Why Four-Hour Batteries Now Beat Gas Peakers on Cost

The number that reset the argument

Wood Mackenzie's analysts concluded that four-hour battery storage is now cheaper than gas peaking capacity on a global basis, a finding picked up by Renewables Now and circulated through the trade press within days. For an industry that spent a decade arguing about whether storage could compete at all, the question has quietly changed shape. The new question is not whether batteries beat peakers on cost. It is what the cost comparison actually measures, and who gets to convert that measurement into a functioning power plant.

Fact: the levelized-cost comparison now favors four-hour lithium-ion storage over open-cycle gas turbines for peak-shaving duty in most major markets, according to Wood Mackenzie. Interpretation: a cost advantage on paper is a necessary condition for deployment, not a sufficient one. Plenty of cheap technologies have died waiting for grid connection.

Why four hours became the default unit

The four-hour duration is not a law of physics. It is a procurement artifact. California wrote the country's first state storage mandate in 2013, and most of the large-scale batteries it ended up with run for four hours. The design spread from there. Once procurement documents, financing models, and warranty templates all assumed 240 minutes of discharge, the industry had little incentive to deviate. Four hours became the industry's standard shipping configuration the way the 20-foot container standardized global freight.

That standardization is now doing real economic work. Maryland's first bulk storage procurement landed at 440 MW and 1,760 MWh, which is exactly four hours per unit of capacity. When a state auction produces a clean multiple like that, it is not coincidence; it is the default hardening into specification. Scale manufacturing follows standardization, and standardization follows the standard.

Some jurisdictions are already moving past it. Bangladesh approved an eight-hour storage requirement for a 442 MW solar project, a duration that makes sense for a grid with heavy evening demand and limited dispatchable backup. South Africa's energy minister has prioritized battery storage alongside gas-to-power, with the Integrated Resource Plan targeting 4.6 GW of battery capacity. The four-hour standard is holding at the center of the market while the edges stretch in both directions.

What the cost curve leaves out

A levelized cost of storage figure captures capital cost, round-trip efficiency, and assumed utilization. It does not capture the queue. Building a battery in a location with available interconnection capacity and building one without are, economically, two different products, and the queue is where clean energy goes to wait. Wood Mackenzie's global average implicitly assumes a project that gets built on schedule. Real projects do not.

There is also the capacity-value question. A gas peaker can run for as long as fuel supply permits. A four-hour battery cannot cover a six-hour evening peak without curtailing something else. Grid planners in ERCOT and the UK have spent years quantifying this difference, and the answers depend heavily on local load shape, weather correlation, and how much solar is already installed. A battery is not a peaker with different fuel, just a shifting device competing for the same procurement line item.

Prediction: expect the next two years of procurement to reveal a widening gap between headline cost figures and awarded prices, with the spread concentrated in congested regions. Interconnection costs, not cell costs, will decide which projects survive contract award.

The backlash has a name problem

Opposition to storage projects has grown loud enough that the Washington Post ran an opinion column calling the battery-storage backlash a costly case of mistaken identity. Brian Deese, who directed the National Economic Council from 2021 to 2023, and Anna Pasnau make the case that batteries are being blamed for grid strain that data centers create. Storage soaks up cheap electrons and shaves the evening peak, then gets zoned alongside the data centers it was meant to relieve.

That framing is defensible and incomplete. Fire risk in containerized lithium installations is a real engineering problem with real incident history, and treating every objection as misdirected NIMBYism has not worked well for the industry's public standing. The more useful distinction is between a complaint about the technology class and a complaint about a particular site's setback distances, fire-suppression design, and transformer noise. The first is hard to answer. The second is a design brief.

The operational side has its own unresolved issue. A paper published this month in Nature Reviews Electrical Engineering on transferable predictive maintenance for battery energy storage systems found that drift in duty cycles and monitoring pipelines undermines diagnostic models unless they are adapted, which matters as fleets age and operators swap cell suppliers. Cheap cells and cheap diagnostics are not the same procurement.

What a cost win actually buys

For readers tracking the wider picture, our complete guide to energy storage and the grid covers how these assets interact with transmission planning and capacity markets. The short version: a cost advantage accelerates procurement only where the surrounding infrastructure exists. Australia's Enerven is building a 400 MW battery project in a market that has spent a decade learning to absorb storage. That learning curve, not the cell price, is the scarce resource.

Wood Mackenzie's finding is genuine, and it will be cited in hundreds of procurement decks before the year ends. The honest reading is narrower than the headline. Four-hour storage has become the cheapest way to buy two hours of reliable evening capacity plus two hours of optimism, in markets where the transmission and interconnection work has already been done. Where that work has not been done, the gas peaker has a competitor on price and a monopoly on delivery dates.