Singapore Strategic Petroleum and Gas Reserves: Days of Supply Estimation

A government deliberately opaque on its own reserves

Since the closure of the Strait of Hormuz on March 1, 2026, the question of how long Singapore can sustain its energy needs has moved from a niche analyst concern to a national policy debate. Yet anyone trying to answer it runs into the same wall: the Singapore government does not publish its strategic reserve levels for crude oil, refined products, or natural gas.

Minister-in-charge of Energy Tan See Leng has repeatedly stated that fuel stockpiles are sufficient "for several months" without providing exact figures, citing national security. This stance contrasts sharply with peer economies in the region. Japan's reserves are reported in detail by JOGMEC at around 254 days of consumption. South Korea, the United States, and several European countries publish granular data through the IEA and JODI databases. Singapore is not an IEA member and does not submit comparable inventory figures to JODI.

The opacity is not just an external observation. The Order Paper of the Singapore Parliament dated April 7, 2026, contains multiple questions from MPs explicitly asking how many months national fuel reserves can sustain electricity generation if the Strait of Hormuz remains closed, what the government's resilience assessment is for prolonged oil price shocks, and what additional stockpiling measures are being implemented. The fact that elected representatives themselves are asking these questions in writing confirms that the information is not in the public domain, even at the parliamentary level.

The Coordinating Minister for National Security K. Shanmugam, who chairs the new Homefront Crisis Ministerial Committee activated in March, has publicly stated that emergency stockpiling, particularly for jet fuel at Changi, would be "costly but necessary." This admission, made without quantifying existing volumes, is itself a signal: if the government is openly preparing to expand stocks during a crisis, current levels are likely tighter than the "several months" framing suggests.

Given this lack of transparency, any serious analysis must rely on OSINT estimation. This article lays out a methodology and produces defensible ranges for crude oil, refined products, jet fuel, and natural gas.

Methodology

The estimation rests on three categories of public data that, when triangulated, produce credible bounds without requiring access to classified inventory levels.

Physical infrastructure. Storage capacities are publicly disclosed by operators (SLNG Corporation, JTC Corporation for the Jurong Rock Caverns, and major commercial terminals such as Vopak, Universal Terminal, Oiltanking, and Tankstore). These figures define the upper bound of what could physically be held in country at any given time.

Annual flow data. The Energy Market Authority (EMA) publishes the Singapore Energy Statistics annually, including imports, exports, refinery throughput, and net stock changes year over year. Stock changes give the direction and magnitude of inventory movement but not the absolute starting point.

Comparative and historical reference points. The 2013 US Energy Information Administration country report on Singapore remains the most recent third-party estimate of national petroleum reserves. While dated, it provides an anchor against which subsequent capacity expansions can be projected.

Cross-referencing these three layers produces ranges rather than point estimates. The ranges are wide because the stratification between strategic and commercial stocks is itself ambiguous in Singapore (more on this below), but they are tight enough to flag where the real vulnerabilities lie.

Crude oil and refined products

Storage capacity (verifiable infrastructure)

Consumption baseline

The headline BP figure of 1.49 million bpd for Singapore (2024) is misleading for reserve calculations because it includes international marine bunkers (around 900k to 1 million bpd) and international aviation fuel uplift. For days-of-supply purposes, what matters is inland consumption plus the fuel needed to run essential services.

Days of supply matrix for petroleum

The viral "245 days" figure that circulated on social media in early March 2026 corresponds roughly to scenario 3. It is technically derivable from public data but assumes that all commercial stocks on Jurong Island are available for national consumption, which is not the case absent emergency requisition.

Scenario 4 explained: refineries running at full capacity
This scenario assumes the three Jurong Island refineries (ExxonMobil, Shell, Singapore Refining Company) keep operating at their combined nameplate throughput of around 1.38 million barrels per day, drawing down stored crude as feedstock. The relevant drawdown rate here is therefore not domestic inland consumption but total refinery intake.

The distinction matters because Singapore is not primarily a final consumer of oil. It is a transformer. The country imports crude, refines it on Jurong Island, and re-exports the bulk of the resulting products as gasoline, diesel, jet fuel, and naphtha to regional markets including Indonesia, Vietnam, the Philippines, and Australia. Maintaining refinery throughput is what preserves Singapore's role as a regional refining hub, its export revenues, and the supply of refined fuels to neighbouring countries that depend on Jurong Island for their own energy security.

This is precisely the dilemma the government faces under the current Hormuz closure. Running refineries at full capacity exhausts crude stocks in around 65 days. Slowing refinery runs (as Reuters reported in mid-March, with two of the three refineries already cutting output) preserves crude inventories but disrupts regional supply chains and domestic export earnings. Scenarios 1 to 3 in the table above measure how long Singapore can supply itself. Scenario 4 measures how long Singapore can keep functioning as the refining hub of Southeast Asia.

Natural gas (LNG)

This is the most calculable and arguably the most concerning bottleneck.

SLNG Terminal capacity

A second LNG terminal based on a floating storage and regasification unit (FSRU) of 200,000 m³ is under construction at Jurong Port but is not expected operational before late this decade.

Consumption baseline

Days of supply matrix for natural gas

The Indonesian West Natuna pipeline contract expires in 2027, which means the 50% pipeline cushion is structurally degrading on a multi-year horizon regardless of the current crisis.

Jet fuel at Changi

Changi consumed approximately 75 to 90 thousand barrels per day of jet fuel before the crisis. There is no publicly disclosed strategic jet fuel reserve. The fact that Minister Shanmugam announced in late March 2026 that emergency stockpiling for Changi would be "costly but necessary" is a strong indicator that pre-crisis dedicated reserves were below 30 days. Industry reports from Energy Intelligence in April confirm Singapore is actively building emergency jet fuel stocks for the first time, having delayed the introduction of its sustainable aviation fuel levy by at least six months in the process.

The strategic versus commercial distinction matters

Unlike Japan, where JOGMEC manages a clearly delineated national strategic petroleum reserve separate from industry stocks, or the United States with its Strategic Petroleum Reserve in Gulf Coast salt caverns, Singapore does not maintain a dedicated national stockpile clearly carved out from commercial inventories.

The bulk of physical storage on Jurong Island is commercial, owned and operated by private actors. ExxonMobil, Shell, and Singapore Refining Company hold feedstock and product inventories tied to refining operations. International trading houses (Vitol, Trafigura, Mercuria, Glencore) hold cargoes in transit, in storage for arbitrage, or pre-sold to regional buyers. The Jurong Rock Caverns are government-owned but most of the leasable space is rented to commercial users.

In a severe crisis, the government has legal levers to mobilize parts of these stocks. The Petroleum (Amendment) Act and the Essential Goods Control Act provide authority for requisition, allocation, and price control. Singapore has also implemented provisions allowing temporary suspension of export contracts under emergency conditions. However, these powers have never been tested at scale, and their application raises complex legal and diplomatic issues that would not arise with a clean sovereign reserve.

This ambiguity is precisely why headline figures like "90 million barrels" or "245 days" can be technically defensible while being practically misleading. The government's deliberate vagueness on what constitutes the "strategic" portion of these stocks compounds the analytical difficulty.

OSINT indicators used in this analysis

The estimation above draws on the following public sources:

1. Energy Market Authority (EMA) Singapore Energy Statistics, annual publication. Provides supply, demand, imports, exports, and stock change data at the national level. Latest edition reflects 2024 data.

2. Ministry of Trade and Industry (MTI) speeches and parliamentary responses, including the MTI Committee of Supply Debate 2026 speech by Minister Tan See Leng on March 2, 2026.

3. Singapore Parliament Order Papers, including the April 7, 2026 sitting which contains explicit MP questions on national fuel reserves and electricity generation resilience.

4. SLNG Corporation public disclosures on terminal infrastructure (capacity, throughput, jetty configuration).

5. EIA Singapore country analysis (2013 reference report), the most recent third-party estimate of strategic petroleum reserves.

6. JODI-Oil database for cross-referencing peer countries (Singapore does not submit complete inventory data).

7. Maritime AIS tracking data (via Vortexa, Kpler) for verifying tanker arrivals at Jurong Island and SLNG terminal calls. This is a real-time leading indicator of physical inflows.

8. USEP wholesale electricity prices as a proxy indicator for natural gas supply tightness. The reading of SGD 169.23 per MWh in late March 2026 was the highest of the year.

9. Industry trade publications, including Energy Intelligence, Argus Media, Reuters energy desk, and S&P Global Commodity Insights, for cross-confirmation of operational decisions (refinery run cuts, force majeure declarations, emergency procurement).

Synthesis: defensible ranges

The most fragile link is natural gas. Even a fully filled SLNG Terminal can cover at most two weeks of total gas demand if both Indonesian and Malaysian pipelines were simultaneously disrupted. With 95% of Singapore's electricity generated from natural gas, this translates directly into electricity supply risk. The "several months" framing used publicly by Minister Tan most plausibly refers to backup diesel for power generation in stress scenarios, not to gas itself.

The most resilient link, on paper, is crude oil and refined products, given the sheer storage capacity of Jurong Island. The practical question is what fraction of those 90 million barrels would be available for domestic use in a sustained crisis, given that they are owned by private actors operating across multiple jurisdictions. The honest answer is that nobody outside the Singapore government knows, and the parliamentary record suggests even MPs are not fully briefed.

Caveats and limitations

This analysis is bounded by several important constraints. First, all calculations rely on pre-crisis consumption baselines. Demand reduction measures, rationing, and price-induced demand destruction would all extend days of supply in a sustained scenario but are not modeled here. Second, the working volume of LNG tanks is treated conservatively at 70% of nameplate, but actual operational floors depend on heel maintenance and boil-off management practices that are not publicly disclosed. Third, the ranges assume no incoming cargoes, which is the worst-case for stress testing but unrealistic over a multi-month horizon. In practice, Singapore continues to receive non-Gulf cargoes from Russia, West Africa, the United States, and Australia.

Finally, this entire exercise would be unnecessary if the Singapore government published its inventory levels at the standard JODI granularity. Until it does, OSINT triangulation is the only available method, and ranges rather than point estimates are the only output.