Samsung SDI has locked in 16 GWh of orders, how many more years can the "compliance premium" for U.S. energy storage cells be left for Chinese manufacturers?
Samsung SDI has signed a long-term LFP supply agreement with a U.S. energy storage customer. As FEOC compliance thresholds rise year by year, the compliance premium window for Chinese manufacturers in North America is roughly 2027-2029, and is expected to narrow thereafter.
Foresee Energy notes that the transaction announced by Samsung SDI on November 16 has straightforward figures: its U.S. subsidiary has signed an agreement with Jupiter Power, an affiliate of BlackRock, to supply 16GWh of LFP battery cells from 2029 to 2031. The contract is estimated to be worth about 5 billion RMB, all produced at the StarPlus plant in Indiana. Combined with previous agreements, the total locked volume between the two parties is about 27GWh.
What is complex is the time structure. The LFP production line at StarPlus only completed mass production conversion in October, but Jupiter Power has locked in the output three years in advance.
Jupiter Power is not a developer short of battery cells. It has 19.7GWh of operating or under-construction capacity, and another 110GWh of projects in the development stage. A buyer with a pipeline of over 110GWh of reserve projects is willing to pay for forward commitments from a plant that has just been put into operation for two months. What it buys is not the current battery cell price, but the supply qualification that can pass compliance reviews after 2029.
The per-unit selling price premium of Chinese energy storage manufacturers in North America is attached to this qualification. The scarcer the qualification, the higher the premium. The day when the qualification is filled by Korean local production capacity, the nature of the premium will change.
01
It is not production capacity that is locked
but a compliance ticket that tightens year by year
StarPlus Energy was originally an EV battery joint venture between Stellantis and Samsung SDI, with a planned total investment of over 6.3 billion U.S. dollars and a total of 67GWh across two phases. After North American electric vehicle demand slowed down, Samsung SDI converted three of its four production lines for energy storage use, and two of them were switched to LFP.
Financial signals are more worth watching than production capacity signals. Samsung SDI's battery division posted an operating profit of 361.9 billion won in the third quarter, compared with a loss of 630.1 billion won in the same period last year, with revenue of 4.41 trillion won. As of October, the company's cumulative energy storage battery orders obtained in the U.S. in 2026 have exceeded 60GWh. The idle capacity of EV production lines has regained pricing power in the face of energy storage orders.
But what made Jupiter Power bet three years in advance is not the speed of the production line, but the year-on-year increase in the proportion of "equipment costs from non-Chinese suppliers" required by FEOC rules.
Modo Energy gave a specific tiered structure in an analysis in November: for projects starting construction in 2026, the requirement for the proportion of equipment costs from suppliers not owned or controlled by Chinese entities is 55%, rising to 60% in 2027 and 75% in 2030. Battery cells account for 52% of the cost of energy storage equipment.
For a project using Chinese battery cells, under the rules for projects starting construction in 2027, it is impossible to meet the 60% compliance requirement only through localization in other links.
This means that for U.S. energy storage projects starting construction after 2027, the battery cells themselves must be "clean". The 2029-2031 delivery window locked by Jupiter Power falls in the phase when the compliance ratio climbs from 60% to 75%. What it signed is not a procurement contract, but a ticket that tightens year by year.
Samsung SDI emphasized in its announcement that it is "the only non-Chinese prismatic BESS battery manufacturer in North America". Data from Modo Energy adds a footnote to this statement: among the 12 BESS battery factories in North America, only 7 are currently clearly compliant with FEOC rules, and all 7 are owned or participated in by Korean enterprises.
02
The cost gap in the Korean material chain is real
but it disappears very slowly
The cost of the Korean "de-Sinification" supply chain is real, but not large enough to make the compliance premium disappear immediately.
POSCO Future M itself admits that at current cost levels, its LFP cathode materials are more expensive than Chinese products. The company is betting on the "precursor-free process" to turn the tide, with the goal of achieving lower costs than Chinese products after 2028. L&F supplies cathode materials to Samsung SDI's Indiana plant, and the capacity expansion of its Daegu plant from 30,000 tons to 60,000 tons is scheduled for the first half of 2027.
When the cost of this material chain is transmitted to the ex-factory price of battery cells, it is diluted by two factors. Cathode materials account for about 40% of the cost of LFP battery cells. After the premium at the material end is transmitted upward, the gap at the battery cell level will narrow. More critical is the absorption capacity at the system level.
Calculations from Modo Energy show that U.S.-made battery systems cost about 215 U.S. dollars/kWh, while Chinese-supplied ones cost about 115 U.S. dollars/kWh, with a price difference of about 100 U.S. dollars/kWh. A 30% ITC can cover up to 125 U.S. dollars/kWh of additional equipment costs on the cost of a typical project.
For projects that need to retain tax credits, the material premium of Korean battery cells is absorbed at the system level.
The real problem is not how much more expensive the Korean material chain is, but how slowly it gets cheaper. The long-term LFP cathode agreement signed between POSCO Future M and Samsung SDI covers 2027 to 2032, with an amount of about 6 trillion won. The capacity ramp-up pace of the Korean material chain roughly corresponds to 2027 to 2028. Before that, the scarcity of compliant battery cells will not be alleviated by "planned production capacity".
03
Three paths are stuck at different positions
Chinese manufacturers' responses to the North American compliance premium can be broken down into three different bets.
CATL separates technology from entities. LFP battery cells at Ford's BlueOval Battery Park in Michigan have rolled off the production line, using CATL's technology licensing. Ford owns the plant, CATL holds no shares, and the licensing fee is estimated to be about 7 U.S. dollars per kWh. This path avoids the FEOC review of "foreign entity ownership or control", at the cost of CATL giving up the profits from the battery cell manufacturing link. The technology licensing model can be replicated to U.S. manufacturers like Ford that have existing plants but lack LFP mass production experience, but developers like Jupiter Power that need to embed battery cells into their own project reserves may not take this path.
Eve Energy has placed its bet on the capacity pace in Malaysia. The structure of its 206GWh framework agreement with Fluence is very illustrative: the rigid delivery in 2027 is only 16GWh, and the 190GWh from 2028 to 2031 is reserved capacity, with specifications, prices and delivery dates subject to subsequent orders.
Fluence is willing to pay an option for a "potentially compliant" capacity, but is not willing to sign a long-term lock-in agreement like Jupiter Power did with Samsung SDI. There is currently no definite answer as to whether Eve Energy's Malaysian capacity will be recognized as a "non-foreign entity product" under FEOC rules after 2028.
Hithium has taken the most pragmatic and most restricted path. The 10GWh plant in Mesquite, Texas is a module and system assembly capacity, which completed the first batch of shipments in August 2025, with battery cells procured from external sources. This plant solves the localization identity at the system level, but FEOC rules focus on battery cell and critical mineral traceability. The compliance space that can be obtained through localization of the assembly link is not equal to the compliance qualification at the battery cell level.
SEIA and Benchmark have raised their forecast for cumulative U.S. energy storage installed capacity by 2030 to 708GWh, with 48.3GWh of new additions in the first three quarters of 2026, a year-on-year increase of 20%. The market is indeed growing. However, Chinese battery cells still dominate the ESS systems installed locally in the U.S. The simultaneous expansion of the market size and the inability of Chinese players to capture the full share is happening right now.
The expansion pace of Korean LFP energy storage capacity in North America is that LG Energy Solution will exceed 50GWh by the end of 2026, and further increase to over 70GWh in 2027. According to Modo Energy's statistics, the announced non-FEOC compliant battery cell capacity in North America will reach about 100GWh by the end of 2027. If four more plants are approved under pending guidelines, the upper limit will be about 135GWh.
Chinese manufacturers' overseas capacity will be released intensively around 2028. CATL's 14GWh plant in Thuringia, Germany is already in operation, and its plant in Debrecen, Hungary has started trial production with a planned capacity of 100GWh.
When the timelines of the two sides are superimposed, we can roughly draw a judgment that the compliance premium window for Chinese energy storage manufacturers in North America falls between 2027 and 2029.
The tax-included price difference will not disappear suddenly in 2027. The CIF price of Chinese 314Ah LFP battery cells in North America in September is 61.99 U.S. dollars/kWh, 20.1% higher than the ex-factory price of 51.62 U.S. dollars in China. After levying the maximum 40.9% tariff, the CIF cost is about 87 U.S. dollars/kWh, which is still lower than the system cost of U.S. local manufacturing.
However, the "compliance component" of the premium will narrow as Korean production capacity ramps up and FEOC proportional requirements rise. After 2029, what Chinese manufacturers face in North America is more likely to be a price difference caused by freight and delivery rhythm, rather than the current structural premium where projects cannot pass financing reviews without your supply.
The length of the window depends on two variables: when POSCO Future M's "precursor-free process" will actually bring Korean material costs close to the Chinese level, and whether FEOC rules will further tighten the identity recognition of Chinese-owned capacity in Malaysia and Hungary after 2027. The former is the Korean players' own schedule, and the latter is Washington's schedule.
This article is from the WeChat official account "Foresee Energy", published with authorization by 36Kr.