The Ombudsman Just Recovered $5.3 Million in Aged Care Wages — And Funding Wasn't the Problem
The Ombudsman Just Recovered $5.3 Million in Aged Care Wages — And Funding Wasn't the Problem
Thirteen of 22 providers investigated had underpaid staff. The leading cause was classification — which is not a funding issue, it's a data issue.
On Friday the Fair Work Ombudsman published the results of its targeted aged care compliance campaign. More than $5.3 million is being back-paid to nearly 3,600 direct care employees. Of 22 residential and home care providers investigated, 13 were found non-compliant with workplace laws in one or more areas, eight were compliant, and one remains under investigation.
Every one of the 13 non-compliant providers had underpaid employees.
The number that should interest a finance leader isn't the $5.3 million. It's the average: $1,478 per employee. The investigation doesn't make findings about intent, and it isn't the job of this post to guess at any. But an average of that size, spread across nearly 3,600 people, is the shape you get when a rule is applied slightly wrong, quietly, across a large number of pay runs — not the shape you get from a decision not to pay people.
The timing is the story
This landed four weeks after the final tranche of the aged care work value case took effect on 1 August 2026 — the last instalment of a multi-year process that lifted Nurses Award minimums by between 1.45% and 4.43% in this tranche alone, on top of increases already delivered in March 2025 and October 2025.
So the sequence for the sector over the past twelve months reads like this: absorb the largest deliberate increase in your own award rates in a decade, then have the regulator demonstrate that the harder problem was never funding the rates. It was applying them correctly.
Those are two entirely different problems and they sit with two entirely different people. Funding the increase is a Board and government conversation. Applying it correctly is a payroll configuration exercise that usually gets three weeks of attention around the effective date and none afterwards.
|
$1,478
Average underpayment per employee across the campaign. Small enough to survive a payslip check, large enough to matter at scale.
|
13 of 22
Providers investigated that were found non-compliant in one or more areas. All 13 had underpaid employees.
|
Four causes, and they are all the same kind of error
The FWO named the leading causes of breach explicitly: incorrect classification of employees, incorrect calculation of overtime, incorrect application of broken shifts, and incorrect payment of minimum engagement periods.
Read those four together and a pattern appears. Not one of them is a decision about how much to pay someone. Every one of them is a question about which rule applies to a particular person on a particular shift — and every one of them requires two systems to agree with each other. Classification requires the HR record and the payroll record to agree. Overtime, broken shifts and minimum engagements all require the roster or timesheet and the payroll record to agree.
Underpayment of this shape isn't a pricing failure. It's a reconciliation failure between systems that were never designed to be reconciled. That doesn't mean funding pressure is imaginary — it means a fully funded organisation can still generate every one of these four breaches, and several in the campaign did.
Every rate change is a re-classification event
Here is the mechanism that makes wage tranches dangerous rather than merely expensive.
When an award changes rates, most payroll systems handle it by updating the dollar value attached to each classification code. That part almost always works. What the update doesn't do is re-test whether each employee is sitting on the right code in the first place. And when a work value case also introduces a revised classification structure — as this one did — the codes themselves shift underneath the people.
The result is a compounding error. An employee sitting on a slightly wrong classification was underpaid by a small margin before the tranche. After the tranche, the gap between the code they're on and the code they should be on has widened, because both codes moved and they didn't move by the same amount. The error doesn't stay still. It grows every time the sector gets a pay rise.
That is why the periods either side of each rate change are the ones worth looking at, and why an audit done once, in the month of the increase, tells you very little about the eleven months that follow.
Where AI actually earns its place here
A retrospective rate-application check across three tranches and thousands of shifts is a genuine data-matching problem, and it is exactly the shape of work where AI-assisted tooling is useful rather than decorative.
The useful jobs are narrow and specific. Pulling classification, rate and hours out of three different systems — HR record, roster or timesheet, and payroll — and putting them into one comparable structure. Flagging every instance where the three don't agree. Identifying employees whose classification changed, or should have, at a tranche date. Surfacing the shifts that carry a broken-shift or minimum-engagement rule and testing whether the rule was actually applied. Producing the exception list.
The jobs it must not be given are just as specific. It does not decide what classification a person should hold — that is an interpretation of the award against a real job, and it belongs to a human who can defend it. It does not decide whether an exception is a genuine breach or a legitimate variation. And it does not sign anything.
The practical framing I'd offer: use it to shrink a population of 40,000 shifts down to a list of 300 that need a person to look at them. That is a real and defensible use. Asking it to tell you whether you're compliant is not.
Three things worth having on a page
1. A classification census. Every employee, their current award classification, the date it was last reviewed, and who reviewed it. Most organisations cannot produce this quickly, which is itself the finding.
2. A rate-change calendar with a look-back attached. Not just the dates the rates changed, but a scheduled check of the two pay runs either side of each one. Three tranches means six pay runs that deserve a second look.
3. An exception rate you actually track. The proportion of shifts each month where roster, timesheet and payroll don't agree. If nobody owns that number, nobody knows whether it's getting better or worse — and "we haven't had a complaint" is not a control.
The Ombudsman was clear about what it expects, and it went out of its way to commend the providers that had run their own audits and back-paid staff beyond the scope of the investigation. That is the posture being asked for. It is also, incidentally, much cheaper than the alternative — the FWO has already moved on to inspecting a further 30 employers on aged care cleaning and catering staff.
Could you produce your classification census this week?
If the answer involves three exports and a fortnight, the reconciliation problem is real and it isn't going to fix itself at the next tranche. PFL provides senior-level outsourced finance, management reporting, and AI automation for Australian NFP, NDIS, and SME organisations — including the unglamorous work of making payroll, roster and HR data agree with each other.
Talk to PFL →Fair Work Ombudsman — Aged care investigations recover $5.3 million for underpaid workers (28 August 2026)
Australian Ageing Agenda — $5.3m recouped for underpaid aged care workers
Fair Work Ombudsman — Aged Care Work Value Case: Changes to awards
Department of Health, Disability and Ageing — Better and fairer wages for aged care workers
Aged Care Insite — Aged care nurses receive final wage increase
Comments
Post a Comment