Funding the activist state: where public money backs left-wing causes

A common complaint on the right is that taxpayers now fund one side of politics. The money, the argument goes, goes to activists, lawyers, peak bodies and causes that most voters never asked for, and some that they voted against.

That complaint deserves a proper test, not an echo. This page lists what we could document: what was spent, by whom, from what year, and what any auditor found. It then asks two separate questions. Is this waste, in the sense of money lost or poorly managed? Or is it a policy choice that people are entitled to disagree with? Those are different charges, and mixing them up is how good arguments get dismissed.

This page adds to two others. The spending ratchet covers the big structural costs, such as the NDIS, health, aged care, defence and interest. The public service rebuild covers consultants and the 39,000 extra public servants. We do not repeat those figures here.

How much did the Voice referendum cost?

The Australian Electoral Commission's final cost for the October 2023 referendum was $411.3 million. The figure came out at Senate Estimates under questioning from Liberal Senator James McGrath. Before the vote, the AEC had estimated about $450 million.

The 2023 Budget set aside $364.6 million over three years for the whole exercise. Of that, $336.6 million went to the AEC, including $10.6 million to print and post the Yes and No pamphlets. Another $12 million went to the National Indigenous Australians Agency and the Museum of Australian Democracy for neutral civics education. A further $10.5 million funded mental health support for First Nations people during the campaign, and $5.5 million went to the NIAA for policy work (AAP FactCheck, June 2023).

The proposal was rejected nationally, with about 60% voting No, and in every state.

Two things are true at once. First, most of the $411.3 million is the fixed cost of asking every enrolled voter a question. Running any national vote costs hundreds of millions, whatever the cause. Second, the government chose to spend it on a question that the electorate clearly did not want. Senator Jacinta Nampijinpa Price said the AEC figure was "just the tip of the iceberg", pointing to advertising and tax deductions for donations. We could not find a published total for those, so we do not add one.

One popular claim needs correcting. Taxpayers did not fund the Yes campaign. The government money went to the AEC, the official pamphlets for both sides, and a neutral education program (RMIT FactLab, 2023).

What has Victoria's treaty cost so far?

The Victorian Government told AAP FactCheck in October 2025 that $326.7 million had been spent supporting treaty in state budgets since 2020-21. Media reports put the total since 2016 at about $380 million.

The ongoing bill is now written into law. The Statewide Treaty Act 2025 sets the maximum yearly funding for Gellung Warl, the permanent body that grew out of the First Peoples' Assembly. The cap is $3 million in 2025-26, $23.8 million in 2026-27, $71 million in 2027-28 and $72.1 million in 2028-29. After that it rises by 2.5% a year. On top of that, $36.8 million is set aside for capital costs between 2026 and 2029 (AAP FactCheck, 15 October 2025).

A claim that the treaty costs "$2.34 billion per year" circulated on social media in 2025. AAP FactCheck found it was false. The real cap is about 3% of that.

The stronger objection is not the price. It is what the money buys. Our page on Treaty and the First Peoples' Assembly sets out the powers the body now holds, how few people voted for its members, and the fact that the treaty itself was never put to Victorian voters. Those are questions of democratic consent. They are fair to raise in an election year, and they do not depend on calling the spending waste.

What are the climate programs costing?

The Net Zero Economy Authority. The 2024-25 Budget committed $399.1 million over five years from 2023-24 to help workers and regions through the net zero transition. Another $616.8 million follows from 2028-29 to 2034-35, then $93.4 million a year ongoing. Within that, $209.3 million over four years, and $53.3 million a year ongoing, goes to setting up and running the Authority itself (Net Zero Economy Authority, 2024-25 Budget measures).

Is that ideology or industry policy? It is honestly both. The Authority exists because the government has chosen a net zero target and wants to manage the closure of coal power stations. If you reject the target or the pace, you will see the Authority as the bureaucracy of a policy you oppose. But part of the money is help for workers losing jobs at power stations, which most governments would fund in some form. Our pages on the real cost of the transition and the Safeguard Mechanism deal with the target itself.

International climate finance. The Department of Foreign Affairs and Trade reports that Australia delivered $3.9 billion in climate finance from 2020 to 2025, above a $3 billion commitment. About $1.3 billion went to the Pacific, mostly to help countries adapt, and $997.1 million to Southeast Asia (DFAT, Australia's Climate Finance 2020-2025). That is roughly $780 million a year on average. The period spans both the Morrison and Albanese governments.

The claim that Australia is "sending billions overseas for climate change" needs care. DFAT says the commitment was met "largely through existing development assistance". In other words, most of it is the aid program, counted again under a climate label. A Pacific sea wall or drought-resistant crop scheme can be both aid and climate finance. Relabelling is not new spending. If you think the aid budget is too big, that is a separate and fair argument. But this is not $3.9 billion of extra money.

What did the Auditor-General find about climate grants?

This is the one item in this audit with a formal finding against it.

Before the 2022 election, Labor promised 400 community batteries. In October 2022 the new government committed $222.43 million over four years. The Australian National Audit Office reported to Parliament on 12 March 2026 (Report No. 24, 2025-26).

Its overall conclusion was that delivery was "largely effective". The detail is less flattering:

• The program's costs and benefits came from "modelling commissioned by the Australian Labor Party when in opposition".
• The department "did not provide the Australian Government with advice on the appropriateness of the pre-determined locations announced during the election".
• "A higher percentage of marginal electorates and Labor-held electorates received community batteries in pre-determined locations."
• Both agencies had "deficiencies in the management of conflicts of interest".
• The department did not assess applications against an agreed cost-per-kilowatt-hour test, "potentially impacting value for money".
• As at January 2026, 58 batteries were operating, against a target of 400. Grant agreements had been approved for 491.

The auditor did not find money stolen or lost. It found an election promise placed with a tilt towards marginal and Labor-held seats, without the department testing whether those were the right places, and a value-for-money test that was skipped. That fits the ordinary meaning of poor value. It is a process failure of a kind both sides of politics have been found guilty of before.

The strongest case against this spending is not that it is breaking the budget. It is that a government should not pay people to lobby it, or place grants by the electoral map.

Why did Australia restore funding to UNRWA?

UNRWA is the UN agency for Palestinian refugees. Australia has funded it every year since 1951. From 2011 it gave $20 million a year, except in 2020-21, when the Morrison government halved it to $10 million. The Albanese government restored the $20 million in late 2022 (Parliamentary Library, 22 May 2025).

On 16 January 2024, the government announced an extra $6 million for UNRWA as part of a Gaza aid package. Days later, Israel alleged that UNRWA staff took part in the 7 October 2023 attacks. Foreign Minister Penny Wong paused the payment on 27 January. On 15 March 2024 the government lifted the pause and paid the $6 million, with stricter conditions on staff neutrality.

In August 2024 the UN's internal investigators reported on the 19 staff accused. In nine cases, the evidence "indicated that the UNRWA staff members may have been involved in the armed attacks". They were dismissed. In ten cases, the evidence was missing or insufficient (Parliamentary Library, quoting the UN Secretary-General's office).

So the government released the money five months before that finding. Critics argue it should have waited. The government said its new agreement imposed stricter conditions on staff neutrality and on the security of supply lines. This is a foreign policy disagreement, and a serious one. It is not waste in the audit sense. The dollar sum is $20 million a year, plus the one-off $6 million.

Does the government fund the Environmental Defenders Office?

Yes. The Environmental Defenders Office is a legal centre that runs cases on environmental law, often against resource projects. Its federal funding was cut in 2013, under the Coalition. The Albanese government restored it in its October 2022 Budget, with $8.2 million over four years to 2026 (National Indigenous Times, 9 February 2024; PS News, 22 May 2024).

In January 2024 the EDO lost a Federal Court case it ran for Tiwi Islands elders against Santos's Barossa gas pipeline. Justice Natalie Charlesworth was scathing about expert evidence prepared for the case, calling it "so lacking in integrity that no weight can be placed" on it. Environment Minister Tanya Plibersek asked her department to check whether the EDO had broken its grant terms.

On 22 May 2024, the department said it was "satisfied that the EDO's conduct described in the judgement… were not in breach of the grant agreement". It found no evidence of fraud.

That answer settles the legal question and leaves the political one. The grant may have been spent within the rules. But a government funding a body that sues to stop projects the same government has approved is a real tension. Before the 2025 election, the Coalition promised to defund the EDO. We could not confirm whether the federal grant has been renewed past 2026.

Does the government fund groups to lobby it?

Sometimes, and it says so openly. In March 2025 the government announced "$6.2 million in grants to support leading homelessness peak bodies to strengthen their advocacy" and to give policy advice to government (ALP, March 2025). In February 2026, core funding was reinstated for National Shelter and Homelessness Australia, worth a combined $1.6 million a year for three years (Shelter NSW, February 2026).

A peak body is an organisation that represents a whole sector, such as charities working on homelessness. Peak bodies do useful coordination work. They also lobby, and governments of both colours have funded them. According to Labor, the Coalition cut funding to the homelessness peak bodies in 2014. Labor has been putting it back.

Can we total all federal grants to advocacy groups since 2022? No, and anyone who quotes a precise figure should show their working. GrantConnect, the federal grants register, has no "advocacy" category. A grant to a peak body for "sector development" looks the same as one for running a service. Building a reliable total would mean coding thousands of grants by hand. We have not done that, so we give examples rather than a sum.

What about DEI training and consultants?

Consultant spending is covered in detail on the public service rebuild. The short version is that the government cut consultants and hired permanent staff instead, which may cost more over time.

On diversity, equity and inclusion training, including "unconscious bias" courses, we could not find a published, verifiable figure for Commonwealth spending since 2022. Individual contracts appear on AusTender, but there is no agency-wide total that we could confirm. We therefore make no claim about the size of it. If a figure is produced at Senate Estimates, we will add it.

Putting it in proportion

In 2025-26 the Commonwealth spent $786.6 billion (Final Budget Outcome, 28 September 2026). That is about $2.15 billion a day.

On that scale, the $411.3 million Voice referendum is about four and a half hours of federal spending. The recurring federal items on this page add up to roughly $77 million a year. That is the Net Zero Economy Authority's $53.3 million ongoing, UNRWA's $20 million, about $2 million a year to the EDO, and $1.6 million in peak body core funding. It comes to about one cent in every $100. We leave out climate finance because most of it is existing aid, and we leave out Victoria because it is a different budget.

We have not added the one-off items to the yearly ones. Mixing a single referendum with annual grants and a five-year aid figure produces a big number that means nothing.

So if the goal is to fix the budget, these items will not do it. That job sits with the five big drivers covered in the spending ratchet. The case against this spending is about something else: whether public money should take sides in contested debates, and whether grants are placed on merit.

What did the government get right?

Three things deserve credit. The referendum cost came in under the AEC's own estimate of about $450 million. The government checked the EDO's grant compliance after the court case rather than ignoring it. And it put conditions on the restored UNRWA money and spread Gaza aid across other agencies, including $4 million to UNICEF.

The verdict: testing the claims

Claim: taxpayers are funding left-wing activist groups. Partly supported. The EDO and several advocacy peak bodies receive federal money, and one grant is described as being to "strengthen their advocacy". But the sums are small, the EDO was cleared of breaching its grant, and both parties have funded peak bodies at different times.

Claim: the Voice cost taxpayers half a billion dollars. Partly supported. The AEC's final cost was $411.3 million. Other costs exist but have not been published as a total. Taxpayers did not fund the Yes campaign.

Claim: Victoria's treaty costs billions a year. Not supported. The legislated cap reaches $72.1 million in 2028-29. About $326.7 million has been spent since 2020-21.

Claim: Australia is sending billions overseas for climate change. Not supported as stated. $3.9 billion was delivered over five years, mostly through the existing aid budget.

Claim: Labor restored UNRWA funding without waiting for the facts. Supported. The extra $6 million was released in March 2024, five months before UN investigators found nine staff may have been involved in the 7 October attacks.

Claim: an Auditor-General found climate grants were badly run. Supported, with care. The ANAO found the community battery sites skewed to marginal and Labor-held seats, no departmental advice on those sites, a skipped value-for-money test and only 58 of 400 batteries running. Its overall rating was still "largely effective".

For the big drivers of federal spending, read the spending ratchet. For consultants and staffing, read the public service rebuild. For money kept outside the headline budget, read off-budget funds.