The Energy Transition
Why the Grid Is the Hardest Technology Problem of Our Lifetimes
We've figured out how to make cheap clean electricity — the part we haven't figured out is what to do with it when no one wants it.
The Idea
The energy transition has a dirty secret: the hard part was never generating renewable electricity. Solar and wind have collapsed in cost faster than almost any technology in history — faster than mobile phones, faster than computers. The hard part is matching supply to demand on a grid that was designed around the assumption that power plants could be turned up or down on command. The sun doesn't negotiate with the morning commute. Wind doesn't care about a cold snap in January. So as renewable penetration rises, grids face a structural mismatch: moments of wild oversupply (sunny weekend afternoons when factories are quiet) and moments of frightening scarcity (cold, still evenings when demand peaks and generation doesn't). Engineers call this the 'duck curve' — the shape you get when you plot net demand across a day as solar generation rises, then drops sharply at sunset right as people get home and turn everything on. Storage is the obvious answer, and batteries are getting cheaper fast. But there's a crucial distinction between short-duration storage — smoothing out a few hours of imbalance — and the truly difficult problem: seasonal storage. Northern Europe in winter. California in a high-pressure system with little wind and short days. The gaps can last weeks. No battery chemistry currently handles that economically at scale. What starts as an engineering challenge becomes a civilisational one: how do you reliably power an industrial economy on energy sources that depend on the weather?
In the World
California offers a near-perfect case study in both the promise and the pathology. By 2023, the state regularly generated more solar electricity than it could use on spring afternoons — so much that grid operators paid neighbouring states to take the surplus, and still sometimes had to curtail the panels entirely, effectively wasting clean energy that cost almost nothing to produce. Meanwhile, on a handful of evenings per year, the grid came within minutes of rolling blackouts as the sun set and air conditioners ran full blast. The state's response has been instructive. California accelerated the rollout of utility-scale battery storage more aggressively than anywhere else on earth — and it worked, in a narrow sense. The batteries absorbed afternoon surplus and discharged it into the evening peak, shaving off the most dangerous hours. By 2024, California had more grid-scale battery capacity than the rest of the world combined had a decade earlier. But the batteries helped with the duck curve, not with January. A two-week high-pressure winter anticyclone — cold, dark, still — would drain them in hours. So California, like Germany, like the UK, quietly maintains gas-fired plants on standby. They run rarely, but they have to exist. The uncomfortable truth embedded in that fact is that the transition isn't just a matter of building more clean generation — it requires solving storage at timescales that current technology doesn't yet reach.
Why It Matters
Understanding the grid problem reframes a lot of energy debates that otherwise generate more heat than light. It explains why nuclear keeps appearing in conversations that seem, on the surface, to be about solar and wind — not because it's cheap or fast to build, but because it generates dispatchable, weather-independent power that plugs directly into the gap seasonal storage hasn't yet filled. It also reframes how you think about electrification at home. An electric vehicle, charged overnight on cheap surplus renewable electricity, is genuinely different — environmentally and economically — from one charged during peak evening demand. The same electrons, the same car, but a completely different grid impact depending on timing. 'Going electric' is less a single decision than an ongoing negotiation with the grid's rhythms. And it puts a sharper edge on what 'solved' actually means in energy policy. Building cheap clean generation was the first act. The second act — matching it reliably to the way modern life actually uses power — is where the real engineering, economic, and political complexity lives. Anyone speaking with total confidence about timelines probably hasn't looked closely enough at the duck curve.
A Question to Ponder
If the hardest part of the energy transition turns out to be storing electricity across weeks and seasons rather than hours, which existing or emerging technologies — hydrogen, pumped hydro, advanced nuclear, something else entirely — do you think is most likely to fill that gap, and what would need to be true for it to get there?
Get a new one of these every morning.
Start learning with Thinkable