The managed decline of Australia’s fuel supply
March 2026 was not a surprise. It was the predictable consequence of a deliberate policy architecture.
On 11 March 2026, the national average price of diesel hit 240.2 cents per litre. In three weeks, diesel had risen 64 cents, a 36% increase. Petrol was up 49 cents. Regional service stations ran dry. The Prime Minister went on television and told Australians to stop panic buying. The ACCC called every major fuel retailer into a meeting. The government set up a fuel supply task force.
And through all of it, not one major media outlet published a breakdown of where the 240 cents actually goes.
The public debate has been about price gouging, Middle East wars, panic buying, and government inaction and policy failure. Commentators blamed petrol stations. Petrol stations blamed Singapore. The government blamed consumers, then the Middle East and the bad orange man, then quietly released 762 million litres from strategic reserves and simultaneously reduced the mandatory minimum stockholding requirement. Australia responded to running low on reserves by officially requiring less.
But as in all things, if you really want to know what is going on, you have to follow the money. And if you do that, two things become clear: the crisis was not a surprise, and the debate happening right now is asking the wrong questions.
Key takeaways:
Australia is permanently one maritime disruption away from a fuel crisis. By design
Australia holds 26–29 days of petrol and diesel reserves. The International Energy Agency (IEA) requires 90 days. The government has known about this gap for over a decade and described the cost of closing it as “economically unviable.”
The government takes more out of every litre than the refiner, shipper, wholesaler, and retailer combined
Excise plus GST equals approximately 60c/L, roughly a third of the normal pump price. The public debate about industry margins has focused on the wrong margin-taker.
The Strait of Hormuz matters to Australia. Just one step removed
86% of Australia’s diesel is imported. Approximately half of that comes from South Korea and Japan, which run on Middle Eastern crude.
The value chain
Crude Oil
Australia is almost entirely a price-taker on crude oil
Australia produces approximately 220,000 barrels per day of crude oil and condensate. It exports more than 85% to refineries in South Korea, Japan, and Singapore that pay a premium for it.
The reason is a decades-old paradox. Australian crude is light and sweet, the premium end of the market, commanding a 5–10% price premium in Asian markets. Domestic refineries were configured decades earlier for heavier crude from the Middle East and Southeast Asia. So it is genuinely more profitable to export Australian light crude at a premium and import cheaper, heavier grades suited to domestic configurations. So that is what happens.
The crude that isn’t exported (roughly 8–16% of what the two domestic refineries actually process) supplements the imported oil they run on. Every movement in the global oil price, amplified by a falling Australian dollar, flows directly into the cost of running those plants. The exchange rate alone adds 3–4c/L for every 10% the dollar falls.
In normal conditions, crude oil accounts for approximately 36c/L of a 180c/L pump price. In March 2026, as global oil prices rose 75% in three months and the dollar weakened, that component roughly doubled.
Domestic Refining
Two refineries remain. Both exist because of a government subsidy that barely fires.
Australia had eight refineries in 2005. It now has two. Lytton (Ampol, Brisbane) and Geelong (Viva Energy). Together, they refine approximately 1,242 ML per month. Enough to cover roughly 20–25% of domestic petrol and diesel demand.
The refiner margin, the spread between the cost of crude and the value of refined products, is the thin line between viability and loss. Lytton’s margin ranged from USD 7.44/bbl in the first half of 2025 to USD 15.14/bbl in Q4 2025. At the current scale, that translates to approximately 5–8c/L under normal conditions.
The government introduced the Fuel Security Services Payment (FSSP) in 2021, designed to keep both refineries open when market economics alone would close them. It is a subsidy that is triggered when refining margins fall below a set floor. In five years, it has paid out exactly once.
Lytton lost A$42.3m in FY2024, but did not meet the threshold. The mechanism is there; it just doesn’t fire. But without it, the commercial case for both refineries collapses, and 80% import dependency becomes 100%.
Imports and Storage
86% of Australia’s diesel never sees a domestic refinery. It goes straight from an overseas tank to a local one.
The import terminals are the actual fuel system. The refineries are a supplement.
In January 2026, importers brought in 1,032 ML of petrol (against 477 ML of domestic production) and 2,662 ML of diesel (against 428 ML domestic).
Import share: 68% for petrol, 86% for diesel.
Those imports arrived primarily from South Korea (29%), Singapore (19%), and India (a rapidly growing source, up approximately 50% year-on-year, with 55% of Indian refinery inputs being Russian crude)
The product moves from ship to bulk storage terminal. Those terminals hold approximately 26–29 days of petrol and diesel cover against an IEA obligation of 90 days.
The gap has been acknowledged by successive governments and accepted as “too expensive to close” for over a decade. The cost of closing it is approximately A$20 billion. The government has described this as economically unviable. The cost of a 30-day supply disruption under the current configuration is estimated at A$20–50 billion in lost economic output. This comparison has not featured prominently in the policy debate.
The IEA is an intergovernmental body founded in 1974 after the oil crisis, originally to coordinate energy policy among developed nations. Australia is a member.
Wholesale and distribution
Singapore sets your pump price. In March 2026, retailers repriced yesterday’s cheap fuel at today’s crisis prices.
Singapore Mogas 95 is the number that moves Australian retail petrol prices. Under normal conditions, it accounts for approximately 42% of the pump price. Singapore Gasoil 10ppm does the same for diesel.
The mechanics: Terminal Gate Prices are the published spot price at bulk distribution terminals and are recalculated daily based on a rolling Singapore price average. The usual lag between a Singapore price movement and its appearance at the TGP is 7–10 days.
In March 2026, that transmission was instantaneous. As the conflict escalated and Singapore prices spiked, retailers repriced existing inventory. The fuel they had purchased the previous week at pre-crisis prices was set at crisis-inflated prices.
Singapore doesn’t produce oil; it refines it, almost entirely from Middle Eastern crude shipped through the Strait of Hormuz. Roughly 20% of global seaborne oil passes through that single chokepoint. When the conflict threatened that route, crude prices spiked globally, and Singapore refined product prices spiked further because the market was pricing in both the crude shortage and the risk that Singapore refineries wouldn’t receive their next shipment.
Singapore Mogas 95 rose 37c/L in three weeks. Retail prices rose 48.8c/L. The ACCC was direct: “petrol retailers increased prices at the pump when they were selling fuel they had bought before the conflict at cheaper prices.” The difference (~7c/L) was margin extracted from pre-purchased inventory.
The distribution failure was less visible but more damaging for communities outside the capital cities. Regional independent distributors reported receiving 10% of their normal daily allocations from major wholesalers, who prioritised their own retail networks. Fuel sat in bulk terminals in Sydney and Melbourne while rural Queensland and Western Australia ran dry. The physical distance between a bulk terminal and a country service station is measured in days, not hours. The system has no mandated obligation to serve regional independents during declared emergencies.
Retail
The retailer earns 3c/L on fuel. The shop is the actual business.
The Gross Industry Retail Difference (GIRD is the ACCC’s official term for the retail margin), the spread between what the station pays for fuel and what it charges you, was 17.9c/L in the December quarter 2025. After wages, rent, electricity, and card fees, net fuel profit is approximately 3c/L per litre.
The government takes 60c. The retailer keeps 3c.
The meaningful margin in Australian fuel retail is not at the bowser. It is in the shop. On the Run (OTR), acquired by Viva Energy, generates more than 70% of its EBITDA from non-fuel. This is the arithmetic of the chain: petrol is traffic. The business is coffee and sandwiches. Any executive in adjacent retail, convenience, QSR, or grocery should note that the most contested real estate in Australian fuel retail is not the forecourt. It is the 200 square metres behind it.
This didn’t happen to us. We built it.
The question behind the question: how did a country with abundant reserves, a functioning refining industry, domestic crude production, and full IEA membership end up holding less than 30 days of fuel reserves, exporting its own oil while importing refined product, and pricing its fuel entirely from a benchmark in Singapore?
The answer is four policy decisions, made over 40 years, each defensible in isolation, damaging in combination.
Import parity pricing (Fraser Government, 1970s)
In the mid-1970s, the Fraser Government introduced import parity pricing for domestically produced crude oil. The rationale was straightforward: price domestic crude at world market rates, and producers will invest in exploration rather than accept artificially suppressed prices. John Howard defended the logic in 1990: “By taking the world price we encourage investment in the industry. If you start artificially suppressing the price you won’t have the same amount of investment.”
It worked. Production grew. Australia reached approximately 70% domestic crude self-sufficiency through the 1980s and 1990s.
But import-parity pricing had a structural consequence that its architects did not adequately weigh. Once domestic crude is priced at world market rates, producers can, and do, export it to Asian markets at premium prices rather than supply domestic refineries at the same cost. The policy, designed to encourage upstream investment, simultaneously created incentives that channelled production away from domestic processing.
The crude quality mismatch
Our premium quality Australian crude is ideally suited to Asian mega-refineries that pay a 5–10% price premium over benchmark grades. Domestic refineries, built in the 1950s and 1960s, were configured for heavier crude. That is what was available and cheap when they were designed. Australia’s major oil fields weren’t discovered until later: Bass Strait in 1965, the North West Shelf through the 1970s. By the time it was clear that domestic production would be light and sweet, the economics of retrofitting the refineries made no sense when the alternative, exporting the premium crude and importing cheaper heavy crude to run through the existing plant, was more profitable. Nobody made a dumb decision. The decisions were each rational. The problem is that nobody connected the dots when circumstances changed, and no government ever required them to do so.
Import parity pricing institutionalised this arrangement. The deregulation of crude exports in the mid-1980s formalised it. By the 1990s, Australia was systematically exporting its premium domestic crude and importing cheaper heavy crude for domestic processing. A rational market outcome built on a structural policy foundation.
Six refinery closures - commercial pressure, policy acquiescence (2013–2021)
Australia had eight operational refineries in the early 2000s. Between 2013 and 2021, six closed. Shell’s Clyde in Sydney (2013), Caltex’s Kurnell in Sydney (2014), BP’s Bulwer Island in Brisbane (2015), BP’s Kwinana in Perth (2021), ExxonMobil’s Altona in Melbourne (2021). Each closure was commercially justified: Asian mega-refineries, 15 times larger and operating in lower-cost jurisdictions with newer technology, had made small Australian facilities uneconomic. A Singapore refinery processes 1.1 million barrels per day. Clyde processed 75,000.
But the government had the power to intervene and consistently chose not to. Julia Gillard, on the 2012 closure of Kurnell: “ageing capital... really quite a small refinery... by the standards of the world.” Resources Minister Martin Ferguson, on Clyde: the closure was “a matter for the company to decide commercially.”
This is the critical tell: in 2021, the Morrison Government introduced the Fuel Security Services Payment specifically to prevent the closure of Lytton and Geelong. The FSSP proves that government intervention was always possible. Six refineries closed before it existed. Two survived because of it. The governments that accepted closures as inevitable in 2012 and 2014 were the same governments that intervened in 2021.
Singapore pricing (gradual drift, 1990s–2000s)
Australia’s integration into Singapore-based fuel pricing benchmarks was not a single decision. It accumulated: import parity pricing in the 1970s, crude export deregulation in the 1980s, Singapore benchmarks adopted progressively through the 1990s and 2000s, and the ACCC endorsing the arrangement as competitive and transparent. By the early 2000s, Australian retail fuel prices were fully tied to Singapore MOGAS95. No domestic price anchoring remained. Australia had made itself a complete price-taker on the world’s most geopolitically exposed commodity.
Simultaneously resource-rich and fuel-insecure
These four decisions share a common logic: the conviction that market integration and specialisation produce better outcomes than domestic capacity and sovereign resilience. Each step appeared rational. Together, they dismantled the infrastructure of energy independence.
The paradox is stark. South Australia may hold 200 billion barrels of shale oil reserves, a resource endowment that would place Australia among the top three countries globally. The refineries that could process it have been closed. The policy framework to develop it has not been built. Australia is resource-rich and fuel-insecure simultaneously.
That is not a market outcome. It is a policy outcome.
EVs are not the answer. Not yet.
When pressed on fuel security in March 2026, the government reached for the clean energy transition. Electric vehicles will reduce fuel dependency. Net zero by 2050 is the structural fix. But the maths does not support this as a near-term policy response.
Australia has approximately 454,000 electric vehicles, 2.1% of a fleet of 21.7 million registered vehicles. New EV sales reached 8.3% of total new vehicle sales in 2025, up from 7.4% in 2024. An encouraging trajectory, but starting from a very small base.
Fleet turnover is slow. The average Australian vehicle is 10.1 years old. Approximately 4.4% of the fleet retires each year, totalling around 950,000 vehicles. Achieving a 20% reduction in national fuel consumption requires roughly 36% of the passenger vehicle fleet to be electric, which is equivalent to replacing roughly 5.3 million of 14.7 million passenger vehicles. Under CSIRO’s Progressive Change scenario (reflecting existing policy settings), that milestone arrives somewhere between 2045 and 2048. Under the more optimistic Bloomberg NEF trajectory, 60% of new sales electric by 2030, arriving around 2041 to 2044.
Freight is the harder problem. Trucks account for 22.6% of Australia’s national fuel consumption. There is no commercial electric alternative for long-haul articulated freight at scale today. Mining and agricultural vehicles are 10–20 years from meaningful electrification. EVs will not contribute materially to freight fuel demand before 2035 under any credible scenario.
The net zero transition has genuine long-term energy security logic. An EV fleet running on domestic renewable electricity represents real sovereignty. No Singapore benchmark, no Strait of Hormuz, no Middle East price shock. The argument is valid. But it creates a different dependency. EV battery supply chains are dominated by Chinese manufacturing. Lithium processing is geographically concentrated. The geopolitical exposure is just a different vector.
EVs are a 2045 solution to a 2026 problem. The additional two to three million electric vehicles required to materially reduce Australian fuel demand cannot be manufactured and deployed within any timeframe relevant to the current geopolitical risk.
Singapore sets your price. The government takes the biggest cut.
At 180c/L under normal conditions, the pump price breaks down approximately as follows:
Crude oil: ~36c/L
Domestic refining margin: ~6c/L
Shipping, insurance, wharfage: ~4c/L
Terminal storage and handling: ~5c/L
Singapore-linked wholesale component (Mogas 95 benchmark plus distribution): ~69c/L
Retail gross margin (GIRD): ~18c/L (net fuel profit: ~3c/L after costs)
Excise and GST: ~60c/L
The government takes more from every litre than the refiner, shipper, terminal operator, wholesaler, and retailer combined. In March 2026, at 240c/L, the excise component is fixed at 50.8c/L per litre; the GST component rose with the price. The government’s per-litre revenue increased as the crisis deepened.
The public debate about whether fuel companies are gouging is not irrelevant. The ACCC found that retailers extracted approximately 7c/L in crisis margin from pre-purchased inventory. That is worth scrutinising and, if a pattern, worth addressing. But it is a second-order issue sitting atop a first-order structural problem: a supply chain built over 40 years to be entirely dependent on global markets with no sovereign buffer.
The task force is asking the wrong questions
The government’s new fuel supply task force was established in response to empty pumps and public anger. The policy debate has centred on two proxies: whether fuel companies are gouging (the margin question) and whether the Middle East conflict is the cause (the geopolitics question).
Both are real. Neither addresses the structural vulnerability that made March 2026 inevitable.
Should Australia hold more strategic reserves?
The 90-day IEA standard costs approximately A$20 billion to meet through domestic storage infrastructure, the number successive governments have described as “economically unviable.”
But there is a lower-cost pathway. Formalising agreements with allied nations, Japan, South Korea, and the United States, to hold fuel on Australia’s behalf. It is the energy equivalent of keeping emergency savings in a joint account. The IEA emergency response mechanism already exists. Australia’s participation in it could be contractualised at a fraction of the cost of domestic storage. The question is whether the government will treat this as an infrastructure investment or a budget line item to defer.
Should the FSSP be redesigned?
The FSSP expires in June 2027. Both Lytton and Geelong need policy certainty to justify continued capital investment. But the current mechanism is a margin subsidy: it pays when refining margins collapse, provides no minimum stockholding requirement, and demands nothing in return. A redesigned FSSP that requires refineries to maintain minimum fuel stocks and contribute to a national supply buffer would convert a passive subsidy into an active strategic asset. The government is currently reviewing the mechanism. The outcome of that review will determine whether Australia still has two domestic refineries in 2030.
Should regional fuel distribution be treated as an essential service?
The March 2026 crisis was not a national supply shortage. It was a logistics failure in which major wholesalers prioritised their own retail networks and allocated 10% of normal volumes to independent regional distributors. The supply sat in terminals in capital cities. The shortage was in country Queensland and rural WA. Mandating minimum supply allocations to independent regional distributors during declared fuel emergencies is a low-cost regulatory fix for the most politically visible failure in the system. It requires a regulation, not an infrastructure budget.
Schedule the next crisis
Australia ran its fuel system on a just-in-time model for two decades and called it efficiency. Four policy decisions, each rational in isolation, built a supply chain that is entirely dependent on global markets, prices its fuel from a benchmark in Singapore, holds less than 30 days of reserves against a 90-day international standard, and exports its own crude oil while importing refined product.
March 2026 was not a surprise. It was the predictable consequence of a deliberate policy architecture.
The question is not whether the government knew the risks. The government’s own reports acknowledged them. The question is whether a single bad month, empty pumps, crisis pricing, emergency reserve releases, and a task force are enough to change 40 years of policy logic.
Or whether it will be quietly filed away, and the system will return to exactly where it was before the next one.
Dates to watch
June 2027: FSSP expiry for both refineries. The government needs to signal terms well before mid-2026 for companies to plan capital investment.
June 9, 2026: ACCC decision on Ampol’s acquisition of EG Australia. If approved, Ampol controls approximately 40%+ of the national retail network with material implications for supply allocation during future declared emergencies.
Q2 2026 Budget: Whether fuel security infrastructure investment appears as a budget line item, or whether the task force delivers a report, and the system returns to normal.
May 2026 APS release: February data will show whether post-crisis reserve rebuilding has begun or whether the emergency release simply reduced the denominator.