Revised 8 August 2026 · v3.0.0 · Correction: v2.0.0 stated that nobody publishes terminal-level LNG throughput against capacity. PNGRB does, in its 2030-40 demand-projection study, and the figures change the reading of India’s 47% utilisation — that claim is withdrawn and Exhibit 5 now carries the terminal-by-terminal data · earlier: India section and source-coverage map added; figures re-sourced from the full IGU report

The scale

Europe is now the market that global LNG is priced around.

The EU depends on imports for more than 90% of its gas. Since 2022 it has replaced Russian pipeline supply largely with LNG, which now accounts for 47% of total gas supply. In 2025 EU LNG imports reached a record 146 bcm, roughly 1,700 TWh, and the EU passed China and Japan to become the largest LNG importer in the world.

Two things drove the 2025 record, and only one of them was demand. Gas need was higher, but the bigger factor was that Europe entered the year nearly empty: EU storage sites were 34% full on 1 April 2025, against 59% a year earlier. A refill from that base is a purchase obligation, not a choice, and the market knew it.

Germany is the structural change. It imported no LNG at all in 2022 and now runs a terminal network built inside three years. Belgium and Italy added the most volume in 2025. The only declines were in Finland, Portugal and Croatia — three small markets.

The concentration

One supplier, 58% of the cargoes.

ACER is blunt about this: the EU’s LNG portfolio has become increasingly concentrated, with US LNG playing the central role. At 84 bcm, the United States supplies 58% of EU LNG imports and 25% of all EU gas consumption — a figure ACER reports directly, roughly a quarter of everything Europe burns, from one country, arriving by ship.

For scale from the other side of the trade, the IGU puts total US LNG exports at 110.7 million tonnes in 2025. On a standard conversion that is about 151 bcm, so the EU absorbs roughly 56% of everything the United States ships. The dependency runs both ways, which is a form of security, but it is not diversification.

Why alternatives are hard to buy

Four to six years, and most of it already sold.

Securing additional LNG by contract is slow. New volume generally depends on new export capacity, and ACER puts the lag from final investment decision to first cargo at four to six years. Meanwhile most LNG output is committed under long-term contracts, so the share of new capacity that ever reaches the spot market is limited. Global liquefaction has been running at about 92% utilisation — there is no idle plant to call on.

The consequence is that Europe’s incremental demand has gone to the spot market by default. In 2025 54 bcm of EU LNG imports were spot volumes — over a third of the total. ACER’s price assessment drew on a record 980 reported spot transactions covering 81 bcm of trade, and confirmed the Dutch TTF as the reference index for 74% of them.

The chokepoint

A 2026 problem that is already in the price.

ACER’s report is dated May 2026 and is dominated by an event Europe did not cause. Renewed Middle East tensions have made the Strait of Hormuz — the export route for Qatari and Emirati LNG — the central variable in the 2026 balance, and 17% of Qatari LNG capacity has been damaged by drone attacks.

ACER models it as a fork:

The swing between those branches is 47 bcm. Set against EU annual imports of 146 bcm, the difference between the Strait reopening and not is about a third of everything Europe imports in a year. TTF intraday prices have already been pushed above €70/MWh by the tension alone.

The second-order effects are visible already. In Asia, governments are reintroducing emergency demand-reduction measures last used during the pandemic, and some are burning more coal to limit LNG consumption. Inside the EU, Italy, the Netherlands and Germany have increased coal use for power generation, and the Commission has called for demand reduction, faster renewables rollout and earlier storage injections.

The India angle

Europe bought more. India bought less, having just built more capacity.

2025 split Asia from Europe, and India sits on the wrong side of the split. While EU imports hit a record, India’s LNG imports fell 1.5 Mt to 24.60 Mt. Pakistan fell too, down 0.8 Mt to 6.46 Mt — the IGU reads this as constrained South Asian demand, and the mechanism is not mysterious: when Europe is refilling empty storage at any price, price-sensitive buyers step back.

The part that should trouble Indian planners is the denominator. India ended 2025 with 52.5 Mtpa of regasification capacity across eight terminals, the largest regas market in Asia after China and Japan. Against imports of 24.60 Mt, that is 47% utilisation. And it got worse during the year: India added 7.1 Mtpa in 2025 — commissioning Chhara LNG and expanding Dabhol after its long-delayed breakwater was finally completed — while imports fell. On the prior capacity base of 45.4 Mtpa, 2024 utilisation was about 57%.

So India built a tenth more capacity into a year when it bought less gas, and drove utilisation down ten points. A further expansion of Dahej — at 17.5 Mtpa the country’s only ultra-large terminal and the sixth largest in the world — is under construction. The constraint on Indian LNG has not been the ability to land it for some years; it is the delivered price against domestic alternatives, and the downstream pipeline and CGD offtake to move it.

Who reports what

Three bodies, three different slices, no single source.

Anyone assembling an Indian LNG picture has to work across three agencies that do not overlap as much as you would expect. This mapping is worth having explicit, because a figure that is missing from one is usually present in another under a different unit.

It is also worth doing before concluding that a number is unpublished. The 47% national utilisation above is an average, and an average across eight assets says nothing about whether the shortfall is spread evenly or sitting in a few specific terminals. PNGRB publishes the breakdown — not in a statistical release, but inside a demand-projection study — and it turns out to matter a great deal.

There is a serious case for building anyway, and PNGRB makes it in the same document. The board projects Indian gas demand to roughly double to about 200 MMSCMD by 2030, with LNG imports more than doubling to bridge a supply gap that domestic production will not close — KG-D5 comes on stream, but decline at Mumbai High and Bassein offsets it. It expects the wave of US liquefaction reaching the market from 2026 to make long-term contracts cheap enough to lock in, which is what IOCL and BPCL did with ADNOC and GSPC with TotalEnergies at India Energy Week 2025. Regas capacity takes years to build and demand does not wait for it, so some idle capacity is the price of being ready.

That case rests on demand arriving. Its weak point is the one PNGRB names itself: Indian gas demand is a function of affordability, and imported LNG competes against domestic gas the government prices administratively — the APM formula is 10% of the Indian crude basket, floored at US$4 and capped at US$6.50 per MMBtu. Where a buyer can get domestic gas at that ceiling, spot LNG has to fall a long way to win the business. The second constraint PNGRB points at is physical rather than commercial: it wants terminals “well-integrated with demand centers” and flags evacuation planning specifically, which is the regulator’s way of saying that a terminal with no pipeline to a buyer is not a supply asset. Kochi and Ennore, the two that have never averaged above 13% over their first eight years, are the standing illustration. Neither of those constraints is fixed by adding a ninth terminal.

What happens at the end of 2026

The last Russian cargo, and a concentration that gets worse before it gets better.

Russia remains the world’s fourth-largest LNG producer at 43 bcm, and the EU still buys some of it. Under the Russian gas phase-out regulation the EU ends Russian LNG imports by the end of 2026. That is the right policy on its own terms and it has an arithmetic consequence worth stating plainly: removing a supplier from a concentrated portfolio makes it more concentrated, not less. The US share of EU LNG rises from 58% by construction.

The mitigation ACER points to is time and capacity. 2025 was a record year for final investment decisions — 68.4 Mtpa on IGU’s count, about 90 bcm on ACER’s — and most of it American. But at four to six years from FID to first cargo, that capacity lands in 2029 to 2031. It does not help with 2026, 2027 or 2028, and when it arrives it will make the portfolio more American still.

Sources. European figures — Agency for the Cooperation of Energy Regulators (ACER), Analysis of the European LNG market developments: 2026 Monitoring Report, published 13 May 2026, 46pp: EU import total, Member State breakdown, US share, spot volumes and transaction counts, storage fullness, liquefaction utilisation, Russian production and the Strait of Hormuz scenarios, the last based on ICIS LNG Edge and S&P Global data. Global figures — International Gas Union, World LNG Report 2026, 88-page report, published 7 July 2026: global trade volume and growth, US export total, European import growth, liquefaction and regasification capacity, and 2025 final investment decisions. Unit reconciliation between the two is the author’s calculation at 1 Mt LNG ≈ 1.36 bcm and is shown in the text rather than applied silently to either source’s figures. Indian figures — Petroleum & Natural Gas Regulatory Board, India’s Evolving Natural Gas Landscape: NG Demand Projections for 2030 and 2040, from the board’s knowledge-sharing page: terminal-by-terminal capacity and utilisation (attributed by PNGRB to PPAC), the 2030 and 2040 demand scenarios, APM and HPHT gas pricing, and the CGD and pipeline authorisation totals. India’s 2025 import volume and national regasification capacity are IGU’s. No figure on this page comes from a commercial data vendor, and the arithmetic check in the text on one such vendor’s European import total is the reason.

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.