NDIA Accountability Project

Analysis of NDIS payment architecture and regulatory oversight

‹ PrevPage 1 of 11 · Source p. 1Next ›

SUBMISSION TO THE JOINT COMMITTEE OF PUBLIC ACCOUNTS AND AUDIT

Inquiry into the administration of the National Disability Insurance Scheme

MAGNA NDIS Consultant

Submitted by: Loma Naser Founder & Principal Consultant Date: 28 April 2026 Contact: Loma@ndisconsultant.au

MAGNA — JCPAA Submission — 28 April 2026

TERMS OF REFERENCE — SCOPE CONFIRMATION

  1. This submission is made in response to the Committee’s inquiry into the administration of the National Disability Insurance Scheme (NDIS), specifically in relation to the six Australian National Audit Office (ANAO) reports identified in the Committee’s Terms of Reference:
  • ANAO Report No. 22 of 2024–25: Audits of the Financial Statements of Australian Government Entities
  • ANAO Report No. 25 of 2024–25: Audits of the Performance Statements of Commonwealth Entities
  • ANAO Report No. 39 of 2024–25: Interim Report on Key Financial Controls of Major Australian Government Entities
  • ANAO Report No. 41 of 2024–25: Effectiveness of the Board of the National Disability Insurance Agency
  • ANAO Report No. 48 of 2024–25: National Disability Insurance Agency’s Management of Claimant Compliance
  • ANAO Report No. 2 of 2025–26: Effectiveness of the NDIS Quality and Safeguards Commission’s Regulatory Functions
  1. Every factual claim in this submission is sourced from one or more of the reports listed above, or from the NDIA’s own published quarterly data (Supplement E National, available on the NDIA website). No claim relies on modelled or unverified external data. Citation format throughout: [Report No., paragraph reference].

EXECUTIVE SUMMARY

  1. Three findings emerge from reading the six ANAO reports before the Committee together.

  2. First: the NDIS payment architecture was, in the NDIA’s own words recorded by the Auditor-General, “catastrophically weak.” The NDIA’s September 2023 Crack Down on Fraud business case projected that losses from fraud, non-compliance and payment error could reach $3.6–6.0 billion per year by 2027–28 without remediation. The Commonwealth committed more than $495 million to repair the architecture between 2021–22 and 2028–29. As at the conclusion of ANAO Report No. 48 (tabled June 2025), the repair was not yet complete.

  3. Second: the NDIS Quality and Safeguards Commission does not have visibility of the payment channels it is responsible for overseeing. In Q4 2024–25, 94 per cent of active providers were unregistered. Plan managers paid $7.3 billion in that quarter alone, of which 42 per cent went to those unregistered providers — a proportion that has risen since 2021–22. The Commission’s own April 2025 statement to the ANAO confirmed it “does not have visibility of payment arrangements for self-managed participants” [Report No. 2, paragraph 1.10].

  4. Third: the NDIA Board has not directed strategic reporting on reforms sufficient to assure itself that the NDIA will fulfil the Government’s commitments to moderate growth in scheme expenses [Report No. 41, paragraph 16].

  5. On 22 April 2026 — one day before the Committee’s public hearing — the Minister announced a reform package proposing to reduce total participant numbers from

2

MAGNA — JCPAA Submission — 28 April 2026

739,414 (at 30 June 2025) to approximately 600,000 by 2030. The modelling underpinning that target, the pathway, and the cost-transfer implications have not been published.

  1. This submission places four questions before the Committee — one directed to the NDIA (payment architecture), one directed to the NDIA (participant modelling), one directed to the NDIS Commission (oversight gaps), and one directed to the ANAO (proportionality of the reform package relative to audit findings). The evidential basis for each question is set out in Sections 1–4 below.

SECTION 1 — THE PAYMENT ARCHITECTURE GAP

Finding

  1. ANAO Report No. 48 of 2024–25 assessed the NDIA’s management of claimant compliance and found it “partly effective.” The report discloses the following findings from the NDIA’s own September 2023 Crack Down on Fraud business case:
  • Prior to reform, 98 per cent of NDIS claims were immediately paid with no automated pre-payment validation against legislative, regulatory or administrative rules or known risk flags [Report No. 48, paragraph 2.20].
  • Only 0.02 per cent of total NDIS claims were reviewed pre-payment per year; 0.04 per cent post-payment per year [Report No. 48, paragraph 2.20].
  • The NDIA projected that losses from fraud, non-compliance and payment error could increase to $3.6–6.0 billion per year by 2027–28 without remediation [Report No. 48, paragraph 3.23].

  1. Manual pre-payment review, introduced as an interim measure, now covers 0.4 per cent of NDIS outlays by dollar value. Of the claims reviewed, 53.7 per cent were cancelled for non-compliance [Report No. 48, summary box]. This means that nearly 99.6 per cent of scheme expenditure by value continues to be processed without pre-payment review.

  2. The NDIA’s own integrity campaign on plan-manager reimbursements (October– December 2023) reviewed $836,000 across 44 providers. Of the claims reviewed, 18.9 per cent self-corrected or were cancelled and a further 38.3 per cent were referred for further investigation — 57.2 per cent of reviewed claims were flagged [Report No. 48, Table 2.2]. The NDIA recorded that plan-manager reimbursements were being used “to withdraw cash from the NDIS in non-compliant or fraudulent way.”

  3. The Commonwealth has committed more than $495 million to addressing NDIS fraud and non-compliance between 2021–22 and 2028–29. The Crack Down on Fraud Tranche 2 status was rated ‘amber’ as at April 2025, with two June 2025 milestones at risk due to procurement delays. The target completion date for Tranche 2 was December 2025 [Report No. 48].

  4. Reports No. 22, 25 and 39 collectively identify the NDIA as a high-risk engagement across financial statements, performance statements and key financial controls audits. Report No. 39 recorded four unresolved moderate findings relating to IT and access controls as at the interim audit.

3

MAGNA — JCPAA Submission — 28 April 2026

Government response

  1. The Minister’s 22 April 2026 reform package proposes reducing participant numbers as a primary mechanism for moderating scheme expenditure. In his National Press Club address of 22 April 2026, the Minister also announced that the NDIA “has no visibility of evidence for 90 per cent of claims that are made by plan managers or by providers directly—around 600,000 claims every day without supporting evidence”, and that enrolling providers in a digital payments system would enable the NDIA to “see evidence from every single provider and ensure that they’re paid directly”. The reform package has not, on the published record, quantified: (a) the projected savings from this digital payments measure; (b) the projected savings from the $495 million Crack Down on Fraud commitment already funded; or (c) the relationship between the two — including whether the proposed digital payments system is in addition to, or absorbs elements of, the Tranche 2 IT uplift that was scheduled for completion by December 2025.

Question 1 — for the NDIA

  1. The Committee’s public hearing of 23 April 2026 having concluded, the Committee may wish to put the following question to the NDIA in writing on notice:

Can the NDIA provide a disaggregated savings estimate showing, separately: (a) projected savings from completing the payment architecture repair already funded under the $495 million commitment; and (b) projected savings from reducing participant numbers to approximately 600,000 by 2030 — and does the NDIA contend that (b) represents a more efficient and economical approach to scheme sustainability than completing (a)?

SECTION 2 — THE MODELLING GAP

Finding

  1. The Minister’s reform package proposes reducing NDIS participant numbers from 739,414 (at 30 June 2025) to approximately 600,000 by 2030. No modelling has been published showing: (a) the annual exit trajectory required to achieve this target; (b) the assumptions about future entries; or (c) the programs to which exiting participants would transition and the fiscal cost of those transitions.

  2. MAGNA has applied financial modelling to the NDIA’s own published quarterly data (Supplement E National, Q2 2025–26) to estimate the scale of exits implied by the announced target. The model highlights the following:

  • Current active participants (30 June 2025, per NDIA dashboard): 739,414
  • Gross entries per quarter (Q2 2025–26, Supplement E, Tables E.1 and E.2): approximately 29,341 per quarter, or 117,364 per year
  • Projected entries over four years (2026–2030): approximately 469,456
  • Required exits to reach 600,000 by 2030: 739,414 + 469,456 − 600,000 = 608,870 total exits
  • Current exit trend (NDIA Supplement E, annualised from Q2 2025–26): approximately 154,960 over four years
  • Exits above current trend required: 608,870 − 154,960 = approximately 454,000 excess exits above natural attrition

4

MAGNA — JCPAA Submission — 28 April 2026

  • This represents approximately 4.1 times the current four-year exit trend

  1. Put differently: achieving the announced target requires exiting approximately 454,000 participants above the current attrition rate over four years — in addition to everyone who would leave the scheme under current settings. This figure is derived entirely from the NDIA’s own published data using arithmetic assumptions.

  2. The NDIA’s own quarterly data (Supplement E National) shows that the developmental delay exit rate has already accelerated sharply: from 7,270 exits in the year ending December 2023 to 19,412 in the year ending December 2024 to 40,900 in the year ending December 2025. This acceleration occurred before any legislative change to eligibility criteria.

  3. ANAO Report No. 41 found that “the Board has not directed strategic reporting on reforms to assure itself that the NDIA will fulfil the Australian Government’s commitments to moderate growth in NDIS expenses” [paragraph 16]. This finding means the NDIA’s accountable authority does not, on the ANAO’s assessment, have adequate visibility of whether the reform pathway will deliver the committed fiscal outcomes.

Government response

  1. No modelling has been published. The Minister’s announcement of 22 April 2026 stated a target of approximately 600,000 participants by 2030. The pathway — in terms of which cohorts would be exited, under what criteria, and at what rate — has not been disclosed.

Question 2 — for the NDIA

  1. The Committee’s public hearing of 23 April 2026 having concluded, the Committee may wish to put the following question to the NDIA in writing on notice:

Can the NDIA publish the modelling underlying the proposal to reduce participant numbers from 739,414 to approximately 600,000 by 2030, including: (a) the assumed annual entry and exit rates; (b) the cohorts from which exits are projected; (c) the programs into which exiting participants are expected to transition; and (d) the estimated fiscal cost to the Commonwealth and to State and Territory governments of those transitions — and has the NDIA Board received and considered that modelling in accordance with its obligations under the NDIS Act?

SECTION 3 — THE OVERSIGHT GAP

Finding

  1. ANAO Report No. 2 of 2025–26 assessed the NDIS Quality and Safeguards Commission’s regulatory functions and found them “partly effective.” The report identifies three overlapping gaps in the Commission’s oversight of the market it regulates.

  2. Provider registration gap. In Q4 2024–25, 94 per cent of active NDIS providers were unregistered (254,018 unregistered providers against 16,363 registered) [Report No. 2, paragraph 9]. Unregistered providers may be used only by self-managed and plan-

5

MAGNA — JCPAA Submission — 28 April 2026

managed participants. Of the 254,018 unregistered providers, 181,938 received plan- managed payments.

  1. Payment visibility gap. The Commission does not have quality assurance processes for compliance activities, and regulatory decision-making is not guided by a risk-based strategy [Report No. 2, paragraphs 8–9]. More fundamentally, the Commission’s own April 2025 statement to the ANAO confirmed: “The NDIS Commission does not have visibility of payment arrangements for self-managed participants” [Report No. 2, paragraph 1.10].

  2. Scale of unmonitored flows. Plan managers paid $7.3 billion in Q4 2024–25 alone, of which 42 per cent — approximately $3.1 billion in that quarter — went to unregistered providers. The proportion of plan-managed payments going to unregistered providers has risen from 39 per cent in 2021–22 to 42 per cent in 2024–25 [Report No. 2, Table 1.1]. The Commission has no visibility of self-managed payment flows at all.

26A. Plan-manager float. Within plan-managed flows, the Architecture publishes no data on the period between NDIA payment to a plan manager and the plan manager’s onward payment to the registered or unregistered provider. Money paid by the NDIA may be retained, disputed, partially paid, paid late or — in cases of plan-manager or provider insolvency — never reach the provider at all, for reasons that are not necessarily related to fraud. Neither the NDIA nor the Commission publishes the scale, destination or duration of those holdings.

  1. Trust deficit. In the 2024 NDIS Commission stakeholder survey of 10,949 respondents, only 24 per cent trusted the Commission “a lot” or “completely” [Report No. 2, paragraph 2.60].

  2. ANAO Report No. 48 provides additional context. The NDIA’s own integrity campaign on plan-manager reimbursements (October–December 2023) found that 57.2 per cent of claims reviewed were flagged as self-corrected, cancelled or referred for investigation [Table 2.2]. The Commission was not the primary vehicle for detecting this non- compliance — the NDIA conducted its own campaign because the Commission lacked visibility of the relevant payment channel.

Government response

  1. The Commission has provided no public update since ANAO Report No. 2 was published (3 September 2025) on the steps it has taken to obtain visibility of self-managed payment arrangements or to implement the ten recommendations made by the ANAO in Report No. 2.

Question 3 — for the NDIS Commission

  1. The Committee’s public hearing of 23 April 2026 having concluded, the Committee may wish to put the following question to the NDIS Commission in writing on notice:

What specific steps has the Commission taken since September 2025 to obtain visibility of payment flows across all NDIS providers — registered and unregistered, plan-managed and self-managed — and how will the Commission’s oversight be integrated with the digital payments system the Minister announced on 22 April 2026, which is intended to require every provider, regardless of registration status, to enroll in order to receive payment?

6

MAGNA — JCPAA Submission — 28 April 2026

For the avoidance of doubt, MAGNA does not propose that all providers be required to register; rather, that all NDIS payments — whatever the provider’s registration status — should be visible to both the NDIA and the Commission.

SECTION 4 — THE PROPORTIONALITY QUESTION

Finding

  1. Reports No. 22, 25, 39, 41, 48 (2024–25) and Report No. 2 (2025–26) taken together describe a scheme whose primary accountability vulnerabilities sit in its payment architecture and its regulatory oversight framework — not in the number of participants per se.

  2. The NDIA’s own business case (September 2023) identified the payment architecture as “catastrophically weak” and projected $3.6–6.0 billion in annual losses by 2027–28 [Report No. 48, paragraph 3.23]. The Commonwealth has appropriated $495 million to repair that architecture. The repair is not yet complete.

  3. The ANAO has identified, across multiple reports, that: (a) the NDIA Board has not assured itself that reforms will deliver committed fiscal outcomes [Report No. 41, paragraph 16]; (b) the Commission does not have visibility of the largest unregistered payment flows [Report No. 2, paragraph 1.10]; and (c) the NDIA’s management of claimant compliance remains “partly effective” despite significant public investment [Report No. 48].

  4. The Government’s announced reform — reducing participants by approximately 139,000 by 2030 — is presented as a fiscal sustainability measure. The efficiency of that measure relative to completing the already-funded payment architecture repair has not, on the published record, been assessed or disclosed.

  5. The Thriving Kids proposal, which would redirect developmental delay and autism early- intervention participants from the NDIS to State-delivered services, would involve fiscal transfers of a scale that the Government’s own fiscal analysis has not quantified on the published record. MAGNA’s modelling, based on NDIA and published literature data, estimates the whole-of-government cost of this transition at $86.2 billion over ten years — compared with continuing current NDIS early-intervention arrangements.

Government response

  1. The Minister’s 22 April 2026 announcement framed the reform package as a fiscal sustainability measure and included a proposed digital payments system intended to give the NDIA visibility of every claim. The Government has not published: (a) the projected savings attributable to the proposed digital payments measure; (b) the projected savings from payment integrity measures already underway under the $495 million commitment, or how those savings interact with the savings from the digital payments measure; (c) the exit-rate assumptions embedded in the 600,000 target; or (d) the cost to other Commonwealth and State programs of the proposed participant reductions.

7

MAGNA — JCPAA Submission — 28 April 2026

Question 4 — for the ANAO

  1. The Committee’s public hearing of 23 April 2026 having concluded, the Committee may wish to put the following question to the Auditor-General in writing on notice:

In the ANAO’s assessment, does the Government’s announced reform package — taken as a whole, including the proposed digital payments system, the proposed reduction in participant numbers to approximately 600,000 by 2030, and the redirection of developmental delay and autism early-intervention participants to State-delivered services under Thriving Kids — represent a proportionate, efficient and economical response to the vulnerabilities identified across the six ANAO reports before this Committee; and, if so, has the ANAO seen modelling that demonstrates this proportionality, or does such modelling remain unpublished?

CONCLUSION

  1. The evidence before the Committee establishes four gaps between what the ANAO has found and what the Government has published:
  • The payment architecture repair is funded but not complete, and its projected savings have not been separated from the savings attributed to participant reduction.
  • The modelling underpinning the 600,000 participant target has not been published, including the exit rate assumptions and the cost-transfer consequences for other programs.
  • The NDIS Commission does not have visibility of the largest unregistered payment flows, and has provided no update on steps taken to remedy this since September
  • The NDIA Board has not directed strategic reporting on reforms sufficient to assure itself that committed fiscal outcomes will be achieved.
  1. MAGNA does not advocate for a particular participant number outcome. MAGNA’s submission is that the pathway to scheme sustainability — whatever its form — should be demonstrably proportionate to the vulnerabilities the ANAO has identified, demonstrably more efficient than completing the repair already funded, and should disclose the cost transfers to other government programs that any participant reduction entails. None of these conditions has been satisfied on the published record. This report is not financial advice nor legal advice.

Loma Naser Founder & Principal Consultant, MAGNA NDIS Consultant 28 April 2026 Loma@ndisconsultant.au

8

MAGNA — JCPAA Submission — 28 April 2026

ATTACHMENTS

Attachment A — MAGNA Fiscal Brief: Thriving Kids Cost Analysis (Summary)

Summary of MAGNA’s whole-of-government cost modelling for the Thriving Kids early- intervention proposal versus continued NDIS early-intervention arrangements, over a ten-year horizon. The full methodology, source-file appendix and bibliography are available upon request from MAGNA NDIS Consultant, April 2026.

9

ATTACHMENT A

MAGNA Fiscal Brief Summary

Thriving Kids — Whole-of-Government Cost Analysis

Supplementary submission to the Joint Committee of Public Accounts and Audit Inquiry into the administration of the National Disability Insurance Scheme

Prepared by Loma Naser MAGNA NDIS Consultant 28 April 2026

Purpose

This Attachment supports paragraph 35 of the accompanying submission. It sets out MAGNA’s estimate that the Thriving Kids proposal — to redirect children with developmental delay and autism (early intervention, NDIS Act section 25) from the NDIS to State-delivered services — would impose whole-of-government costs of approximately $86 billion over ten years above the cost of maintaining the existing NDIS early-intervention pathway. Cohort and unit-cost inputs are drawn from published government sources (NDIA Participant Dashboards, Productivity Commission Report on Government Services 2026, AIHW, Department of Education, Services Australia). Probability weights are MAGNA’s modelling, anchored at the conservative end of peer- reviewed literature ranges. The full methodology, source-file appendix and bibliography are available on request from MAGNA NDIS Consultant.

Headline findings

Finding 1 — Children with developmental delay are already being removed from NDIS individualised support faster than they are being added. Implied annual exits have roughly doubled each calendar year: 7,270 (year ending December 2023), 19,412 (year ending December 2024), 40,900 (year ending December 2025). No legislation has been passed; this is occurring under existing administrative rules.

Finding 2 — The Government’s published fiscal case for Thriving Kids is a saving on a single NDIS Budget line of approximately $1 billion per year. Applying the Government’s own published per-child unit costs across the systems that absorb unmet early-intervention need (health, education, child protection, youth justice, income support), MAGNA’s modelling projects a whole-of-government cost increase of approximately $86.2 billion over ten years (2026–2035). The Government has published no whole-of-government cost projection.

Finding 3 — In a national survey of 1,535 families and young people with disability (CYDA, August–September 2025): 76 per cent worried, 51 per cent scared or confused, 79 per cent reported the proposed implementation timeline as inadequate or unsafe.

Ten-year cost projection (2026–2035, nominal)

Scenario 10-year cumulative
Maintained NDIS early intervention (status quo) $65.6 billion
Thriving Kids (total cost across all systems) $151.7 billion
Difference — additional whole-of-government cost under Thriving Kids +$86.2 billion

How the figure is built

Each system-level cost is the product of three components: the Government-published unit cost per user of that system, a probability weight converting that to a per-cohort-child figure, and the published annual growth rate for that system. The cohort is 135,722 children aged 0–8 with autism or developmental delay at 30 June 2025 (NDIA Participant Dashboards, Table 8). Unit costs and growth rates are all verified from published Government sources. Probability weights are either verified from a single published linkage study (BOCSAR Bulletin No. 254, 2023, for the 5× youth- justice over-representation) or set at the conservative end of the published peer-reviewed range (health, education, OOHC, adult DSP/Carer Payment).

Stress test: simultaneously halving every modelled probability weight in the analysis — and additionally halving the verified BOCSAR youth-justice over-representation from 5× to 2.5× — still leaves the whole-of-government cost transfer above $50 billion over the decade. The Government’s NDIS-line saving over the same ten years is under $11 billion.

Four premises are each established on the public record:

  • (a) Parents do not want it (CYDA n = 1,535: 76% / 51% / 79%).
  • (b) The peer-reviewed evidence base does not support pooled, site-based delivery at $2,063 per child per year as equivalent to individualised section 25 supports.
  • (c) Children are already being removed under administrative rules alone — implied developmental delay exits rising from 7,270 to 19,412 to 40,900 across three consecutive calendar years, under legislation that has not been introduced to Parliament.
  • (d) The Government has published no whole-of-government cost projection. On the Government’s own published unit costs, the reform increases whole-of-government cost by approximately $86 billion over ten years above maintaining current arrangements.

Full file available on request

The complete MAGNA Thriving Kids cost analysis — including the full three-layer methodology (cohort definition; published unit costs by system; probability weights with row-level peer-reviewed evidence; published growth rates by system), the full ten-year scenario tables, the source-file and verification appendix tracing every figure to a named published file, table and row, and the full{bibliography — is available upon request from MAGNA NDIS Consultant.

Contact: Loma@ndisconsultant.au.

Prepared by Loma Naser, Founder & Principal Consultant, MAGNA NDIS Consultant • April 2026