US fossil-fuel power investment overtakes China's for first time in decades

On 12 July 2026, the Financial Times published a tally that broke a trend line two decades in the making: capital spending on fossil-fuel power generation in the United States has, by the FT's count, eclipsed the comparable figure inside China for the first time since the early 2000s. The story was carried onto trading floors within hours by Unusual Whales, the market-intelligence account that flagged the FT data on X at 15:01 UTC the same day.

The inversion is small in dollar terms but large in symbolism. For a generation of energy analysts, China has been the gravitational centre of global power-sector investment, the country that built coal fleets while the West retired them, and then built solar and wind at a pace the rest of the world combined struggled to match. That the FT is now reporting a US lead, however provisional, in fossil-fuel capex is a marker that the energy transition is no longer a one-way sprint. It is a parallel-track contest, and Washington has chosen to run hard on one of those tracks even as Beijing runs hard on the others.

The shape of the crossover

The FT's framing, as relayed by Unusual Whales, is narrower than the word "overtake" implies. The crossover is in fossil-fuel power investment: the dollars flowing into new gas turbines, coal-plant life extensions, and the supporting transmission and storage that makes a fossil fleet dispatchable. It is not a claim about total energy investment, where China still leads on most credible tallies by a wide margin, nor about clean-energy investment specifically, where Beijing's annual run-rate in solar manufacturing capacity alone dwarfs anything on the US side of the Pacific.

That distinction matters. A reader who skims the headline will hear "China is losing the energy race." A reader who parses the FT's wording will hear something closer to: the United States, after a long stretch of power-sector stagnation, is once again writing large checks for thermal generation, and those checks are now bigger than Beijing's, which has spent the last decade moving its own marginal dollar into renewables, nuclear, and grid build-out. The first reading is a victory lap. The second is a structural reallocation on both sides of the Pacific.

Why the US is moving

Three forces are doing the work in Washington. First, the data-centre boom tied to artificial-intelligence training and inference has pushed hyperscalers and utilities into long-duration firm-power contracts that intermittent renewables cannot fully underwrite. Second, the Inflation Reduction Act's manufacturing tax credits have steered private capital into domestic solar, battery, and EV supply chains, but the grid that connects those factories, and the baseload that stabilises the grid when the sun sets, still leans on gas and, in some restructured markets, on coal units that were once slated for retirement. Third, permitting reform and a friendlier posture toward liquefied natural gas exports have shortened the build cycle for new thermal capacity to a degree that would have seemed implausible five years ago.

The result is a US power sector that is, for the moment, building both sides of the ledger: a clean-energy build-out funded largely by the private sector chasing tax credits, and a fossil build-out funded by a mix of utility balance sheets and industrial demand from AI compute. The two tracks are not, in practice, in direct competition for the same dollar; they are competing for interconnect capacity, transformer supply, and skilled construction labour. On the second constraint, both are losing.

The Chinese counter-read

The Chinese framing of the same data set, as carried in outlets from Caixin to the Global Times, treats the FT headline as a story about the United States catching up on a category China has already deprioritised. Beijing's official energy posture for the second half of this decade, formalised in successive Five-Year Plans, has been to freeze new coal capacity at the utility scale while continuing to permit coal as a chemical-feedstock and grid-stability asset. The marginal yuan of state-directed credit has gone into ultra-high-voltage transmission, pumped hydro, nuclear (including the third-generation fleet and a pilot fourth-generation programme), and solar manufacturing capacity at a scale that has driven global module prices to levels most Western installers treat as the baseline.

Two structural facts underwrite that posture. First, China's per-capita electricity consumption is still rising as the country electrifies transport, heating, and industrial process heat, and the marginal kilowatt-hour of that growth is being met increasingly by non-fossil sources, not by new coal. Second, China's manufacturing share of the global clean-energy supply chain, from polysilicon to battery cells to permanent magnets, is a function of state-directed capacity build-out that began in the late 2000s and is now compounding. Any honest read of the FT data has to acknowledge that China is not slowing down; it is reallocating.

That reallocation has costs the Chinese system is also beginning to acknowledge. The grid integration of record renewable build-outs has strained provincial dispatch and pushed curtailed solar output into double digits in some regions. Coal still does the balancing. The official line from the National Energy Administration continues to be that coal's role is shrinking as a share of the mix; the unofficial line, repeated by industry analysts in Beijing and Hong Kong, is that the absolute tonnage of coal burned will not fall meaningfully until storage and transmission catch up.

What the crossover does not mean

It is worth saying plainly what the FT data, on its own terms, does not establish. It does not establish that the United States is decarbonising more slowly than China; emissions intensity per unit of GDP continues to fall faster in Beijing than in Washington, and the cumulative emissions gap is, in any case, not closeable inside a single decade. It does not establish that US capital is deserting clean energy; private clean-energy investment in the United States has set fresh records in each of the last three years on the back of IRA credits and hyperscaler power-purchase agreements. And it does not establish that China's energy build-out is slowing; if anything, the most recent quarterly data show acceleration in nuclear and grid spending.

What it does establish is that the energy transition, as an investment story, has stopped being a story about one country building and another retiring. It is now a story about two large industrial systems, both of them under political pressure to deliver cheap electrons to data centres, electric vehicles, and electrified industry, both of them building everything they can afford to build, and both of them running into the same physical constraints: transformers, transmission rights-of-way, skilled trades, and the slow physics of balancing intermittent supply against firm demand.

Stakes for the rest of the decade

The practical consequence of the crossover is not who wins the energy race. It is who controls the supply chain for the equipment the rest of the world will buy. If the United States is building fossil-fuel capacity at a faster clip, that capacity will be supplied by US-domiciled gas-turbine manufacturers, US-based EPC contractors, and US-sourced LNG. If China is building renewables and nuclear at a faster clip, that capacity will be supplied by Chinese polysilicon refiners, Chinese battery cell makers, and Chinese reactor builders. Other countries, from the Gulf monarchies to the EU member states to the African Union's emerging power pools, will increasingly be choosing which supply chain to underwrite with their own orders.

That choice is already being made. It is being made in Brasília, where the next round of utility-scale tenders will reveal whether Chinese or Western OEMs take the majority share. It is being made in Riyadh, where the Public Investment Fund is co-investing in both US LNG infrastructure and Chinese solar manufacturing. It is being made in Brussels, where the Net-Zero Industry Act is, in effect, a supply-chain industrial policy with a climate label. The FT's tally on 12 July 2026 is the first widely-cited data point in a debate that will define energy diplomacy for the rest of the decade.

The Monexus desk flagged this piece as a counter-read of the FT's framing: the headline registers a US lead in fossil-fuel capex, but the structural story is reallocation on both sides of the Pacific, and the stakes sit in the supply chain that wins the next round of utility tenders outside the two superpowers.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://x.com/unusual_whales/status/HM7YUhPbMAA0nb9
• https://x.com/polymarket/status/HM7YUhPbMAA0nb9