The protein system
Australia's protein economy is built for export. In 2024 the country produced approximately 2.6 million tonnes carcase weight of beef and veal and exported 1.3 million tonnes shipped weight, remaining the second-largest beef and bovine meat exporter behind Brazil.[4] Sheepmeat exports reached a record 614,389 tonnes shipped weight in the same year, and Australia remains the largest sheepmeat exporting country.[5] The 2024-25 wheat harvest came in at 34.1 million tonnes, of which 23.7 million tonnes were exported.[6] Milk production was 8.47 million tonnes in 2023, with exports equivalent to 22 per cent of output.[2] Poultry meat, at about 1.5 million tonnes a year and almost entirely chicken, is the most consumed meat at roughly 50 kilograms a head, and is produced at home behind strict biosecurity rules.[7]
On FAO Food Balance Sheet data for 2023, Australian production covers 314 per cent of domestic beef supply, 318 per cent of sheepmeat and goat meat, 104 per cent of poultry meat and 97 per cent of eggs.[2] The widely repeated claim that Australian farms could feed 75 million people is recorded here as a secondary estimate, unattributed to any primary source; competing versions cite 80 million.[22] Consumption sits inside the production envelope everywhere except the sea and the pig shed: imports supply 73 per cent of seafood domestic supply by product weight and 38 per cent of pigmeat, the latter concentrated on the United States, which took 41 per cent of boneless fresh and chilled pig meat imports in 2023 and 51 per cent in 2024.[2][12] The dependence that matters runs the other way, through inputs. Net imports met 83 per cent of nitrogen nutrient use, 47 per cent of phosphate use and 100 per cent of potash use in 2023;[10] paraquat, the double-knock partner to glyphosate in no-till cropping, is now produced primarily in China;[14] and domestic refineries account for around 20 per cent of national fuel supply.[15]
Durable domestic capacity: 88
Quantitative score 88, no adjustment applied
Australia is self-sufficient across the board, and the score reflects it. It runs structural surpluses in beef, sheepmeat, wheat, pulses, dairy and poultry simultaneously, and the base is durable: the great majority of ruminant protein grows on rangeland pasture, drawing on rainfall rather than irrigation, and national water stress is under 5 per cent.[1][2][8]
Two qualifications keep the score below the low 90s. First, seafood: domestic wild-capture and aquaculture output covers barely more than a third of demand, and the most recent harvest figure the Tasmanian regulator has published remains the 2020-21 observation of more than 83,000 tonnes of salmonids.[2][19] Second, volatility: the 2018-19 drought pushed the sheep flock down 7 per cent to 66 million head, its lowest since 1905, and forced mass destocking of cattle.[9] We apply no adjustment: the indicators already reflect both the abundance and the seafood gap, and the pillar holds at 88.
Resource headroom: 73
Quantitative score 78, adjusted down 5 points
On the static numbers Australia has more headroom than almost any entity we assess. Freshwater withdrawals ran at 4.67 per cent of available resources in 2022, far below the global stress threshold.[8] The cattle herd of 30.4 million head grazes land that competes with no human food crop, cropland per person is among the highest in the world, and the feed grain supporting intensive livestock comes from domestic surpluses.[4][6]
The trend is less comfortable than the stock. ABARES modelling estimates that changes in seasonal conditions over 2001 to 2023, measured against the 1951 to 2000 baseline, reduced average annual broadacre farm profits by 18 per cent, about $28,500 per farm.[24] Floods killed up to 500,000 cattle in 2019, more than 100,000 in 2025, and left more than 48,000 dead or missing in northwest Queensland in early 2026.[22] The protein mix is drifting towards feed intensity: grain-fed cattle accounted for 38 per cent of adult cattle slaughtered in 2024, the pig and poultry sectors consumed 712,000 tonnes of imported soya bean cake in 2023 and 944,000 tonnes in 2024, 96 per cent of it Argentine in both years, and salmon farming relies on wild-caught marine ingredients, with Tasmanian salmonid beach value easing to $1,281 million in 2023-24 on lower production.[4][12][31] Against those pressures the resource base does recover: the sheep flock reached 79.1 million head at June 2024, its largest since 2007.[5]
Two further checks establish that the headroom which remains is real rather than nominal. Agricultural credit is deep and getting deeper: the FAOSTAT orientation index for credit to agriculture, forestry and fishing, which normalises the sector's share of lending against its share of GDP, stood at 3.79 in 2024, nearly four times the parity level of 1.00, and never dropped below parity at any point from 2013 to 2024; the dollar value of that credit rose from US$57.8 billion to US$84.7 billion over the same span.[53][54] Land under crop expanded 33.2 per cent in absolute terms between 2000 and 2024, from 23.5 to 31.3 million hectares, though population grew faster and cropland per person eased from 1.24 to 1.18 hectares.[53] Irrigation is the narrower part of the picture: the area equipped for irrigation covers 6.9 per cent of arable land in 2024, down from 10.0 per cent in 2000, because Australia's expansion has run through rainfed cropping and pasture rather than water infrastructure.[53] Where public money has tried to force expansion through irrigation at the frontier, the return has been poor: the Ord River scheme in the Kimberley has returned about seventeen cents for every public dollar invested since inception, and its post-2009 expansion added 1,600 hectares at a cost of roughly $227,500 a hectare, against a smaller and better-returning national pipeline of water-security projects still adding new irrigable land at the margin.[55][56] The indicators put this pillar at 78; we deduct 5 for the documented, recurring degradation trend that static land and water figures understate, taking the pillar from 78 to 73.
Import exposure: 85
Quantitative score 85, no adjustment applied
Australia's protein import exposure is low. Every major terrestrial protein runs a large net surplus, poultry imports are 0.4 per cent of domestic supply, and beef exports move to a customer base in which no single market exceeds 29 per cent.[2][4] Exposure is confined to the two substitutable niches above, seafood and processed pork.[2][12] The indicators put this pillar at 85. No adjustment applies: Australia's exports and its niche imports both travel long sea lines through Southeast Asian waters, a geography risk the indicators already reflect, and the pillar holds at 85.
Upstream dependence: 43
Quantitative score 40, adjusted up 3 points
This is the pillar that separates Australia from sovereignty, and its distance from every other pillar is the widest in this assessment. FAOSTAT records no domestic production of urea or of any other fertiliser product: 3.2 million tonnes of urea were imported in 2023, and about 3.85 million tonnes on 2024 UN Comtrade data.[11][32] Incitec Pivot closed the country's only urea plant, at Gibson Island in Brisbane, in January 2023, having failed to secure an affordable long-term gas supply agreement.[27] When Beijing restricted urea exports in late 2021, Australia faced a simultaneous fertiliser squeeze and a shortage of AdBlue, the urea-based exhaust fluid without which the modern truck fleet cannot legally or mechanically run; ASPI's assessment is that the country was within weeks of shutting down its national road transport fleet.[18]
That dependence did not ease, and in 2026 it was tested for real. UN Comtrade data confirms that 68 per cent of Australia's 2024 urea imports came from five Gulf producers, the United Arab Emirates, Qatar, Saudi Arabia, Oman and Bahrain, at the upper end of the 65 to 70 per cent estimate this assessment had previously drawn from a single ASPI piece.[25][32] When conflict broke out in late February 2026, outbound fertiliser shipments through the Strait of Hormuz to destinations outside the Persian Gulf came to a standstill for around four months. Domestic granular urea, priced fca Geelong, rose from about A$830-840 a tonne immediately before the conflict to a peak of roughly A$1,430-1,440 a tonne on 7 May 2026, a rise of more than 70 per cent, before a ceasefire reopened the strait in June and prices retreated close to their pre-conflict level.[33][37] ABARES's June 2026 crop report attributes a forecast 21 per cent fall in national winter crop production, to 54.5 million tonnes, to a combination of dry soil moisture and elevated input costs, and reports that growers had enough fertiliser to plant but faced real uncertainty over paying for July-August urea top-dressing.[36] One chokepoint was traded for another, the second one closed for four months, and the closure showed up in the paddock as a cost and a planting decision rather than as an empty shed.
The pattern repeats across the input basket. Potash is entirely imported, with 335,541 tonnes of muriate of potash brought in during 2023 against no recorded domestic production.[10][11] Phosphate is the more comfortable case: domestic output of 413,250 tonnes of P2O5 covered rather more than half of the 772,092 tonnes used in 2023,[10] but that share rests on a single operation at Phosphate Hill. Poultry production is built on breeding strains from overseas genetics companies, which displaced local breeders after the ban on importing genetic material was lifted in the 1990s.[7] Ninety-nine per cent of the volume of Australia's goods trade moves by sea, on a national flagged fleet of about nine commercial ships, down from a peak of more than 100.[20]
The mitigations are real, and one of them is firmer than it was. Feed grain is sovereign and abundant,[6] gas is sovereign and the failure to convert it domestically has been economic rather than geological, and phosphate rock and ruminant genetics are domestic strengths. Perdaman's Project Ceres at Karratha, a US$4.5 billion, 2.3 million tonne a year urea plant with a twenty-year Woodside gas agreement and an Incitec Pivot offtake, now carries a firm commissioning target of March 2027 and first urea from June 2027, with modules on site since late 2025 and its renewable-power stage under construction from mid-2026.[18][38] The indicators put this pillar at 40, once the confirmed Gulf concentration and the demonstrated absence of any interim buffer are weighed against the position available before the shock; we add 3 for Ceres, since the ground remains admissible, but the plant sat outside the window that mattered in 2026, and a firmer date eleven months out is worth less than an operating plant would have been, taking the pillar from 40 to 43.
Access and affordability: 84
Quantitative score 84, no adjustment applied
Australia scores strongly on every national-average indicator here. The FAO/World Bank Cost and Affordability of a Healthy Diet dataset puts the share of the population unable to afford a healthy diet, the indicator carrying 40 per cent of the pillar's weight, at 3.2 per cent in 2024, around 0.9 million people, against a diet cost of 3.23 international dollars at purchasing power parity a day.[41] The same series holds 3.2 per cent for 2025, so the reading is stable rather than a single-year low. Undernourishment sits below the FAO's 2.5 per cent measurable floor, and has done for every three-year average FAOSTAT publishes through 2023-2025, the standard reporting convention for developed countries.[42][51] Child stunting has no current national survey: the only estimate on record, in the UNICEF/WHO/World Bank Joint Child Malnutrition Estimates series, is 1.9 per cent from 2007, retained here as background only.[43]
Food price volatility has been moderate rather than severe. On the ABS monthly CPI indicator the food and non-alcoholic beverages group peaked at 9.5 per cent in the year to December 2022, stood at 7.9 per cent in the year to May 2023, and had settled to around 3 per cent by early 2025, with fruit, vegetables, meat and seafood the most volatile sub-groups.[44] What keeps this pillar out of the low 90s is physical access, and the gap is wider than a national average suggests. A 2026 systematic scoping review of nine Australian pricing studies puts food in outer regional, remote and very remote areas 30 to 160 per cent above metropolitan comparators, averaging 48.8 per cent, with freight and store operating costs the principal drivers;[46] the National Indigenous Australians Agency puts the remote premium at about 39 per cent on supermarket supplies.[52] The distributional consequence is measurable: 41 per cent of Aboriginal and Torres Strait Islander households reported food insecurity for lack of money in 2022-23, rising to 50.7 per cent in remote areas against 40.4 per cent elsewhere, a gap the national average does not show.[45] We apply no adjustment, since the physical access indicator already carries that finding within the pillar's indicators, and the score holds at 84.
Shock endurance: 66
Quantitative score 69, adjusted down 3 points
Australia's endurance is anchored by the deepest passive buffer in the index. With beef and sheepmeat production each running at more than three times domestic supply, export redirection covers domestic protein demand several times over, with no rationing mechanism required beyond price.[2] Wheat ending stocks for MY2025-26 stand at 5.61 million tonnes against domestic use of 9.60 million tonnes, a stock-to-use ratio of 58 per cent, though the MY2026-27 forecast takes stocks back to 3.19 million tonnes.[6] The country came through COVID, the 2018-19 drought and the Ukraine price shock without rationing a staple, and fiscal capacity to outbid on world markets is strong.[1][9]
The 2026 Strait of Hormuz closure was the first real test of that fiscal capacity applied to an input shock, and it moved faster than the 2021 precedent. Parliament passed the Export Finance and Insurance Corporation Amendment (Strategic Reserve) Bill 2026 on 1 April 2026, roughly five weeks after the conflict began, giving Export Finance Australia new powers to secure strategic materials.[40] Backed by an A$7.5 billion Fuel and Fertiliser Security Facility, the government had underwritten about 340,000 tonnes of additional urea across ten shipments with CSBP, Incitec Pivot and Summit Fertilizer by the Fertiliser Supply Working Group's 26 June 2026 communique.[34][35] That is a funded, legislated, delivered mechanism where none existed before, though at under a tenth of one year's urea imports it is a bridge, not a reserve proportionate to the trade it was built to cover.
What the surplus and the new facility together could not fully buy was time for the input system, and the strain still showed. Diesel cover reported under the Minimum Stockholding Obligation averaged 32 days in the March quarter of 2026 and stood at 39 days on 28 July 2026, with petrol at 43 days and jet fuel at 34.[15] Australia last complied with the International Energy Agency's 90-day stockholding obligation in 2012 and has been in breach since, having committed to returning to compliance by 2026.[16] Diesel obligation levels sit at 20 days for refiners and 32 for importers, with a temporary 20 per cent reduction extended to 30 September 2026 in exchange for supply commitments.[17] There is still no strategic grain reserve and no crop chemical stockpile, and the December 2021 AdBlue episode, which ran the country close to exhaustion with no comparable institutional response at the time, remains the sharper historical warning.[18]
The 2026 shock ran for about four months, from late February to a June ceasefire that reopened the strait; roughly 200,000 tonnes of urea moved through it in the following weeks and more than 380,000 tonnes were in transit, and prices had retreated close to pre-conflict levels by early July.[37] Fuel volumes held throughout, with stocks staying above pre-conflict levels and only 65 of 8,118 retail sites out of diesel at 31 July 2026, even as Singapore gasoil traded 80 per cent above its pre-conflict level in late July.[15] ABARES's June 2026 crop report puts the input side of the cost in agricultural terms: a forecast 21 per cent fall in winter crop production, to 54.5 million tonnes, still 4 per cent above the ten-year average, with growers reporting enough fertiliser to plant but rationing application and shifting area toward canola and barley rather than wheat at prevailing prices.[36] That is a materially better outcome than the 2021 AdBlue scare or the 2024 H7 avian influenza outbreak, in which about 1.8 million birds were destroyed,[29] more than 600,000 laying hens were culled by 17 June 2024 against a national flock of more than 21 million,[28] and egg purchase limits of two cartons per customer followed in every state except Western Australia.[21] Both of those episodes produced visible physical scarcity within days; the 2026 fertiliser shock arrived through price and through next season's input bill, not through an empty tank or an empty shelf.
Crisis purchasing power cuts both ways. Foreign exchange reserves cover only about 1.5 months of imports, thin by the standard applied to countries that must defend a managed exchange rate.[47] That signal matters less for Australia than it would elsewhere, because the dollar floats freely and the sovereign is rated AAA by all three major credit rating agencies, with S&P reaffirming on 22 October 2025.[48] Fiscal space is real and was demonstrated in 2026: general government gross debt sits at about 51 per cent of GDP,[49] and the A$7.5 billion Fuel and Fertiliser Security Facility was raised and disbursing within weeks of the shock beginning.[34][35] Social protection has already proved it can scale fast and large in a crisis: the JobKeeper wage subsidy paid out A$88.82 billion to 1,068,856 entities, supporting an average of 3.6 million individuals a month across the original April to September 2020 period.[50] On balance, genuinely thin reserves outweigh, by a small margin, strong fiscal and social-protection capacity.
The indicators put this pillar at 76 before crisis purchasing power, reflecting the demonstrated policy response captured in the Strategic Reserve legislation and the Fuel and Fertiliser Security Facility. AAA credit access and the JobKeeper social-protection scale-up are real, but reserve cover of only 1.5 months of imports is thin, which brings the indicator-based score to 69. We deduct 3 for the continued absence of strategic reserves proportionate to Australia's actual import volumes and for the ongoing IEA stockholding breach, taking the pillar from 69 to 66.
Trajectory
Policy is moving, and 2026 supplied the first real test of it. The Feeding Australia national food security strategy, funded with $3.5 million in the 2025-26 Budget[23] and supported by a skills-based National Food Council of eleven inaugural members appointed in November 2025,[30] was the first attempt to treat food security as national policy rather than agricultural policy, but on its own it bought a strategy document rather than a stockholding regime. The Strait of Hormuz closure forced the harder decision: Parliament passed Strategic Reserve legislation within five weeks of the conflict starting, and an A$7.5 billion Fuel and Fertiliser Security Facility had delivered about 340,000 tonnes of urea by late June.[34][35][40] That is a genuine, funded response, arriving after the shock rather than before it, and still small next to the roughly 3.85 million tonnes Australia imports in most years.[32] The consequential investment remains physical: Project Ceres now carries a firm commissioning target of March 2027 and would on completion move the upstream pillar several points,[18][38] while the strategic fleet programme has secured the first of three pilot vessels against a pledge of twelve, with the Transport Minister citing the closure of the Strait of Hormuz as the reason for new urgency.[20] Diesel obligation levels were eased rather than raised through 2026.[15][17]
The downside risks are climatic and structural, and 2026 was a reminder that they are recurring rather than one-off. The 18 per cent erosion of broadacre farm profit is a trend, and each drought cycle now starts from a warmer baseline.[24] A closure at Phosphate Hill would move phosphorus dependence from 47 per cent towards the potash position of 100 per cent.[10] The clearest lever from here is completion of Ceres, a stockholding regime for diesel, AdBlue and crop chemicals sized to actual import volumes rather than to a single crisis, and credible onshoring of crop protection manufacturing. Australia's path to a stronger score requires no change to what happens on its farms, only to what arrives at its ports, and 2026 showed that when a port closes, the country can legislate and fund a response quickly, even if it cannot yet stockpile its way around one.