Lithium demand from stationary energy storage jumped 71% in 2025 and is forecast to grow a further 55% in 2026, with Fastmarkets raising its global ESS shipment forecast by more than 60% to 750 GWh. The structural shift is now visible in pricing, in cathode chemistry selection, and in how producers and analysts are rebuilding their demand models. For the first time in the lithium cycle, energy storage is not a secondary narrative: it is a co-equal demand driver with consequences that extend from the mine to the grid.
Introduction
For most of lithium's modern history as an industrial commodity, the demand story was straightforward: count the electric vehicles, multiply by battery size, add a modest increment for consumer electronics, and the market balance followed. That framework is no longer sufficient. In 2025, global battery demand for stationary storage grew at 51%, roughly double the 26% growth recorded for EV-related demand, according to Benchmark Mineral Intelligence. The energy storage sector has not merely accelerated; it has structurally separated itself from the transportation cycle that originally defined lithium's commercial trajectory.
The numbers that frame 2026 are more emphatic still. Fastmarkets has raised its global energy storage systems shipment forecast by more than 60%, from 460 GWh to 750 GWh, citing AI data-centre construction and the pace of the global energy transition as the primary demand vectors. JPMorgan, working from an independently constructed model, has placed ESS at 32% of global lithium carbonate equivalent demand in 2026, rising to 38% by 2030. These are not marginal revisions. They represent a fundamental reclassification of what drives the lithium price.
Building on my analysis of the demand dynamics underlying lithium carbonate's recovery to $21.94 per kilogram in August 2026, this article examines the structural architecture of the second pillar: where the growth is coming from, what is sustaining it, how it is reshaping cathode chemistry selection across the upstream supply chain, and what the aggregate data imply for the medium-term supply-demand balance.
The Scale of the Shift: 2025 Baseline and 2026 Trajectory
The headline figures from 2025 establish a baseline that would have appeared implausible two years ago. Global lithium-ion battery demand rose 29% to reach 1.59 TWh, according to Benchmark Mineral Intelligence. Within that aggregate, stationary storage was the standout performer: global battery energy storage system installations reached approximately 315 GWh, a 51% increase from approximately 205 GWh in 2024. For the first time in the industry's history, battery storage expansion clearly outpaced transportation as a growth driver for lithium consumption.
Albemarle's data capture the same structural divergence from a different angle. The world's largest lithium producer recorded ESS demand surging more than 80% year-over-year in 2025, against global EV sales growth of just 21%. InfoLink Consulting, working from cell shipment data rather than installation capacity, calculated global energy storage cell shipments at 612.39 GWh in 2025, nearly doubling from the prior year. The methodological differences across these datasets reflect the complexity of tracking a market that spans utility procurement cycles, corporate power purchase agreements, and sovereign grid investment programmes, but the directional message is unambiguous.
S&P Global Energy CERA analysts project 55% growth in stationary storage lithium demand in 2026, following the 71% jump recorded in 2025. JPMorgan, in a December 2025 research note led by analyst Lyndon Fagan, raised its 2026 global ESS production forecast by 17% to 900 GWh from 770 GWh previously, with a further 30% increase projected for the post-2026 period. Iola Hughes, Head of Research at Benchmark Mineral Intelligence, has set the firm's 2026 new operational capacity forecast at over 450 GWh, compared with 315 GWh in 2025. These forecasts converge on a single conclusion: 2026 is not a continuation of the 2025 trend; it is an acceleration of it.
The most striking data point in the near-term picture comes from Albemarle's Q1 2026 earnings presentation. Global ESS battery production surged 117% year-over-year in the first quarter of 2026. Over the same period, global EV sales grew just 3% on a gigawatt-hour basis, with larger battery sizes in new models offsetting lower unit volumes. The divergence in growth rates between the two demand pillars in a single quarter is the clearest available signal of how rapidly the centre of gravity in lithium consumption is shifting.
AI Data Centres, Utilities, and the Structural Durability of ESS Demand
The drivers behind stationary storage demand differ categorically from those that animate the EV market, and that distinction matters for how analysts model demand durability. EV adoption curves are sensitive to consumer sentiment, policy incentive structures, and competitive dynamics in the automotive sector. Stationary storage demand, by contrast, is anchored in capital expenditure programmes executed by utilities, technology companies, and sovereigns operating on procurement timelines measured in years rather than quarters.
Raju Daswani, CEO of Fastmarkets, stated at the firm's Global Lithium, Battery and Critical Materials Conference in Las Vegas on June 24, 2026: "The period of market overcorrection is over. Energy storage has become a primary driver of growth in this market." He added that Fastmarkets estimates lithium demand for battery storage systems is growing at 40% per year, characterising the shift as "a fundamental change" that adds "a robust foundation if you compare it to a far-more volatile consumer-driven electric vehicle demand picture."
The AI data centre dimension of this demand is still in its early innings but is already material. Ganfeng Lithium executives stated in a filing to the Shenzhen Stock Exchange in April 2026: "The ESS sector has entered a period of explosive growth, and this trend is expected to continue into 2026. At the same time, AI computing and data centres, the new power consumption scenarios, are driving a constant increase in demand for ESS." Industry research places US AI data centre power expansion alone at approximately 160 GWh of battery storage requirements. The GGII report projects that AI data centre ESS shipments will reach 61 GWh by 2027, a 114% year-on-year increase from the 12 GWh recorded in 2025, before surging to 272 GWh by 2030.
Regional data reinforce the breadth of the demand base. China installed 167 GWh of new battery energy storage capacity in 2025 and is expected to add a further 203.5 GWh in 2026, according to InfoLink Consulting. The United States added 52.1 GWh in 2025, with the SEIA reporting that Q1 2026 US BESS deployments reached 9.7 GWh, the largest first quarter in history and a 32% year-over-year increase. Saudi Arabia has emerged from effectively zero to the world's third largest market in months, deploying approximately 11 GWh in Q1 2026 alone. The geographic distribution of this demand is itself a structural feature: unlike EV adoption, which remains concentrated in China, Europe, and parts of North America, grid storage investment is dispersing across sovereign grids at different stages of energy transition, with India's Central Electricity Authority projecting storage requirements of 411 GWh by fiscal year 2031-32.
LFP Dominance and the Reshaping of Battery Metal Demand
The ascent of energy storage as a demand pillar is not chemically neutral. LFP (lithium iron phosphate) accounts for over 90% of battery energy storage systems globally, according to IEA data, and more than 99.9% by some industry estimates. This concentration of ESS demand in a single cathode chemistry is reshaping how battery metal intensity is distributed across the upstream supply chain in ways that the simple lithium demand aggregate does not fully capture.
Iola Hughes at Benchmark Mineral Intelligence described the situation plainly: "It very much is the story of LFP right now," pointing to recent innovation and cost declines that have made LFP "the best chemistry" for most storage applications. The economics are decisive. LFP battery prices fell by more than 15% in 2025 compared with less than 5% for NMC alternatives, making LFP on average more than 40% cheaper. By omitting nickel and cobalt from the bill of materials, LFP achieves approximately 20-30% lower cell-level costs, removes the supply chain due diligence burden associated with cobalt sourcing, and delivers cycle life characteristics suited to the twenty-year revenue models that storage developers require.
LFP passed 55% of global EV battery deployment in 2025, according to IEA data, a remarkable trajectory from less than 10% in 2020. The consequence for nickel and cobalt markets is already visible in revised demand forecasts. The IEA has stated that switching to LFP chemistries has cut cobalt demand forecasts by over 10% compared to previous estimates. Oxford Energy Forum research, citing Atkinson, found that LFP batteries have cut global nickel and cobalt demand per battery by more than half in just a few years. As the ESS segment continues to grow as a share of total battery demand, reaching what JPMorgan projects at 38% of global LCE demand by 2030, the directional pressure on nickel and cobalt from cathode chemistry selection will intensify.
LFP's dominance does not, however, reduce total lithium requirements. The kWh volume growth in both EV and ESS applications offsets the lower lithium intensity per unit in LFP relative to higher-nickel chemistries. Approximately 900 tonnes of lithium carbonate equivalent is required per GWh of grid-scale storage capacity installed. At 750 GWh of ESS shipments projected for 2026, that implies roughly 675,000 tonnes of LCE attributable to storage alone, a figure that underscores why the structural shift in ESS is now a primary variable in the global lithium balance.
The LFP supply chain concentration is itself an emerging policy issue. The IEA reports that over 98% of LFP cathode material and LFP battery cells are produced in China, compared with less than two-thirds of nickel-based cathode material. China also produces 75% of the world's purified phosphoric acid, essential for LFP production. The IEA anticipates a purified phosphoric acid deficit as early as 2030, a potential bottleneck that supply chain planners outside China have not yet fully addressed. The dominance of LFP in ESS therefore concentrates geopolitical risk in feedstocks distinct from the nickel and cobalt chains that have historically received greater analytical attention.
Price Discovery, Supply Balance, and What the Data Imply
The structural growth in ESS demand is now embedded in the lithium price signals that producers, traders, and analysts are navigating. Lithium posted a gain exceeding 22% in the first half of 2026, ranking as the top-performing commodity across major tracked indices during that period. Morgan Stanley projects a deficit of 80,000 metric tons of LCE in 2026; UBS places the deficit at 22,000 tonnes. Both estimates contrast with an anticipated surplus of 61,000 tonnes in 2025, a market that was itself already tightening faster than consensus models had anticipated.
The deficit projections from major banks reflect the same analytical revision that has driven forecast upgrades across the sector. JPMorgan's updated model identifies a sustained shortfall equivalent to 4% to 7% of total demand, which the bank characterises as a "prolonged incentive pricing environment" requiring higher prices to stimulate new supply. JPMorgan simultaneously upgraded its 2026 and 2027 lithium price forecasts from $800 per tonne to $1,100 per tonne and $1,200 per tonne respectively, with the long-term price revised upward from $1,100 to $1,300 per tonne. These revisions are consequential: several miners have indicated they will not restart idle capacity unless the market consistently trades at $1,200 to $1,500 per tonne.
Albemarle's full-year 2026 guidance assumes lithium demand growth of 15% to 40%, with current consumption tracking toward the upper end of that range at 37%. The company reported global lithium demand reaching 1.6 million metric tons in 2025, up more than 30% year-over-year. For 2026, Albemarle projects total global lithium demand of between 1.8 million and 2.2 million metric tons. The width of that range reflects genuine uncertainty in the pace of ESS procurement deployment, which, while more durable than EV demand cycles, can be affected by grid interconnection timelines and permitting constraints that are difficult to forecast with precision.
On the supply side, the context established in my earlier analysis of CATL's Jianxiawo mine shutdown remains relevant: Benchmark Mineral Intelligence had modelled a roughly 78,000-tonne LCE global surplus for 2026, but disruptions in Jiangxi province and the sustained maintenance shutdown at the world's largest proven lepidolite deposit have materially narrowed that cushion. ESS demand growth arriving faster than most models anticipated is exerting pressure from the demand side simultaneously. The interaction between supply disruptions and accelerating ESS consumption is precisely the structural dynamic that is reshaping the price distribution for H2 2026 and H1 2027.
Strategic Implications for Producers, Developers, and Policy
The recognition of ESS as a co-equal demand pillar is changing the strategic calculus for lithium producers, battery manufacturers, and policymakers at several levels. For producers, the geographic diversification of ESS demand is a material commercial benefit. Eric Norris, Chief Commercial Officer at Albemarle, noted at a June 2026 conference that grid storage is "much more evenly distributed around the world" than EV demand, describing it as "an interesting demand driver" that provides steadier volume visibility than the lumpy, regionally concentrated EV purchase cycle. Albemarle's Q1 2026 earnings presentation reinforced this view, with management highlighting that battery producers serving ESS applications were reporting full order books through early 2027.
For battery manufacturers, the ESS segment offers a second growth engine at a strategically important moment. CATL's Q1 2026 revenue rose 52.5% year-on-year to 129.1 billion yuan, in part reflecting the company's growing ESS exposure. BYD reported a 55.4% year-on-year decline in net profit over the same period, reflecting the margin pressure of China's domestic EV price war. The contrast between these two results illustrates why ESS is becoming a strategic priority for Chinese manufacturers seeking volume growth beyond a commoditising EV market, particularly as overseas grid-scale demand continues to expand.
For policy, the implications intersect with the supply chain concentration risks identified in the LFP analysis above. The US SEIA projects that American BESS deployments will reach 70 GWh in 2026 and exceed 110 GWh annually by 2030. The UK has prioritised 34.5 GW of battery projects for grid connection. India's CEA is projecting storage requirements of 411 GWh by fiscal year 2031-32. These are sovereign-level infrastructure commitments that create durable, policy-underwritten demand floors. Yet the LFP supply chain that serves them remains overwhelmingly concentrated in China, from cell production to purified phosphoric acid refining. The policy gap between storage deployment ambition and domestic LFP supply chain development is one that no major Western economy has yet credibly closed.
Rio Tinto CEO Simon Trott captured the producer perspective on July 30, 2026, noting the miner was seeing "a shift in demand for lithium from electrification, with grid storage demand stronger than expected." Jérôme Pécresse, Head of Rio Tinto's Aluminium and Lithium business, stated in June 2026 that lithium demand "in the next two years is going to be much more balanced between EVs and energy storage." These public statements from one of the world's largest mining companies, a company with direct commercial interest in the accuracy of its demand modelling, carry the weight of capital allocation decisions behind them. When Rio Tinto characterises ESS as stronger than expected, the implicit message is that prior supply planning was calibrated to a market that no longer exists.
Conclusion: A Demand Architecture Built to Last
The data assembled across the ESS growth story in 2025 and 2026 do not support a characterisation of stationary storage as a cyclical supplement to EV demand. The growth rates, the capital underwriting from utilities and technology companies, the policy commitments from sovereign governments, and the explicit revisions from Fastmarkets, JPMorgan, Benchmark Mineral Intelligence, Albemarle, and S&P Global collectively describe a structural demand shift that is still in its early stages.
The 71% jump in stationary storage lithium demand in 2025, followed by a projected 55% expansion in 2026, is not noise around a trend. It is the trend. Fastmarkets' 750 GWh ESS shipment forecast for 2026, up from 460 GWh, is itself a conservative estimate relative to JPMorgan's 900 GWh projection. At 900 tonnes of LCE per GWh of grid-scale installation, the arithmetic of ESS demand growth rapidly becomes a primary variable in any credible supply-demand model for lithium through the end of the decade.
The cathode chemistry implications compound the structural story. LFP's dominance in ESS applications has decoupled the growth in storage-driven lithium demand from equivalent growth in nickel and cobalt consumption, reshaping how investors and analysts map battery metal exposure to the energy transition. The phosphoric acid and LFP cell supply chain concentration in China represents an emerging geopolitical chokepoint that is the upstream analogue to the downstream concentration risks already well understood in rare earths and processed lithium.
Raju Daswani's characterisation at the Las Vegas conference, that the period of market overcorrection is over and that ESS has become a primary demand driver, is supported by the full weight of the data. The lithium market that emerges from 2026 is one where no analyst, producer, or policymaker can construct a credible forward model without placing energy storage at the centre of the demand architecture, not as a secondary consideration, but as a pillar of equal structural importance to the electrification of transport that originally defined this commodity cycle.