16/07/2026
A long but insightful read......
I have recently discovered something deeply inconvenient.
Australia’s government is actually… quite good.
According to the World Bank’s Worldwide Governance Indicators, Australia scores extraordinarily highly compared with other wealthy countries.
Sweden: 572
Australia: 563
Netherlands: 561
Singapore: 550
Canada: 549
Germany: 549
United Kingdom: 526
United States: 472
These scores look at things like government effectiveness, political stability, regulation, rule of law, corruption and whether citizens get any meaningful say in how they are governed.
Australia scores highly across nearly all of them.
We have a professional public service. We collect taxes effectively. Our institutions are stable. Our elections are peaceful. Government ministers generally leave office through voting rather than being chased out of the palace by an angry crowd carrying AK47s.
Corruption exists, obviously, because humans are involved. But internationally speaking, Australia is relatively clean, stable and competent.
The United States is the obvious outlier here mostly because it scores much lower than the other countries on political stability, rule of law, institutional trust and corruption.
Sweden, meanwhile, appears to have completed government on expert mode.
So this raises an awkward question.
If Australia is so well governed, why do so many things still feel completely cooked?
How can we have world-class government effectiveness and also housing that requires either wealthy parents, three investment properties or a minor role in the mining industry?
How can we have a sophisticated healthcare system while carers report appalling levels of exhaustion and distress?
How can we have an extremely capable bureaucracy while disability families spend their lives filling out forms, coordinating providers and trying to locate a support worker who has not suddenly moved to Queensland?
The answer is that these governance rankings measure whether government is capable of implementing policy. They do not measure whether it always chooses good policy.
A government can be extremely competent at carrying out a questionable idea. In fact, competence may allow it to carry out the questionable idea with tremendous efficiency.
This is where the NDIS becomes so interesting.
The NDIS is often presented as evidence that Australian government is hopeless. But I’m starting to think it may be the exception that proves the rule.
Australia is generally very good at running large, broad, rules-based systems.
Medicare. The PBS. The Age Pension. HECS. Compulsory super. The tax system, which will find your forgotten $27 of bank interest from 2019 even if you have personally lost all memory of ever opening the account. These systems work because government largely sets the rules, collects the money and pays benefits or funds established services.
The NDIS asked government to do something very different. It asked government to invent an entirely new market for highly personalised disability care.
Government became the funder, planner, regulator, price setter, fraud investigator and market designer.
Families became employers, recruiters, schedulers, contract managers, quality-control inspectors and emergency workforce planners.
People with disability were described as “consumers,” as though they were strolling through a pleasant marketplace comparing the relative merits of occupational therapists.
The theory was choice and control.
The reality can be spending three months trying to find someone willing to work on Friday afternoon. Why is it that no one is ever available on a Friday??
Care is not like buying groceries.
Markets work best when buyers understand the product, quality is easy to compare, suppliers can enter and leave easily, and customers can simply walk away from a bad provider.
Disability care often has none of those features.
Some participants cannot communicate whether a worker is treating them well. Families may be exhausted and desperate. Specialist workers are scarce. In many regional areas, there may be one provider or no provider.
You cannot exercise meaningful consumer choice between “the only person available” and “nobody.”
The NDIS also managed to combine central planning and free markets in a way that preserved the drawbacks of both.
Prices are regulated. Plans restrict spending. Provider entry is partly controlled. Government determines what counts as reasonable and necessary. Not in any kind of consistent way, but in a let’s see where the VIBES are at this time sort of way. But services are still expected to appear through market competition.
So it is not quite a public system. It is not quite a free market.
It is a sort of policy centaur. Half government program, half commercial marketplace, galloping around generating invoices.
The scheme also grew much faster than the institutions around it.
The workforce was not ready. The regulatory systems were not ready. The IT was not ready. Mainstream services were not ready. I suspect several government departments are STILL not emotionally ready.
Meanwhile, states, schools, hospitals and community services increasingly discovered that whenever they failed to provide something, they could point vaguely toward the NDIS. The NDIS became the place people went because everywhere else had already said no.
It was originally meant to be one part of a broader disability support system. Instead, it gradually became the system.
Michael is a good example.
We did what everyone says you should do. We enrolled him in a mainstream public school. The school tried. The teachers genuinely cared. But they simply couldn’t keep him safe or provide the level of support he needed. Eventually we had to withdraw him.
Here’s the interesting question. Was that an NDIS problem? Not really.
It was an education problem. If there had been enough genuinely specialised school places for children with very complex needs—or if schools had been resourced to provide the level of support those children require—Michael would have been learning at school instead of being at home.
But there aren’t many places where a child with Michael’s level of disability can receive that intensity of support all day. A child who needs near-continuous supervision is extraordinarily difficult for any school to accommodate. Families often end up filling the gaps themselves, with support workers funded through the NDIS helping around the edges because the education system simply can’t meet all of the child’s needs.
That’s what happened across the disability system more broadly.
The NDIS was designed to sit alongside health, education, housing and other services. Instead, whenever one of those systems couldn’t quite cope, the question gradually became:
“Can the NDIS help?”
And often the answer was yes.
Eventually the NDIS wasn’t just funding disability supports. It was quietly compensating for the limits of lots of other systems as well.
Then everyone seemed surprised by how expensive it became.
This is where Sweden offers an interesting comparison.
Sweden also uses private providers in some areas, and its system is certainly not perfect. But much more disability support is organised or directly provided by local government. The municipality carries more responsibility for making services exist.
The family is less likely to be handed a budget and told, “Congratulations on your empowerment. You are now the unpaid chief executive of a small disability-services company.”
Imagine running schools the way we run parts of disability support.
Every parent receives an individual education plan. Thousands of education companies compete. Parents recruit teachers separately. They compare hourly rates. They negotiate cancellation policies. They find someone else when the maths provider goes bankrupt.
If nobody is available, the parent teaches Year 8 algebra themselves while also appealing the funding decision.
We do not generally organise schools this way because we recognise that education is a public service requiring infrastructure, workforce planning and government responsibility.
Yet in disability, we often assume the family should become the project manager.
Parents recruit workers. Coordinate rosters. Train staff. Negotiate with providers. Chase invoices. Organise therapies. Manage crises. Fill every gap.
And when the whole structure collapses at 4:30 on a Friday afternoon, the family is the provider of last resort. And a lot of the time saying ‘family’ is another way of saying: the mother.
I recently read a comment from a woman whose mother had received excellent community palliative care. What she remembered most was not simply the shower chair, wheelchair or overnight nursing. It was that someone else absorbed the mental load.
You called one number. Someone worked out what was needed. The system organised itself around the family, rather than requiring the family to organise the system.
I wonder whether this is one of the great lessons we should take from Sweden.
A good disability system does not merely hand families funding.
It creates actual services. It trains workers. It maintains backup capacity. It coordinates care. It notices when a family is drowning before asking them to complete a 46-page form proving they are wet.
Australia’s problem may not be that government is incapable. The international evidence suggests almost the opposite. Australia has a highly capable state.
The harder question is what we ask that state to do.
Do we want government simply to finance care? Or do we want it to build the invisible architecture that makes care possible?
Michael Sandel would probably ask: what is government for?
And I think that is the real question here.
Because Australia is apparently capable of doing difficult things once it decides they matter.
The problem is not always capacity. Sometimes it is priorities. Sometimes it is institutional design. And sometimes it is the strange belief that if we call a person with profound disability a “consumer,” and possibly click your heels three times, a functioning market will magically appear around them.
