Where should Golar's fourth FLNG go?

Golar in one page

Golar sold its shipping and FSRU businesses and is now a pure FLNG company. It converts LNG carriers into floating liquefaction plants, moors them next to offshore gas fields, and charters them out on 20 year contracts. The customer brings the gas. Golar turns it into LNG for a fixed fee, and on its Argentine contracts also takes 25% of the LNG price above $8/MMBtu.

Hilli finished eight years in Cameroon in July 2026 with 156 cargoes and full economic uptime. That record matters more than any brochure: Golar is the only company that has run FLNG as a leased service for years, and buyers who need gas exported fast pay for that.

What one MKII unit earns

FLNG #4 costs about $2.45 billion fully delivered. If it is chartered on the same terms as its sister vessel Esperanza, Golar gets $400 million of EBITDA a year for 20 years, plus 25% of the FOB price above $8/MMBtu, which management puts at $40 million a year for every dollar above the threshold. Management targets a capex to EBITDA multiple of 5 to 6 times.

The fixed fee is a floor. When LNG prices fall below $8, Golar still gets its $400 million and the gas owner carries the loss. When prices rise, Golar shares the gain.

Assumes capex spread 15/30/30/25% over 2026 to 2029, first EBITDA in 2030, 30% of 2.5% US CPI from year six, no tax, no residual value, unlevered.

One more conversion worth knowing: a 3.5 MTPA unit needs roughly MMscf/d of feed gas. That is the first filter for any field. Anything much smaller cannot fill a MKII, which is exactly why Golar left Cameroon.

Why there is a window

Buyers want new LNG quickly and from more places, after supply shocks hit several large exporters. Floating units are faster to build than onshore plants and can be moved if a field disappoints. The bottleneck is yards. Samsung Heavy builds most newbuild FLNGs and has orders queuing from Eni, Delfin and Western LNG. Its only serious rival, Wison in China, was put under US sanctions in January 2026.

Most of the other projects are owned by the gas producer: Eni owns Coral North in Mozambique and the Congo units, and Delfin took FID in June 2026 on its own unit for 2030. Golar's model is different. It owns the unit and leases it, so a producer can export without putting $2 to 3 billion of its own capital into a plant. With delivery at the end of 2029, Golar estimates FLNG #4 is one to two years ahead of anything else that could be ordered today.

Six possible homes

Each candidate gets a score from 1 to 5 on five criteria. The scores are my own judgement from public information, and the reasoning is under each name. Move the weights to see how the ranking changes.

Lead with GTA Phase 2, keep Argentina as the fallback

1. GTA Phase 2, Mauritania and Senegal

bp and its partners are studying a gravity based structure for 2.5 to 3 MTPA. A GBS is a concrete platform built for one site, which takes years and puts the capital on the partners' balance sheets. Golar can offer something they have already watched work: Gimi sits on the same field and produced 15% above its contracted volume in Q2 2026. Same operator, same gas, same crew model, one more unit, and no upstream capex for the plant.

The pitch would rest on three points: earlier first gas than a GBS, no capital tied up in liquefaction, and an operator with a track record on that exact field. The main risk is timing, since the two host governments are still negotiating Phase 2 terms with bp and Kosmos.

2. SESA expansion, Argentina

The surest volume. Vaca Muerta has the gas, Golar already owns 10% of SESA, and the offtake tender showed demand well beyond the 2 MTPA sold to SEFE. The downside is concentration. Three of Golar's four units in Argentina would tie most of the company's value to one country's currency and politics, and a third unit needs more pipeline capacity on the San Matías route.

3. Keep warm

Leviathan is the largest prize but Chevron is running its own FEED and leans towards owning the unit. Nigeria has a signed development agreement but has missed its own FID date by almost two years. Both are worth tracking, neither should hold up a charter for #4.

What I would do in the first weeks

1. Map the GTA Phase 2 decision process: who in bp, Kosmos, Petrosen and SMH decides, and what the GBS concept is estimated to cost and when it could start.
2. Build a side by side of GBS versus a second FLNG on first gas date, partner capex and cost per tonne.
3. Work out how much extra pipeline capacity a third Argentine unit needs, so the fallback has a real price on it.

• Golar LNG, interim results for the period ended 30 June 2026
• Golar Q2 2026 earnings call transcript
• LNG Industry: Golar orders fourth FLNG unit
• World Oil: 20 year charter with Southern Energy
• Kosmos targets 36 GTA cargoes in 2026
• Energy Intelligence: Mauritania and GTA Phase 2
• TradeWinds: Leviathan FLNG dual FEED
• LNG Industry: NNPC and Golar project development agreement
• Capmad: Hilli leaves Cameroon
• PGJ: Delfin FID
• Samsung Heavy FLNG orders and Wison sanctions