Energy Storage (Utilities)

Energy

Energy Storage (Utilities)

The sun sets and the wind drops. Storage is the answer to the only serious objection anyone has left, and its cost fell 89% in thirteen years.

🌱Seasons

The effect compounds within years. Put it in place and it keeps working.

Origins

Electricity has one property that makes it uniquely awkward among the things humans trade: it has to be used at the instant it is made.

You cannot warehouse it. Every second of every day, on every grid on Earth, generation must exactly match consumption, or the frequency drifts and things break. This is why grid operators exist, why there are power plants that sit idle for most of the year waiting for a cold evening, and why the entire architecture of the electricity system is built around the ability to turn generation up and down on command.

Which is precisely what the wind and the sun will not do.

The oldest answer is pumped hydro, and it is beautifully simple: when you have spare electricity, pump water uphill into a reservoir; when you need it, let it fall through a turbine. It is over a century old, it is still the overwhelming majority of the world’s storage, and it works. It also requires two lakes at different altitudes, which is a rather specific geographic demand.

The lithium-ion battery was developed for camcorders and laptops. Nobody in the 1990s was thinking about grids. It scaled because of consumer electronics, then because of electric cars, and only then — almost as an afterthought, riding a cost curve driven entirely by other industries — did it become cheap enough to hold up a power system.

The grid is being rebuilt on the back of a battery invented for a Walkman.

What it actually is

Storage does several distinct jobs and they are frequently confused.

Shifting hours. This is the main one, and it is what batteries are extremely good at. Take the mid-day solar glut, hold it for four hours, release it into the evening peak when everybody comes home. This single function is what makes a solar-heavy grid work, and lithium-ion does it cheaply now.

Stabilising seconds. Grids need instantaneous frequency response, and batteries are dramatically better at this than any thermal plant — they respond in milliseconds, where a gas turbine takes minutes. The first big grid battery, installed in South Australia in 2017, made a great deal of money and embarrassed a great many people simply by being fast.

Bridging days and seasons. This is the hard one, and it is where honesty is required. A week of still, overcast weather in a northern winter — the Germans call it Dunkelflaute, the dark doldrums — cannot be covered by lithium batteries at any plausible cost. Holding a week of a country’s electricity in batteries would be economically absurd.

Long-duration storage is therefore an unsolved problem, and the candidates — pumped hydro, compressed air, iron-air batteries, thermal storage, hydrogen — are all either geographically constrained, immature, or inefficient. Anyone telling you this is solved is not being straight with you.

What is solved, and solved decisively, is the four-hour problem. And the four-hour problem is most of the problem.

The numbers

The cost collapse. Battery storage costs fell roughly 89% between 2010 and 2023 (IRENA). This is the single most consequential number in the energy transition after the solar learning curve, and it was driven almost entirely by electric vehicles, not by grids.

Firm renewables are now competitive. IRENA’s 2026 analysis found solar-plus-storage delivering round-the-clock electricity at US$54–82/MWh in high-resource regions, down from over US$100/MWh in 2020, and projected to fall a further ~30% by 2030 and ~40% by 2035, dropping below US$50/MWh at the best sites. In good regions, storage-firmed renewables already beat fossil benchmarks.

What is still true. Pumped hydro remains the overwhelming majority of installed storage globally — over 90% of it — because it is cheap per unit of energy stored and it works. It just needs the right two hills.

The honest gap. Multi-day and seasonal storage remains genuinely unsolved. Lithium-ion is excellent for four hours and hopeless for four weeks. This is the real technical challenge remaining in the electricity system, and it is where research money should be going.

The speed. A grid battery responds to a frequency event in milliseconds. A gas turbine takes minutes. This is not a marginal advantage; it is a different category of capability.

Why it matters

Every serious objection to renewable energy now reduces to a single sentence: what happens when the sun goes down and the wind stops?

It is a completely fair question. It was, for decades, an unanswerable one. And it has become, over about ten years and almost without anybody noticing, a question with an increasingly boring answer: you put it in a battery, and the battery costs 89% less than it did, and in the sunniest places round-the-clock solar power now competes with gas.

There is something quietly moving about this. The problem that was supposed to be fundamental — the physics-based, unarguable reason we could never run on sun and wind — is being dissolved by a manufacturing learning curve. Not by a breakthrough. Not by genius. By people making an enormous number of batteries, and each batch being slightly cheaper than the last, for fifteen years.

That is how most real progress happens, and it is far less glamorous than the story we like to tell about invention.

The remaining hard problem — the dark, still week in a northern winter — is real, and we should say so. But it is a much smaller problem than the one we started with, and the people working on it are the ones who will hand the next generation a grid that runs on daylight and weather, which is a thing our grandparents would have considered a fairy tale.

What it actually takes

Solving long duration, which is genuinely unsolved. Four hours is done. Four days is not. Iron-air, compressed air, thermal, pumped hydro, hydrogen — all are candidates and none is clearly the answer. This is where research funding belongs and it is chronically underfunded relative to its importance.

Not pretending lithium can do everything. The industry has an incentive to imply that batteries solve the whole problem. They solve most of it. The last part is the hardest part and it needs a different technology.

Mining, honestly. Lithium, cobalt and nickel come out of the ground, and the human and environmental cost of that is real. Cobalt in the DRC is the sharpest example. This is not a reason to abandon batteries — the alternative is coal, whose mining record is far worse and whose product also poisons the air — but it is a reason to demand traceable supply chains, to fund recycling, and to move to chemistries that need less of the worst materials. Lithium iron phosphate, which uses no cobalt, is already displacing older chemistries fast.

Recycling, before the wave arrives. The batteries being installed now will retire in fifteen years. Building the recycling industry before then, rather than after, is the difference between a circular material and a mountain of waste.

Market design. Storage makes money by arbitrage and by providing grid services, and in many markets the rules were written when storage did not exist and do not let it be paid for what it does.

Where it matters most

Australia is the world’s laboratory. Extraordinary rooftop solar penetration, big grid batteries, and a grid being rebuilt in public. The Hornsdale battery in South Australia demonstrated that a battery could stabilise a grid faster and cheaper than a gas plant, and it changed the conversation globally.

California and the American Southwest have the duck curve in its purest form — enormous mid-day solar, sharp evening peak — and are deploying grid batteries at a scale nobody predicted five years ago.

The North German Plain and northern Europe are where the Dunkelflaute problem is most acute: still, dark, cold weeks where neither wind nor solar delivers. This is the hardest storage problem in the world and it is where the long-duration answer will have to be proven.

Norway and the Alps hold Europe’s pumped hydro, and function as a continental battery for everyone else — which is a remarkable geopolitical fact that almost nobody discusses.

The East African Rift and off-grid South Asia, where a small battery beside a solar panel is not grid infrastructure. It is the difference between light after dark and no light after dark.

How to tell it’s being done well

Is it four-hour or long-duration? These are completely different problems and conflating them is the most common way this subject is misrepresented. Lithium does four hours brilliantly and four weeks not at all.

Where did the materials come from? Cobalt from the DRC has a genuine human cost. Traceability is possible and cobalt-free chemistries exist and are growing fast.

Is there a recycling plan? These batteries retire in fifteen years. Building the industry before the wave is far cheaper than after.

Can it get paid for what it does? Storage provides frequency response, capacity and arbitrage. Many markets only pay for one of them, and that is a policy failure, not a technology failure.

What you can do

Anyone

  • The what-about-when-the-sun-goes-down objection is now largely answered by economics rather than by argument. Battery costs fell 89% between 2010 and 2023, and in sunny regions solar-plus-storage now competes with fossil around the clock.
  • The remaining hard part is a still, dark week in a northern winter. Anyone who tells you that is solved is overselling.

Homeowners

  • A home battery roughly doubles the value of rooftop solar, because it lets you use your own power in the evening instead of selling it at midday for very little.
  • Aggregated home batteries are becoming virtual power plants, and in some markets you can now be paid for participating.

Policymakers

  • Fund long-duration storage research. It is the genuine remaining gap and it is underfunded relative to its importance.
  • Reform market rules written before storage existed. In many markets, batteries cannot be paid for the services they actually provide.
  • Build the battery recycling industry now, before the retirement wave arrives in fifteen years.

Business and investors

  • Long-duration storage is the largest unsolved problem in the electricity system and therefore the largest opportunity in it.
  • Battery recycling is an industry that must exist and barely does. The feedstock is already installed and its retirement date is known.

Who is working on this

We are researching which organizations in our directory of 8,493 actively work on this solution, and we only list an organization once we have verified it. That research is ongoing. In the meantime, search the directory yourself:

Search the directory for “Energy Storage (Utilities)” →

Questions

Why does electricity need storage at all?

Because it has to be used at the instant it is made. Generation must match consumption every second on every grid on Earth, or the frequency drifts and equipment fails. The entire architecture of the power system was built around generators that can be turned up and down on command, which is exactly what the sun and the wind will not do.

How much have batteries fallen in cost?

Roughly 89% between 2010 and 2023. It is the second most consequential number in the energy transition after solar's learning curve, and it was driven almost entirely by electric vehicles rather than by grids. The grid is being rebuilt on the back of a battery originally scaled for consumer electronics.

Does storage actually solve intermittency?

It solves most of it. The four-hour problem, shifting midday solar into the evening peak, is now decisively solved and cheap. IRENA finds solar-plus-storage delivering round-the-clock power at US$54-82/MWh in good regions, competitive with fossil fuels.

What is still unsolved?

Multi-day and seasonal storage. A still, overcast week in a northern winter, what the Germans call Dunkelflaute, cannot be covered by lithium batteries at any plausible cost. Long-duration storage is a genuinely open problem, and anyone claiming otherwise is not being straight with you.

What about the mining?

It is a real cost and it should be stated plainly. Lithium, nickel and especially cobalt have significant human and environmental costs, and cobalt in the DRC is the sharpest example. This is not a reason to abandon batteries, since the alternative is coal, whose mining record is far worse and whose product also poisons the air. It is a reason to demand traceable supply chains, fund recycling, and move to chemistries like lithium iron phosphate that use no cobalt at all.

Isn't pumped hydro old technology?

It is, and it still accounts for over 90% of the world's installed storage, because it is cheap per unit of energy and it works. Its only real limitation is that it requires two reservoirs at different altitudes, which is a fairly specific geographic demand. Norway and the Alps effectively function as a battery for the rest of Europe.

Sources

The solution taxonomy follows the framework popularised by Project Drawdown. The analysis above is our own; for their carbon modeling and rankings, visit them directly.