Battery Storage Beats Gas Peakers on Cost in All 43 Markets

Gas peakers have long been the cheap fallback for evening demand spikes. That assumption no longer holds on cost. Wood Mackenzie’s global levelised cost analysis, released on 8 October 2026, found that four-hour battery storage is cheaper than new open-cycle gas peakers in all 43 markets where both were modelled.

The timing matters because the two cost curves are moving in opposite directions. Gas turbine order books stretch to 2030 or later, while storage pack prices keep falling.

If you hold thermal generators, renewables developers or battery materials producers, the gap affects you differently in each case. Here is what the data shows about where the gap is widest, which drivers are structural and which are cyclical, and what that means for your energy and materials exposure.

What does the 43-market result actually say about regional costs?

The headline is sweeping. Four-hour storage costs about US$120/MWh in 2026, and Wood Mackenzie forecasts a 33% fall to roughly US$80/MWh by 2035. The source ties that trajectory to the Middle East and Africa in particular.

The Wood Mackenzie press release confirms the finding spans Europe, North America, Latin America, Asia Pacific, and the Middle East and Africa, so the cost advantage is not confined to one grid or policy regime.

The margin of the win varies widely by region. In the United States, a spokesperson told Utility Dive that four-hour batteries reaching commercial operation in 2026 are 65%-75% cheaper than new open-cycle peakers. The range depends on whether state carbon pricing applies, and the figures include federal tax credits. Wood Mackenzie did not publish precise absolute prices, so the US comparison exists only as a percentage.

China sits further out still, with storage costs more than 55% below the Asia Pacific average. Its manufacturing scale makes it the global cost benchmark.

Analyst view Principal analyst Ahmed Jameel Abdullah described the shift as decisive and widening. He attributed it to turbine shortages and fuel volatility pushing peaking costs up while battery manufacturing growth pulls storage costs down.

The regional spread tells you the investable story differs by market. In North America the advantage is driven by policy, in China by scale, and in the Gulf and Africa by solar-plus-storage. One global headline covers three different theses.

Why levelised cost is the right lens, and where it stops

Levelised cost of electricity (LCOE) is the average cost of producing one megawatt-hour over an asset’s full life, covering construction, financing, fuel and operation. Wood Mackenzie frames its comparison in $/MWh, not the battery pack prices quoted in $/kWh.

The comparison is fair here because a peaker and a four-hour battery do the same job: they cover a few hours of high demand, usually in the evening. On that task, comparing cost per megawatt-hour is reasonable.

A peaker and a four-hour battery do the same job on the demand curve, but the way battery storage earns revenue differs, with energy arbitrage, capacity payments and ancillary services all stacking onto the cost advantage.

The two assets still differ in one important way. A gas peaker can run for as long as it has fuel. A four-hour battery runs out of energy after four hours.

What LCOE leaves out

That difference is why the finding applies to peaking and should not be stretched further. The metric misses several things:

• Capacity value and resource adequacy: whether the system has enough firm supply when it needs it most.
• Multi-day reliability: extended low-renewable periods such as dunkelflaute (still, dark weather) or long heat waves still need gas, hydro, nuclear or long-duration storage.
• Degradation and augmentation: lithium-ion cells lose capacity over time, and the cost of adding or replacing cells can be understated in averages.
• System flexibility value: a single cost metric does not price how an asset supports the wider grid.

Read the result as a verdict on the economics of new-build peaking. It does not forecast the disappearance of gas generation.

How the gas turbine squeeze is lifting peaker costs

Falling battery costs explain only part of the gap. On the gas side, costs are being pushed up by a manufacturing queue that keeps lengthening.

Wood Mackenzie expects a North American gas generation supply deficit through the late 2030s, driven by data centre load growth. The three major manufacturers hold backlogs that, in Wood Mackenzie’s description, range from 35 GW to 116 GW each.

Note: GE Vernova figures are not like-for-like. Utility Dive reported 100 GW in Q1 2026, while July 2026 earnings materials cited 83 GW of backlog and slot reservations on a different basis.

Some units now carry lead times out to 2029-2031 or beyond. Pricing has followed. William Blair estimated GE Vernova was booking orders at about $2,500/kW in early 2026, and Ainvest put new US combined-cycle plant costs at a minimum of $2,400/kW. Both figures reflect full project or order pricing.

The figures from BNP Paribas are a different measure. Its July 2026 estimates of $790-$1,800/kW apply to turbines by class. Wood Mackenzie projects turbine prices reaching $600/kW by end-2027, a 195% rise since 2019, also on an equipment-only basis.

The backlog data tells you turbine makers are likely to hold pricing power for years. That supports the value of existing thermal assets while making new peakers harder to justify.

Policy, solar and wind: what shapes the storage cost curve

The long-run storage story is encouraging. BloombergNEF’s December 2025 survey put stationary storage packs at $70/kWh, down 45% in a year, against an average pack price of $108/kWh and LFP (lithium iron phosphate) packs at $81/kWh.

The battery storage cost curve behind these figures reflects years of manufacturing scale and LFP adoption, and the same forces explain why packs fell to $70/kWh in a single year.

The near term carries more friction, particularly in the US. The Inflation Reduction Act created a standalone investment tax credit (ITC) for storage, along with the 45X manufacturing and 45Y production credits. Wood Mackenzie expects the ITC phase-out to begin in 2038, followed by a cost spike, with storage LCOE then falling 10% by 2060.

Trade policy adds further exposure. ESS-News warned on 9 October 2026 that US tariffs on imported battery and solar components threaten near-term gains. Foreign-entity-of-concern (FEOC) guidance for grid storage procurement is still evolving and has not been settled.

Solar and onshore wind remain the cheapest new build. Single-axis tracker solar leads in 43 of 48 modelled markets, and Gulf solar is on track to fall below US$20/MWh by 2033. North America faces headwinds: utility-scale module prices are expected to rise about 5% a year through 2030, despite 168 GW of safe-harbored capacity. Onshore wind LCOE is expected to fall 16% by 2060.

Structural versus cyclical drivers

• Structural: manufacturing scale, LFP adoption, and mature four-hour lithium-ion technology.
• Cyclical: turbine bottlenecks, data centre demand, and volatility in fuel and carbon prices.

This tells you the US advantage is partly created by policy. Exposure to US storage names carries regulatory risk that Chinese or Gulf exposure does not.

What this means for thermal, renewables and battery materials investors

Each sector reads the result differently, and the materials layer may move furthest from the expected direction.

Thermal: New open-cycle peakers face a shrinking niche. Existing combined-cycle plants may shift into lower-utilisation mid-merit roles, though turbine scarcity supports incumbents.

Renewables: Cheaper four-hour storage covers evening peaks, which strengthens the case for high wind and solar penetration. Markets such as California, Texas and Australia are often cited as examples. These cases are illustrative only and have not been independently verified in the underlying research.

Australian GenCost analysis reaches a similar conclusion from a national modelling base, showing cheaper batteries steadily eroding the case for new gas generation in a market with high wind and solar penetration.

Battery materials: China, the Middle East and Africa are the clearest demand centres. Higher demand, however, does not guarantee higher prices.

Key price point Lithium carbonate (EXW China) declined through Q3 2026 to around RMB 127,750/tonne (about $19,025/tonne), according to Benchmark Mineral Intelligence.

Lithium, nickel, cobalt and graphite have all been volatile in 2026. Read the finding as support for storage and renewables volumes, not as a guarantee of materials prices.

Reading the cost gap without overreaching

Batteries now win the cost contest for four-hour peaking, while gas keeps a role in reliability and long-duration supply. Turbine scarcity is likely to widen the divergence for years. Three variables decide how much of that reaches your portfolio:

1. US tariff and FEOC outcomes, which set the size of the policy-made advantage.
2. The turbine price trajectory, which shows whether scarcity eases or persists.
3. Metals pricing, which tests whether demand translates into producer margins.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors.