"IRENA: Global Clean Energy Additions Will Fall in 2026, the First Drop in Five Years"
The head of the International Renewable Energy Agency said something on 27 October that should unsettle anyone who assumed the energy transition only moves in one direction: the world will add less clean energy capacity in 2026 than it did in 2025, the first annual decline in five years. Francesco La Camera, IRENA's director-general, told Reuters on the sidelines of Singapore International Energy Week that the slowdown is not a story about demand, or about the cost of solar panels and wind turbines. It is a story about wires.
The contrast with last year is stark. Clean power installations, including hydropower, surged by a record 692 gigawatts in 2025, up 15.5 percent, taking total installed renewable capacity to 5,149 gigawatts — roughly half of all the world's power capacity. Growth had accelerated every year since 2021, when the pandemic disrupted supply chains, and governments had pledged to go further still, partly because the war with Iran choked energy supplies and inflated prices. On that trajectory, a down year was not supposed to happen yet.
What changed, La Camera said, is the mismatch between the capacity to generate electricity and the capacity to distribute and transmit it. China, the world's largest renewable energy producer, has been hit hardest by curtailment — clean power switched off because the grid cannot take it. In the first half of 2026 alone, China turned away enough renewable electricity to power Mexico for a year. Networks in Australia and Japan have also failed to keep pace with a surging buildout, and trade restrictions and supply-chain difficulties have slowed the cross-border flow of components. La Camera added that he did not expect growth elsewhere in the world to offset the slowdown in China, though he declined to give a numerical forecast for 2026.
| Indicator | 2025 figure |
|---|---|
| Global clean power additions | 692 GW |
| Growth in total clean capacity | 15.5% |
| Total installed clean capacity | 5,149 GW |
| Share of world power capacity | About half |
I find this the most important energy story of the year precisely because it is boring in the way infrastructure stories are boring. We learned how to build solar and wind at extraordinary scale and speed, and we did not learn, at the same pace, how to move the electricity they produce to where it is needed, when it is needed. A curtailed wind farm is not a failed technology; it is a working technology attached to a grid that was designed for a different era, in which a few large plants sent power in one direction. Every gigawatt curtailed is clean energy that was paid for, built, and then thrown away, and curtailment on China's scale is a warning that the constraint on the transition has moved downstream, from the factory to the network.
The timing matters, because the climate arithmetic has not softened. The United Nations Environment Programme's Emissions Gap Report, released on 21 October, found that solar and wind together abated 4.1 billion tonnes of carbon dioxide in 2025, and that renewable additions between 2013 and 2025 ran three times the level anticipated in 2015. Yet under current policies the world is still on course for warming below 2.8 degrees Celsius by the end of the century, and even full implementation of national climate pledges would only bring that to within 2.6 degrees. The report's message and La Camera's warning point the same way: the transition is no longer mainly limited by how much clean power we can build, but by whether grids, storage and interconnection can absorb it.
That is also why I expect IRENA's full-year figures, when they are published, to confirm the decline — additions in 2026 below the 692 gigawatts of 2025. The year was already more than three-quarters over when La Camera spoke, the Chinese curtailment was already in the books for the first half, and he was explicit that the rest of the world would not make up the difference. A fourth-quarter surge could still surprise me, and the exact definition IRENA uses — including hydropower, as in the 2025 record — will matter at the margin. But the direction of travel is now set by concrete and copper, not by announcements. If the dip proves to be a one-year pause while grids catch up, it will be remembered as a useful scare. If it becomes a pattern, it will mean the energy transition has a bottleneck that no amount of cheaper panels can fix, and that the next race is not to build more, but to connect what we have already built.