Marginal Barrel

Where Asian LNG goes after Hormuz

Change five assumptions and the model re-prices the market to 2036. Every combination has been run in full; nothing here is a guess drawn on a chart.

New LNG plants start
Hormuz back to normal by
Henry Hub gas from 2027

Barely moves Asian prices while US plants are full; matters once they have spare room.

Brent oil from 2027
LNG demand growth

One point a year faster or slower. Bites while new plants are late; spare capacity absorbs it later.

Asian LNG price, USD per MMBtu, actual 2024 to Sep 2026 and modelled to 2036
Asian spot LNG, USD per MMBtu, monthly. Model months start Oct 2026.

Why the model prices December below the market

JKM quoted in early November is the price for December cargoes. For December 2026 the model's fundamentals price is $23.07. Here is how that gets to the screen price.

  • Oil is higher than the model assumes for the winter. Brent averaged $114 in September; the base case uses $105 for the rest of 2026. Asian utilities can burn oil instead of LNG, so dearer oil lifts what they will pay. Worth about $1.9.
  • Shipping is dearer in a crisis. The model uses a $60,000/day charter rate. At $150,000/day a US cargo to Japan costs more to deliver. Worth about $0.5.
  • The rest is buying for safety, not for burning. The model only buys gas that will be used that month. To justify $32 on its numbers, buyers would have to take roughly an extra 4–5 million tonnes in December, about 55–60 cargoes or a tenth more than a normal month. That is what stockpiling against a longer closure looks like. Worth about $6.5, and the model cannot predict it.

The last few million tonnes move the price a lot. In the model, the first 2 Mt of extra buying adds about $4 and the next 4 Mt adds another $7. A tight market has a steep demand curve, which is why JKM has been the most volatile energy price of 2026.

JKM reference: S&P Global reported JKM at $30.76/MMBtu on 22 October 2026; ~$32 in early November. JKM is a Platts assessment; this page does not reproduce it, and the chart's actual line is the IMF monthly Asia LNG price.

What is inside the model

  • Supply. 22 exporting regions with their plant capacity, new projects on announced start dates, and the Qatar and Hormuz outages. US plants run at their real output, which has been above design capacity since 2025.
  • Demand. 28 importing regions. Each buys a must-have volume, then more as the price falls, in steps from $35 down to $8. The steps move with the oil price.
  • Shipping. Distance, speed, charter rate, canal tolls and boil-off for every route.
  • Price. Each month the model sends every cargo to the buyer who values it most. The price is what the last cargo needed was worth.
  • Calibration. Tested against actual prices from January 2025 to September 2026, it missed by $2.26/MMBtu a month on average. It reads too high in January and February, because it has no gas storage, and too low in September 2026, when buyers were stockpiling.

Data: Henry Hub, Brent and the IMF Asia LNG and EU gas prices from FRED; US LNG exports from the EIA; outage timing from the IEA, OIES and S&P Global reporting. The base case assumes Hormuz normalises in April 2027, Qatar's two damaged trains return in mid-2029 and North Field East starts in July 2027.