Data-driven investigation

They're misleading you about autistic kids

Autism is 42.6% of NDIS participants - but only 21.4% of spending. Children aged 0-8 - the early intervention cohort the government wants to cut - account for just 2.5% of total NDIS spending.

59,848

Kids aged 0-8

$1.23B

Actual spending (0-8)

2.5%

Of NDIS spending (0-8)

Autism 0-8 is the largest cohort in the 0-8 age band by total budget - but that's driven by volume, not per-child cost. Each child costs 65% less than the NDIS average.

Power in Numbers is data professionals and NDIS families. Every number on this site traces to official NDIA data. powerinnumbers@proton.me

Scroll for receipts

The Big Mislead

"40% of NDIS participants have autism"

This framing conflates participant share with spending share.

✕ Misleading

"40% of NDIS participants have autism"

Butler said 40% of participants have autism. True - except they only cost 21.4% of spending. And the 0–8 year olds they keep talking about?

2.5%

That's what autistic kids 0–8 actually cost. 59,848 children. $1.23B of $48.96B.

Source: NDIS Participant Numbers and Plan Budgets, Dec 2025 · NDIS Payments Data, Dec 2025 · dataresearch.ndis.gov.au

The Math They Won't Show You

Cut the kids. Save nothing.

Truth is: Save $616M now, pay $2B later.

Current Spending

$1.23B

Cut 50% of Kids

Save $616M

% of Total NDIS

1.26%

Future Adult Cost *

$2B

* Based on 3.3× cost increase from 0–8 ($30k) to 25–34 ($100k) age band. NDIS Plan Budgets, Dec 2025.

✓ Brutal Reality

Cutting 50% of autistic kids saves 1.26% of NDIS budget

The claim: cutting kids fixes the budget. Reality: cut 29,924 children and you save 1.26%. That's not reform. That's noise.

Source: NDIS Payments Data, Dec 2025 · Autism 0-8 total payments $1.231B ÷ 2 = $616M · Total NDIS payments $48.96B · $616M / $48.96B = 1.26%

Here's the real bill: Cut kids now. They show up later - hospitals, psych, 24/7 support care. Research shows early intervention returns between $4.10 and $11.30 for every $1 invested (Synergies/AEIOU). Cutting $616M in early intervention = increased future adult support costs.

Sure, the number depends on the cohort - but imagine only 50% of those cut kids come back as adults. That's still $1.5B per year. Saved $616M. Still pay $1.5B later. Try not to think about what happened to that 50%, by the way.

They Said What?

Politicians vs. Real Numbers

Every major claim, fact-checked.

Mark Butler 21 Aug 2025, 2GB Radio Source →

"Seven in 10 people who joined the service in the past year did so for autism. One in six boys is now on the NDIS."

True, but Headcount isn't cost. Autistic kids 0-8 = only 2.5% of spending

◆ NDIS Payments Data & Participant Numbers, Dec 2025 · dataresearch.ndis.gov.au

Mark Butler 20 Aug 2025, National Press Club Source →

"About 23 per cent are younger than nine, and about 40 per cent have a primary diagnosis of autism."

This stat is used to blame kids for adult costs.

📊 The Numbers

▸ True: 40% of all participants have autism (all ages)
▸ Also true: autism across ALL ages accounts for 21.4% of actual spending - half the 42.6% participant share
◆ NDIS Payments Data & Participant Numbers, Dec 2025 · dataresearch.ndis.gov.au

Bill Shorten Dec 2023 Source →

"Almost half of participants in the Scheme are children."

Half the people. Not half the money. Math is hard.

📊 The Numbers

▸ Autism kids 0-8: $30,000 average allocated budget - actual spending $20,574 (28% goes unused due to provider shortages)
▸ Overall NDIS average: $85,000 per person
▸ Kids cost 65% less than average
◆ NDIS Participant Numbers and Plan Budgets, Dec 2025 · $30k autism 0-8 vs $85k all participants · dataresearch.ndis.gov.au

Follow The Money

Kids aren't expensive. Adults are. Assisted living is. 24/7 care is.

What these kids will cost as adults after they're kicked off NDIS doesn't tell the full story either.

Kids 0-8 (allocated budget)

$30,000

Adults 25-34

$100,000

Increase

3.3×

✓ They Know This

Cut kids now = pay 3.3× more later

Source: NDIS Participant Numbers and Plan Budgets, Dec 2025 · Autism avg budget: 0-8 $30k, 25-34 $100k, 65+ $156k · dataresearch.ndis.gov.au

The actual cost breakdown: Ages 0-8 average $30k allocated ($20.5k actual spending). Ages 19-24 average $81k. Ages 25-34 average $100k. Ages 65+ average $156k. You want to cut the cheapest cohort? Explain that math.

The Appeals Racket

$34.8 million fighting disabled people. Losing.

When plans get cut, participants appeal. The NDIA spends tens of millions on AAT matters to fight them. And loses. Their solution? Remove the right to appeal total plan funding.

NDIA spent on AAT matters

$34.8M

Cases changed in participant's favour

73%

NDIA's fix

Remove right to appeal total funding

✕ Exposed

They don't lose gracefully. They remove the game.

In 2020-21, the NDIA spent $34.8 million in total on AAT matters - fighting participants at the Administrative Appeals Tribunal. Of that, $17.3 million went to external law firms. The remainder covered internal legal staff and associated costs. The result? 73% of plan-related cases were changed in the participant's favour (NDIS Quarterly Report Q3 2024-25). They spent $34.8 million to be told they were wrong - seven times out of ten.

They lost 73% of cases at the tribunal. So they changed the law.

The ART can't change your plan funding anymore. It can only send you back to the NDIA for another assessment - using the same tool that got it wrong the first time. The needs assessment itself (s 32L) isn't even a reviewable decision under s 99. You can't challenge the thing that decides your funding.

The NDIA confirmed at Senate Estimates this saves them money. Of course it does. It's cheaper when people can't fight back.

✓ Follow the money

The cost is climbing - because the decisions are getting worse

NDIA external legal spend: $18.4M (2018-19) → $29M (2019-20). That's a 58% increase in one year. Not because participants got greedier. Because plans got worse, more people had to fight, and the NDIA hired more lawyers instead of making better decisions. Every dollar spent on lawyers is a dollar not spent on therapy, equipment, or support workers.

Senate Estimates Dec 2025 Source →

Under the new planning model, participants will no longer be able to appeal the total amount of plan funding to the ART.

73% of plan-related cases were changed in the participant's favour. Their response: remove the right to appeal total plan funding.

Sources

▸ $34.8M total AAT spend: FOI disclosure, Dec 2021 (Right to Know)
▸ $17.3M external legal firms: FOI disclosure, Dec 2021
▸ 73% of plan-related cases changed in participant's favour: NDIS Quarterly Report Q3 2024-25
▸ Appeal rights removed: Senate Estimates, Dec 2025
▸ $29,899 per 2.5-day hearing: IER Evaluation Report, Oct 2023

"Unsustainable"

Denmark spends 4.7%. We spend 1.7%. We're the cheap ones.

The government says the NDIS is unsustainable. By international standards, Australia spends less than average on disability. Not more. Less.

Denmark

4.7%

OECD Average

2.2%

Australia (NDIS)

1.7%

✕ Misleading

"The NDIS is unsustainable"

Public spending on disability as a percentage of GDP, by country: Denmark 4.7%, Norway 4.5%, Sweden ~3.4%, OECD average 2.2%, UK 1.3%, Australia's NDIS 1.7%. Australia is below the OECD average. On disability services specifically, Australia spends under 0.5% of GDP. Denmark spends 7.1%. The NDIS isn't a blowout. It's a catch-up - and we're still behind.

Before the NDIS existed, Australia spent just 1% of GDP on disability - well below comparable countries. The Grattan Institute confirms Australia "lagged behind many comparable countries" before 2013. The NDIS was built to close that gap. Cutting it now doesn't return us to normal. It returns us to last place.

✓ In context

Australia's total social spending is in the bottom third of the OECD

Total social spending as % of GDP: Denmark 28.7%, Sweden 26.3%, OECD average ~20%, UK 22.3%, Australia ~17%. We spend less on social services than almost every comparable country. The NDIS isn't generous. It's the bare minimum - and they want to cut that too.

Data sources

Sources

▸ Disability spending by country: OECD Social Expenditure Database (SOCX)
▸ NDIS at 1.7% GDP: Federal Budget 2025-26
▸ Australia lagged pre-NDIS: Grattan Institute, "Saving the NDIS" (2024)
▸ Denmark 7.1% on disability services: Hwang (2024), Australian Journal of Social Issues

About

Who we are

Power in Numbers is maintained by a small group of data professionals and NDIS families. We work in data science, policy analysis, and disability advocacy. Some of us are parents of children on the NDIS.

We built this because the numbers being cited in Parliament don't match the numbers published by the NDIA's own data portal - and we can prove it.

We remain anonymous because some of us are NDIS participants. Publicly challenging the agency that controls your child's funding is not a risk we can take. The data speaks for itself.

For media: We're available on background. We can walk you through every calculation and point you to the source data. We won't waste your time.

Corrections: If you find an error, tell us. We'll fix it and note the correction. Getting this right matters more than being right first.

powerinnumbers@proton.me

For Media

Story angles - verified, ready to use

Every angle below is independently verifiable from official NDIS data, AEC results, Senate Estimates, and FOI disclosures. Full methodology: methodology page.

1. The headcount-vs-spending disconnect
Butler says 40% of participants have autism. True - but autism accounts for 21.4% of actual spending, not 40%. And autism children aged 0-8 - the cohort being targeted for cuts - are only 2.5% of total NDIS spending. The government is citing headcounts to imply costs. The data shows otherwise.
Verify: NDIS Payments Data + Participant Numbers, Dec 2025 quarter.

2. AEIOU Foundation collapse
Australia's largest specialist autism early intervention provider closed all centres in March 2026 after average funding packages were cut 70%. 120 children lost therapy overnight.
Verify: ABC News, 22 April 2026

3. $34.8M fighting participants at tribunal - then removing the right to appeal
The NDIA spent $34.8M in total on AAT matters in 2020-21 ($17.3M to external law firms). 73% of plan-related cases were changed in the participant's favour. Under the new planning model (mid-2026), participants will no longer be able to appeal the total amount of plan funding.
Verify: FOI disclosure Dec 2021 (Right to Know), NDIS Quarterly Report Q3 2024-25, Senate Estimates Dec 2025.

4. Four marginal Labor seats where NDIS voters alone could flip the result
AEC preference flow analysis shows Bullwinkel (WA), Solomon (NT), Petrie (QLD), and Forde (QLD) can be flipped by disability voter defection to the Greens, independents, or other parties — with Liberal winning on preferences.
Verify: AEC Tally Room 2025, NDIS Participants by CED data.

5. Australia spends less on disability than the OECD average
NDIS at 1.7% of GDP is below the OECD average of 2.2%. Denmark spends 4.7%. The "unsustainable" narrative doesn't survive international comparison.
Verify: OECD SOCX Database, Federal Budget 2025-26.

They're misleading you about autistic kids

"40% of NDIS participants have autism"

Cut the kids. Save nothing.

Politicians vs. Real Numbers

Kids aren't expensive. Adults are. Assisted living is. 24/7 care is.

AEIOU Foundation: 120 children lost therapy overnight

$34.8 million fighting disabled people. Losing.

Denmark spends 4.7%. We spend 1.7%. We're the cheap ones.

Share the real numbers

Who we are

Story angles - verified, ready to use