning, because it moves last.
Yet insolvency counts are what most finance teams actually watch, because they're what gets reported in the news and they arrive in a tidy monthly number. Meanwhile the indicators that move first — payment behaviour, tax debt, ageing drift — sit in systems the finance team already owns and generally doesn't monitor systematically.
The June data is a near-perfect illustration. If you had been watching insolvency counts through FY26, hospitality looked like it was recovering. If you had been watching payment defaults, you would have seen risk rebuilding from May. Same sector, same period, opposite conclusions — determined entirely by which indicator you chose.
Three indicators that move before the failure does
Payment behaviour change, not payment lateness. The signal isn't that a customer pays at 45 days. It's that a customer who reliably paid at 30 now pays at 45. Absolute lateness is often just how a business operates; deterioration is information. Most ageing reports show the former and hide the latter, because they present a snapshot rather than a trend per customer.
Partial payments and part-payment patterns. A customer who starts paying invoices in instalments, or settles the oldest invoice while letting newer ones age, is managing a cash constraint. That behaviour typically appears months before anything shows up in external credit data, and it's visible only in your own ledger.
The $100,000 ATO threshold. Worth understanding precisely because it's asymmetric information. Debts above that level get disclosed to credit bureaus; debts below it don't. So a clean credit report is not evidence of a healthy tax position — it's evidence of either a healthy tax position or a debt under $100,000. In a small-business customer base, that distinction matters.
The sole trader concentration nobody prices in
One figure in the June index deserves more attention than it got. Of the 35,361 businesses carrying ATO debts above $100,000, 53.8 per cent — 19,024 of them — are sole traders.
Sole traders run on tighter margins and thinner cash buffers, and they carry an additional feature worth understanding: there is no corporate veil. Their business and personal financial positions move together, which makes their failure mode faster and less orderly than a company's.
That matters differently depending on which side of the relationship a sole trader sits on. Where a sole trader is your customer — someone who owes you money — the concentration above is a straightforward credit-risk read: extending trade terms to unincorporated debtors carries more tail risk than extending them to companies. But this is not only a customer-side issue. Any organisation whose supplier or subcontractor base skews toward individual operators — allied health practices engaging independent clinicians, NDIS providers working with sole-trader support workers, community services organisations contracting individual specialists — carries a related but different exposure. There, the sole trader isn't a debtor; if they fail, the risk to you is service continuity, not an unpaid receivable. Worth knowing what proportion of your supplier spend goes to unincorporated entities, and treating that as a continuity question rather than a credit one.
Where AI turns this from a project into a routine
The reason most finance teams don't monitor payment-behaviour deterioration isn't that they don't understand its value. It's that doing it properly across a few hundred customers means tracking a per-customer payment-days trend month over month, and nobody has time to maintain that by hand.
That's an unglamorous, high-volume pattern-detection task — the kind AI handles well precisely because it requires consistency rather than judgement. The useful output isn't a risk score. It's a short list: the customers whose payment behaviour changed materially this month, ranked by exposure. Ten names, not a dashboard.
The discipline that makes it work is keeping the AI on the detection side and the human on the decision side. Flagging that a customer's average payment days moved from 31 to 52 is a computation. Deciding whether to tighten terms, pause supply, or pick up the phone is a commercial judgement that depends on relationship history the model doesn't have.
What this changes in the management reporting pack
Most management reporting packs carry a debtors ageing summary: current, 30, 60, 90+. It's been the standard for decades and it answers the wrong question. It tells you how much is late. It doesn't tell you who is getting worse.
Three additions worth making, none of which require new systems:
A deterioration list. Customers whose average payment days increased materially versus their own three-month baseline. Five to ten names with the exposure attached.
Concentration by counterparty type. What share of receivables sits with sole traders and unincorporated entities. A single number, tracked over time.
A named-exposure line. Your largest three receivable balances and how long each has been outstanding. Sector-level risk commentary is interesting; concentration risk is what actually causes a problem.
The honest test of a risk report is whether it would have changed a decision. An ageing summary rarely does, because everyone already knows the total. A list of customers who are quietly getting worse changes what you do this week.
Could you name the five customers whose payment behaviour got worse last month?
If not, your reporting is telling you what already happened rather than what's coming. PFL provides senior-level outsourced finance, management reporting, and AI automation for Australian NFP, NDIS, and SME organisations.
Talk to PFL →
Comments
Post a Comment