The Trump administration has been calling on its partners and allies to lift defence spending to 3.5 per cent of GDP. NATO has agreed to that benchmark, with another 1.5 per cent thrown in for vaguely defined national resilience and critical infrastructure programs. Of course, nearly all NATO members have got some way to go to get to that extremely ambitious target.

So where does Australia stand against that benchmark? And is it a worthwhile target for us anyway?

While the Albanese government has boasted of its record peacetime increases to defence spending, that “record” is more due to inflation and the steady growth of our population and economy than deliberate policy decisions to invest a greater share of national resources in Australia’s military capability.

Rather, Australia’s defence spending under the Albanese government and its predecessor has been stuckin a narrow band between 1.9 per cent and 2.1 per cent for seven years.

Even after its 2023 Defence Strategic Review amplified the bleak strategic assessment of the Morrison government’s 2020 Defence Strategic Update, the dial has barely moved.

Under the current spending plan, the defence budget will only grow (marginally) past 2.1 per cent of GDP in 2027-28, the last year of the government’s second term.

Even the ultimate target of 2.33 per cent in 2033-34 is a further two parliamentary terms away and still only two-thirds of the US administration’s expectation.

In light of those facts it was going to be a hard to win a debate with the Trump administration on whether Australia was an ally doing enough heavy lifting.

So Deputy Prime Minister and Defence Minister Richard Marles has changed the key definitions. If we use NATO’s definition of defence spending, he suggested last month, Australia is actually spending 2.8 per cent of GDP. That’s not an insignificant difference; it would represent 36.6 per cent, or $21.4bn, more than the figure in the 2025-26 budget papers. In one athletic leap from $59bn-$80.4bn, we’ve gotten more than halfway to Trump’s 3.5 per cent.

Unfortunately, Marles didn’t say how he’d reached that figure.

NATO’s methodology for defining defence spending includes military pensions, so presumably the revised number includes that. But what else, if anything?

We can turn to an informed third party, the Stockholm International Peace Research Institute, for adjudication.

SIPRI publishes defence spending figures for every country in the world. It’s advertised methodology is similar to NATO’s and includes pensions. Not surprisingly, its figures for NATO countries are similar to the alliance’s own published numbers. But SIPRI puts Australia at under 2.0 per cent, nowhere near Marles’s new figure. So until the government shows how it got to the new number, we should stick to Australia’s traditional way of counting, which has us stubbornly mired around 2.0 per cent.

The government is on firmer ground in arguing that percentages of GDP are irrelevant, the real question is whether you are spending enough to acquire and operate the military capabilities you need.

That said, percentages of GDP are a clear signal of the government’s understanding of geopolitical risks and its willingness to manage them, regardless of what it acquires.

Of course, spending increases have to translate into capability. So which capabilities should Australia be acquiring and what will they cost? There are myriad views on what military capabilities Australia needs. Moreover designing a military is not an exact science, particularly if you are seeking to deter an adversary; effective deterrence is just as much an exercise in psychology or even mind reading as it is a rigorous calculus.

Nevertheless, we can form some clear judgments on what we need. First, if we no longer have 10 years of warning time of conflict (and in the view of many analysts, much less) we need to ensure that the existing force is ready to fight.

But all the anecdotal evidence, confirmed by the Australian National Audit Office’s recent report on the state of the navy’s amphibious fleet, is that Defence’s sustainment budgets – the money that maintains and repairs equipment and supports training – are being harvested across the defence force. It’s being funnelled into the rapidly-growing cost of acquiring nuclear submarines, which will exceed $3bn this year alone. We can’t let the opportunity cost imposed by the future force consume the current force.

Which leads to the second judgment, namely you can’t acquire and operate exquisite nuclear capabilities on a budget of two-and-a-bit-per cent of GDP. We have a relevant analogy that shows the impact of doing that, namely the UK. Not only has supporting its naval nuclear capability meant that it’s run down its army and air force, but its navy now has an historically small number of surface combatants. The irony is that very often it can’t get a nuclear-powered attack submarine to sea.

Third, with all of the acknowledged risks around the AUKUS SSN program, Australia needs to invest in mitigations should the SSN be delivered late (or not at all) or not be the answer to a maiden’s strategic prayer that they are often made out to be. There are different views on what those could be, from stealth bombers to swarms of drones, but currently any discussion of the issue is dismissed as undermining AUKUS and indeed the Alliance. Whatever the solution, it will take additional money.

Fourth, the ADF needs to build its resilience and be able to stay in the fight beyond the initial exchange of fire. As the conflict in Ukraine has confirmed, war between modern industrialised nations can last years. That requires investing in Australia’s industrial and technological base in ways that are uneconomic by any peacetime measure. It may be easier in peacetime to acquire weapons overseas, but if they are not available when you need them in time of conflict, it’s a failed approach.

Finally, one common argument against GDP spending benchmarks is that we should first determine what capabilities we need and then fund them. Analysis has consistently identified many of the capabilities Australia needs – but they have been ejected from Defence’s acquisition plans due to lack of funds. Air and missile defence and maritime mine clearance are essential for the government’s military strategy of denial, but their funding lines have evaporated. Funding is needed to restore them to the defence acquisition program.

Addressing these five gaps might add up to 3.5 per cent of GDP. That would represent an annual budget of $100bn in current terms. It could be less, but if doubts around the reliability of the US as our security guarantor continue to grow, it may need be more.

Either way, it will be significantly more than the current 2.05 per cent or indeed 2034’s target of 2.33 per cent. Whatever the number, it will be a very large pill for any government to swallow; it’s not surprising that changing the definitions of the debate has been more palatable.

But a proper debate about what we need to spend starts with acknowledging the mismatch between the government’s strategic assessment that we are in the most uncertain and dangerous times since World War II and the current defence funding level that falls well below what Australia spent in that era of extended peace.

Dr Marcus Hellyer is head of research at Strategic Analysis Australia. This article was first published in the Australian.