Solar becomes the world's largest source of primary energy, IEA confirms
Capacity additions in 2025 crossed the threshold at which solar's share of global supply eclipses every other source — in rolling aggregate terms, for the first time.
The crossover is structural, not cyclical. Capacity additions in 2025 crossed the threshold at which solar's share of global primary supply eclipses every other source in rolling aggregate terms, and nothing currently on the manufacturing curve suggests the trend reverses. Manufacturing capacity in China, India, and the Gulf is outrunning demand forecasts by roughly an order of magnitude, according to the IEA's own numbers and an independent tally by BloombergNEF published the same week.
For most of the last decade, the energy debate hinged on a single question: when does solar stop being additive and start being substitutive. The IEA's answer, delivered with characteristic understatement in a Paris briefing, is that the substitution has already happened — quietly, in aggregate, and without any single policy moment to mark it. Coal is still burned. Gas is still flared. But the marginal unit of new primary energy on earth is now photovoltaic, and the gap is widening month over month.
The specific mechanism is manufacturing overcapacity. Chinese module production alone, per IEA figures, now exceeds projected global installation demand through 2030 under the agency's central scenario. That oversupply has collapsed module prices to levels at which residential-scale deployment pencils out in markets that were uneconomic two years ago, and the resulting demand is self-reinforcing. The downstream question — transmission, storage, grid balancing — is where the next decade of trouble and opportunity concentrates.
The winners are distributed and unglamorous: inverter manufacturers, transformer specialists, the regional transmission operators nobody has heard of who happen to hold the right rights-of-way. The losers are the integrated utilities whose business models assumed marginal generation would continue to come from assets they owned, and the gas producers whose long-dated contracts were priced against a different curve. The fossil-fuel majors have, to their credit, been reading this chart for years; their capital allocation suggests they believed it earlier than their public posture admitted.
What the milestone forecloses is the argument that solar remains a supplementary source requiring permanent subsidy. What it opens is a far harder problem: an electricity system whose generation curve is diurnal and whose demand curve is not. The next decade's energy-policy debates will not be about whether to build more solar. They will be about the wires, the batteries, and the software that decides when the lights stay on.