Disabilities, markets and rights: The limits of a marketised national disability insurance scheme

This article considers the compatibility of Australia’s marketised National Disability Insurance Scheme and the United Nations Convention on the Rights of Persons with Disabilities. The author recognises that market-based disability support is largely compatible with Australia’s human rights commitments but resists the idea that it is capable of serving the interests of all Australians with disabilities. Markets can and do fail, and market stewardship, the government’s proposed antidote, is only a partial solution. As such, we should embrace a pluralist disability support system that sees non-market mechanisms – like block funding – as playing an important role in disability support systems.

In October 2022, the federal government commissioned an independent review into the design, operation and sustainability of Australia's National Disability Insurance Scheme ('NDIS', 'Scheme'). Despite mounting evidence of problems attributable to the Scheme's market-based design, 1 the review's terms of reference signal that a marketised NDIS is here to stay. 2 This article asks whether the government is trying to tighten the screws of a train that is on the wrong tracks. I argue that while market-based systems like the NDIS are largely compatible with the Convention on the Rights of Persons with Disabilities ('CRPD'), 3 markets should not be seen as the sole mechanism for delivering disability support in Australia. In some circumstances, non-market mechanismslike block funding providers in remote locations where markets are likely to failwould better promote and protect the rights of individuals with disabilities.

Assessing compatibility
There are many standards against which a disability support scheme could be measured, including its aims, 4 the terms of a social contract, 5 frameworks drawn from disability scholarship, 6 and the CRPD. 7 This article nominates the last of these. There are practical reasons for doing so, including that the CRPD provides more explicit guidance than broadly expressed goals in government strategy documents. More to the point, non-compliance with the CRPD can amount to a breach of international law. Additionally, the CRPD derives substantial moral force from its status as a 'landmark achievement for inclusive law-making': 8 its terms were drafted byand reflect extensive consultation withindividuals living with disabilities. 9 Finally, Australia's federal, state, territory and local governments recently endorsed the NDIS as being delivered 'in line with Australia's commitments under the … Convention on the Rights of Persons with Disabilities', 10 suggesting that they, too, attach significance to whether the NDIS upholds the norms embodied in and created by the CRPD. In these circumstances, we should be interested in what the CRPD has to say.
If we can agree on the appropriateness of this yardstick, we must then consider what to measure with it. Our instinct might be to gather data on participants' experience in the NDIS. By this logic, if the NDISa market-based schemehad failed to promote and protect rights enshrined in the CRPD, it could be concluded that a market-based NDIS is incompatible with the CRPD. However, this approach risks conflating system design issues with implementation issues. It is possible that problems with the NDIS are not attributable to its marketbased design but to its imperfect administration. With this in mind, I propose to address whether market-based disability support schemes would likely advance the aims of the CRPD if implemented effectively. This might be called an assessment of 'theoretical compatibility'. While this method depends on defeasible assumptions about how markets and government welfare schemes behave, it enables usto the extent that we accept those assumptionsto assess compatibility between market-based systems, including the NDIS, and the CRPD without making attribution errors.
But to what extent could a finding of 'theoretical compatibility' hold its ground against overwhelming evidence that a scheme is failing in practice? Could we legitimately say that the scheme's design is 'theoretically compatible' with the CRPD notwithstanding this evidence, or should we instead revisit our theories about how market-based disability support systems behave? It depends on the extent and apparent causes of failure. So long as it remains plausible that the failings of a scheme are attributable to administration issues rather than its market-based design, there is no tension between a finding of theoretical compatibility and failure in practice; there is nothing contradictory about the idea of a good plan that has been poorly executed. If, however, an inquiry found that a market-based scheme was failing despite near-perfect implementation, we should wonder whether our assumptions about how markets behave are problematic.
I hasten to add that this is arguably a moot point in Australia: the NDIS has not been an overwhelming failure from a human rights perspective. The most recent quarterly report to disability Ministers stated that more than 77 per cent of NDIS participants over 25, and 70 per cent aged under 25, agreed with the proposition that 'the NDIS helped you have more choices and more control over your life'. 11 Participants also reported significant increases in participation in community and social activities. 12 These are not indicators of a flawed model incapable of promoting and protecting the rights of persons with disabilities. But nor do these figures prove its compatibility with the CRPD, so let me begin my assessment.
Is market-based disability support 'theoretically compatible' with the CRPD?

The promise of markets
Market-based disability support schemes typically involve a participant, or their agent, receiving a government 'funding package'. This package is then used to obtain goods and services offered by private providers who have a commercial incentive to meet the needs and preferences of participants. This differs from models in which the government runs services directlyan increasingly rare approach 13or funds non-government organisations to deliver services on its behalf (known as block funding).
There are obvious points of compatibility between market-based support and the CRPD. The Convention attaches great significance to the autonomy and choice of persons with disabilities, 14 which aligns with the central promise of markets: 'service, innovation, responsiveness and choice'. 15 Markets also encourage rejection of problematic assumptions that jar with the Convention's recognition of the 'diversity of persons with disabilities': 16   This is not to suggest that it has the unqualified support of all persons with disabilities, of course; but the CRPD arguably speaks with greater moral authority than a particular scholar or organisation. 10 See, eg, Department of Social Services (n 4) 5. 11 National Disability Insurance Agency, NDIS Quarterly Report to Disability Ministers (Q3) (Report, 2022) 43. These figures may, however, be slightly more positive than reality: the data was captured in the presence of NDIS staff, and it is plausible that some participants did not feel comfortable providing candid responses in that context. Even still, the scheme seems largely successful. providers of goods and services were to adopt the problematic 'vulnerability zeitgeist' of the 21st century, 17 which is full of 'dangerous new categorisations and binaries', 18 they would fail. Providers must understand and respond to participants' diverse needs and preferences to succeed. It may seem, therefore, that markets and the CRPD are naturally aligned, but there is more to the story. I will now explore how the goals of disability support systems can come into conflict with markets. I begin by imagining how a 'real' disability market might work to highlight tensions between market mechanisms and disability rights. I then consider whether 'tempering' markets through stewardship and regulation would adequately address these tensions while preserving the promises identified.

Delivering disability support through conventional markets
Let us imagine it is 2065, and the Australian government has decided to reform its disability support system. Guided by values like 'choice' and 'competition', it has changed the Scheme's name to the 'Real Disability Marketplace' ('RDM') and introduced four key policies: 1. Provide support to participants through lump-sum payments, which they can use however they like. 2. Implement tax incentives for providers entering the disability market. 3. Cut 'red tape' that creates barriers to providers entering and operating in the disability service market. 4. Remove intermediaries (ie, disability support agents) except where the participants clearly cannot make decisions without support.
This model, the fictitious government insists, will increase economic efficiency. It believes that providers will compete to offer more affordable services, and that everyone will benefit.
Sarah is experiencing chronic pain caused by a rare genetic condition and has obtained funding under the RDM. 19 She searches the internet for providers who can offer therapy and home adjustments. However, she is unsure which providers are most likely to provide highquality services. She opts for The Pain Management Company, which targeted her through a digital marketing campaign. Unbeknown to her, the company had decidedsome 18 months earlierto lay off its best practitioners and spend heavily on a digital marketing agency known for aggressive sales tactics. This was intended to increase its revenue and profit margin. The agency floods Google with fake reviews and creates a strong social media presence. The provider tells Sarah that its services are in high demand and that they may not be able to fit her in unless she enters a long-term arrangement, so she checks the provider's online reviews. Pleased by what she finds, Sarah enters into a sixmonth service contract. She is then assigned an inexperienced practitionerunder the old scheme, this practitioner would have been required to work under supervisionand is notified that her home adjustments will take several months to complete.
Meanwhile, James is recovering from a brain injury. Sometimes he can think clearly, while at other times he feels 'confused' and 'hazy'. As advised by his doctor, James would like to enrol in a vocational training course and re-enter the workforce. However, he cannot use a computer well enough to access information about education providers or job opportunities. James gives up on the plan until a neighbour, Tony, offers him assistance. However, as far as Tony can see, there are no on-campus training providers within 400 km of their regional town.
The hypothetical RDM scheme highlights several ways market-based schemes can fail to 'promote, protect and ensure the full and equal enjoyment of all human rights and fundamental freedoms by all persons with disabilities'. 20 First, in relation to Sarah's situation, it is arguable that her right of access to information about support has been abrogated. 21 This rests on the view that the government's duty to provide information extends to suppressingor at least 'cutting through'misinformation. For the avoidance of doubt, an expansive interpretation of the right does not rely on the false idea that people with disabilities are always more 'vulnerable' to misinformation; misleading marketing occurs in all sectors and affects all market participants 22 . The material point is that the stakes are often higher when it comes to disability support. There is a substantial difference between, for example, a person who books a bad restaurant based on fake reviews and a person whoalso based on fake reviewslocks themselves into a long-term contract for services needed to work and socialise (such as therapy).
Some might argue that markets will address the issue of imperfect information of their own accord. For instance, start-up companies like Provider Choice 23intermediaries that claim to help NDIS participants make optimal decisions and manage their plansmay address information asymmetry between providers and participants. While intermediaries could indeed provide helpful advice, there is a risk that they, too, could be conflicted (ie, paid to promote and endorse providers) in an underregulated market. Further, it seems implausible that intermediaries alone could address the growing risk of misinformation driven by competition and digital marketing. The 'red tape' that has been cut by the hypothetical government would have offered more robust rights protection in these circumstances.
One might also doubt whether the state has discharged its obligation to 'adopt all appropriate legislative, administrative and other measures for the implementation of [rights]', 25 which extends to enabling 'full and effective participation and inclusion in society'. 26 For instance, it is difficult to see how it is appropriate and consistent with CRPD obligations to lower the standards expected of practitioners. It is also difficult to justify the dissolution of an audit systemlike the NDIS Quality and Safeguards Commissiondesigned to detect the excessively profit-centred operating model of The Pain Management Company.
The RDM also failed James, who had difficulty conducting research and making decisions about his support package. Perhaps the government would argue that the appointment of an agent is unnecessary, as Tony's intervention shows. However, not everyone can rely on friends or family. Besides, under the CRPD, the governmentnot a neighbouris obligated to 'provide accessible information to persons with disabilities'. 27 Supply failure is another problem experienced by James. In a perfect market, the interests of participants and providers would fall into alignment: the interest of the participant is to obtain the goods and services required to achieve their goals, such as employment; the interest of the provider is to turn a profit; and the only way for the provider to turn a profit is to deliver affordable goods and services that will help the participant meet their goals. In reality, though, markets do not always conform to this model. 28 For James, living in a regional location prevented him from accessing services that would be relatively easy to obtain in an urban marketplace (in his case, education). 29 This discrepancy conflicts with Australia's commitment to making training and other opportunities accessible 'in both urban and rural areas'. 30 To assert the theoretical compatibility between schemes like the RDM and the CRPD, it would be necessary to accept implausible assumptions about markets (such as that they can always be relied upon and that all market participants engage with them in the same way). I will now consider whether quasi-markets promise greater theoretical compatibility.
Quasi-markets: All the pros with no cons?
One could be forgiven for thinking that the Australian government intended to create a scheme akin to the RDM. In a 2016 report, the National Disability Insurance Agency (NDIA) 31 indicated that it did not want the NDIS to become 'another government-controlled welfare scheme' 32 and argued in favour of marrying disability support with values like 'choice', 33 'consumer power', 34 'growth potential' 35 and 'incentives'. 36 However, the NDIS is more accurately characterised as a 'quasi-market'. Like conventional market schemes, quasi-markets purport to put individuals 'in control of their own service needs by leveraging market-mechanisms' 37 and enable a 'more empowered relationship with the state'. 38 However, in quasi-markets, governments are more proactive in '[addressing] market deficiencies, such as thin markets, market gaps or other market failures' 39 (initiatives usually described as 'market stewardship'). Governments also play a more proactive role in regulating conflicts of interest, 40 'facilitating exchange of [information] about provider quality', 41 and providing 'service intermediaries, such as plan management, and support coordination agencies'. 42 An average consumer would be surprised to find a government employee at the entrance of the Apple Store to explain Apple's products and negotiate with its salespeople on behalf of consumers, 43 but not so when it comes to goods and services available on a quasi-market. Moreover, the NDIS 25 CRPD (n 3) art 4 para 1(a). 26 Ibid art 3 para (c). Agency relationships are common in conventional markets. Consider, for instance, the relationship between real estate agents and vendors. However, this kind of agency relationship differs from the hypothetical agent in the Apple Store. Whereas real estate agents perform tasks that their clients lack expertise or time to perform themselves, and do so at the behest of the principal, the hypothetical agent is paid by a government body to protect the interests of an ostensibly vulnerable consumer. The former is common in conventional markets, while the latter is not. Quality and Safeguards Commission provides an accreditation framework, an audit process, and a complaintshandling process. 44 It may seem as though market stewardship and regulation, if implemented effectively, could reduce the risks created by markets while preserving their benefits. Perhaps individuals could benefit from competition and choice without being exposed to the issues Sarah and James encountered, including low-quality service, misrepresentations and dependence on family or friends in place of government-funded agents. However, it remains to be considered whether the central pillar of quasi-market theorythe idea that market deficiencies can be addressed through 'stewardship'holds water. 45 Market stewardship is often described as having three primary aspects: 1. Monitoring: 'observing the NDIS marketplace and assessing whether it is achieving its outcomes'. 46 2. Facilitating: 'actions that directly influence demand in the NDIS marketplace and indirect actions to improve the functioning of the NDIS marketplace', including 'setting prices' and 'facilitating supports diversity of supplier business models'. 47 3. Commissioning: 'direct sourcing of supports or establishment of preferred provider arrangements supported by controls and "rules" that must be complied with to participate in the NDIS marketplace'. 48 In regional locations, for instance, the government might establish a 'provider of last resort'. 49 While these proposals sound attractive at first glance, the issue of how governments should steward markets remains a vexed question. 50 When analysing market failures within a market paradigm, governments find themselves in the midst of an ideological conflict: it is difficult to intervene in a marketto 'steward' a marketwithout losing the benefits of having a market. Carey et al go further, suggesting that this tension is not only related to the fear of 'introducing rigidity into the market through government intervention' and thereby 'undermining the supposed gains of public sector markets' but to a 'belief in the vision of a fully de-regulated disability service market'. 51 This tension makes general formulationslike the NDIA's vision for market stewardshipmore contentious than they appear at first glance.
Consider, for example, what the government should do when a service provider withdraws from the market, as Anglicare did in Tasmania in January 2023. 52 Should it put faith in the market to address the needs of the 420 affected clients, 53 block fund a new provider of last resort (that is, embrace a non-market mechanism), or offer the provider an incentive to stay? On close inspection, even the government seems conflicted. In 2011, the Productivity Commission contemplated an amalgam of market-based support and block funding -'[funding] service organisations to provide support to a number of people' 54as needed. 55 This idea was later endorsed in the 2016 NDIA report. 56 Yet the term 'block funding' andmore importantlythe pragmatic attitude it signifies is conspicuously absent from the most recent NDIS Rural and Remote Strategy. 57 The strategy appears to express a dogged desire to make markets succeed rather than accept their limitations. It focuses on 'working with providers to identify incentives to deliver supports' (which includes 'use of remote travel provisions'), 'setting up processes to grow supply', and 'recognising the need for greater innovation'. 58 In places where one might expect the government to unequivocally state that it will not hesitate to block-fund organisations to guarantee a base level of support in fulfilment of an obligation to uphold the right to access goods and services, 59 it instead commits to taking 'a considered approach to determine how to intervene, and with what levers, to support market development without negatively impacting on other sectors.' 60 Apparent aversion to initiatives that depart from market principles is also reflected in the latest NDIA report on the experience of NDIS participants in remote locations. Discussing solutions to 'thin markets', the NDIA once again seemed determined to make markets succeed, stating that it would continue to 'publish data and insights to inform the market', 'improve price control arrangements to enable providers to recover the costs of delivering supports in 44 remote locations', 'undertake provider communication and education activities' and 'manage provider exits for continuity of supports for participants'. 61 Given the NDIA's finding that 'the proportions of participants living in Remote and Very Remote areas who said [the] NDIS has helped were notably lower than other areas', 62 it is surprisingperhaps even startlingnot to see a concession that quasi-markets are failing in certain locations and that the government is prepared to focus its attention and resources on other disability support models.
If the term 'market stewardship' denotes a system in which the government is prepared to embrace a mixture of market and non-market mechanisms to deliver disability support, then we can be optimistic about the compatibility between quasi-markets and the CRPD. But if the term now denotesas seems to be the casethe means by which governments attempt to make markets succeed at all costs, it seems improbable that quasi-markets will promote and protect the rights enshrined in the CRPD. The problem is that while governments are figuring out how to make markets work, rights violations like those surveyed in the hypothetical scenario above can and do occur. 63 The solution to failed markets is not always a better market, and reluctance to depart from market paradigms can come at the cost of fulfilling human rights obligations.
One might be reluctant to enthusiastically embrace non-market mechanisms, like block funding, that the Productivity Commission decried as ineffective. 64 Yet even the Commission recognised the need for pluralism. 65 Framed in terms of the CRPD, the critical point is that even if we continue to pursue market-based models that provide choice and harness the benefits of competition, 66 we must show greater preparedness to concede their limitations.

Concluding remarks
The concept of a marketised NDIS is not fundamentally flawed; indeed, it has an important role to play in promoting and protecting the rights of Australians with disabilities. But it seems both unnecessary and futile to rely on market mechanisms if we are going to fulfil the obligation of upholding the rights of all Australians with disabilities. The government must be prepared to break its apparent ideological commitment to the idea that a market-based system can be all things to all people.
It is a live question whether the burden of providing diverse forms of support should rest on the shoulders of the NDIS. As the federal Minister for the NDIS reminded us recently, the Scheme was never meant to be the 'only lifeboat in the ocean'. 67 Perhaps other governmentslocal or stateshould play a more prominent role in creating safety nets that run in tandem with the NDIS. In principle, it does not matter whether disability support originates from the NDIS or a collection of federal, state and local government systems running in parallel; what matters is whether Australia lives up to its human rights commitments. Until legitimate alternatives to the NDIS emerge, ideological commitment to market-based support may have to soften.