The Dirty Secret of Vietnam's Coal Boom: 25-Year Fuel Contracts Written Against a Narrowing Tariff Moat

Vietnam is not short of clean-power ambition; it is short of electricity. Coal supplied 54.5% of national output in the first half of 2026, up from 46.2% ten months earlier, while the country's 22.5 GW LNG target for 2030 is only 7.3% achieved. The result is that seaborne thermal coal becomes the marginal energy source for the next three years of industrial load. Every additional megawatt of Vietnamese demand is now, in practice, a coal procurement decision.

Vietnam's utilities have started buying coal on terms that outlast the politics which created the demand. State utility EVN signed an agreement with state coal group Vinacomin to supply Indonesian coal to plants running on imported coal until 2047, part of a centralisation of fuel procurement that also lifts Vinacomin's own import book from about 10 Mt in 2025 to 13 Mt in 2026 and 17 Mt in 2027. At plant level the contracts run longer still: Vung Ang 2 is being supplied 3.6-4.0 Mt a year from Indonesia and Australia on 25-year terms.

Vietnam is not building coal plants because coal is cheap. In July 2026 Newcastle thermal coal traded around $115-122 a tonne, roughly $19-20 per MMBtu, while imported LNG for Vietnamese power generation sat at $14.05 per MMBtu. At a typical 38% coal plant efficiency that is roughly $45-53 of fuel per megawatt hour from coal against about $14 from a combined-cycle gas plant. The build-out is an energy-security and permitting decision, and it is being financed with contracts that lock those economics for two decades.

The premise that Vietnam is the tariff-free workshop is out of date. On February 20, 2026 the US Supreme Court held that the International Emergency Economic Powers Act does not authorise tariffs, and US Customs stopped collecting the IEEPA duties on February 24, taking with them the 20% reciprocal rate Vietnam had negotiated and the 40% transshipment surcharge that policed the China+1 assembly model. What replaced them is narrower and more origin-focused: on July 24, 2026 USTR finalised Section 301 duties of 10% and 12.5% on 60 economies, with Vietnam among the 38 economies facing 12.5% on roughly 37% of its exports to the US, and Vietnam's own forced-labour import ban effective from July 22.

For the coal chain this matters more than for the tariff headlines. The 17-19 Mt of imports and the 5,300 MW of proposed coal capacity are justified by manufacturing load that came to Vietnam to dodge tariffs, and the tariff architecture that made that relocation rational was dismantled in February and replaced with rules that penalise precisely the Chinese-component assembly Vietnam's export base depends on. The fuel is contracted for 21 years; the load that justifies it is a policy variable.

American coal reached Vietnam in June 2026, but only as a pilot. Union Pacific, Ferromex and Peabody Energy moved Powder River Basin coal from the North Antelope Rochelle mine in Wyoming to the Mexican port of Guaymas and loaded it onto a Panamax vessel bound for Vietnam, opening a Pacific export lane with no volume behind it yet. Core Natural Resources, the country's largest coal exporter, has been taking orders from what its chief executive calls surprise markets, Indonesia and Vietnam, with exports expected to rise 10% this year. The tonnage is trivial next to domestic US burn, so the earnings effect lands on the exporter, not on the largest mine.

Two groups cash the build-out without carrying the fuel risk. Doosan Enerbility signed a KRW 837.97 billion EPC contract with PVN for the 1,155 MW O Mon III plant, equal to 4.91% of its KRW 17.06 trillion in sales and running to October 2030, and booked about KRW 3.87 trillion of Vietnamese combined-cycle awards in 2026 across five projects. Mitsubishi Corp supplies the fuel instead, holding a 40% stake in the Vung Ang 2 vehicle that now imports coal from Indonesia and Australia on 25-year terms, having walked away from Vietnam's Vinh Tan 3 coal project in 2021 on decarbonisation grounds. The factory floor, meanwhile, is paying for the strategy in outages.

For the companies at the top of the chain this is a volume story; for the ones at the bottom it is a cost story that no amount of factory relocation can hedge. Samsung Electronics shows the gap between the two: second-quarter operating profit of KRW 89.5 trillion against KRW 4.7 trillion a year earlier means a fortnight of rationed power in northern Vietnam is a rounding error at group level, but it is a permanent tax on the China+1 assembly model that the tariff regime was built to reward.

Here is the trade in one paragraph. Vietnam needs 17-19 Mt of imported coal in 2026 and has contracted the fuel through 2047 because coal is the only thing that can be built and dispatched at the speed its factories require, and because gas, hydro and the grid cannot close the gap on the same timetable. The same supply chain that guarantees the fuel is pricing it at levels where Vietnamese buying is weakest, and the tariff regime that created the factory load has been narrowed twice in 2026. Own the contracted fuel and the contracted engineering; do not own the coal price and do not own the relocation premium. The demand case behind 21 years of fuel is the least certain part of it.