Subject to Government Approval

Transcript of the published episode, checked against the episode script. Figures are written as numbers. Spot an error? Email auditthismedia@gmail.com.

00:00 Twenty days

Twenty days before Australia cancelled its new submarine program, the man in charge of it sent the Secretary of Defence some good news.

Twelve submarines. The largest defence project the country had ever taken on. And it was still, he wrote, inside its original cost estimate, though it would need some of its contingency.

The first boat was forecast to arrive between August 2033 and February 2035, at 80 per cent confidence. An independent analysis had tested that schedule 10,000 times.

And the French shipbuilder’s offer for the next stage of the design was, he wrote, affordable and acceptable.

His team, he would later tell a Senate committee, had been kept separate from what the government was weighing.

On 16 September 2021, at one minute past seven in the morning, the Prime Minister of Australia appeared on a screen at the White House, alongside the President of the United States. The Prime Minister of Britain appeared on another.

Good morning from Australia, he said. Australia would be getting nuclear powered submarines. The French ones were cancelled the same morning.

It was not, the Prime Minister said, a change of mind.

It was a change of need. There had been, he said, game changing developments in the strategic circumstances of our region.

By Defence’s count, the bill came to $3.4 billion. Australia did not get a submarine.

Nothing in this story was stolen. Nobody was charged with anything.

At a Senate hearing, a senator asked what taxpayers had got for the biggest part of that money. A lot of it, an official explained, was design work. Part of the answer was two words.

Intellectual property. It wasn’t of use, he said, in constructing that particular conventional submarine. But, he added, there was a lot of knowledge, know-how and experience that was gained through the program.

01:50 The helicopters

Every dollar of that bill was subject to government approval.

In June 1997, Australia’s Defence department signed up for eleven helicopters, for the Navy’s new frigates.

To save money, they weren’t new. The deal was to take old American navy airframes and rebuild them. Seven of the eleven had first been built between 1963 and 1965. One had already made 32,837 landings.

They were called Super Seasprites. The contract was worth about $662 million, and the last helicopter was due in May 2002.

In May 2001, the United States Navy retired the Seasprite. Australia was still waiting for its first one.

Defence provisionally accepted the first one in October 2003. In what it called an interim configuration, with a list of faults. Over the next two years, nine of the eleven were accepted that way. Two never were.

Then came the restrictions. From late 2004, the Navy wasn’t allowed to fly them off ships. Which was what they were for.

They weren’t allowed to fly into cloud on purpose. And some aircrew had to leave the squadron, because they were, in the words of the Auditor-General’s report, anthropometrically unsuitable.

In other words, they were the wrong size for the cockpit.

And then there were the hard overs. The flight control system pushing a control all the way to the end of its range, when the pilot hadn’t asked it to. By March 2006, there had been four of them, in about 1,600 hours of flying.

The airworthiness regulator couldn’t work out how likely another one really was. So it used a default figure. On that figure, the helicopter’s failure rate was assessed at 2,000 times the accepted standard.

The Navy stopped flying them that month. In May, the type certificate was withdrawn. The Navy never flew a Super Seasprite again.

The manufacturer later said its changes addressed every known cause of the hard overs. Defence’s view, after its own testing, was that the first round of fixes hadn’t resolved every safety issue.

03:44 Less than fifty million back

In March 2008, a new government cancelled the project. The Defence Minister told parliament it meant flushing $1.36 billion of taxpayers’ money down the drain.

The Auditor-General later put the total at more than $1.4 billion. All in, about $127 million a helicopter.

Under the settlement, the manufacturer took back all eleven, with the simulator and most of the spare parts, and guaranteed Australia $39.5 million.

About three cents in the dollar. And the Commonwealth agreed, in writing, not to disparage the helicopters.

So we won’t. Anyway. Five years later, New Zealand bought ten of the same airframes, with a simulator and missiles, for NZ$242 million.

Its government said a key issue had been Australia’s own specification. A crew of two, and a flight system that could fly the aircraft with no hands on the controls. New Zealand would use a crew of three. And it would not adopt the no hands requirement.

Between the middle of 2016 and January 2024, New Zealand flew them for more than 7,000 hours. More than four times as long as the Australian Navy ever had.

By 2023, on average, fewer than one in five was serviceable. In August 2025, New Zealand named its preferred replacement.

The helicopter Australia’s Navy bought after it gave up on the Seasprite.

The Auditor-General’s report put the big question in one sentence. How the project was allowed to continue for so long at significant expense to Government. Part of its answer was a degree of optimism, and a reluctance to make firm decisions.

05:21 The next story

Here’s the thing about the Seasprite. The project failed, and a parliamentary committee said so in those words. When a project fails that visibly, you can see where the money went.

In plain dollars, the next story cost more than twice as much. And according to the people running it, it was working.

Australia spends $3.4 billion on a submarine program and never gets a submarine, on this episode of Audit This.

06:18 Ruled out

Start with the submarines Australia already had. The Collins class. Six of them, built in Adelaide and commissioned between 1996 and 2003. Each was designed to last 28 years. Without an extension, they were due to leave service between 2026 and 2036.

Submarines take decades to replace. So in 2009, the government of the day wrote down a plan.

Twelve new submarines. Double the fleet. Assembled in South Australia. A program spanning three decades, and in the words of the Defence White Paper, Australia’s largest ever single defence project.

And one more sentence. The Government has ruled out nuclear propulsion for these submarines.

Four years later there was a new White Paper, a new Prime Minister, and very nearly the same sentence. The Government has ruled out consideration of a nuclear powered submarine capability.

In 2015, Defence ran a competition. Three bidders. Germany. Japan. And France.

The French design was named after a fish. The Shortfin Barracuda. An indigenous species of the Barracuda, the company said, found in Australia’s Great Barrier Reef. Which is, professionally speaking, how you name a bid.

It was a new design for a conventional submarine, built, the company said, with technology from its larger cousin, the French Navy’s Barracuda nuclear-powered attack submarine. You cannot convert a nuclear submarine to conventional power, the head of Defence’s program later told a Senate committee. This was a new boat.

On 26 April 2016, the Prime Minister stood at the shipyard at Osborne, in Adelaide, and announced that the French had won.

Australian built, Australian jobs, Australian steel, he said. Here, right where we stand.

$50 billion. 2,800 jobs. Twelve regionally superior submarines. The first in the early 2030s. The last in the late 2040s, or around 2050.

A week later, the Prime Minister was asked whether the chance to switch to nuclear power one day had been part of the reason for choosing the French design. That issue of nuclear propulsion, he said, did not form any part of our considerations.

08:27 Two numbers

The paperwork came in layers. A design contract with the French company in September 2016, and on the same day, Lockheed Martin Australia, chosen to design and integrate the combat system. A treaty between the two governments that December. And in December 2018, the first boat got a name. HMAS Attack.

The master agreement was signed in Canberra in February 2019, nearly sixteen months later than planned. By then there was a new Prime Minister. He called it the greatest partnership between Australia and France since Nancy Wake was let loose on the Nazis.

Now. Here’s the thing about the money, because that’s where most people think this story went wrong.

Defence counts a project like this two ways. One is in today’s dollars, as if you paid for everything at once. The other is in the dollars of the years you actually pay, with three decades of inflation and exchange rates added on.

In 2016 dollars, the submarines cost $50 billion. In the second kind, by 2020, the same submarines cost $89.7 billion.

Both numbers were true. Defence’s own documents didn’t always say which kind of dollar the original $50 billion had been. But on its constant dollar measure, Defence said, the cost had not changed. The Secretary of Defence later told the Senate the out-turned figure was the same estimate that the department took to government in 2016, adjusted for foreign exchange rate variations. There was no cost blowout, he said.

The schedule was another matter. In January 2020, the Auditor-General reported that the design was running nine months late. Defence had spent $396 million on design, and, in the auditor’s words, could not demonstrate it had been fully effective in reaching its first two major design milestones. The audit also recorded that in 2018, Defence had found 63 studies for an earlier stage of the design that the French company hadn’t completed. In 2019, Defence had written to it about a deepening concern.

10:28 The other option

Defence’s answer was that the delivery of the first submarine had not been delayed.

The Secretary of Defence later set out the timeline for the Senate. In March 2020, after a request from the Prime Minister’s office, Defence began a preliminary look at whether Australia could have nuclear powered submarines after all. It was an option, he said, that was not available to Australia in 2016.

A small dedicated team worked on it. Its study went to the offices of the Prime Minister and the defence ministers that December. Between that December and June 2021, Defence advised the government that significant risks had been identified in the French program. In February 2021, Defence was directed to run a formal review that would take in the option of nuclear powered submarines.

Meanwhile, the program carried on. In January, it had entered its next big design review. The plan was to finish it by April. It didn’t.

At Senate estimates on 2 June 2021, a senator asked the Secretary of Defence about a plan B.

He said Defence was doing prudent contingency planning in case we are unable to proceed. Asked when that planning had started, he said he’d been thinking about it for quite some time.

He said it had become clear to him that Defence was having challenges with the Attack class program over the last twelve months, and that the government was absolutely committed to trying to make it work. Asked whether the government had made any decisions about alternatives, he said no. The government’s final decision, he later told the Senate, came in the day or two before the announcement.

On 12 June, by Defence’s later account, the Prime Minister, the American President and the British Prime Minister met at the G7 summit in Cornwall, and agreed that officials should work out a way ahead on nuclear powered submarines.

The program’s own team was not part of any of it. The head of the program later described it to a Senate committee: We were aware of contingency planning, but we were not aware of the considerations or deliberations that went into informing the decision of government, nor were we part of that decision making process.

12:27 The email and the first letter

We were kept separate from that.

Which brings us back to August. On the 23rd, the French company sent in its updated offer for the next phase of the design. On the 27th, the head of the program sent the email you heard at the start. Inside the cost estimate, though it would need some of the contingency. 10,000 runs of the schedule. Affordable and acceptable. He wrote that there are no extreme program strategic risks.

A joint statement by the governments of Australia and France, dated 30 August, came out of high level talks between the two. It had one sentence about submarines. Ministers underlined the importance of the Future Submarine program.

Australia’s foreign minister later told a Senate committee that at those talks, Australia had raised its review of whether conventional submarines would be capable enough in the region. But, at that time, she said, there had been no final decision taken on these matters. The government’s final decision, the Secretary of Defence told the Senate, came in the day or two before the announcement.

On the 31st, the Secretary of Defence wrote back to the head of the program. He would make sure, he said, that the good progress to date is part of the advice we take to Government.

On 8 September, the program head’s deputy, a navy commodore, wrote to the French company, setting out how that design review could be signed off. Six conditions. The first had been successfully completed. The second remains open. The other four are redacted in the copy Defence released.

And the letter was clear about who decided what came next. The Commonwealth’s agreement to sign off the review, and any start on the next stage of the design, were both subject to government approval.

14:01 The second letter

On the 15th, the commodore wrote to the company again: I confirm the exit of the Functional Ship System Functional Review has been achieved as required under the Submarine Design Contract.

The same letter said it did not provide any authorisation to continue work beyond what was already agreed. Defence’s records show it was first put in a shared online folder for the company at five past twelve, Canberra time.

The head of the program later told a Senate committee: We were aware an announcement would be made. We were not aware of what the content of that announcement was. We continued to work in good faith with Naval Group until government’s decision was announced. The letter, he said, was part of our ongoing contractual arrangement with Naval Group. They had been simply clarifying that the exit criteria for our systems functional review had been met, and the decision to enter the next design phase rested with government.

On the 15th, he said, his team learned the announcement would be a joint one with Britain and the United States. He couldn’t say whether that was before or after the letter. Asked separately, leaving the letter aside, whether he knew of the announcement that morning or that afternoon, he said, the late afternoon. His team, he said, had not been part of the decision. We were kept separate from that.

The senator asking the questions made a point of it. I have no question, she said, that you would always operate in good faith.

The Prime Minister later said he told the President of France on the night of the 15th, at about half past eight.

That night, the Secretary of Defence spoke to the head of the French company, and the leaders of its Australian arm. They were, he said later, surprised and disappointed, understandably.

15:33 The announcement

At one minute past seven the next morning, it was announced. In Perth, it was one minute past five.

The Prime Minister’s reasons were the region, and the technology. Nuclear submarines, he said, were faster and stealthier, and his government’s statement said they could stay submerged for months. And their next generation reactors would not need refuelling during the life of the boat, so Australia would not need a nuclear power industry of its own.

This in no way reflects, he said, in any way, shape or form on the Attack class submarine, the Naval Group and the commitment of the French Government. The Secretary of Defence later put it plainly to the Senate. The contracts were terminated, he said, because our requirements have changed, not because of the poor performance by either Naval Group or Lockheed Martin Australia.

He said he had instructed the Secretary of Defence to operate in total good faith in our dealings with Naval. He also said: We’ve invested $2.4 billion in the Attack class programme, and I say all of that investment, I believe, is further building our capability.

France recalled its ambassadors from Canberra and from Washington. According to the French foreign ministry, as reported at the time, it had never done that to the United States before.

Three days after the announcement, the Prime Minister put the choice more plainly. The question was whether the submarines being built would be able to do the job Australia needed when they went into service. His government’s strategic judgment, he said, based on the best possible intelligence and defence advice, was that they would not. And once Australia could get what he called a supreme submarine capability, it would have been negligent for us not to.

The termination notices, the head of the program told a Senate committee, had gone out on the day of the announcement. Six weeks later, Defence counted about 560 people employed directly on the program by the two contractors. A union official told a Senate committee that some Australians had packed up everything at home and landed in France three days before the announcement. About ten more engineers, he said, had been due to fly out the following Monday.

In written answers to the Senate, Defence said 370 companies in Western Australia had registered their interest in working on the program. It pointed out these were simply expressions of interest and not contracts. Asked in writing whether the government had consulted small businesses before cancelling it, Defence answered: No, due to the classified nature and connection to national security.

17:57 The bill

Now, the bill. Walking away at that point didn’t trigger a break fee, Defence said. There are no break payment fees under the terms of our contracts, the head of the program told a Senate committee. The first, he said, would have come at a design milestone in 2023. But there would be, he added, costs associated with termination. And the French company, Defence later told the Senate, had reserved their rights.

Then there was an election, and a new government. On a Saturday in June 2022, the new Prime Minister announced a settlement. €555 million. About A$830 million. Fair and equitable. Some of the details, he said, would remain confidential.

Asked whether that was the end of any legal action, he said, Yes. Yes.

$2.1 billion of the 3.4, Defence told the Senate, had been booked as an asset, on paper still under construction. When the program was cancelled, it was written off. A senator pointed out at estimates that in Defence’s budget papers, it appeared as a footnote. In 8-point Times New Roman, he noted.

The senator asked the obvious question. What did the taxpayers get for $2.128 billion? What did they get, apart from not a sub?

A senior Defence official explained that a lot of it was design work. Activity by the two main contractors and their subcontractors towards creating the design for twelve regionally superior conventional submarines.

So these are blueprints and other... the senator began. Intellectual property.

Intellectual property that’s now of no use to Australia? It is not of use in constructing that particular conventional submarine, the official said, but there was a lot of knowledge, know-how and experience that was gained through the program, by both people working in Defence and the industry participants themselves that are still resident in Australia.

All up, by Defence’s count, the program had cost $3.4 billion. And every dollar of it, Defence told the Senate, came out of the provision the government had approved for the program.

Subject to government approval. Even paying the settlement turned into an argument. The money came from funds Parliament had given Defence for building things, not for running costs. Parliament’s audit committee said paying it from that money should not have occurred and was against the spirit, if not the letter, of the Appropriations Acts. Government lawyers, asked after the payment, had advised that a court would, on balance, find no breach of the Constitution. And the government replied that it had come from a legally valid appropriation source.

20:46 The bridge boats

Here’s the part almost nobody followed. Remember the Collins boats. The ones due to start retiring in 2026. They were the bridge. They had to keep going until the new submarines arrived, so Defence planned to extend their lives. And in 2019, it decided how.

The Collins upgrade would use the same equipment, the same suppliers and the same technology chosen for the French submarine. That way, the thinking went, it would cut costs, and take some of the risk out of the French program. A new main motor. New diesel generators. Chosen to match a submarine that didn’t exist yet.

A Defence risk assessment in May 2020 put the trade off in writing. Because the motor had been chosen to match the French program, it said, we must accept an element of, we get what we get.

Between December 2020 and June 2021, Defence advised the government of significant risks in the French program, and the government agreed to work on alternatives. The people choosing the motors and generators for the Collins, the Auditor-General found, were not informed that alternatives to the Attack class submarines were being examined or that there was a risk to the continuation of the Attack class program. And Defence, it found, did not advise the government of the likely implications for the Collins.

In September 2021, the French submarine was cancelled. The plan to fit the Collins with its equipment wasn’t.

22:05 Changing course

That November, the government owned company that maintains the Collins recommended the project’s scope be comprehensively reassessed. Defence reviewed it, and decided to go ahead with the motor and generators chosen for the French submarine anyway. Its own internal reviewers, the Auditor-General reported, later described that review as a missed opportunity.

In February 2022, five months after the cancellation, Defence signed the design contract. It started at $125 million. It was amended 53 times. It reached $813 million.

By early 2026, $693 million had been spent on design and equipment. Four of the five key design reviews hadn’t been met on schedule. And the first boat wasn’t on track for its upgrade in June.

In May 2026, after what the Auditor-General called ten years of planning and design activity, the government agreed to change direction. For up to five of the six boats, Defence would stop replacing the old systems, and refurbish them instead.

Defence’s explanation is in its response to the audit. Its primary focus, it said, had been to avoid a capability gap, and that had, in practice, limited how fully it could reassess.

In other words, the old submarines couldn’t stop. So neither could the plan.

Here’s the thing nobody puts in a press release. A decision like that doesn’t end at the contract. The French submarine was cancelled in a morning. But it was already built into other plans, by people who hadn’t been told it might go. Changing course took more than four and a half years. For one of the six boats, the design work for the new equipment is still going.

$3.4 billion was the bill for the submarine. It was never the bill for the decision. As far as we could find, nobody has published that one.

The new plan for the Collins comes with a new number. Up to $11 billion over the next decade, to keep them going. In 2021, the number was up to $6.4 billion. To be fair, each figure covers something different.

24:03 The next bill

As for the nuclear submarines, the plan arrived in March 2023. From the early 2030s, pending the American Congress, Australia would buy three Virginia class submarines from the United States. The first submarine built in Australia would follow in the early 2040s.

The government’s own measure of the cost is around 0.15 per cent of GDP per year. The Parliamentary Budget Office, using Defence’s figures, put the upper bound to the mid 2050s at $368 billion. Including a contingency of $123 billion.

The money has started moving. US$2 billion to the United States’ submarine industry in 2025, of US$3 billion promised. £2.4 billion promised over ten years to expand Rolls-Royce in Britain, which will build the reactors. $8.5 billion so far into a new shipyard at Osborne, of about $30 billion. $6 billion of the first nine, the defence industry minister said in 2023, would come from money left over from the French program.

In 2025, the new American administration reviewed the whole arrangement. The review hasn’t been published. The President said they were going full steam ahead. In May 2026, the three defence ministers welcomed a proposal. Instead of a mix of new and used Virginia class boats, Australia would buy three that are already in service.

In September 2026, the government’s submarine agency marked five years of AUKUS. The next milestone, it said, would come in 2027, when American and British submarines would start rotating through HMAS Stirling, on Garden Island, off Rockingham. Its statement said ministers had reaffirmed that the plan remains on track. It didn’t mention the cost.

No Auditor-General, as far as we could find, has ever audited the decision to cancel. A Senate committee reported on it in 2022, and split along party lines. Beyond that, the only official finding about the money that we could find is the one about how the settlement was paid for.

26:09 The ledger

Which leaves the numbers.

Twelve submarines planned. $3.4 billion spent, by Defence’s count. $2.1 billion of it written off.

€555 million to settle. Submarines delivered. None. Contracts ended because requirements changed. Not for performance.

For every person in Australia, about $122. The Collins upgrade contract. 53 amendments. $125 million, to $813 million.

The Super Seasprite. More than $1.4 billion. Less than $50 million back.

The nuclear submarines. About 0.15 per cent of GDP a year, by the government’s measure. Up to $368 billion, the Budget Office’s upper bound, including contingency.

HMAS Rankin, the youngest of the Collins class. Commissioned in 2003. Now planned to retire in 2048, 45 years after it was commissioned.

27:12 One more document

HMAS Farncomb, the boat that was due to retire in 2026, is now planned to serve until 2038.

There is one more document, and it isn’t Australian. Under American law, before the first Virginia class submarine can be handed over, the President of the United States has to certify to Congress, at least 270 days ahead, that, among other things, the transfer will not degrade the United States undersea capabilities.

So after the White Papers and the competition, the contracts and the treaty, the letters, the settlement and the write off, the first nuclear submarine Australia is due to receive is subject to government approval.