NDIA Accountability Project

Integrity measures and ethical oversight within the National Disability Insurance Scheme

‹ PrevPage 1 of 26 · Source p. 1Next ›

Integrity of the National Disability Insurance Scheme Inquiry

To: Committee Secretary, Joint Standing Committee on the NationaL Disabilitv Insurance Schemc

Re: Integritv of the Natioual Disability Insurancc Schcmc Inquirc

Dear Committec Secrctary,

I wclcome thjs inquiry anu its recognition uf impacts un participants.

This submission is made in mv personal capacity und as an individual but takes a systemic approach tu the analysis of integri ty issues and associated recommendations.

It addresses the terms of reference thrugh a more indepth exploration of sume intcgri tv themes introduced in Submission 7 to this Committce’s Annual Report No. 2, with particular focus ou recommended amendments, along witb numerous other matters (some only recently in tbe public domain) which fall within the scope of this inquiry.

Where possible, I have also touched on themcs in the terms of rcference I proposed for a longer integrity inquiry in Submissiun 7. These largely fali under imputs on particiapants or adequacy of policies, but doing these issues justice will require a supplementary submitton in due course.

As I said in Submiton 7, I ask the Committee to remember that sustainability, flexibility and faithfully upholding our human rights can coexist with integrity.

For the purpose of this inquiry, I alsu ask the Commtttee keep one word in mind in its examination of all evidence received, and in making your recommendatious: Dignity.

Lastly, I hope thc Committec wilI kecp asking itself onc key question in considering every angle uf effectiveness, und every propused reform: Does this measure haxve the potential tu discourage ur prevent a participant from accessing support they need, anu to whicb they are entitled?

I would welcumc the opportunity to discuss any of the issties I have raised or solutions propossd at au hearing, and wouid particularly value tbe chance to work through some of the unresolved integritv questions and concerns I put to the Agency during my two years un the Participant Reference Group, anu the opportunities I see for addressing them.

Yours sincerely,

Cat Walker

Contents

Recommendations ……………………………………………………………………………………………………….. 3 Approach to terms of reference ……………………………………………………………………………………….. 5 The nature and extent of non-compliance, including fraud and sharp practices, in the National Disability Insurance Scheme …………………………………………………………………………………………… 5 The impacts of non-compliance on NDIS participants and their families …………………………………….. 5 Effectiveness and adequacy of successive government policies to improve scheme integrity, safeguard participants, and tackle non-compliance ……………………………………………………………………………. 5 Any legislative or other reforms required to strengthen scheme integrity …………………………………….. 5 How we talk about integrity of the NDIS matters …………………………………………………………………… 6 Addressing the impact on participants starts with grounding integrity discourse in Australia’s international obligations, recommendations of the Robodebt Royal Commission, and holding media accountable to both. …………………………………………………………………………………………………….. 6 How we approach participant-directed integrity measures requires the highest standards of ethics and procedural fairness, and accountability pathways to prevent harm …………………………………………… 9 MISSING: Schrodinger’s Ethics and Human Oversight Framework …………………………………………….. 9 The stricter rules we already have arose as a compliance shortcut. We need to unpack the underlying assumptions and justifications for existing Section 10 rules before we can have a proper discussion about further ‘discipline.’ ……………………………………………………………………………………………… 10 But the debt exists!* “Nothing in the bill changes debt recovery” and other exculpatory narratives as a form of strategic knowledge avoidance …………………………………………………………………………….. 12

  • As it turns out, a bunch of debts did not, in fact, exist. ………………………………………………………… 13 Scrutiny of NDIA interpretation of Section 45 and 46 events is essential to substantiate the ‘nature’ and claimed ‘extent’ of non-compliance ………………………………………………………………………………… 15 The integrity conversation cannot be one-sided, and nor can accountability: We must be able to interrogate the integrity and ethics of NDIA decision-making, reporting and administration if we are to avoid adverse outcomes. ……………………………………………………………………………………………… 18 Nudging or prompting sounds reasonable on the surface, but red flags are already emerging. Without guardrails, we cannot assume the manner in which behavioural insights were weaponised in Robodebt will not creep in. …………………………………………………………………………………………………………. 19 Reimagining integrity measures through a principles-based, CRPD-conforming approach to Section 10 ………………………………………………………………………………………………………………………………. 23 The Data Lake, and the inputs and outputs that matter ………………………………………………………… 23 What else is missing from the Integrity inquiry we need? ………………………………………………………. 25

Cat Walker 26 March 2026 Page 2 of 26

Recommendations

  • Recommendation 1: The Committee consider and report on evidence received in this inquiry with particular consideration of the impact of public narratives, and assess potential non-compliance with Article 8 in communications about NDIS integrity.

  • Recommendation 2: The Committee inquire into and report on processes for capturing and responding to the unique and intersectional vulnerability indicators and accessibility needs of NDIS participants in compliance activities.

(See also recommended amendments to Section 118)

  • Recommendation 3: An Ethics and Human Oversight Framework be enshrined in the NDIS Act, with direct reference to relevant obligations under the CRPD.

  • Recommendation 4: That the Committee require the NDIA and Fraud Fusion Taskforce partners to table all documentation forming the complete Ethics and Human Oversight Framework, in its current, draft, or abandoned form, and any associated NDIA operational guidance.

  • Recommendation 5: The Committee request all substantive material and legal advice from the historical debt review be tabled as a matter of urgency and public trust.

  • Recommendation 6: The Committee consider recommendations and amendments proposed in Submission 7 to Annual Report No.2 of the 48th Parliament regarding Section 10 for inclusion in legislative amendments explored during this inquiry.

  • Recommendation 7: Insert specific review rights for Section 45 and 46 determinations into Section 99.

  • Recommendation 8: Insert an additional function of the NDIA at s118(1)(bb) of the NDIS Act to ensure all activities relating to s118(1)(ba) comply with the CRPD, are ethical, and uphold the highest standards of procedural fairness in all processes relating to participants, with regard to disability-related procedural accommodations and vulnerability indicators.

  • Recommendation 9: That the Committee require the NDIA to table all current operational guidance and policies regarding participant debt determination and management.

  • Recommendation 10: Any statutory ethics framework must encompass the unique ethical risks and vulnerabilities arising from implementation of behavioural insights, choice architecture, nudging, and associated integrity ‘risk profiles’ are accountable to disability-related needs, including suicide risks relating to specific cohorts and intersectional risk factors.

This should incorporate a positive duty of care on the Agency.

  • Recommendation 11: A principles-based definition of NDIS support under Section 10 should be developed with the disability community through a parliamentary inquiry.

  • Recommendation 12: An inquiry to define a community-endorsed, principles-based definition of NDIS support could simultaneously develop an ethical compliance and outcomes framework to be operationalised through existing data lake infrastructure and self-service options in the participant portal and myNDIS app.

This compliance and outcomes framework should be based on a holistic range of qualitative data points which link funded supports to CRPD obligations and meaningful, participant-led outcomes aligned to the objects and general principles of the NDIS Act.

Recommendation

####### ######## ######### ########## ########### ############ ############# ################ #################

Approach to terms of reference

The nature and extent of non-compliance, including fraud and sharp practices, in the National Disability Insurance Scheme

  1. This submission primarily focuses on the nature and extent of non-compliance. This is not intended to detract from the adverse impacts of fraud and sharp practices, or my support for efforts to address those aspects of NDIS integrity. My focus on non-compliance is directly related to the need to define its nature and extent with utmost care, and with due regard for the intrinsic relationship between accurately answering the questions posed, and adequacy (or absence) of statutory review rights, procedural fairness and other forms of scrutiny which are presently lacking.

  2. This focus is also firmly grounded in the moral (if not legal) duty of care expected of integrity measures directed at cohorts swimming in vulnerabilities comparable to those which led to the most tragic outcomes in Robodebt. To put it bluntly, this is also about suicide prevention.

  3. I also want to clearly state that I appreciate the willingness of many NDIA staff across these teams to hear tough feedback and acknowledge that many concerns raised carry psychosocial risk for staff. I am also grateful to those who have answered the questions they can or otherwise assisted in addressing concerns. I urge the Committee to recognise the psychosocial risk to staff should key recommendations not be actioned, and the harms I seek to prevent be realised. The most important questions and concerns that remain unanswered are those put to senior executives, one of whom I have named in this submission. I am nonetheless happy to provide specific examples upon request, in-camera if necessary, to illustrate the issues, cognitive biases, and strategic non-knowledge this submission addresses, if this would assist the Committee.

The impacts of non-compliance on NDIS participants and their families

  1. I address impacts throughout. However, I place particular emphasis on ensuring integrity measures are fair, just, ethical, accountable, and not the source of adverse impacts on participants.

Effectiveness and adequacy of successive government policies to improve scheme integrity, safeguard participants, and tackle non-compliance

  1. I address adequacy of government policies by interrogating the underlying assumptions or justifications, evidence for and against their claimed effectiveness, and whether they ultimately safeguard participants, cause harm, or fail to prevent it in the pursuit of tackling non-compliance.

  2. My commentary and recommendations to this effect should be understood as supportive of integrity measures which are sufficiently transparent, accountable, and open to scrutiny, to ensure we are able to verify the integrity, accuracy, fairness, and ethics of those measures.

Any legislative or other reforms required to strengthen scheme integrity

  1. I propose legislative and other reforms throughout. Fundamental to these are the need for NDIA integrity functions to be grounded in the CRPD, open to scrutiny of statutory review rights for participant-directed compliance activities, and bound by a statutory ethics framework.

How we talk about integrity of the NDI matters

Addressing the impact on participants starts with grounding integrity discourse in Australia’s international obligations, recommendations of the Robodebt Royal Commission, and holding media accountable to both.

While the intent for this Committee to conduct any NDIS integrity inquiry was mentioned in the Senate on 23 March 2026, the first indication of its scope and purpose was, once again, delivered via the media,

citing anonymous “sources” to establish the desired vibe in public consciousness two days before terms of reference were published and submissions opened (emphasis added):

Labor wants an inquiry to hear fresh evidence of fraud in the $50 billion National Disability Insurance Scheme as it builds a case for stricter rules and tighter spending on the scheme in the May budget…

McAllister’s request on Monday suggests the government wants to go even further in its crackdown… With the scheme forecast to surpass $100 billion in a decade, Butler has said urgent improvements to discipline and integrity would be the most essential actions to deliver cost savings… He flagged in the middle of last year that he wanted to bring spending growth down to 5 or 6 per cent by the end of the four-year forward estimates. Annual growth was 10.3 per cent at the end of December.

Sources said the government also believed it must do more to weed out fraud so that the scheme retains its social license while delivering on outcomes for its 760,000 participants.

McAllister has asked the committee, chaired by Labor MP Libby Coker, to report into the nature and extent of non-compliance in the NDIS, including fraud, and the impact this has on families. She also wants the committee to scrutinise how effective policies have been to date in improving integrity and protecting participants from harm.

As part of its work, the committee will be asked to propose new laws or other reforms that will improve the scheme’s integrity. McAllister has asked it to report back by June 19.

Let us be clear about this much: No one cares more about funds being directed towards outcomes and ‘weeding out’ legitimately fraudulent behaviour than NDI participants.

Sadly, no one has done more to erode social licence for the NDIS in recent years than Government,

in the pursuit of public consent for cuts both sides remain dissatisfied by, while pulling focus from measuring the outcomes that matter most for protecting social licence. As The Saturday Paper reported in 2024:

More than a year ago, the federal government commissioned Labour-aligned research and political strategy outfit RedBridge Group to conduct a series of focus groups and “narrative” tests about potential and actual NDIS reform. Ever since, Shorten has been adopting the language of its findings…

[When] we presented rorts, fraud, and unreasonable pricing as posing an existential threat to the NDIS, we were able to create an environment in which respondents were amenable to reforms designed to counter these things.

A separate element of the NDIS reforms – reducing supported independent living (SIL) funding so disabled people with significant, 24/7 support needs will be forced to live in groups of at least three, counter to a recommendation from the disability royal commission to phase out “group homes” – was also stress tested…

RedBridge found the government could cut services if it had a believable story to tell about protecting participants.

“After presenting respondents with the above proof points that provide vital, believable context of reforms designed to benefit participants, we were able to elicit a degree of qualified tolerance for reforms that would restrict either eligibility for the Scheme (raising the threshold for children with developmental delay) or the amount of support high-need participants would receive (moving to a 1:3 care model for Supported Independent Living participants),” the research says…

It is not that participants don’t want to fight rorting in the scheme. Their concern is that the central agency delivering the disability scheme has weaponised a culture of “hostile complexity” that it now looks to have codified in law.

“In our experience, NDIA decision-making relies heavily on Operational Guidelines even when the [Administrative Appeals Tribunal] has found that the guidelines are inconsistent with the NDIS Act or Rules,” National Legal Aid says.

Indeed, there are Federal Court precedents dating back to 2017 that the NDIA has simply ignored. Now the new legislation seeks to embed this way of doing business.

12. “Extent” is therefore a rather loaded assumption to adjudicate via an inquiry of under three months,

in the absence of review rights to challenge cancelled claims or debts, the ability for the public to scrutinise evidence supporting the claims of the Agency as to any such extent of non-compliance, and the broader social licence backdrop: That social licence for further restrictions relies on social licence to deem our human rights and participation in our communities a luxury and a drain on the economy, while erasing the economic and social returns we now risk losing.

13. It is therefore prudent to begin our analysis with Article 8 of the CRPD, in acknowledgment that

public discussion of this inquiry linked integrity to austerity before it even opened for submissions:

Article 8 - Awareness-raising

States Parties undertake to adopt immediate, effective and appropriate measures: To foster respect for the rights and dignity of persons with disabilities; To combat stereotypes, prejudices and harmful practices relating to persons with disabilities, [in] all areas of life; To promote awareness of the capabilities and contributions of persons with disabilities.

Measures to this end include: Initiating and maintaining effective public awareness campaigns designed:

  • To nurture receptiveness to the rights of persons with disabilities;

To promote positive perceptions [towards] persons with disabilities; …

Encouraging all organs of the media to portray persons with disabilities in a manner consistent with the purpose of the present Convention;

These obligations are also reinforced by Recommendations 10.1 and 11.4 of the Robodebt Royal Commission, which highlight the relationship between stress, stigma and vulnerability.

These are bare minimum standards we should be able to expect of the NDIA, and of Government in setting public narratives about the NDIS and integrity of the Scheme:

Recommendation 10.1: Design policies and processes with emphasis on the people they are meant to serve

Services Australia design its policies and processes with a primary emphasis on the recipients it is meant to serve. That should entail: avoiding language and conduct which reinforces feelings of stigma and shame associated with the receipt of government support when it is needed; … acting with sensitivity to financial and other forms of stress experienced by the customer cohort and taking all practicable steps to avoid the possibility that interactions with the government might exacerbate those stresses or introduce new ones.

Recommendation 11.4: Consideration of vulnerabilities affected by each compliance program, including consultation with advocacy bodies

Services Australia should incorporate a process in the design of compliance programs to consider and document the categories of vulnerable recipients who may be affected by the program, and how those recipients will be dealt with.

Services Australia should consult stakeholders (including peak advocacy bodies) as part of this process to ensure that adequate provision is made to accommodate vulnerable recipients who may encounter particular difficulties engaging with the program.

It is for this reason I repeat my observations from last week to reinforce the international obligations this Committee has a responsibility to enforce and uphold throughout this inquiry and its report.

Ongoing harmful media narratives, including headlines last week that I will not dignify with references, are plainly dog-whistling to all of Australia that people with disability should go back to being shut in and shut out: That equal opportunity is not reasonable and necessary. That the only kind of support people like you can have is support which shrinks one’s world to predefined boxes.

Accepting the premise of such narratives, any ongoing lists-based approach to Section 10, or further attempts to confine the scope of NDIS support – as the SMH was seemingly briefed on while participants remained out of the loop – would amount to endorsement of a dehumanising view of people with disability the NDIS promised to leave in the rearview mirror. Australia needs to be better than this, and that starts with our leaders.

Recommendation 1

The Committee consider and report on evidence received in this inquiry with particular consideration of the impact of public narratives, and assess potential non-compliance with Article 8 in communications about NDIS integrity.

John Delmonico, The media wants disabled people to be locked inside until they die, The Shot, 18 March 2026.

Recommendation 2

The Committee inquire into and report on processes for capturing and responding to the unique and intersectional vulnerability indicators and accessibility needs of NDIS participants in compliance activities.

(See also recommended amendments to Section 118)

How we approach participant-directed integrity measures requires the highest standards of ethics and procedural fairness, and accountability pathways to prevent harm

MISSING: Schrodinger’s Ethics and Human Oversight Framework

  1. Seriously: Is it as dead as the poor cat? Because the previous reference to the Fraud Fusion Taskforce being “governed by a robust Ethics and Human Oversight Framework” was seemingly deleted from the NDIS website in January 2025, just prior to the PRG finally seeing a one-page slide listing 12 ethical principles supposedly informing this elusive framework – which only occurred after the better part of a year pressing the NDIA on this framework, including in the joint submissions my PRG colleague Uli Cartwright (who contributed to development of the twelve ethical principles on the Fraud Working Group) and I made on the 2024 NDIS Bill.

  2. The justification for not publishing the framework or principles was that it involved all Taskforce agencies. Isn’t that even more reason for the Taskforce to be publicly accountable to it, and the NDIA to their role in informing how the Taskforce acts in any matters relating to NDIS participants?

  3. As I repeatedly put to Integrity teams in 2024 and throughout 2025, an ethics framework only exists when it is published and those it applies to able to be held accountable to it. And if it has not progressed beyond unenforceable, unpublished principles – why are participants in the firing line without basic, agreed, public standards to ensure their rights are upheld in all integrity measures?

Recommendation 3

An Ethics and Human Oversight Framework be enshrined in the NDIS Act, with direct reference to relevant obligations under the CRPD.

Recommendation 4

That the Committee require the NDIA and Fraud Fusion Taskforce partners to table all documentation forming the complete Ethics and Human OversightFramework, in its current, draft, or abandoned form, and any associated NDIA operational guidance.

The stricter rules we already have arose as a compliance shortcut.

We need to unpack the underlying assumptions and justifications for existing Section 10 rules before we can have a proper discussion about further ‘discipline.’

20.

In my submission to the 2025 DSS ‘consultation’ on Section 10, I highlighted issues of concern arising from revelations about the genesis of the purported need for legislative changes to insert a definition of support at Section 10, in the context of previously undisclosed raising of unlawful historic debts by the NDIA. While my submission to Annual Report No. 2 focuses on the case for rejecting this approach to Section 10, it is important for us to fully unpack these revelations and place them in proper context of Scheme integrity before engaging with any additional reforms proposed by Government or the Executive.

     In Submission 183 to the second inquiry on the Back on Track No. 1 Bill, my co-authors and I had questioned the real goal of Section 10:

   "One must question whether this is the real motivation for restricting the definition of NDIS support. It would certainly be an attractive shortcut to more convenient compliance interventions, but that path is fraught and unlikely to achieve genuine savings if it denies participants access to many cost-effective and practical supports. What’s the point of the Wedding Tax argument if we are forced into using those charging Wedding Tax just to have some confidence our claims will be paid out in a timely manner?"

    Unsurprisingly, that's exactly what an ANAO report just found: The Section 10 we were sold as reinforcing the constitutional basis by 'implementing the CRPD' was really a bait and switch:

  On 3 October 2024 the National Disability Insurance Scheme Amendment (Getting the NDIS Back on Track No. 1) Act 2024 came into effect, which amended the NDIS Act to strengthen the legislative basis for the NDIA's compliance activities by:

  • Defining NDIS supports (section 10 of the NDIS Act)

                               Why? Because...

     In May 2023, the NDIA received advice that it was limited in its ability to recover payments as debt, due to difficulties in: determining that spending was not in accordance with a participant's plan; and setting a hard limit on plan budgets when supports were described generally in participant plans.

                     What else did the ANAO report reveal?

 Lack of alignment with the NDIS Act (2013), resulting in the raising of unlawful debts...

               I find it quite infuriating to have that confirmed when [we] had not only put forward an alternative in the same submission that took aim at compliance activities (as reproduced in

Cat Walker, Submission No 7 to Joint Standing Committee on the National Disability Insurance Scheme, Annual Report No. 2 of the 48th Parliament (18 March 2026).

Cat Walker 26 March 2026 Page 10 of 26

the 2024 submission appended to this one), but demonstrated we were ready and willing to help build principles-based justifications into the claiming and audit processes.

   What is the goal here? To claw back funds and limit spending for poisonous political
      reasons? Or to uphold the original intent outlined in the Objects and Principles?

    Because you’re doing a fantastic job of making Australia a more hostile place to live if you
   need disability support, and the rhetoric of the last 14 months and the ANAO report strongly
                 suggest it’s the former. This rhetoric is making it unsafe for us.

    These lists and the sick narratives they were sold with have placed targets on our backs by
         deliberately eroding social licence and encouraging the wider community to assume
     everybody is rorting the Scheme, whilst making it impossible for us to get the support we
          actually need to achieve the intended outcomes of full inclusion and participation.

         I get that in some cases the ‘right’ support will cost more. But in many cases, the problem isn’t
   how big a slice of the NDIS pie we get. The problem is that the Government Approved Pie
    Recipe is inedible, overpriced rubbish we’re expected to shut up and be grateful for,
    even if it makes us unwell – or causes a deadly anaphylactic reaction, and even if our
            own recipe would sustain us fabulously for half the price.

21. Again, I emphasise my support for accountability over NDIS expenditure – but I submit that how we approach that matters, which is precisely why I remain as willing as I was in 2024 to explore how a principles-based definition of NDIS support could be operationalised for compliance purposes.

  1. Indeed, this would be easier than ever to implement, using the same technological advances currently being wielded as a blunt instrument to enforce the Section 10 lists, and numerous options I explore in this submission are suggestions already posed to senior Integrity executives and the Crack Down on Fraud team during my time on the PRG. The infrastructure is there.

  2. But before we get to that, let us consider findings of the ANAO report and historical debt review findings in relation to unlawful debts, in the context of new information from the sealed Robodebt chapter and academic research on non-knowledge in integrity settings. And let us ask: Did anyone pause to consider whether changing the scope of what qualifies as ‘non-compliance’ to match what turned out to be unlawful NDIA assessments to that effect, was actually compliant with the objects, principles, and intended outcomes of the NDIS Act? Can that really be called ‘integrity’ if it only makes the NDIA’s job easier, and their approach lawful, at the cost of outcomes?

    Recommendations 5 and 6

    The Committee request all substantive material and legal advice from the historical debt review be tabled as a matter of urgency and public trust.

    The Committee consider recommendations and amendments proposed in Submission 7 to Annual Report No.2 of the 48th Parliament regarding Section 10 for inclusion in legislative amendments explored during this inquiry.

Cat Walker 26 March 2026 Page 11 of 26

But the debt exists!* “Nothing in the bill changes debt recovery” and other

exculpatory narratives as a form of strategic knowledge avoidance

24.

As previously highlighted in Submission 7 to Annual Report No. 2,5 matters in the sealed Robodebt tabled in March regarding APS Code of Conduct referrals (rather than the six NACC referrals) have specific relevance to this inquiry.

25.

Although it remains unknown whether these were among the 12 individuals ultimately found by the APSC to have breached the Code of Conduct or APS Values, a key reason for the Commissioner’s referrals of individuals who subsequently spent years in NDIA executive/SES roles was that they:

…adopted words in a brief to the then Secretary of DHS that attempted to put a positive spin on the alarming results of a pilot for what would ultimately become the Robodebt Scheme and thereby fail to uphold the standard of frankness and candour in the provision of advice that should be expected of a public servant in their position.

26.

My experience and observations of reforms to date is that such ‘positive spin on alarming results’ – or feedback – is a culture so pervasive, and a tendency I have so frequently witnessed and personally encountered, as to suggest this is completely normalised, in spite of the seriousness of the risks many of us have raised to the Department, NDIA or Ministers offices along the way.

27.

Academics Adam Hannah and Linda Botterill explored the nature of such patterns in detail in their paper, ‘Ignoring harm, saving face: non-knowledge, senior public servants and the Robodebt scheme.’6 For example:

Strategies of non-knowledge go beyond an unintended lack of awareness or bounded rationality (see Hannah et al. 2023). Rather, policy actors may strategically cultivate claims to not know or deliberately avoid acknowledgement or inscription of inconvenient or troubling information, to preserve a favourable status quo or to protect organisational or individual reputations (Boswell and Badenhoop 2021; Howlett 2023; McGoey 2007; Stark and Head 2019) …

[We] highlight four specific strategies used to avoid grappling with these fundamental issues:

  1. Avoiding consultation

  2. Knowing what not to know and not to ask

  3. Non-inscription of inconvenient knowledge

  4. Repetition of exculpatory narratives and focus on things that ‘did not matter’.

5 Cat Walker, Submission No 7 to Joint Standing Committee on the National Disability Insurance Scheme, Annual Report No. 2 of the 48th Parliament (18 March 2026). 6 Hannah, A., & Botterill, L. C. (2025). Ignoring harm, saving face: non-knowledge, senior public servants and the Robodebt scheme. Australian Journal of Political Science, 60(2), 149–162. hhttps://doi.org/10.1080/10361146.2025.2486108

Cat Walker 26 March 2026 Page 12 of 26

Considering the role of non-knowledge in the Robodebt case moves us to a murky area of policy practice which sees inconvenient or ‘uncomfortable’ knowledge ignored, obfuscated, hidden, or discredited in policy development and implementation.

   In this way, we contribute to calls for scholars to pay greater attention to the 'dark side' or 'malign' policy-making (Howlett 2020; Legrand 2022; McConnell 2018) and to the issues of strategic ignorance and knowledge avoidance in the policy process (Hannah et al. 2023).

  The case suggests that when individuals - reacting to implicit organisational priorities, as well as interactions with their superiors - perceive the costs of 'knowing' to be high, they will employ a varied range of strategies of ignorance or non-knowledge.
  These range from failing to ask obvious questions and deliberately avoiding those who might offer a critical view, avoiding inscription of 'bad news', and focusing on minutia at the expense of more fundamental issues.

     Importantly, while the results were unethical, in the context of the Departments of Human and Social Services during this period, they can hardly be described as irrational for the individuals deploying these strategies.

   There were few incentives for individual public servants to think seriously about the social costs of their actions, especially at a senior level and especially not compared to the risks of disappointing a Minister or enduring media scrutiny.

   Clearly, the knowledge and experiences of those the two Departments are meant to serve, namely welfare recipients, have been sidelined in favour of convenient narratives around fraud, fanciful projections of budget savings, and obsessive monitoring of media reporting.”

28. I adopt the concerns and examples already raised to the extent they apply to this inquiry, and again suggest the Committee keep such possibilities front of mind at any time evidence from officials or Ministers diverges significantly from community feedback, or dismisses community concerns as ‘anxiety’, misinformation7 or lack of understanding without engaging with substantive concerns.

  1. This is especially important as we explore examples focused specifically on integrity matters.
  • As it turns out, a bunch of debts did not, in fact, exist.
  1. ‘Nothing in the bill changes debt recovery’ and similar statements were repeated ad nauseum during 2024 in an attempt to deflect the valid criticisms of advocates and legal experts.

7 I further suggest that the frequent invocation of ‘misinformation’ by Government in the face of clearly-evidenced criticisms or valid questions about lack of information, undermines public trust more broadly.

Cat Walker 26 March 2026 Page 13 of 26

  1. When pressed at a hearing on the 2024 Bill, officials insisted debts are not raised based on a decision but exist because facts establish the existence of a debt.

  2. Only, the historical debt review subsequently found that almost a quarter of participant debts reviewed should not have existed, leading to their revocation. The lists do not resolve this any more than they ensure valid claims will never be cancelled.

  3. But what is more concerning is the effort put into deflecting valid concerns about compliance events and resulting potential debts during 2024 with strategic narratives that were exactly what Hannah and Botterill describe, at the same time the historical debt review was in progress, and at the same time concerns over debts were deflected and defended, including in co-design settings.

  4. The outcome was buried with positive spin focusing on the debts upheld – with inadequate information about those revoked – and an apology which only acknowledged the harm caused and focused on exactly the human error which necessitates still-absent review rights – when found to have acted unlawfully, repeatedly. Officials had continued to defend the Agency’s right to be judge, jury and executioner over our right to a legal pathway to check their homework the whole time.

  5. Such harm is what advocacy for Section 45 and 46 review rights sought to prevent in future. The Agency has not produced any evidence that the claimed lessons from this review will prevent harm, including in response to concerns repeatedly raised in consultations on debt management in 2025. I am happy to provide further evidence on concerns arising and questions repeatedly unanswered.

Scrutiny of NDIA interpretation of Section 45 and 46 events is essential to substantiate the ‘nature’ and claimed ‘extent’ of non-compliance

36.

I adopt relevant aspects of my 18 March 2026 submission here, where it was made before this inquiry was announced and key sections are directly relevant to addressing the terms of reference of this inquiry or establishing the basis for associated recommendations.

37.

In 2024, my co-authors and I stated:

It is extremely concerning to realise that the pre-payment review plan rolled out by the Scalable Integrity Responses Branch in recent months – and protections against the scenario highlighted in this case study – has specifically been omitted from the new powers inserted at proposed Section 45(4), let alone from having an associated reviewable decision.

At present, we have a situation which allows the Agency to be judge, jury and executioner, potentially leaving participants without supports they may well be entitled to, no recourse to challenge the decision, and the risk of being pursued by private debt collectors on behalf of unpaid providers.

The Agency will report this as the number and value of “non-compliant” claims successfully stopped, without any mechanism for the validity of such claims to be scrutinised… Would the Agency be so quick to call this a success if they waited until the payments were made and reported on the number and value of debts raised against participants?

That seems doubtful, and the fact remains that keeping the Agency’s fingerprints off of debts arising from cancelled claims for support already received could have dire consequences such as those we saw in Robodebt.

As the impacted individual put it:

“In addition this episode has taken a toll on my [partner] who has scrupulously accounted for expenditure over 10 years and now feels like [they are] somehow at fault and also worries if we can’t resolve this we will be saddled with the debt to the Service Provider (into several thousand dollars).”

If the Agency believed in the “integrity“ of its Integrity program, it would have recommended it be legislated.

It is essential that the Committee recommends amendments to enable scrutiny and legal challenge of this approach, including robust review rights for all decisions that are currently taking advantage of a non-reviewable No Man’s Land to avoid scrutiny of compliance activities…

When all current reforms are being shaped to an arbitrary 8% target, a claimed “error rate” of 5% is a very attractive weak spot to squeeze perceived savings from.

We therefore must ask the question: Is this current program about genuine compliance, or is it going harder as part of an overall push to reach this ultimate KPI through any means necessary?

  1. Recent correspondence shared with me demonstrates that Crack Down On Fraud is morphing into Crack Down On Legitimate Claims for Prescribed Disability Shop™ Assistive Technology.

  2. Predictably, the informal ‘reconsideration’ in the absence of statutory review rights confirmed the decision. The participant is now out of pocket. This is exactly what we said would happen:

    At this point, it is worth revisiting the Robodebt testimony of whistleblower Colleen Taylor – awarded an Order of Australia Medal just today – of what she observed when the level of human involvement was restricted in favour of scale and savings, while still in the manual phase…

      Ms Taylor expressed her earnest concern, “…as a Compliance unit, we should not be the ones stealing from our customers.” When asked about that comment in oral evidence, she responded:

     “Well, if we know there’s no debt, and yet we’re sending a debt notice out to someone, isn’t that stealing?” 11

  `redacted`

    Even in the context of cancelled claims rather than debts raised, Ms. Taylor’s perspective is an important one when discussing the Agency’s stated 5% error rate, with no scrutiny of the risk that some of these determinations may well be wrong, or the consequences for participants who are left out of pocket, in debt or without support if their claims are cancelled incorrectly.

  3. It is therefore disturbing to see in DHDA’s Additional Estimates published last month, the note that Other Gains in NDIA figures to the value of more than a billion dollars is attributed to “In-kind services contributions from state and territory governments, Scheme payment cancellations, reversal/write back of provisions.”

  4. As we concluded in Submission 183 in 2024:

    The Government may have recognised her as Colleen Taylor OAM today, but did they understand what she was trying to teach them?

  5. When it comes to Section 46, and approaching two years after the “media firestorm” my co-authors and I referenced in Submission 183, there is still no evidence the NDIA’s decision-making in relation to Section 46 would withstand judicial scrutiny.

  6. I lack time to reference all submissions and oral evidence of legal experts urging this Parliament to enshrine review rights for the Section 46 “compliance event.” In brief, administrative law expert Dr. Darren O’Donovan explained to the Community Affairs Legislation Committee that this compliance event is the “birth of the debt”, among other adverse consequences for participants.

  7. I further refer to Dr. O’Donovan and Rick Morton’s plain language explanation of the equivalent Robodebt situation in Mean Streak,12 with one important exception:

redacted

redacted: s47F - Personal privacy: Disclosure would involve the unreasonable disclosure of personal information about any person (the most common reason for redacted names in these documents).

Ryman’s ‘Right of Review’ and Its Impact on Welfare Recipients

Ryman and his team also introduced a new ‘right of review’ designed to trick a welfare recipient into thinking a Centrelink decision could be re-examined by a (usually) competent Authorised Review Officer (ARO).

Under an ordinary review, a customer could test their case internally and then take it to the Administrative Appeals Tribunal for a fair(er) hearing. This official step was moved even further away from a person, if they even knew how to ask for it properly.

‘The customer has the right to request a reassessment of income details at any time,’ Ryman’s process map says. ‘A reassessment is undertaken as part of the intervention process. The customer still has the right to exercise the formal review and appeal rights after this period.’

Darren O’Donovan explains the chilling effect of this clinical statement.

‘So unless you asked for an authorised review officer to do a review, magic words right, unless you specifically asked for that you were getting checked by some lackey in the compliance unit,’ his said.

These intentional frustrations mirror the maddening dimensions of K’s ordeal in Kafka’s The Trial. In K’s increasingly knotty legal drama there are no charge sheets and no legal records of the case, because the Court has simply deemed it unnecessary to release them either to the public or, indeed, the defendant or his counsel. That any plea could be constructed at all was done entirely by guesswork, based on inferences drawn from various interrogations which were themselves held in obscure locations hidden in a snarl of staircases and corridors.

  1. The exception is that, as Miles Browne of Victoria Legal Aid told the same Committee in 2024, “Well, in Robodebt, at least people could seek a review of their debt.” In other words, the Parliament has “simply deemed it unnecessary” for participants to have any defence whatsoever.

  2. Yet, at the time of writing, and despite my bringing this to the Agency’s attention, there remains a false and misleading claim regarding seeking review of a debt on the NDIS website:13

We have safeguards in place to support participants

We understand that managing an NDIS plan can be complex. That’s why we’ve put safeguards and fair processes in place to support participants before a debt is raised…

We consider individual circumstances before raising a debt. Participants can request a review, waiver, or write-off if needed.

(Bold emphasis added)

  1. This is not a statutory review right. This is exactly the process Dr. O’Donovan was referring to in Mean Streak – and while it is no doubt the result of consistent criticism about the absence of s 46 review rights, it has the express effect of lulling NDIS participants into a false sense of security with no formal review right when they get the same answer.

  2. Mean Streak won the Prime Minister’s literary award, and yet NDIS participants are being exposed to the risk of substantially greater debts – or payment cancellations – without the same rights as Robodebt victims to challenge them.

13 The NDIS supports transition period is ending (3 October 2025). Captured changes available upon request.

Cat Walker 26 March 2026 Page 17 of 26

Recommendation 7

Insert specific review rights for Section 45 and 46 determinations into Section 99

The integrity conversation cannot be one-sided, and nor can accountability: We must be able to interrogate the integrity and ethics of NDIA decision- making, reporting and administration if we are to avoid adverse outcomes.

  1. The aforementioned unlawful historic debts raised against participants – which were not only revealed far too late for debate on the 2024 Bill but ultimately resulted in 24% of the 475 participant debts reviewed being revoked by the NDIA – reinforce that the issues my co-authors and I raised in 2024, alongside numerous legal experts, remain central to the important conversations we still need to have.

  2. This is underscored by the ongoing discourse playing out in a landscape where participants lack any pathway to seek external scrutiny of NDIA determinations of non-compliant claims falling within the statistics cited by the Agency, or indeed, determinations they have misused funds.

  3. As we said in that submission (as quoted earlier), the previous 5% error rate claimed deserves significant scrutiny.

  4. If that estimate has now doubled based on the volume of claims stopped, it requires even greater scrutiny: Participants should not carry the risk or consequences of incorrectly cancelled claims or debts they may not lawfully owe without the right to challenge their basis.

  5. Parliament therefore should not accept such statistics on face value in the absence of dispute pathways and independent external scrutiny. However, such external scrutiny or audits which lack understanding of the relationship between highly individual disability impacts, contexts and support needs and instead rely on reductive, homogenised internal guidance, would nonetheless be inadequate for assurance purposes.

  6. I make this submission in my personal capacity and as an individual, and all sources cited are in the public domain if not part of the Parliamentary record. However, and having now concluded my two- year term on the NDIA’s Participant Reference Group, I make the following comments which are equally applicable to participants impacted by their own mistakes, or those of the Agency, which they currently have no genuine pathway to challenge:

    a. Since we made that joint submission, I have been involved in approximately six sessions on integrity processes, including debts.

    b. I stand by every concern I have previously raised and the lack of procedural fairness inherent in the Agency’s approach to systems, business rules, and processes which are now rolling out at scale without any independent scrutiny to protect the rights of participants.

    c. I have consistently raised important questions in good faith. None of these have ever been adequately answered, including by current and former senior executives.

    d. I do not believe the NDIA can be trusted to self-assess the fairness of any changes made based on the historical debt review. I believe all findings and information it is based on should be made publicly available as a matter of urgency.

  • e. I do not have any faith in the NDIA’s capacity to prevent harm or self-assess compliance with the NDIS Act or ‘lessons’ from the Robodebt Royal Commission.

  • f. I believe review rights for Section 45 and 46 determinations are urgently required and that failure of this Parliament to enshrine them in the NDIS Act will see harm, death, and consequences amounting to far more billions in downstream impacts, including potential class actions should such harm ensue.14

  1. Section 45 and 46 review rights can prevent these predictable outcomes. The further proposed changes to the Agency’s functions at Section 118 would assist in improving accountability of NDIA decision-making enough to render them somewhat less frequently required.

Recommendation 8

Insert an additional function of the NDIA at s118(1)(bb) of the NDIS Act to ensure all activities relating to s118(1)(ba) comply with the CRPD, are ethical, and uphold the highest standards of procedural fairness in all processes relating to participants, with regard to disability-related procedural accommodations and vulnerability indicators.

Recommendation 9

That the Committee require the NDIA to table all current operational guidance and policies regarding participant debt determination and management.

Nudging or prompting sounds reasonable on the surface, but red flags are already emerging. Without guardrails, we cannot assume the manner in which behavioural insights were weaponised in Robodebt will not creep in.

  1. I will preface my comments on behavioural insights and their use in compliance as follows:

  • a. I began studying behaviour modification in my early teens, well over two decades ago

  • b. I was one of the earliest adopters of a highly successful behavioural economics app, developed by researchers whose work continues to inform behavioural insights approaches used by governments, including those at issue

  • c. I have tertiary qualifications in horse behaviour (fundamentals of which are identical across species) and postgraduate-level recognition of my equivalent professional experience shaping behaviour change in humans from the individual and grassroots levels to systemic change

  • d. I have read all relevant Robodebt exhibits and transcripts and do not draw parallels lightly

  • e. I was consulted on various aspects of ‘prompting’ (a.k.a. ‘nudging’) whilst on the PRG, and speculative concerns I raised about potential use-cases have already been implemented

14 Exclusive: Misfeasance case paves way for robodebt action. Rick Morton, The Saturday Paper (31 January 2026).

Section Heading

Subsection Heading

Third-level heading

This text was extracted from an OCR source and may contain errors.

  • This line starts with lowercase ‘e’ followed by space - it’s part of bullet-point list item one.
  • Another bullet point here showing how these lines are formatted in original document as well.
Column One Column Two
Data A Info X
Data B Info Y

Quoted content goes inside blockquotes like this for direct quotations or excerpts.

  1. It is worth revisiting the NDIA behavioural insights story Rick Morton reported back in November 2022 (during the Robodebt Royal Commission), including the comments from Agency sources. While the concerns raised were deflected by the Agency when given the opportunity to respond at the following Estimates, these concerns remain, have not been adequately addressed when I have raised them in relevant meetings, have since come to pass (including in the form of Section 10 implemented), or are now playing out in various forms (bold emphasis added):15

    A draft of that National Disability Insurance Scheme (NDIS) legislation, with tracked comments from agency staff, was obtained by The Saturday Paper. It shows Jason Ryman, who was seconded to the NDIA from DHS after robo-debt began, marking up changes with his approval and questions in January last year.

    One key proposed change was an additional section that sought to outlaw the use of NDIS funds on “ordinary living expenses” or goods and services that the “minister considers are more appropriately funded or provided” through state and territory support systems or services.

    Ryman was enthusiastic about this draft change.

    “Funding not to be used for certain purposes – the clear reference an amount spent in contravention of this subsections is a debt is a positive,” he wrote in annotations attached to the document.

    “The drafting of the rules of what is an ‘ordinary living expense’ or ‘goods and services specified in the rules’ should provide us with a basis for making determinations of non-compliance and remove some of the questions of what is allowable / not allowable with the removal of the term ‘reasonable and necessary’.”

    Ryman was supportive of another change, linked to this section, that would raise a debt against a person who breached these rules.

    “Provides clarity in terms of who the debt is against, (the recipient who receives the money and does not spend in accordance with the Act),” he wrote…

        It was Jason Ryman who authored the June 24, 2014, paper that went to his then boss at the

    Department of Human Services (DHS), Scott Britton, outlining the early concepts for robo-debt. This document, discovered by the Royal Commission into the Robodebt Scheme, shows that a core feature of robo-debt was present from the earliest stages: the reversal of the onus of proof…

    A senior employee of the NDIA who is familiar with the workings of the division where both Britton and Ryman work says hunting down fraud is very different to some conceptions of compliance.

    “The very clear problem to me is that the agency [NDIA] has a long history now of telling participants things that are just not true in the hope that they will not fight a decision,” this official says. “Say, for example, they might deny funding for a particular therapy or for travel, or cap support hours. What we see, if a person decides to take on the very onerous task of

challenging that decision, is that these frequently get overturned at the Administrative Appeals Tribunal. In other words, the agency was wrong to make the decision it did.

“This is the same thinking that underpins compliance. And when we are talking about using behavioural insights to enhance that, we are talking about using psychological tricks to discourage certain behaviours. What I want to know is whether we are discouraging people from claiming support to which they are entitled.”

…

In a statement, the [NDIA] spokesperson said: “Agency compliance activities are ethical and consistent with our commitment to focus on the welfare of participants at all times.”

  1. It has been over three years, and again I ask: Where in the world is the evidence substantiating said claims to ethics, and the unique understanding of disability impacts and vulnerability indicators any ethics framework for NDIS compliance needs?

  2. Because these are questions we also raised in Submission 183 to the second NDIS Bill 2024 inquiry,16 and we are yet to have answers to ethical questions relating to behavioural compliance strategies and guardrails for preventing harm – including suicide – in the NDIS context:

Speaking to the overall failure of Government, the APS and the NDIA to heed the bigger lessons of Robodebt, Kath [Madgwick, our co-author and mother of Jarrad Madgwick] implored all to consider that:

“[The Government] doesn’t have a system that prevents harm in what they’re trying

to do. You don’t go out and take a Mini to tow a caravan. That is a lack of duty of care. If the system can’t cope and pick up vulnerabilities, don’t do it. If you want to do something, the system must be able to prevent harm.”

We would argue that this encompasses all the vulnerabilities examined in Robodebt, but also the more complex types of harm the Government and NDIA must consider in NDIS compliance initiatives.

This is because the implications for NDIS participants don’t stop at debts, and bringing compliance upstream creates substantial additional risk for participants which is playing out in an unlegislated, unreviewable and wholly invisible space, reported only as an unchallengeable error rate of claims successfully stopped.

We understand the Fraud Fusion Taskforce has an Ethics and Human Oversight Framework, but it still has not published this for public scrutiny. Many questions also remain unanswered regarding the ethics of using behavioural insights in NDIS compliance – imported to the NDIA by the same subject matter expert who made this a key feature of Robodebt – in the context of this specific population.

  1. I put Kath’s ‘Mini towing a caravan’ example to the Agency during multiple integrity consultations during 2025, in the context of unresolved questions and concerns about prompting, self-service, and debt themes at issue.

16 Cat Walker, Uli Cartwright and Kath Madgwick, Submission No 183 to Community Affairs Legislation Committee, Inquiry into the National Disability Insurance Scheme Amendment (Getting the NDIS Back on Track No. 1) Bill 2024 (19 July 2024), p.9.

Cat Walker 26 March 2026 Page 21 of 26

61.

And let us be clear: Behavioural insights-informed nudging, or prompting, is the central principle the Taskforce summarises as Making it easier to get it right and harder to get it wrong.

62.

Perfectly sound in theory… until one asks the question the anonymous senior NDIA official posed back in November 2022: Are the Taskforce, NDIA and key aspects of Crack Down on Fraud “discouraging people from claiming support to which they are entitled?”

63.

It would appear so, in the growing uncertainty about whether claims will be paid out now; in one of the use-cases I suspected when consulted on far more basic prompts last year having now been implemented in the form of nudges suggesting a claim looks like a non-NDIS support; and indeed, in the recent assistive technology example disclosed to me, which highlights the risk that even if one proceeds, a potentially legitimate claim may be cancelled.

64.

Which raises another important question: What is the extent of false positive “non-compliant” claims now being cancelled at scale; what is the extent of human oversight (including KPIs, any guidance which may bias discretion towards cancelling a flagged claim, and the nature of claim summaries extracted for approval or rejection); and how can we ever know the difference without statutory review rights to dispute cancelled claims?

65.

Because while these integrity measures are reasonable in theory, we must grapple with the reality of enormous new barriers to accessing supports one validly needs, and the risk that a claim will not be paid out even for something on the “in” list which clearly should be.

66.

When outcomes are already being compromised on a scale from interference or indignity to life- threatening risk or death, the other question is this: Could the same system not be applied to a principles-based approach instead, prompting users to identify the disability-related need and purpose (the “but for my disability” test, where required), outcome, associated plan goal, any required therapist evidence, and so on – from the same plan budget, without ever increasing Scheme expenditure if it comes from the same bucket of approved funding?

67.

If the Executive or Government’s answer to that question is “No,” should we not be asking why? Should we not be asking whether savings are being banked by artificially manufacturing

underutilisation of approved plan funds NDIS participants have already been deemed entitled to but cannot use effectively, or with confidence valid claims will be paid out?

Recommendation 10

Any statutory ethics framework must encompass the unique ethical risks and vulnerabilities arising from implementation of behavioural insights, choice architecture, nudging, and associated integrity ‘risk profiles’ are accountable to disability-related needs, including suicide risks relating to specific cohorts and intersectional risk factors.

This should incorporate a positive duty of care on the Agency.

Reimagining integrity measures through a principles-based, CRPD-conforming approach to Section 10

The Data Lake, and the inputs and outputs that matter

  1. Alternative use-cases for Crack Down on Fraud infrastructure, versus the blunt instrument approach which displaced individual considerations with operationalised Section 10 lists, bring me to the data lake, and some of the questions I raised during a PRG session back in September 2024 (shortly before we, along with the public, saw the final Section 10 instrument for the first time).

  2. If the NDIS were a system of public expenditure where individual needs were irrelevant, the data lake and other approaches to ensuring compliant claiming would be the perfect solution.

  3. But that is not what the NDIS was originally intended to be, and no amount of “discipline” in an attempt to police the manner in which disabled lives can be lived if one requires support to live ones disabled life, based on which supports are permitted or prohibited, can make it so.

  4. Imagine this, though: A data lake which links our goals to the international obligations our funding is intended to implement in real-time, as soon as we update them using self-service features in the app. It pulls in our reports, any peer-reviewed research relevant to our disabilities, and any relevant environmental features, remoteness factors, or other information which differentiates our needs.

  5. The lake allows comprehensive visibility of providers used by participants who take up the self- directed registration category in the form proposed by the Registration Taskforce, without limiting their options to ‘disability’ providers, and connects valid disability uses of mainstream or local solutions to our evidenced disability support needs.

  6. The lake verifies our claims and “but for my disability” justifications against a truly holistic set of data points which place the individual, their disability impacts and the golden thread to their goals and aspirations at the centre of an evidence-based matrix which substantiates the robust link to the claimed support as a tailored and flexible response to their individual needs.

  7. The lake allows the person to report and monitor outcomes themselves, then turns pooled qualitative and quantitative outcomes data into structured, meaningful, Scheme-level data substantiating the “nature” and “extent” of the social and economic value generated by each dollar

Cat Walker 26 March 2026 Page 23 of 26

Section Heading

of Scheme expenditure – while enabling the flexibility to develop innovative, highly individual solutions to our support needs which eliminate the waste of poorly tailored supports and deliver genuine value for money.

  1. For the first time, data could be extracted which captures the diversity of multiple disabilities and the ways they manifest in different contexts, and the outcomes we report, using the data points that make a real difference in our lives, and contributes to emerging evidence of what works, for whom.

  2. Above all, a reimagined business case for the data lake could be the key to a principles-based framework which best supports sustainability and integrity whilst capturing outcomes that matter.

    Recommendation 11

    A principles-based definition of NDIS support under Section 10 should be developed with the disability community through a parliamentary inquiry.

    Recommendation 12

    An inquiry to define a community-endorsed, principles-based definition of NDIS support could simultaneously develop an ethical compliance and outcomes framework to be operationalised through existing data lake infrastructure and self-service options in the participant portal and myNDIS app.

    This compliance and outcomes framework should be based on a holistic range of qualitative data points which link funded supports to CRPD obligations and meaningful, participant-led outcomes aligned to the objects and general principles of the NDIS Act.

    Recommendation 13

    Parliament should amend the NDIS Act to adopt the self-directed registration category in the form proposed by the Registration Taskforce led by Natalie Wade.

    This should not be confined to direct employment/services-for-one arrangements (as implied by subsequent consultation materials): The arrangements proposed by the Taskforce responded directly to exemplary community consultation and ensure integrity and visibility measures do not sacrifice individualised support.

What else is missing from the Integrity inquiry we need?

  1. The remaining integrity inquiry terms of reference I proposed in my Annual Report No. 2 submission,18 but which I have been unable to cover in detail and hope to address at a hearing – or preferably, through an extension to this inquiry which permits adequate submissions on and investigation of such important issues – include the following topics and the opportunity to address them through an inquiry lens grounded squarely in Australia’s obligations under the CRPD:

  • The Crack Down on Fraud program

  • The historical debt review, including public scrutiny of all associated materials and reports (such as the Kordamentha report)

  • Past and current debt determination and debt management policies

  • Participant experience of NDIA integrity activities and decisions engaging s46

  • NDIA media strategies and conformity of NDIA statements to media with Articles 3, 4, and 8 of the CRPD

  • Integrity measures required to address participant experiences of fraud, sharp practices or consumer law breaches which currently lack cross-jurisdictional resolution pathways or are beyond the capacity of the Commission to pursue

  • Conformity of participant-directed integrity measures with the CRPD, objects and principles of the NDIS Act, natural justice and other legal principles

  1. I have touched on elements of Crack Down on Fraud but have more to add arising from aspects I was consulted on, opportunities not taken up which would significantly support efforts to tackle non-compliance, and concerns identified which are beginning to play out.

  2. The minimal public information and immediate attempts to place “positive spin on alarming results” of the historical debt review should be cause for significant scrutiny by the Committee.

  3. The absence of published detail on debt determination policy and practice and the repeated avoidance of substantive engagement on pointed questions to senior executives and business teams throughout my time on the PRG is a concern I hope the Committee will explore carefully: I have significant further evidence to provide should it do so.

  4. Regarding f) and as noted in addressing the terms of reference, sharp practices and issues such as breaches of consumer law are beyond the scope of this submission. However, should I be invited to address the systemic perspectives I have focused on at a hearing, I am also willing to give evidence of my individual experiences in this area in-camera to support the Committee’s recommendations.

Recommendation 14

Streamlined resolution pathways for consumer law breaches in relation to individual participants should be developed in partnership with the ACCC.

This would resolve situations where the burden currently rests on participants to

pursue legal action even against registered providers if provider conduct does not rise to the level of fraud and/or the Commission is not sufficiently resourced to pursue legal action on the behalf of individual participants.

  1. I have given g) my best effort in a short timeframe, in the hopes this submission supports the Committee to view this entire inquiry through the lens of natural justice and the rights of participants under the CRPD.

  2. I nonetheless urge the Committee to extend this inquiry, which requires much more time to ensure the Committee can give it the care it deserves, including by consciously approaching each aspect through this lens and holding media and your fellow Parliamentarians accountable to Article 8.

Recommendation 15

The Committee should extend its inquiry into the integrity of the NDIS to a minimum of twelve months, with interim reports supporting more urgent recommendations, and should incorporate additional terms of reference such as those explored in this submission.

Recommendation 16

The Committee publish a media release addressing Government’s obligations under Article 8 and expectations of media in reporting on this inquiry and NDIS integrity more broadly.

Cat Walker 26 March 2026 Page 26 of 26