Why Seoul Chose Encinal, Texas for Its First US Investment

South Korea has chosen a proposed 6.3-gigawatt gas-fired power complex in Encinal, Texas — a town of about 540 people — as the first deployment of the $350 billion investment package Seoul promised the United States. The project would be built in stages: first 1.4GW of simple-cycle gas turbines, then about 4.9GW of more efficient combined-cycle generation. The White House could announce the selection as early as Wednesday, possibly with President Trump himself, though final US approval is still pending.

The plant is designed for the one customer Washington is most anxious to attract: electricity-hungry data centers and chip fabs. But for now, no customer has been confirmed and no power purchase agreement exists. Korean media reports that signing a PPA is targeted for 2027, and US officials have reportedly asked for a roughly 25% expansion of the project, which would push its cost from about $22.3 billion toward $25 billion.

Under the November 2025 arrangement, Seoul would supply $200 billion in upfront capital for strategic industries, capped at $20 billion a year, plus $150 billion for shipbuilding. Project cash flow would be split 50/50 until Korea recovers principal and interest; after that, Korea's share falls to 10%. South Korean officials project the Encinal plant could generate $43 billion to $45 billion in revenue over 20 years, but they have not shown how that gross figure recovers the capital under the 50/50 structure.

The Revenue Math, Turbine Bottleneck and Hyperscaler Demand Behind Encinal

The $43–45 Billion Revenue Claim Doesn't Match the 50/50 Split

South Korean officials say the gross revenue is enough to recover principal and interest, but the numbers are not straightforward. $43–45 billion over 20 years is about $2.2 billion a year, roughly twice the upfront capital, before fuel, operations, maintenance and financing costs. Since Korea receives only half of cash flow until it is repaid, distributable cash flow would need to be well above the capital invested; Seoul has not published a breakdown demonstrating that. The US-led Investment Committee decides what counts as commercially reasonable, and Trump makes the final call.

Turbines Are the Real Constraint

Large gas turbines are effectively sold out through 2030, and Applied Digital's CEO has warned that orders placed today may not arrive until 2032. Encinal's plan to install 1.4GW of fast-to-deploy turbines first and 4.9GW of combined-cycle capacity later puts it in the same line every hyperscaler is already chasing. Some developers are reportedly turning to boilers and steam systems to get power online sooner, a sign of how tight equipment supply has become.

Goldman's Power Math Explains Why Washington Wants This

Goldman's Carbonomics team now expects behind-the-meter power generation to rise from 40GW to 67GW by 2030, with on-site gas generation meeting about 28% of US data center power demand by 2030. The bank names GE Vernova, Siemens Energy, Mitsubishi Heavy and INNIO as preferred beneficiaries — none of them Korean. That raises an uncomfortable question for Seoul: how much of the equipment spending from a Korean-funded plant will actually go to Korean suppliers.

Japan's Ohio Template Is a Cautionary Tale

South Korea is following Japan, whose opening $36 billion tranche was led by SB Energy's 9.2GW Ohio gas plant. That project has since delayed its IPO and is having issues. The per-kilowatt cost is almost identical: Encinal at about $3,540/kW and Ohio at about $3,590/kW, both well above pre-AI-boom US combined-cycle costs. The bull case for Seoul is that a hyperscaler signs a PPA by 2027; the bear case is that Korea has fronted $22 billion on the assumption one will, while the AI capex cycle may be nearing its peak.

What the Encinal Decision Means for Power Project Developers and Investors

For energy developers, equipment buyers and investors with exposure to US data-center power, the Encinal selection creates several concrete watchpoints:

  • Do not count Encinal's 6.3GW as committed supply. There is no PPA and no confirmed customer; signing is targeted for 2027, and final US approval is pending.
  • Plan turbine procurement around a sold-out order book. Large gas turbines are effectively sold out through 2030, and orders placed today may not arrive until 2032; simple-cycle first could face different lead times than combined-cycle capacity.
  • Evaluate returns on Korea's 50% cash-flow share before repayment. The official $43–45 billion gross revenue over 20 years has not been reconciled with capital recovery, and Seoul receives only 10% after principal and interest are repaid.
  • Treat GE Vernova, Siemens Energy, Mitsubishi Heavy and INNIO as the named turbine beneficiaries. Goldman's list contains no Korean equipment supplier, so any Korean content share remains an open question.
  • Watch for a White House announcement as soon as Wednesday, timed around Xi Jinping's Washington visit; the project still requires final US sign-off.

Risk & Opportunity Assessment

Commercial RiskHighThe project carries $22.3 billion of upfront capital with no confirmed customers and no power purchase agreement, and officials' projected $43–45 billion in gross revenue over 20 years has not been reconciled with Korea's 50% cash-flow share before repayment.
Competitive RiskMediumEncinal would compete with other US gas developments, including SB Energy's 9.2GW Ohio project, while turbine buyers face sold-out order books through 2030; no customer has chosen Encinal yet.
Regulatory RiskHighFinal US approval is pending; the US-led Investment Committee decides what is commercially reasonable and Trump has the final call, with Seoul's objections risking higher tariffs.
Reputation RiskHighKorean officials describe sleepless nights, and the project could repeat Japan's Ohio template, which has delayed its IPO and is facing issues; the absence of a published cash-flow breakdown amplifies vaporware criticism.
Technology DisruptionMediumHeavy-duty gas turbines are sold out through 2030, pushing some developers to steam-based generation, while Goldman raises behind-the-meter power forecasts from 40GW to 67GW by 2030; gas remains central but equipment bottlenecks could alter project timing.
Commercial OpportunityHighIf hyperscalers sign PPAs, Encinal could serve data-center demand that Goldman expects on-site gas to cover 28% of by 2030, but no customer is confirmed.