The aggregate picture

Global oil markets have not collapsed. Three mechanisms explain why, and why that is not the whole answer.

Hormuz carries roughly 20% of global crude supply on a normal day. A sustained disruption of that corridor would be expected, on aggregate figures alone, to produce a severe supply shock. Aggregate figures are not showing one. The explanation sits in three supply-side mechanisms that have been widely noted in commodity analysis: Chinese demand has softened, Saudi and UAE pipeline infrastructure bypasses Hormuz, and Atlantic Basin producers are ramping output. Vortexa, among others, has labelled these the three pillars of rebalancing.

The pillars are real. The aggregate market has genuinely absorbed more than many pre-disruption models suggested. What the aggregate commentary cannot tell you is which specific importers benefit from which pillars, and which importers sit outside all three. The bilateral data can. The answer is that European crude importers are relatively well-covered, Asian crude importers are partially covered at elevated cost, and South and Southeast Asian LPG importers are covered by none of the three pillars at all.

Updated August 2026
Two data corrections since May 2026. First, a routing-engine fix now properly applies Saudi Arabia's and the UAE's real multi-basin export capacity (Yanbu on the Red Sea; Fujairah on the Gulf of Oman) instead of defaulting all Gulf crude to a single Hormuz transit — this brings the crude dependency figures below down materially for every importer in this piece, corroborated against independent reporting on both pipelines' normal-time utilisation (see hormuz.astro method note for detail). Second, LPG (HS 271112/271113/271119) has now been pulled from Comtrade as a genuine data series, replacing the previously unmodeled placeholder figures — and it surfaces a bigger gap than the original Southeast Asia framing: India, Pakistan, and Sri Lanka show LPG dependency as high or higher than anything in Southeast Asia, all outside the three pillars just the same.
01
Chinese demand reduction
EV adoption, economic slowdown, and strategic reserve drawdown have reduced China's net seaborne crude demand. The softening comes primarily from Gulf crude, reducing aggregate competition for non-Chinese importers at the margin.
Who benefits Indirect · global price pressure only
02
Pipeline rerouting
Saudi Arabia's East-West Pipeline (~5mbd capacity, currently ~2–3mbd utilisation) delivers crude to Yanbu on the Red Sea. UAE's Abu Dhabi Crude Oil Pipeline (~1.5mbd) delivers to Fujairah, outside Hormuz. Together they can bypass the Strait for Saudi and UAE crude volumes.
Who benefits Europe via Mediterranean · Red Sea corridors
03
Atlantic Basin surge
US shale, Brazilian pre-salt, Guyanese deepwater, and West African output can collectively add ~5mbd over a 60–90 day ramp window. Light sweet grades suit European refinery configurations. The volume is sufficient to partially offset Gulf crude shortfalls at the aggregate level.
Who benefits Europe · Americas · partial Atlantic-facing Asia

Geographic distribution of benefit

Where the pillars reach, and where they do not, is determined by geography and refinery grade.

The rebalancing is not uniformly distributed. Each pillar has a geography of benefit that is determined by shipping distance, refinery configuration, and pipeline terminal location. The aggregate surplus in Atlantic production does not automatically reach an Indonesian refinery configured for Gulf heavy sour crude. A Saudi tanker that loads at Yanbu and transits the Red Sea costs approximately four days less to reach a Mediterranean buyer than the same crude via Cape of Good Hope, though still a significant premium over normal Suez routing for an Asian buyer. The pipeline does not make Yanbu crude cheap for Japan.

Pillar coverage matrix: by importer region and commodity
Importer / region Hormuz crude dep. Pillar 1 (demand) Pillar 2 (pipeline) Pillar 3 (Atlantic) Net coverage
European crude importers
Greece 38.3% indirect Yanbu → Med Atlantic close Partial · grade + vol constraints
France 9.6% indirect Yanbu → Med Atlantic close Well covered · SPR buffer
Germany 8.0% indirect Yanbu → Med Atlantic close Well covered · low base dep.
Asian crude importers
China 34.6% self-reducing +4–5d Cape +Cape premium Partial · Cape cost + grade gap
Indonesia 20.1% indirect +13d Cape +Cape premium Limited · dual crude + LPG exposure
LPG importers: South & Southeast Asia
Pakistan 99.8% LPG not applicable no LPG pipeline no equiv. supply Uncovered · all three pillars absent
Sri Lanka 91.9% LPG not applicable no LPG pipeline no equiv. supply Uncovered · cooking fuel dependency
India 91.2% LPG not applicable no LPG pipeline no equiv. supply Uncovered · largest absolute exposure
Thailand 75.6% LPG not applicable no LPG pipeline no equiv. supply Uncovered · petrochemical feedstock
Malaysia 52.0% LPG not applicable no LPG pipeline no equiv. supply Uncovered
Vietnam 42.8% LPG not applicable no LPG pipeline no equiv. supply Uncovered
Philippines 33.2% LPG not applicable no LPG pipeline no equiv. supply Uncovered · cooking fuel dependency
Indonesia 28.4% LPG not applicable no LPG pipeline no equiv. supply Uncovered · compound crude + LPG
The coverage gap in one sentence
The three pillars of rebalancing are geographically proximate to European buyers, partially accessible to Asian buyers at a Cape cost premium, and entirely absent for LPG importers across South and Southeast Asia. Aggregate analysis that treats the rebalancing as uniform systematically understates residual exposure in those two groups.

Pillar 2 examined: the pipeline bypass

Saudi Yanbu and UAE Fujairah bypass Hormuz. The question is who can reach the terminals.

Saudi Arabia's East-West Pipeline is the oldest and largest piece of Hormuz bypass infrastructure. It runs from Abqaiq, the world's largest oil processing facility, west to Yanbu on the Red Sea, with a nameplate capacity of approximately 5 million barrels per day. In normal operations it runs below capacity, at roughly 2 to 3 mbd, because Hormuz transit is cheaper. Under sustained Hormuz disruption, Yanbu utilisation would increase, though it would take weeks to ramp, and the output terminates at a Red Sea port whose access involves either the Bab-el-Mandeb or the Suez Canal. Both are separately at risk in the current conflict environment.

The UAE's Abu Dhabi Crude Oil Pipeline terminates at Fujairah, in the Gulf of Oman, outside Hormuz. It has a capacity of approximately 1.5 mbd and is the cleanest bypass: no Red Sea transit required, crude loads directly onto tankers for any destination. Combined with the Yanbu volumes, the pipeline infrastructure can move approximately 4 mbd of Saudi and UAE crude outside the Strait. Against a normal Hormuz throughput of roughly 21 mbd, that is a partial but meaningful offset for the volumes that matter: Saudi Arab Light and UAE Murban, both widely used grades.

What the pipeline does not carry
The pipeline bypass is entirely a crude petroleum mechanism. Qatar's LPG, LNG, and condensate exports have no equivalent pipeline bypass. Qatar is the world's largest LPG exporter and the primary supplier to Southeast Asian importers. Qatar has no overland pipeline access to a non-Hormuz terminal. Every barrel-equivalent of Qatari LPG for Thailand, Indonesia, or Vietnam transits Hormuz or does not move at all. The pipeline rebalancing story is entirely a Saudi-UAE crude story, and it should be analysed as such.

Pillar 3 examined: the Atlantic Basin

Atlantic supply can surge. The benefit concentrates where the tankers are already pointed.

Atlantic Basin crude production (US Gulf Coast, Brazilian pre-salt, Guyanese deepwater, West African) has grown substantially over the last decade and continues to grow. Vortexa's assessment puts the surge potential at approximately 5 million barrels per day over a 60 to 90 day ramp window, drawing on spare capacity, inventory drawdowns, and accelerated field output at existing projects. That is a large number. It is also a number whose benefit distribution is highly uneven.

Where Atlantic surge reaches at a premium
Asian refineries face grade mismatch and Cape economics.
Chinese, Japanese, Korean, and Southeast Asian refineries are predominantly configured around Middle Eastern heavy sour crude. Atlantic Basin output is mostly light sweet, a different grade that requires either refinery adaptation or blending, neither of which happens at speed. Beyond grade, the economics are punishing: an Angolan VLCC to South Korea transits the Cape of Good Hope, adding roughly 15 days over a Hormuz routing. The same 5 mbd Atlantic surge that is normal economics for Rotterdam is a $1.4 million per voyage premium plus refinery-adjustment cost for Busan.

The LPG gap

All three pillars fail for LPG. The commodity is outside the rebalancing story entirely.

The three pillars of rebalancing are, in their different ways, crude petroleum mechanisms. Chinese demand reduction reflects crude import cuts. The pipeline bypass carries crude. The Atlantic Basin surge is crude production. LPG sits outside all three in a structural sense that is not a modelling artefact: it reflects the actual supply geography of the global LPG market.

Qatar is the anchor supplier of LPG to Southeast Asia, providing the bulk of imports to Thailand, Indonesia, Vietnam, and the Philippines. Qatar's LPG exports transit Hormuz. Qatar has no pipeline bypass to a non-Hormuz terminal. The Atlantic Basin has no large-scale LPG production equivalent to the crude surge (US propane exports are growing but are fully committed to existing contract customers under a sustained disruption scenario). Chinese LPG demand has not softened in the same way as crude: LPG is a cooking and petrochemical feedstock, not a transport fuel being replaced by EVs.

Eight nations outside all three pillars
Pakistan (99.8% of LPG supply via Hormuz), Sri Lanka (91.9%), India (91.2%, $14.7B annual exposed, the largest absolute LPG value of any importer in the model), Thailand (75.6%), Malaysia (52.0%), Vietnam (42.8%), the Philippines (33.2%), and Indonesia (28.4%, on top of 20.1% crude) are Tier 1 bilateral exposures: real country-pair trade flows from CDM v3, a multi-vintage blend of 2022 and 2024 UN Comtrade data. For each of these importers, LPG supply disruption via Hormuz cannot be offset by demand reduction elsewhere, rerouted via pipeline, or replaced by Atlantic Basin production. These are the nations most systematically underanalysed in the current rebalancing commentary, and the South Asian members of this group — Pakistan, Sri Lanka, India — are absent from almost every existing writeup of the LPG gap. The analysis that leads with the three pillars and stops there will miss them entirely.

What the bilateral data adds

Aggregate rebalancing is not the same as individual importer coverage. The bilateral data distinguishes them.

The commentary challenge is that both things are true. The aggregate oil market has partially rebalanced: the three pillars exist and are working to some degree. And specific importers remain materially exposed in ways the aggregate does not capture. These findings are not contradictory. They operate at different levels of analysis, and conflating them produces errors in either direction: excessive alarm if you only look at the bilateral exposures, false comfort if you only look at the aggregate rebalancing.

Aggregate rebalancing vs. bilateral exposure: what each analytical layer provides
Analytical question Aggregate / three-pillar analysis Narrows bilateral model
Has the global market partially rebalanced? Yes: three pillars are working Not the primary instrument
Which importers benefit from pillar 2 (pipeline)? Cannot answer: no bilateral resolution Europe via Med; Asia pays Cape premium
Does the Atlantic surge reach my refinery grade? Cannot answer: no commodity granularity Light sweet (Atlantic) vs heavy sour (Gulf) by refinery
Are LPG importers covered by any pillar? Cannot answer: LPG absent from pillars No · 8 nations across South & Southeast Asia uncovered · bilateral Tier 1
What is my counterparty's residual Hormuz exposure? Cannot answer Dep% at (importer, chokepoint, commodity) level
What does partial coverage cost this importer? Cannot answer Cape premium by vessel class; grade substitution lag
Methodology and data sources
Bilateral dependency figures derived from the Narrows Chokepoint Dependency Model v3 (rebuilt August 2026), a multi-vintage blend of 2022 and 2024 UN Comtrade bilateral data, computed via basin-aware routing that weights each exporter's real port geography across all matched basins rather than defaulting to a single dominant basin — this is what brought the crude figures above down from the May 2026 baseline (see hormuz.astro method note for detail). LPG figures were rebuilt in August 2026 from a dedicated authenticated Comtrade pull covering HS 271112 (propane), 271113 (butane), and 271119 (other petroleum gases), aggregated into a single synthetic LPG commodity class and blended across vintages using the same methodology as every other CDM v3 commodity; this series was previously scoped but never completed, and the India/Pakistan/Sri Lanka figures above are new in this update. LPG and crude petroleum figures are Tier 1 bilateral data: real country-pair flows. Pipeline capacity figures: Saudi East-West Pipeline nameplate capacity ~5mbd (Aramco); Fujairah ADCOP ~1.5mbd (ADNOC). Atlantic Basin surge estimate: Vortexa, May 2026 scenario analysis. Rerouting cost estimates use representative vessel-class rates: VLCC bunker $30k/day, charter $85k/day; Cape detour +12–15 days by route. Grade mismatch characterisation: Middle East exports are predominantly medium/heavy sour; US shale, Brazilian and West African grades are predominantly light sweet; Asian refinery configurations sourced from IEA Refinery Survey 2023. Bab-el-Mandeb risk caveat sourced from public US DoD reporting. Three pillars framework: Vortexa Commodity Analytics, May 2026. These figures describe structural, normal-year exposure, not a live tracker of the Strait's actual closure since March 2026. All model assumptions are substitutable. Contact fysh@narrows.io to run a scenario against your specific inputs.