The claim
At a press conference during a visit to a Queensland defence industry business, Opposition Leader Peter Dutton sought to burnish his party's credentials on defence spending.
"When we were in government, we had a record spending on defence," he said.
"Under Labor it had been down to the lowest level since 1938 and we have a lot to do in the defence space."
Did the Coalition have a record spending on defence after it dropped to the lowest level since 1938 under the previous Labor government? RMIT ABC Fact Check investigates.
The verdict
Mr Dutton is incorrect.
When measured as a share of GDP, defence spending under Labor fell to 1.52 per cent, which is the lowest since 1938-39 when it was 1.5 per cent.
But the high point of spending under the previous Coalition government (2.04 per cent) is far from the highest year on record, 1942-43 (34 per cent).
Even when measured using real spending figures, Mr Dutton's claim does not pass muster.
While spending in the 2020-21 financial year was the highest on record using this measure, most of Labor's tenure saw record spending in real terms at the time. The next lowest years of spending occurred under the Howard government.
While experts preferred GDP for making historical comparisons, they said both of these measures were imperfect for that purpose.
They said policy should be focused on strategy and whether defence spending set at an appropriate level for threats to national security.
Previous claims
Fact Check contacted Mr Dutton's office to ask for the source of his claim but did not receive a response.
Mr Dutton has made the latter half of the claim (regarding Labor's record of spending) before.
"Albo criticising the ADF over the flood response is a bit rich given he cut defence spending to its lowest level since 1938," Mr Dutton tweeted before the 2022 election, when he was defence minister.
That's similar to a claim which Fact Check tested in 2016, from then prime minister Malcolm Turnbull.
"Under Labor, defence spending as a share of GDP dropped to its lowest level since 1938," Mr Turnbull said at the time.
Fact Check assessed Mr Turnbull's claim on the basis of figures supplied by the Department of Defence and the Australian Bureau of Statistics to the year 2013-14.
Defence spending was measured as a proportion of GDP to enable historical comparisons.
On that measure, the year 2012-13 was found to be the lowest year on record since 1938, but Fact Check found that to be not the full story.
That's because under the Howard government, it was only 0.02 per cent higher, and Mr Turnbull had left out Labor's best year, which was better than any year under the Howard or Abbott/Turnbull governments at the time.
Defence spending as a proportion of GDP
Fact Check also previously noted the imperfect nature of the measure.
While defence spending figures are not revised, the GDP numbers are estimates, not actual figures, and the ABS continues to update them sometimes years after the event.
Perversely, this means that if a data revision shows worse economic performance than previously reported in a particular year, defence spending for that year will then look more favourable.
Nonetheless, this measure is commonly relied on for comparisons over time.
Marcus Hellyer, a senior analyst focusing on defence budgets and capability at Australian Strategic Policy Institute, a non-partisan think tank which is partly funded by the Department of Defence, told Fact Check that share of GDP was superior to nominal or real spending figures when making historical comparisons.
"I think that's a better way of comparing the level of national effort devoted to defence," he said.
Peter Robertson, a professor of economics and dean of the University of Western Australia's School of Business, who has done research in international comparisons of defence spending, agreed.
"... inflation adjusted real spending gives a measure of total military resources and this will tend to increase over time as the economy grows, soldiers become more skilled, better, and technology increases effectiveness of systems," he told Fact Check in an email.
"Hence real economic growth will drive real spending up, making it inappropriate to compare over time — unless you want to measure the real growth of our defence forces due to technological improvements."
How to measure defence spending
As Fact Check has previously explained, there's no publicly available or easily interpretable Australian government source for defence expenditure.
When Fact Check tested Mr Turnbull's claim, historical data from 1901-02 from the ABS was used in combination with more recent data supplied by the Department of Defence.
To check a more recent claim about the level of defence spending under Labor made by then shadow defence minister Brendan O'Connor, Fact Check used a combination of defence annual reports and defence portfolio additional estimates statements to create a series between the 1996-97 and 2012-13 financial years.
The portfolio additional estimates statements from later budgets can be used to extend this series to the present.
However, from 2018-19, when the Australian Signals Directorate was established as a statutory agency within the Department of Defence, its spending was reported separately from consolidated defence spending.
As ASD funding was reported as part of consolidated spending before this date, Dr Hellyer told Fact Check it should be added back in from that year to maintain a consistent series.
It should also be noted that the historical dataset and the more recent sources for defence spending are not entirely consistent in what they include and exclude.
Nonetheless, they offer a rough comparison over time and mirror the approach ASPI has previously taken to historical comparisons, which was contained in a spreadsheet Dr Hellyer shared with Fact Check.
Annual reports and PAESs are not available online before the 1997-98 financial year, which contains the 1996-97 data.
Thus, Fact Check has graphed defence spending as a proportion of GDP below, using historical figures to 1995-96 and defence PAES statements afterwards, noting the series break.
Figures for GDP were sourced from the Australian Bureau of Statistics. When Mr Dutton made his claim, the latest release was dated June 1, 2022, but the bureau has since updated these figures.
Fact Check has used the GDP estimates released in June as they were available when Mr Dutton made his claim.
Using this measure, defence spending in the year 2012-13 at 1.52 per cent is the lowest since 1938-39, when it was 1.5 per cent.
But more recent defence spending under the Coalition is far from a record on this measure.
The most recent year of spending available, 2020-21, is the high watermark of the Coalition's defence spending.
The following graph shows this relative to the Howard and Rudd-Gillard years.
At 2.04 per cent of GDP, this is vastly outdone by spending during the world wars. In 1942-43, defence spending reached a peak of 34 per cent of GDP.
What about real spending?
Experts spoken to by Fact Check preferred share of GDP over real spending to make historical comparisons.
Professor Robertson told Fact Check that ideally real spending would be calculated using a defence price deflator.
The ABS publishes such a deflator in its GDP series, but Professor Robertson cautioned it would take into account spending which is not necessarily covered by figures taken from the budget papers or historical ABS data.
Thus, Fact Check has deflated defence spending figures using a combination of the Consumer Price Index and its predecessor, the Retail Price Index, obtained from the Reserve Bank of Australia.
As the RBA's series begins in 1922, Fact Check has calculated real defence spending from that year onwards.
Indeed, in 2011-12 dollars, the most recent figure of $35,667m is larger than any other year in the historical or budget paper series.
But every single one of Labor's years of defence spending is greater than the level of spending in 1938-39 also. The next lowest years of spending were during Coalition prime minister John Howard's tenure.
This is consistent with calculations previously published by ASPI in 2011, which calculated real defence spending in 2011-12 dollars between 1901 and 2009 (with projections to 2014).
The highest level of defence spending in ASPI's series was during WWII.
The data once again dispels the implication in Mr Dutton's claim that Labor's defence spending is historically low while the Coalition's is historically high.
Dr Hellyer said this comparison shows why GDP is better for historical comparisons than real spending.
"Just by looking at that comparison between today and World War II in real dollars, it says we're spending more now than we were at the height of World War II, which is ... a pretty ludicrous way of trying to assess defence spending," he said.
Are historical comparisons even valid?
Though Dr Hellyer agreed that GDP was better than real spending for historical comparisons, he argued that these long-term comparisons shouldn't be made at all.
That's because government spending and societal expectations of government spending change over time.
"So if we go back a few centuries the state coffers essentially did two things. One was to support the household of the king, and the other one was to have an army and that would have been the vast, vast bulk of government expenditure," he said.
Dr Hellyer noted that defence as a proportion of government expenditure was now in the single digits.
"So those longer term comparisons are not particularly useful," he said.
Andrew Carr, a Senior Lecturer at Australian National University's Strategic and Defence Studies centre argued for the decoupling of GDP and defence spending almost a decade ago.
As Fact Check has previously reported, this became the stated policy of the Coalition government upon the release of the Defence White Paper in 2016.
Dr Carr told Fact Check he still holds the view that the share of GDP measure "is a very poor way to measure defence spending".
"Australia’s spending as a percentage of GDP has gone considerably up the last few years, yet every expert believes we are far more at risk than five years ago. So GDP simply doesn’t capture what’s important for understanding our security or the appropriateness of government responses to the threats we face, " he said.
But he cautioned that even looking at the level of spending could paint a misleading picture also.
"“How much you spend only loosely translates into how much national security we’re purchasing. Governments can tighten budgets but be more effective, or spend wildly and waste money on poorly chosen capabilities or priorities.
“At the very least the public should expect consistency in how politicians talk about levels of spending. Though they’d be far better off if the argument was about strategy and what we’re doing, rather than assuming the numbers are the main story.”
Dr Hellyer said that defence spending needed to be "meaningful against the threat".
"So, if there's low threat and you're spending 5 per cent of GDP on defence, you're wasting public money. You're raising taxes that don't need to be raised. You're increasing the debt. You're sucking revenue money out of useful public services like schools and hospitals and infrastructure. So a high number isn't necessarily a good thing if there's there's no threat.
"So the question is not: what are we spending now compared to a different point in time? The question is: are we spending what is needed in our current strategic circumstances?"
Professor Robertson agreed:
"The obvious point to make is that the strategic environment has changed over time, so spending on defence needs to be appropriate for the strategic environment. Perceived threat levels have changed as have perceived alliances," he said.
"Not having high spending in 1938 might be a very different proposition to not have high spending in 2008-2013.”
Sources
- Peter Dutton, Doorstop interview, September 2, 2022
- Peter Dutton, Tweet, March 8, 2022
- ABC Fact Check, Fact check: Did defence spending under Labor drop to its lowest level since 1938?, February 25, 2016
- RMIT ABC Fact Check, Brendan O'Connor says defence spending under the Howard and Rudd-Gillard governments was similar, on average. Is that correct?, April 20, 2022
- Australian Bureau of Statistics, Defence expenditure over the century, January 25, 2001
- Department of Defence, Annual reports 1997-98 and 1998-99
- Department of Defence, Budget documents, PAES 1999-00 to 2020-21
- Australian Bureau of Statistics, Australian National Accounts: National Income, Spending and Product, Table 34 Key Aggregates and Analytical Series, Annual, June 1, 2022
- Reserve Bank of Australia, Measures of consumer price inflation, Historical series and explanatory notes, July 28, 2022
- Australian Strategic Policy Institute, Australian defence almanac 2011-2012, July 19, 2011
- Andrew Carr and Peter J. Dean, The funding illusion: The 2% GDP furphy in Australia's defence debate, Security Challenges, 2013
