A backlog cleared is not a backlog solved
A royal commission found 42,000 unallocated veterans' claims and set a deadline. DVA cleared them, early. Twenty-eight months later there were 32,925 unallocated claims again. The surge worked and the system did not change.
Steps this would change
Rationale
This is the most important cautionary finding in this database, and it applies to almost every other page on this site.
The sequence. The Royal Commission into Defence and Veteran Suicide found 41,799 unallocated claims as at May 2022, and said in terms that long waits for entitlements damage veterans' mental health and in some cases contribute to suicide and suicidality. It made eliminating the backlog an urgent recommendation with a 31 March 2024 deadline. DVA cleared the last unallocated claim at the end of February 2024, ahead of the deadline.
That was a genuine achievement by the people who did it and should be recorded as one.
As at 30 June 2026: 103,074 claims on hand, of which 32,925 unallocated. Average time to process an MRCA permanent impairment claim across all claims on hand had risen to 364 days from 315 the year before.
Nothing here suggests bad faith. It is arithmetic. A queue empties when clearance exceeds arrival, and refills when it does not. Surging staff raises clearance temporarily. It does nothing to arrival, nothing to the work each claim requires, and nothing to the share of claims that arrive incomplete or that generate disputes.
What changing the inflow would actually mean:
- Take the evidence from Defence rather than from the veteran. Service and service-medical records are already held by the Commonwealth. See Stop making veterans carry records between two Commonwealth agencies.
- Automate determinations where a Statement of Principles is clearly satisfied. The Statements of Principles already reduce the central question to published factors. Where a service history plainly meets them, the determination is an application of a known rule, and rules that can be stated can be applied consistently at volume.
- Publish the unallocated count monthly. It is the leading indicator. The 2026 figure appears in annual reporting; a queue that is only visible once a year can regrow for eleven months before anyone outside the department knows.
- Measure the department on time-to-determination for recently received claims and on the unallocated count, not on backlog milestones. A milestone is met once. A rate has to be held.
Who acts: the Commonwealth alone.
What is being traded away
Automation in this domain has an unusually bad worst case. A wrongly refused veteran is exactly the person the royal commission was concerned about, and an automated refusal at volume would be catastrophic in a way an automated approval would not. Robodebt is the obvious and correct comparison, and anyone proposing automated decision-making in a Commonwealth benefits system has to answer it before saying anything else. The only defensible version is automation that can grant but never refuse — clearing the clear cases into the yes pile and routing everything else to a human, so that the machine's failure mode is a delayed approval rather than a wrongful denial.
Faster inflow processing may increase inflow. A system known to pay promptly attracts claims that would not otherwise be made, including entirely legitimate ones from veterans who had given up. That is a good outcome and it is also a budget consequence, and pretending otherwise would make this proposal easy to dismiss by anyone holding the appropriation.
The published numbers do not decompose. DVA publishes claims on hand, unallocated counts and processing times. It does not publish arrival rates, rework rates, or the share of claims requiring further evidence — which are exactly the numbers needed to know whether inflow can be changed at all. This proposal is therefore an argument from the shape of the queue, not from a diagnosis of it, and it should be treated as weaker than the confidence of its title suggests.
And the general warning, which is the real point of this entry. Several reforms on this site would produce a one-off improvement in a queue. Hospital gateway reviews, the tribunal, the aged care assignment queue — each could be surged down and each would refill on the same arithmetic. DVA is the worked example of what happens when a backlog is treated as an event rather than a rate, and it should be read alongside every other proposal here.
Precedent
The 2024 clearance is its own precedent, in both directions: it proves the department can move at volume when resourced, and it proves that moving at volume does not alter the arrival rate.
Sources
- Royal Commission into Defence and Veteran Suicide, interim report 2022: 41,799 unallocated claims as at May 2022; urgent recommendation to eliminate the backlog by 31 March 2024
- DVA, confirmation the unallocated backlog was cleared at the end of February 2024
- DVA, published claim processing times as at 30 June 2026: 103,074 claims on hand, 32,925 unallocated; MRCA permanent impairment average time to process 364 days, up from 315