Australia set a 2028 closing date for its child sexual abuse redress scheme — 60 percent of 63,995 applications were still without an outcome in July 2025, and named institutions can decline to join

What is happening?

Australia opened the National Redress Scheme on 2018-07-01, acting on a recommendation of the Royal Commission that examined institutional responses to child sexual abuse between 2013 and 2017. The scheme was legislated as a ten year arrangement. It stops taking applications on 2027-06-30 and closes on 2028-06-30. The maximum payment is 150,000 Australian dollars, against the 200,000 the Royal Commission had recommended.

Seven years in, the arithmetic of the time remaining is the problem. As at 2025-07-04 the scheme had received 63,995 applications, and about 60 percent of them, roughly 38,000, had not yet received an outcome. Cumulative payments stood at 1.7 billion Australian dollars. For applications that had been completed, the average processing time as at 2025-06-30 was reported at 16 months.

Demand and completion are not moving at the same rate. Eleven months earlier, in August 2024, the scheme had received 44,342 applications, of which 36 percent had resulted in a payment and 41 percent had reached a determined outcome. Applications grew by about 44 percent across those eleven months and the resolved share did not keep pace. Further back, at 2021-06-19, the totals were 10,936 applications, 6,670 determined outcomes, 5,920 payments and 502.6 million Australian dollars paid.

The auditor reached the same conclusion in its own terms. Auditor-General Report No. 9 of 2025-26, tabled on 2025-11-24, assessed the administration of the scheme by the Department of Social Services as partly effective and said processing would have to accelerate for the scheme to meet its 2028 deadline. The same audit found that repeated recommendations to improve the consistency, transparency and fairness of redress decisions had not been fully implemented, including a consistent decision-making framework that had been promised in 2021.

The second half of the problem sits on the institution side. Joining the scheme is voluntary. An institution named in an application that does not join within six months of being notified faces one consequence: it becomes ineligible for Commonwealth grant funding, and its name is published. At least eight institutions had declined as at May 2021, three more were named in June 2021, and further institutions were named in December 2024 and October 2025. The naming has continued into the final years of the scheme, and no source opened here gives a cumulative total of institutions that have never joined.

Whose problem is this?

The body that sets the closing date is not the body that processes the applications, and neither of them is the institution that decides whether a given survivor has a scheme to apply to at all. The people for whom the deadline is decisive have no lever on any of the three.

Where does this problem end?

The boundary here separates a promise from its delivery. The Commonwealth accepted that survivors were owed redress and built an instrument to provide it, so what is measured in this document is not whether the obligation exists but how much of it will have been discharged when the instrument expires.

What is the state now, and what should it be?

Now

Needs a new measurementthe target state: no source opened here states how many of the outstanding applications the scheme intends to resolve before it closes, what average processing time it is working toward, or what share of named institutions it expects to have joined by 2028. The audit says processing must accelerate and does not name a rate. The statute sets an end date and, on this record, no throughput obligation to go with it. Every source describes the distance between the caseload and the deadline, and none of them states the standard that distance was supposed to meet.

How big is it?

Between about 38,400 and 63,995 people. The lower bound is the number of applications that had not received an outcome as at 2025-07-04, obtained by applying the reported share of about 60 percent to the 63,995 applications received by that date. The upper bound is that application total itself, being everyone who had entered the scheme and was therefore subject to its payment cap, its decision consistency problems and its 2028 closing date.

The two bounds answer different questions. The lower one counts the people who were waiting on a decision at a single moment. The upper one counts everyone the scheme has taken in over seven years, including those already paid, because the scheme they applied to is the one whose consistency the audit found wanting and whose closing date is now fixed.

Three limits sit on the interval and they do not all push the same way. The count is of applications rather than of people, and no opened source states whether a person can lodge more than one, so the figure could overstate people to an unknown degree. It also excludes survivors who never applied: the Royal Commission estimated about 60,000 people would be eligible for redress, on a basis that no opened source dates, and applications had already passed that figure by 2025, so no residual of eligible non-applicants can be derived from this record and the upper bound understates the eligible population rather than bounding it. Finally both bounds are dated 2025-07-04 while applications remain open until 2027-06-30, so the values at the close will be larger than either.

One quantity that would sharpen this is missing entirely. No opened source states how many applications name an institution that has not joined the scheme, which is the number that would separate people waiting on administrative capacity from people waiting on a decision that is not the scheme to make.

Under what conditions does it arise?

1. The obligation to participate does not exist, so the sanction has to substitute for it. An institution named by a survivor is invited, not required, to join. The single consequence of refusing is ineligibility for Commonwealth grant funding, which is a penalty only for institutions that seek Commonwealth grants. An institution funded by its own assets, donations or fees can absorb the naming and carry on, and the survivor who named it has no route through this scheme at all.

2. A closing date is not a rate of work. The statute fixes when the scheme ends and does not fix how much it must have done by then. Nothing in that arrangement converts a rising caseload into an obligation on anyone, which is why the audit could find the administration partly effective and recommend acceleration without any figure being attached to the word.

3. A deadline pulls applications forward while the caseload is already outstanding. Applications grew by about 44 percent across the eleven months to July 2025, and the last lodgement date of 2027-06-30 gives every remaining survivor a reason to apply in the final window. The demand peak and the capacity ceiling are scheduled to meet at the same point.

4. Waiting costs more for this cohort than the average delay statistic suggests. The abuse in question occurred in childhood and in many cases decades ago, so the applicants include people who are elderly or seriously ill. A 10,000 dollar advance payment for those applicants was introduced in 2021 in recognition of that, which addresses the consequence of the wait rather than the wait.

5. Discretion without a published framework produces variation that nobody can audit. A consistent decision-making framework was promised in 2021 and the audit of 2025 found the consistency, transparency and fairness recommendations not fully implemented. Where the standard for a decision is not written down, two applications with the same facts can produce different outcomes and neither applicant can demonstrate it.

6. Reviews accumulate faster than changes do. A second anniversary review reported 38 recommendations in 2021, an amendment act commenced three years later carrying five changes, an audit reported in 2025, and two more processes are due to report in 2026. Each of them restates the position and none of them alters the closing date.

What has been tried?

Every attempt on this list addresses either the speed of the scheme or the terms of a payment. None of them changes the condition that an institution named by a survivor may decline to take part, and none of them moves the date on which the whole arrangement expires.

What was found?

Why is it still unsolved?

Enforcement absent — the scheme exists, the obligation it was built to discharge is accepted, and nothing in its design compels either the institutions it names or the administration that runs it to finish in the time it has.

The participation rule is the clearest case. A survivor identifies the institution where the abuse occurred, the scheme invites that institution to join, and the institution may say no. The only lever attached to the refusal is eligibility for Commonwealth grants, which selects precisely the wrong target: it reaches institutions already inside the government funding relationship and misses the ones outside it. Naming works on institutions that care about being named. The record shows both that the lever works when it bites, with 27 institutions joining in early 2019 and the processable caseload nearly quadrupling in three months, and that it stops working once the remaining refusers are those it cannot reach, with fresh namings still occurring in 2024 and 2025.

The second reason is that the deadline runs in only one direction. Parliament fixed when the scheme ends and did not fix what it must have delivered by then, so an outstanding caseload does not breach anything. An audit can find the administration partly effective, recommend acceleration and name no rate, because there is no rate to measure against. That is why the position can deteriorate on the numbers, with applications up 44 percent in eleven months against a resolved share that did not follow, while every institution involved remains compliant with every requirement that binds it.

The third reason is that the passage of time removes the problem without solving it. Applications close in 2027 and the scheme closes in 2028. Applicants are people who were harmed as children, in many cases decades ago, so the cohort is ageing and some of them are seriously ill. Advocates state that applicants have died while waiting for an outcome, and no source opened here publishes a count of them, which means the most direct measure of the cost of delay is one that nobody produces. A backlog that expires with its scheme leaves no unfinished file to point at.

The fourth reason, which is what makes the pattern stable rather than merely slow, is that reviewing the scheme is the response that is always available. There has been a second anniversary review with 38 recommendations, an amendment act carrying five of the changes it wanted, a performance audit, a parliamentary inquiry and an eighth anniversary review, and the two most recent are due to report in the final years of an arrangement whose end date none of them can move. Each round produces a document that describes the gap accurately. None of them alters the two facts that produce it, which are that institutions may decline and that no one is obliged to finish.

What observation would mean it is solved?

Candidates — (a) the outstanding caseload is resolved before the scheme closes, with the share of applications awaiting an outcome falling toward zero by 2028-06-30 (b) participation ceases to be optional, so that every institution named in an application is answerable through the scheme or through a route that replaces it (c) the average time from application to outcome falls and stays down, measured against a published target.

(a) can be reached by the calendar rather than by the work. An application that is refused, withdrawn or never lodged also leaves the outstanding count, and the count necessarily goes to zero when the scheme closes because there is no longer a scheme for anything to be outstanding in. Read on its own this indicator cannot distinguish a caseload that was worked through from one that ran out of time, which is exactly the distinction that matters here. It would have to be read against the number of applications finalised with a payment, and against what happens to applications lodged shortly before the 2027 deadline.

(b) is the observation that would show the design had changed, because it addresses the condition rather than the symptom. It is also the least likely of the three, since compelling participation means reaching institutions that hold no Commonwealth funding relationship and, on this record, no proposal to do so appears in any source opened here. A weaker version would count as partial evidence: a published cumulative figure for institutions that have never joined, and a published figure for how many applications name one, would at least make the size of that gap visible for the first time.

(c) alone measures the wrong population. An average processing time is calculated over applications that have been completed, so it says nothing about the ones still waiting and improves whenever the simplest cases are finalised first. This record already shows how unstable the measure is, with 16 months and 14.1 months both attached to the same date. The three have to be read together, and (a) has to be read against the closing date that will produce it regardless.

What is it connected to?

Fills with researchcivil litigation for institutional abuse in the state courts and the limitation reforms that opened it, which is the alternative route a survivor may take instead of applying to the scheme, and the effect that route has on scheme application numbers · redress arrangements for institutional abuse in Ireland, New Zealand and Scotland as a comparison for the voluntary participation design · implementation of the wider Royal Commission recommendations on safeguarding, of which redress was one part · the implementation record of the Disability Royal Commission as a second case of recommendations awaiting action. Relation type and evidence grade were not confirmed in this round.

What these sources do not say

• How many applicants have died while waiting. Advocates state that survivors have died before receiving an outcome, and no source opened here carries a count. This looks like an absence at the source rather than an unsearched question, in that the scheme publishes application, outcome and payment totals and does not appear to publish this one.
• How many institutions have never joined. The public naming happens in batches, and the batches are reported individually — at least eight, then three, then further institutions in December 2024 and October 2025. No opened source adds them up, and the list page that would give the current total did not open, so the cumulative figure is unavailable rather than absent.
• How many applications name an institution that has not joined. This is the number that would say how much of the outstanding caseload is waiting on administrative capacity and how much is waiting on a decision the scheme does not control. No opened source states it, and the 2019 figures for processable applications are the only indication anywhere in this record that the two are different quantities.
• What the average payment is. A figure of 89,408 Australian dollars appeared in search results during research, attributed to a scheme performance publication for June 2025. That publication did not open, no other opened source carries the figure, and it is therefore recorded here as untraced and is not used as data anywhere in this document.
• Which processing time figure is the published one. The opened legal journal report gives 16 months as at 2025-06-30. Research recorded 14.1 months for the same date in search results whose source did not open. Neither figure states its counting basis, and no opened source acknowledges the other, so the disagreement is left visible rather than resolved.
• What happens to an application that is unresolved when the scheme closes. No source opened here describes the treatment of applications lodged before the 2027 deadline but undecided at the 2028 closing date, nor whether any residual arrangement exists after that date.
• Why processing has not accelerated. The audit records that it must and that the consistency recommendations were not fully implemented. Nothing opened here carries an explanation from the administering department of what the constraint is, whether it is staffing, institutional response times, evidentiary requirements or something else.
• Almost the entire primary record, because it did not open. The audit report and its landing page, the parliamentary inquiry page, the scheme performance publication, the list of institutions that have not joined, the second anniversary review material, a ministerial release on participating institutions and a legal note on the sanctions all failed to open during research. Eight rows below are blank for that reason. The audit findings that carry a substantial part of this document reach it through a single report of the audit rather than from the audit itself.

See the evidence

No primary scheme document opened. The two departmental media releases are the only government records here read at their own address, and they carry the participation rule, the six month period, the consequences of not joining, the 2019 effect of institutions joining on the processable caseload, and the scheme totals as at 2021-06-19, which is why those rows are the high-grade items in the findings table apart from the absences. Every page of the scheme itself and of the audit office attempted in research timed out, and the parliamentary inquiry page returned a refusal, so the two documents this problem most depends on — the performance audit of November 2025 and the current list of institutions that have not joined — reach this document only through reporting of them or not at all. Eight rows above are blank with the reason recorded rather than filled with a substitute link. Where the record does not settle a value the disagreement is left visible rather than resolved: the average processing time is given as 16 months by the opened source and as 14.1 months for the same date in search results that could not be traced, and an average payment of 89,408 Australian dollars appears only in an untraced summary and is used nowhere in this document. The cumulative number of institutions that have never joined is absent from every source opened, and that absence is recorded as a finding rather than filled. This is a Path A output, so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range 38,397–63,995 As of 2025-07-04

Sensitivity The interval spans 38,397 to 63,995, and its width is the difference between people waiting on a decision at one moment and everyone the scheme has taken in across seven years. Both ends are counts of applications rather than of people. No source opened here states whether one person can lodge more than one application, or how many applications were withdrawn or found ineligible, so both bounds could overstate people by an unknown amount. In the other direction the interval excludes survivors who never applied, and no residual can be derived for them: the Royal Commission estimate of about 60,000 people eligible for redress was already exceeded by the 63,995 applications received, so subtracting one from the other would produce a negative figure rather than a count of non-applicants, and the eligibility estimate carries no stated base year in any source opened here. Both bounds are dated 2025-07-04 while applications remain open until 2027-06-30, so the values at the close of the scheme will be larger than either. One quantity that would sharpen the low end is missing entirely: no source opened here states how many applications name an institution that has not joined the scheme, which is the number that would separate people waiting on administrative capacity from people waiting on an institution that has declined to take part and cannot be compelled to.

Regional breakdown Every figure opened for this document is national. The scheme is a Commonwealth arrangement covering the whole of Australia, and no opened source breaks applications, outcomes, payments or processing times down by state or territory, nor states where applicants live or where the institutions they name are located. The performance publication that would be the most likely carrier of such a breakdown did not open during research. Distributing the national totals across states in proportion to population would be invention, not measurement, and is prohibited here.