On 22 April 2026, Mark Butler stood at the National Press Club and announced the most significant overhaul of the NDIS since the scheme began. Mandatory registration is being expanded to cover most of the market. Plan management, support coordination and Supported Independent Living will all be commissioned over the next few years. The justification, repeated throughout the speech, was integrity. The NDIS, in his words, being treated as an "ATM for shonks, grifters, fraudsters and crooks."

Some of these reforms are the right policy. Some are the right policy for the wrong reasons. And one of them, commissioning SIL, is the wrong policy entirely. It is the most dangerous part of the package for participants who actually live in supported accommodation.

‍

Mandatory registration: the right call, but stop selling it on fraud

The push to make registration mandatory has been gathering momentum for a couple of years now, and the loudest argument for it has always been fraud. Headlines about rorts, dodgy operators, and billions lost to bad actors have made registration feel like the obvious fix. Lock the gate, and the cowboys can't get in.

I support mandatory registration. I think it should apply to every form of direct support. But the fraud narrative is one of the worst arguments anyone could be making for it, and mainstream coverage that leans on it is doing the sector a disservice, including the participants it claims to be protecting.

Registration does not stop fraud. A registered provider can rort the scheme just as effectively as an unregistered one. Probably more effectively, because they are better at the paperwork. Registration is not a quality control mechanism. It is a visibility mechanism. And visibility is genuinely valuable, but for completely different reasons than the fraud narrative suggests.

I'm a pharmacist. There is no such thing as an unregistered pharmacist in Australia. If you are dispensing medication, AHPRA knows you exist, and that is the entire point. AHPRA cannot suspend, ban or take action against a pharmacist it does not know about. The regulator's power to remove bad actors only works if everyone is in the system in the first place.

That is the real argument for mandatory NDIS registration. Right now, the NDIS Commission can deregister/ban a registered provider, but its tools for dealing with unregistered ones are limited. You cannot remove someone from a system they were never in. Mandatory registration doesn't guarantee quality; it makes accountability mechanically possible. It doesn't deliver good services, but it creates the conditions in which quality can be prioritised, monitored and enforced.

This distinction matters because if we sell registration as the cure for fraud, we are going to be disappointed. Or worse, we are going to be tempted to pile more compliance onto registered providers as proof that registration is "working", and that creates a different problem entirely.

‍

Why participants are walking away from registered providers

There is a question mainstream coverage rarely asks: why do so many participants actively prefer unregistered providers?

The honest answer is that a lot of registered providers have become compliance robots. They are so focused on demonstrating risk mitigation, ticking boxes, and protecting themselves from regulatory scrutiny that they have stopped feeling like services run by humans for humans. Shifts get rostered around what looks good on paper. Documentation takes priority over relationships. Workers churn because the job has been engineered to be replaceable rather than relational.

Independent and unregistered providers, the good ones, often deliver something completely different. They show up. They actually know the participant. They flex, change and adapt when life gets messy. The relationship matters more than the case note.

When you ask those independent workers and small operators what they are worried about under mandatory registration, this is what they say. They do not want to become the robots they got into this work to avoid. They have watched what compliance culture does to good people, and they are scared of it.

That fear is legitimate, and it is addressable, but only if registration is designed as a floor rather than a cage. The Commission has to be honest about the trade-off and proactive about not killing what works. If the regulatory regime that follows registration is shaped by the fraud panic, we will end up with the worst of both worlds: a compliance-heavy sector that still has fraud, and a missing layer of small, human, flexible providers who quietly left.

‍

Commission what shouldn't be a market: plan management and support coordination

Two of Butler's commissioning announcements are correct, and overdue.

Plan management should be commissioned. Plan management, stripped of the marketing, is paying invoices. It is an administrative function. The argument that participants need a competitive market of providers to choose between to process their claims has not aged well. We have automation now. We have AI now. We have the technical capacity to process claims accurately, quickly and cheaply at scale. Instead, the sector pays a meaningful slice of every plan-managed budget for what is, for most participants, invoice processing with a friendly customer service layer (sometimes). That is not a market delivering value, it's a market extracting it.

Support coordination is harder, but the same direction is right. Support coordination was supposed to be the navigational layer of the scheme: the people who would help participants understand their plans, find good providers, and exercise choice and control meaningfully. In practice, the quality has been wildly inconsistent since day one. A lot of support coordinators have inadvertently steered participants toward bad providers, missed funding opportunities they should have spotted, or simply not understood the scheme well enough to do the job. There has been real harm done. Not because most coordinators are malicious, but because the role was never properly defined, trained or quality-assured. A commissioned function with proper standards, accountability and clear conflict-of-interest rules is a better answer than the current market.

These two reforms also remove a structural conflict that has corroded the scheme since it began: intermediary services owned by, or commercially aligned with, the providers they are supposed to help participants choose between. You cannot run a participant-side function as a profit-seeking arm of provider-side businesses and expect it to advocate for the participant. Commissioning these layers separates them.

‍

Don't commission SIL. It's a return to what the NDIS was built to replace.

This is where the package goes seriously wrong.

Supported Independent Living is fundamentally different from plan management or support coordination. Plan management and support coordination are administrative and navigational. They are back-office functions that sit between the participant and their supports. SIL is the support. It is the actual delivery of intimate, personal, daily care to people in their own homes. It is help with showering, cooking, getting dressed, going to bed. It is who walks through your front door at 6am and who is in your kitchen at 9pm. It is one of the most personal, relational, high-stakes services any human being can receive.

Butler's plan is to commission SIL through a panel of "accountable, quality providers" from which participants can choose. The framing is that the SIL market "isn't working." That framing has some truth to it. SIL is concentrated, conflicted, under-regulated and difficult to exit. But the response is wrong, and the people most exposed to it are the participants who already have the least leverage.

A commissioned SIL panel will work like this in practice. The providers who win panel positions will be the ones with the resources to compete for tenders: the legal teams, the corporate infrastructure, the compliance departments, the policy people. That is not a description of a small, person-centred provider. That is a description of large institutional providers. The same large institutional providers that the NDIS was specifically designed to break the monopoly of.

We have done this before. The pre-NDIS system was block-funded supported accommodation delivered by a small number of large rigid providers. There was no meaningful choice. There was no real control. People with disability lived where they were placed, with co-residents they did not choose, supported by workers they did not select, in a system they had no leverage over. The Disability Royal Commission's findings on violence, abuse, neglect and exploitation in disability services were, in large part, findings about exactly that model.

The central insight of the NDIS, the reform that made it different from everything that came before, was that individual budgets gave participants the ability to leave a bad provider and take their funding with them. That single mechanism is what gives a person with disability any structural leverage at all over the people who come into their home every day. Commissioning SIL removes that leverage in the area where it matters most.

If the only providers you can choose from are the ones the government has put on a panel, and those providers are the same large institutional operators with the same culture and the same incentives, you do not have choice and control. You have the appearance of choice and control inside a closed shop. That is the system the NDIS was built to dismantle, and it is the system Butler's reform would re-create, at exactly the point in a participant's life where the stakes are highest.

There is a coherent answer to the genuine problems in the SIL market, and Butler already has it on the table: mandatory registration, proper Practice Standards, real audits, worker screening, serious enforcement powers for the Commission and that super expensive PACE system the NDIA still haven't effectively rolled out. That builds visibility, raises the floor, and gives the regulator the ability to remove bad actors, without surrendering the participant-side leverage that the whole scheme was built on.

Butler said in the Q&A after his speech that choice and control "has got to be at the centre of this scheme." He is right. Commissioning SIL is incompatible with that statement.

‍

The inefficiency loop no one wants to name

There is one more piece of this that the fraud narrative completely misses, and it sits underneath the entire commissioning push. A large portion of waste in the system is not theft. It is structural inefficiency, and a meaningful share of it is generated by large registered providers forcing participants to fit their service model rather than adapting their model to the participant.

The pattern goes like this. A participant has a plan. The plan, as written, does not quite fit how the provider operates. Instead of changing the rostering, the support model, or the way they do business, the provider pushes for the plan to be resubmitted to the NDIA so it lines up with what they can deliver. The NDIA reassesses. The plan gets reshaped. Sometimes funding rises to make the provider's model viable. Sometimes the participant ends up with a service that is a worse fit than what was originally designed.

Why do large providers not just adapt? Because the regulatory burden of changing internal practice is enormous: rosters, training, documentation, incident frameworks, audit prep. It is cheaper and operationally easier to push the problem back onto the participant and the planner than to redesign the service.

So we end up in an infinite inefficiency loop. Plans get rebuilt to suit providers. Providers get bigger and more rigid. Compliance pressure goes up. Smaller providers who could have flexed and adapted get squeezed out. Participants are left with fewer real options. And then we look at the cost growth, point at it, and call it fraud.

It is not fraud. It is a system rewarding rigidity. And commissioning SIL will accelerate every part of it.

‍

Get the policy right. Drop the framing.

Mandatory registration is the right policy. Commissioning plan management is the right policy. Commissioning support coordination, done carefully, is the right policy.

Commissioning SIL is not.

The first three reforms strengthen the scheme by making providers visible, removing conflicted intermediaries, and raising the floor on quality. The fourth removes the one mechanism that the entire NDIS was built around: the participant's ability to leave. It hands the most personal, most relational, highest-stakes part of the system back to the institutional providers the scheme was created to break free from.

Butler is right that the SIL market isn't working. He has the right tool to fix it sitting on his desk: registration, standards, audits, enforcement, multi million dollar systems. He should use it. He should not use the fraud narrative to justify dragging participants back into the closed-shop world the NDIS replaced.

Get the policy right. Drop the framing.

‍

‍

‍