From orbit, the desert north of Ürümqi in China reveals coal storage bunkers, each 300 metres long, arrayed around aluminium smelters and polysilicon plants built in the last five years. These are not relics of Mao-era industrialisation, nor transitional facilities awaiting retirement. The coal they store powers factories that produce most of the world’s solar panels.
The pattern repeats across China’s interior, wherever energy-hungry clean-tech manufacturing has clustered around cheap, reliable power. The geography matters because it exposes a basic misreading of the energy transition: the West has focused on what China deploys, not on how China produces.
The fashionable view holds that China is building the 21st-century economy while America clings stubbornly to the 20th. That framing has hardened into orthodoxy: Washington hawks the fuel of last century while Beijing invests in the infrastructure of the next. The contrast is seductive – electrons versus molecules, future versus past – but it mistakes a temporary industrial strategy for a permanent transition.
China’s installed wind and solar capacity doubled in three years. In early 2025, renewables overtook coal in total generating capacity for the first time. Clean sources met 84% of electricity demand growth in 2024, and in the first half of 2025 renewable output covered all additional demand, forcing fossil fuel generation into retreat. Chinese electric vehicles (EVs) now outsell internal combustion engines.
What emerges is not a green transition but a green arbitrage: carbon burned in one jurisdiction so that virtue may be claimed in another.
But trajectory is not destination. Coal still generated 53% of Chinese electricity in 2024. China burns more coal than the rest of the world combined, building nine of every ten new coal plants. These new “flexible” units may be sold as grid backup, but each represents a 40-year bet on the world’s dirtiest fuel.
What emerges is not a green transition but a green arbitrage: carbon burned in one jurisdiction so that virtue may be claimed in another. The arbitrage works because Western nations have committed to a transition they lack the industrial capacity to deliver. China supplies the difference, at the cost of its own air and soil.
The supply chains tell the story plainly. Indonesian nickel for EV batteries is processed in Chinese-owned industrial parks with coal plants attached like IV drips. Magnesium for lightweight automotive alloys is smelted using the Pidgeon process, a deliberately archaic method that runs on coal. Electrified alternatives exist and once competed; they lost on cost. This is not negligence. It is strategy.
The logic is cynical but persuasive: China will scale dirty to lock in dominance, then electrify once the game is won. Brussels bets that carbon border adjustments will force China to clean up its grid to maintain market access. Washington favours tariffs without rebuilding the industrial base beneath them. Both approaches rest on a shared delusion: that China’s energy system is still being built for Western approval. It is not.
Beijing is building a Green Iron Curtain. On one side: rich, carbon-constrained economies, where energy is high-cost and heavily regulated. On the other: the Global South, where China is exporting its coal-fired manufacturing model to markets that value cheap power over clean molecules. Carbon leakage is not a failure of the system; it is the secret of its success.
Against this strategy, Washington’s response looks half-baked. The United States pumps 13.5 million barrels of crude a day, yet remains impotent against a battery supply chain built on critical minerals rather than oil. Wind, solar, and batteries grow cheaper by the quarter; oil, gas, and coal do not. Even a great power cannot outrun cost curves. Betting on fossil fuel abundance is betting on a wasting asset.
China’s coal dependency is a vulnerability dressed as strength: clean products sold abroad, dirty air kept at home.
Cheap energy is necessary for manufacturing leadership but not sufficient. The web of smelters, refiners, and component suppliers required to produce batteries and solar panels does not exist on American soil in meaningful volume. You cannot drill your way to a solar-electric economy any more than you can farm your way to semiconductor dominance.
China’s coal dependency is a vulnerability dressed as strength: clean products sold abroad, dirty air kept at home. Beijing stopped publishing protest statistics when the annual count exceeded 100,000. Environmental grievances were among the top three causes. America’s fossil fuel abundance is a distraction from its industrial hollowing, a consolation prize mistaken for a winning hand. Neither country has cracked the problem that will determine the next century: how to manufacture without coal.
The race that matters is not between electrons and molecules but between clean production and dirty production. On that measure, neither superpower has yet pulled ahead. China leads in green products even as its factories choke on coal. America leads on artificial intelligence while its grid groans under the load. Both seek to solve the wrong problem and call it grand strategy.
The transition will arrive only when we decarbonise manufacturing itself – not the grid, not transport, but the smelters, furnaces, and chemical plants that make everything else possible. Until the chemistry changes, the carbon remains. The coal bunkers of Ürümqi will keep growing, supplying the world’s solar panels – visible from space, if anyone cares to look.
