Getting a clear view of the cloudy budget for Australia’s nuclear submarine program

Trying to understand Australia's nuclear submarine program budget is like knitting socks from smoke. A one stop budget is needed to run this megaproject and to give the Parliament essential oversight.

Written by

Marcus Hellyer

This brief has been updated since Senate estimates hearings on the Defence portfolio on 14 February 2024 to incorporate discussion at that forum.

What do we know about the budget for Australia’s nuclear-powered submarine (SSN) program? Much commentary has focused on the Government’s estimate of $268-368 billion over the next 30 years, but if the Parliament and public want to understand both the plan to acquire and operate the capability as well as assess how the Government and Defence portfolio are succeeding at implementing that plan, they’ll need more information.

Bottom line up front: the Government’s current disclosure mechanisms aren’t allowing the Parliament and public to do that – and that needs to change fast if this megaproject is going to be run well. A one-stop budget for the program is an essential start. It’s also necessary for the Submarine Agency to do its work effectively.

Let’s look at what we do know. When the Government announced the ‘optimal pathway’ to our future SSN capability in March 2023, it released some numbers in addition to the $268-368 billion figure. It said the SSN program would cost $9 billion over the forward estimates (i.e., the current budget year and the three subsequent ones) and $50-58 billion over the decade (the forward estimates plus a further six years). That first decade is meant to get us to the delivery of the first second-hand US Navy submarine around 2032 or 2033. SAA provided an initial analysis of what those figures meant here.

The October Senate estimates brief

More information is contained in the Australian Submarine Agency’s (ASA) brief prepared for its seniors attending Senate estimates hearings in October last year that was released under a Freedom of Information request. The Estimates brief, which is intended for Defence seniors and is not automatically provided to the Senate or the public, actually contains more information than public documents such as the Portfolio Budget Statements (more on that later).

In the October 2023 Senate estimates brief, the numbers for the SSN program were still $9 billion for the forward estimates but the decade figure had firmed up at the upper end of the range at $58 billion. Interestingly, the funding freed up by the cancellation of the Attack class had shrunk to $24 billion, meaning the gap between the two submarine programs is now $34 billion over the decade, which is a big hole to fill.

The Senate estimates brief broke down the $9 billion over the forward estimates this way:

  • $5.6 billion as outlined in Program 2.16 Nuclear Powered Submarines (presented in Table 37 of the Defence Portfolio Budget Statement 2023-24 (page 88));
  • $3.3 billion reflected against other Defence Programs (per Note A to Table 37 of the Defence Portfolio Budget Statement 2023-24 (page 88)); and
  • $0.3 billion provided to other Government agencies (outlined in Budget 2023-24 Budget Paper No 2 (page 94)).

Each of those figures is broken down a further level in the brief. But there’s no breakdown of the $58 billion or the $268-368 billion.

What spending $58 billion would look like

If the ASA is going to deliver the first SSN on schedule, by its own admission it needs to spend $58 billion over 10 years. If the spend in the first four years is $9 billion, that leaves a huge $49 billion to spend in just the remaining six years of the decade. To get that money away, the ASA will need to hit a spend spread something like this.

Table 1: Indicative annual SSN spend, 2023-24 – 2032-33 (A$ billion)

 2023-242024-252025-262026-272027-282028-292029-302030-312031-322032-33
Indicative annual SSN spend123346891012

The first thing to note is that that represents a very challenging ramp up. Admittedly if we are buying US Navy boats rather than building them, getting money out the door is easier; you sign a check for the boat. But there’s still a lot other spending that needs to occur—on facilities, dry docks for maintenance, the shipyard for the AUKUS boats, new regulatory and safety functions, etc. Second, to achieve $58 billion in the decade, the ASA likely reach a $10 billion in-year spend. That’s a huge chunk of the overall Defence acquisition budget.

The public numbers

So where do we go to find at set of numbers like that in the various budget papers? The unfortunate answer is it doesn’t exist in the public realm, although Defence no doubt will have one. Here’s what we do have.

The 2023-24 Defence Portfolio Budget Statements provided the budget for Program 2.16 Nuclear-Powered Submarines (Table 37, page 88). Those numbers are updated in the 2023-24 Portfolio Additional Estimates Statements released in February 2024 (Table 41, page 54).

However, on 1 July 2023 shortly after release of the 2023-24 budget, the Nuclear-Powered Submarine Task Force became a separate agency within the Defence portfolio called the Australian Submarine Agency. The ASA’s budget is also presented in the 2023-24 Defence Portfolio Additional Estimates Statements (Table 1, page 148). Those funding lines are presented below.

Table 2: Australian Submarine Agency budget, 2023-24 – 2026-27

 2023-242024-252025-262026-27FE Total
PBS (Defence Program 2.16)515.0389.23,722.7977.65,604.6
PAES (Defence Program 2.16)588.7444.43,913.81,045.45,992.3
PAES (Australian Submarine Agency)323.4384.7433.9530.61,672.6
Difference between PAES Program 2.16 and ASA-265.3-59.8-3,479.9-514.8-4,319.7

Source: PBS, PAES

The first thing we notice is that there have been some modest increases to Program 2.16 since the budget. The second is that we now have two different sets of numbers for the SSN program (Program 2.16 and the ASA). Third, none of the sets of numbers reach $9 billion over the forward estimates, so none of them gives the full picture for the SSN enterprise. Finally, the very prominent blob of $3.7 billion funding in 2025-26 has gone from the ASA’s budget line. What’s going on?

It appears that the capital budget for the SSN program doesn’t sit in the ASA’s budget line. Rather, those funds sit within the Department of Defence’s acquisition budget. This is similar to the way the Australian Signals Directorate’s capital budget it handled; it sits in the Defence Department’s acquisition budget (the ‘Integrated Investment Program’) until projects are approved by Government at which point the funds are transferred over to ASD (which is why each year the table of budget measures in the PBS and PAES has a line called ‘Transfer of Projects to ASD’).

That’s why the ASA budget line in the PAES is only $1,672 million for the forward estimates, far short of the $9 billion figure. The ASA budget line really just covers the operating costs of the ASA, which are a very small part of the total SSN program’s costs.

If the ASA’s budget line does cover the full cost of the SSN enterprise, where can we find it in the budget papers? Program 2.16 comes close (at $5,992 million of the $9 billion) but that reporting line will likely go away in future budget documents since that program no longer exists, it’s become the ASA. And unlike Defence programs which state their estimated capital expenditure over the forward estimates (e.g., Table 21 on page 60 has the line for the Navy’s total capital expenditure for each year of the forward estimates) there doesn’t appear to be a similar line for the ASA. Of course, it wouldn’t be hard for the Government to direct ASA to include such a line in the budget papers.

Where did the $3+billion blob go in the ASA’s budget? The blob was to cover, at least in part, the transfer of funds to the US (stated at USD$3 billion) and the United Kingdom (an as-yet unspecified amount) for ‘industrial uplift’, that is, to increase their industrial capacity. Presumably they are part of the capital funds that now sit in the broader Defence acquisition budget, visible for now in the Defence Department’s defunct Program 2.16 but not in the ASA budget.

In sum, there’s no comprehensive spend spread provided to the Parliament or the public for the $9 billion, let alone the $58 billion for the decade. The ASA’s future budget estimates only cover a very small part of planned expenditure on SSNs.

The AUKUS SSN shipyard

There are some other information gaps, such as what we are planning to spend on the AUKUS SSN shipyard and how we will spend it. The ASA’s October estimates brief doesn’t put a figure on the cost of the AUKUS SSN shipyard. The estimates brief refers to ‘Public Debt Interest for the Submarine Construction Yard’ and notes this is covered in Budget Paper 2. That document states the following (page 96):

The Government will also provide Australian Naval Infrastructure Pty Ltd with an equity injection over 5 years from 2023–24 to commence early construction and design works for a submarine construction yard and to design the Skills and Training Academy, in South Australia. The financial implications of this component of the measure are not for publication (nfp) due to commercial sensitivities.

Pu the two together and it looks like the Government will provide Australian Naval Instructure with an equity injection and the ASA will then pay the interest to cover the cost of capital. What that suggests is that the shipyard is not part of the $9 billion or perhaps even the $58 billion, only the ASA interest payments. This was the approach taken with the shipyard that was being developed for the Attack-class submarine until it was cancelled. The cost of that shipyard was also protected by the veil of commercial sensitivities.  However, the cost of the SSN shipyard will be many times greater than the Attack-class yard. It would be good for the Government to clarify whether the cost of the SSN shipyard is included in its $9/58/268-368 billion figures.

Defence funding increases?

While the 2023-24 Defence budget didn’t include any funding increases over the forward estimates, it did state (page 207) that ‘the Government has made a provision in the Contingency Reserve for increased Defence funding over the medium term to implement the Defence Strategic Review.’ A rather opaque figure suggested this was around $30 billion in the decade beyond the forward estimates—a number confirmed by Defence officials.

If the gap between the funding programmed for the Attack class and the SSN program is $34 billion, then entire $30 billion in the contingency reserve (should the Government decide to commit it) will be consumed just by the SSN program. That suggests the Government is not planning on any additional funding for any other defence capabilities in the coming decade.

What does all this mean?

We can draw several observations from this brief:

  • There is no comprehensive public document that provides all relevant, releasable information on the nuclear submarine program’s budget. The information that is available needs to be assembled from several different documents.
  • Much of the information that is public has only been released under FOI requests; it was not provided automatically to the Parliament and public.
  • There are several significant information gaps such as the anticipated spend spread over the forward estimates (let alone the decade) for the entire SSN program, as opposed to the ASA’s operating budget, which is a small fraction of the total cost.
  • It is not clear if the cost of the AUKUS SSN shipyard is part of the $9 billion/$58/$268-368 billion estimate or in addition to it as some ‘off budget’ amount.

Managing the SSN program budget

This leads to some big issues that need to be addressed. First, how will the SSN budget be managed? If it is part of the broader Defence investment budget it is essentially invisible outside of Defence.

But if it is part of the broader Defence budget it runs the risk of distorting Defence’s overall investment plan. Once the SSN program hits annual numbers like $10 billion, if it needs additional funding in a particular year, that will be dragged out of other projects inflicting scope and schedule impacts.

Finding even $1 billion more in a single year will be distorting. Conversely, if too much money is held for the SSN program, the opportunity cost on other capabilities of the rest of the Australian military will be huge. Moreover, this approach runs the risk of the true cost of the SSN program being underestimated or hidden so that it doesn’t appear to suck funding out of the broader investment portfolio.

An alternative approach is to keep the SSN program separate from the broader Defence Integrated Investment Program and indeed hold the SSN program’s budget outside of the Defence portfolio. That way, should it sneeze, the IIP won’t catch cold. The UK has recently decided to adopt an approach that quarantines funding for its nuclear capabilities from the broader Defence investment program. It’s not surprising that as soon as it did that, estimates for the nuclear capabilities immediately went up to reflect their true cost, not what the Ministry of Defence thought would fit within the overall budget.

The second big issue is without greater transparency, it will be extremely difficult for the Government and Department of Defence to win and maintain public support for the SSN enterprise. The public, who are after all footing the bill for the $268-368 billion endeavour, have a right to understand the costs and importantly the risks. The Senate has an obligation to ensure that this vast sum of public money is being spent well. It cannot do this if key information is scattered across multiple documents, particularly if those documents are only made public through FOI requests.

And surely the managers of this Australian government megaproject within the Australian Submarine Agency itself need to understand their total program costs and budget to manage this enormous enterprise. 

A clear one stop budget, accessible to the Parliament and the public, will serve many needs.  But we don’t have anything approaching that now.

SHARE THIS ARTICLE