Four-Hour Battery Storage Undercuts Gas Peakers in All 43 Markets

A cost crossover that now covers every modeled market

On the narrow question of how to meet peak electricity demand, gas has lost its cost lead. Wood Mackenzie's latest global levelized cost of electricity (LCOE) report, released October 8, finds that four-hour battery storage now costs less than open-cycle gas turbines in all 43 markets where the firm modeled both technologies.1113 Open-cycle turbines, often called peakers, are the plants grid operators bring online for a few hours when demand spikes. That is exactly the job a four-hour battery is designed to do.

Ahmed Jameel Abdullah, a principal analyst at Wood Mackenzie, called the shift "decisive and widening." He gave two reasons: turbine shortages and fuel price swings are pushing up the cost of gas peaking, while the growth of battery manufacturing keeps pulling storage costs down.1517 Coverage across the trade press was consistent on the core finding. Outlets in Africa, Australia, Europe and North America all repeated the 43-market figure and the same regional numbers.121620

Many headlines said batteries are now "cheaper to install" than gas turbines. That wording goes further than the analysis. The comparison is based on levelized cost, which spreads capital, operating, fuel and financing costs over a plant's lifetime. It is not a comparison of upfront construction prices.1115 The difference is important. Upfront cost is not where batteries win, and some of the most important caveats come from that gap.

The regional picture: China sets the floor

Costs still vary widely from region to region, and the report shows how much manufacturing scale shapes them. China is the global benchmark for grid-scale storage. Its levelized cost is more than 55% below the average for the rest of Asia Pacific, which stands at $134/MWh. Wood Mackenzie attributes the gap to integrated domestic supply chains and very large production volumes.1117 The firm expects cheaper Chinese supply to bring the rest of the region's average down to $92/MWh by 2036. Japan, Australia and the Philippines are expected to keep paying a premium because of import duties, higher installation costs and policies favoring domestic manufacturing.1120

The Middle East and Africa are where the shift is most visible. Four-hour storage there costs about $120/MWh in 2026 and is projected to fall 33% to $80/MWh by 2035, which would make it cheaper than gas peaking in every gas market in the region.1112 Batteries in the region pair with some of the world's cheapest solar. Single-axis tracker solar costs $37/MWh in 2026 and is forecast to reach $24/MWh by 2035. Saudi Arabia and the UAE are on track to go below $20/MWh by 2033.17

In Latin America, storage is expanding through mandates attached to renewable projects and through standalone procurement for grid stability. Wood Mackenzie expects storage costs there to fall 42% by 2060.11 At the generation level, single-axis tracker solar is now the cheapest new-build option in 43 of the 48 markets modeled, and onshore wind leads in the other five.1518

North America: data centers tilt the math

The US angle has drawn the most attention, mostly because of data centers. According to one summary of the reporting, Wood Mackenzie puts four-hour storage at 65% to 75% cheaper than new open-cycle gas peakers in the United States.14 Demand from AI data center developers has tightened turbine supply. Open-cycle units now take two to four years to procure, and waitlists for combined-cycle turbines reach into the early 2030s.19 Reported manufacturer backlogs range from 35 GW to 116 GW.14

Wood Mackenzie expects investment in gas generation to stay in a supply deficit through the late 2030s, driven largely by data center load growth. That keeps the capital cost of thermal plants high and strengthens the long-term case for storage.1117 There is an irony here. Data center demand is often cited as the reason the US needs more gas, but it is also what makes new gas peakers expensive.

US storage also has policy support and policy risks. Tax credits still give batteries an advantage and partly offset the effect of foreign-entity-of-concern restrictions and supply chain limits. Wood Mackenzie expects a cost spike once investment tax credits begin phasing out in 2038, followed by a 10% decline in storage costs by 2060 as new chemistries and domestic supply chains mature.1112 Other analysts have warned that additional US tariffs could cut 2026 installations, and that extreme tariff levels could push effective battery prices back toward 2023 levels.5

The EV battery connection

The storage milestone grows out of the electric vehicle battery industry, and the two markets are now closely linked. BloombergNEF's December 2025 survey put the global average lithium-ion pack price at $108/kWh, a record low and an 8% real decline from 2024.2 Stationary storage packs averaged $70/kWh, down 45% in a year. That made stationary storage the cheapest battery segment for the first time, below EV packs at about $99/kWh.92 The lowest stationary prices observed were $36/kWh for cells and $50/kWh for packs, both for lithium iron phosphate (LFP) chemistry.9

LFP is the key link. It was first adopted at scale in cheaper EVs, and grid storage now relies on it heavily because it trades some energy density for longer life, better safety and lower cost. LFP packs averaged $81/kWh in 2025, compared with $128/kWh for nickel-manganese-cobalt packs.2 The factories China built for electric cars, and their overcapacity, are now driving down the price of grid batteries. That is a large part of why China's storage costs sit so far below everyone else's.

System prices have fallen with pack prices. BloombergNEF's 2025 survey found that turnkey four-hour systems averaged $110/kWh globally. Regional averages were $73/kWh in China, $177/kWh in Europe and $219/kWh in the US.4 The think tank Ember estimated all-in project capex outside China and the US at about $125/kWh. Roughly $75/kWh of that is Chinese-made core equipment and about $50/kWh is installation and grid connection, which works out to a levelized storage cost of around $65/MWh.7 In February 2026, BloombergNEF reported that the global benchmark levelized cost of a four-hour battery project had dropped 27% in a year to $78/MWh.121

Where the reporting diverges, and the lithium wildcard

The benchmarks do not match exactly. BloombergNEF's $78/MWh is far below Wood Mackenzie's $120/MWh for the Middle East and Africa and its $134/MWh rest-of-Asia-Pacific average. Other academic estimates put the mean levelized cost of storage for new batteries closer to $211/MWh.1115 These gaps mostly reflect different assumptions about financing, utilization, duration and location. The direction is the same across all of them: storage is getting cheaper and gas peaking is getting more expensive. Readers should take the general trend from these figures, not any single precise number.

The main near-term risk is materials. In Europe, turnkey battery capex rose about 2% in 2026, the first increase in three years, after cell prices rebounded around 10% from their 2025 low on stronger demand and higher lithium prices.11 Battery-grade lithium carbonate has roughly tripled from its mid-2025 low.2 Wood Mackenzie still expects European capex to fall 12% by 2031, but it also expects lithium prices to roughly double in 2029 as oversupply clears and demand outpaces new mining and refining capacity.16 Ember points to sodium-ion batteries, which use no lithium, as one way to reduce this exposure.7

What the milestone does and doesn't mean

Levelized cost has limits. It describes lifetime cost, not every service a grid operator needs. A gas turbine can run as long as it has fuel. A battery is limited by its stored energy and its ability to recharge.15 A four-hour battery is a strong replacement for a peaker that runs a few hours on hot evenings. It is not a replacement for firm capacity during a multi-day winter cold snap.

The headline finding still holds up. Different methods all point the same way, and the drivers are structural: falling costs from EV-scale manufacturing, a turbine supply chain under heavy demand, and fuel price volatility that batteries avoid. Abdullah's argument is that solar and storage together are now changing the basic economics of power systems, and that the case for new gas peaking capacity is closing.1718 The crossover is real. What limits further gains is trade policy, lithium prices and grid connection costs, not the batteries themselves.