#72 Why is China Building Nearly Half the World's Reactors?

China cleared 8 More Reactors in a Single Sitting this Month and the Way it Builds Them Cheaply and at Speed is the Part India should be Studying 

Jayhant Mundhra 

 

 

On 1 August, China's State Council sat down and approved 8 new nuclear reactors in one go, a build worth roughly $24bn.¹ Not a plan, not a consultation, not a "study". Eight actual reactors, greenlit in a single meeting, and the story barely made a ripple outside the trade press.

That is the part that stopped me. Not the number, the ease. China now approves reactors the way most countries approve highway stretches, and it has been doing it for years.

Here is the fact I keep coming back to: nearly half of all the nuclear reactors under construction anywhere on Earth right now are Chinese.² Not a quarter. Nearly half. One country.

And India, which needs baseload power more desperately than almost anyone, has 8.78 GW of nuclear capacity across 24 reactors.³ China has 62.5 GW across 63.⁴ Let that gap sit for a second, because the rest of this edition is about how it got that wide, and what we can actually do about it.

The cadence is the story

The scale is stunning, but the cadence is what tells you this is deliberate.

Look at the approvals China has handed out, year by year:

• 2019: 6 reactors

• 2020: 4 reactors

• 2021: 5 reactors

• 2022 to 2025: 10 reactors, every single year⁵

• 2026 so far: 8 more, cleared on 1 August¹

That is not a country reacting to events. That is a country running a production line and setting the annual output in advance. Six identical reactors get slotted onto one coastal site, then the next six on the next site.

The result:

• 63 reactors operating today, about 62.5 GW.⁴

• 38 more under construction, another ~40 GW.⁴

• 118 GW of prospective capacity across announced, pre-construction and building phases, more than the next seven countries put together.²

Sometime in mid-2026, China's operating nuclear fleet quietly slipped past France's to become the world's second largest, behind only the United States.⁴ The 15th Five-Year Plan, approved in March, sets a target of 110 GW of operating capacity by 2030.⁵ On current pace China will pass the US before the decade is out and become the largest nuclear power generator on the planet.

This licence-and-repeat rhythm is exactly the kind of quiet machine I find myself pulling apart most mornings, and it is what we go back and forth on in the Decoding the Dragon WhatsApp group, where I share a smaller China deep-dive like this with thousands of readers every day (t.ly/t7uhs).

 

 

The thing the West cannot copy: cost

Now here is where it gets genuinely uncomfortable for everyone else.

The West has not forgotten how to build reactors. It has forgotten how to build them at a price anyone can afford. The two cautionary tales everyone cites are Plant Vogtle in Georgia and Flamanville in France, both years late and multiples over budget.

The numbers are brutal:

• Vogtle's two AP1000 reactors landed at roughly $8,000/kW on an overnight basis, and closer to $16,000/kW once you count financing and delay.⁶

• China's 8 new units were cleared for about $24bn, roughly $3bn a reactor, which works out near $2,600/kW.¹

Read that again. China is building the same class of large reactor for something like a third of the American overnight cost, and a fraction of the all-in cost. This is the single most important number in the whole story, more important than any target, because cost is what decides whether nuclear scales or stalls.

How does China do it? Four interlocking things, and none of them are magic:

• Localised supply chain. For the newer designs, over 90% of components are made in China.⁴ No waiting on a foreign forging that arrives 18 months late.

• Serial construction. Co-locating up to six identical reactors on one site shares infrastructure and cascades labour savings, cutting cost per kW meaningfully versus a one-off build.²

• A workforce that never stops. Because the pipeline never breaks, the same welders, foremen and project managers walk from a finished reactor straight to the next foundation. Western projects lose this muscle memory between builds, and pay for it.

• New money in the mix. The Jinqimen plant in Zhejiang is now taking private capital alongside state firms, a first for a sector that was wholly state-owned.⁷

The West calls the opposite of this the "nuclear cost curse", where idle specialists and missing parts inflate every project. China simply never let the curse take hold, because it never stopped building.

 

 

From copycat to frontier

For years the lazy take on Chinese nuclear was that it was borrowed technology. That take is now wrong.

The workhorse is the Hualong One, a home-grown 1,100 MW reactor built by CNNC and CGN, and on 1 August the State Council cleared its upgraded Version 2.0 for the Jinqimen and Taipingling sites.⁷ It is the reactor China now sells to the world.

But the more interesting action is at the frontier, where China has quietly moved ahead:

• Gen IV is already commercial. The HTR-PM pebble-bed reactor at Shidaowan began commercial operation in December 2023, the world's first fourth-generation plant. It can cool itself down with no human intervention, the holy grail of "walk-away" safety.⁸

• Small modular reactors. The Linglong One is the first SMR anywhere to clear an IAEA safety review, aimed at heating, desalination and remote grids.⁴

• Thorium and salt. An experimental thorium molten-salt reactor is running in the Gobi, a design that needs almost no water and suits China's dry interior.⁴

• Fusion. China's EAST tokamak held a plasma for a record 1,066 seconds in January 2025, and a successor machine, BEST, is being built to push towards actual electricity.⁹

You do not have to believe every timeline to see the direction. China is no longer importing the future of nuclear. It is increasingly writing it.

 

 

 The export machine, and its dark twin

All of this feeds a foreign-policy engine.

The Hualong One is the designated export product, precisely because it is free of Western intellectual property and therefore free of Western export controls.⁷ China has floated an ambition to build as many as 30 reactors in Belt and Road countries by 2030, and it already has Hualong units running in Pakistan at Karachi.² When you can offer a country a turnkey reactor at a third of the Western price, with the financing attached, you are not selling electricity. You are selling a 60-year relationship.

Then there is the part that keeps Western security officials awake.

• China is running the CFR-600, a fast breeder reactor on Changbiao Island, with Russian help on the fuel.¹⁰

• Fast breeders produce very pure, weapons-usable plutonium as a by-product.¹⁰

• The Pentagon reckons China already holds more than 600 nuclear warheads and could reach around 1,500 by 2035.¹¹

• And since 2017, Beijing has stopped filing its voluntary civilian plutonium declarations to the IAEA, so the world is largely guessing.¹⁰

   

  
  
   

  This is the uncomfortable duality of the whole programme. The same closed fuel cycle that makes China's civilian nuclear so efficient also blurs the line between the power plant and the arsenal. It is genuinely hard to build the first without creating the option of the second, and China has chosen opacity over reassurance.

  The wider machine

  None of this is really about electricity, or not only.

  Nuclear sits at the intersection of three things Beijing cares about most: energy security in a country that still burns a mountain of imported coal and gas, technological self-reliance in an era of export controls, and industrial policy that keeps thousands of state and private firms fed with predictable order books. The Hualong programme alone is said to touch nearly 6,000 supplier companies.⁴

  Approving 10 reactors a year is not an energy decision. It is a way of keeping an entire high-end manufacturing base warm, the same logic China applied to high-speed rail, then solar, then EVs, then batteries. Nuclear is simply the latest sector to get the treatment.¹³

  The West has noticed. The United States passed the ADVANCE Act in July 2024 to speed up its own reactor licensing and cut regulatory cost.¹² It is a sensible law. Whether a lighter regulator alone can close a five-times cost gap against a state that has been building without pause for 20 years is a very open question.

  What this means for India

  Here is where I want to be honest, in both directions.

  The gap is real and it is large. China has 62.5 GW of nuclear and is heading for 110 GW by 2030. India has 8.78 GW and is aiming for 100 GW by 2047.^{3 5} Read those two timelines together and the uncomfortable truth is that India's 2047 goal is roughly China's 2030 goal, a 17-year lag on the destination, before you even count who is more likely to hit it.

  I share these India cuts, alongside the daily China ones, in the same Decoding the Dragon community, and this is exactly the sort of comparison we chew on there (t.ly/t7uhs).

   

  
  
   

  But the lesson from China is not "spend more". It is "build repeatedly". The reason Chinese reactors are cheap is boring and completely copyable: standard design, same site cluster, same crew, no gaps. India has historically done the opposite, building in ones and twos with long pauses, which is precisely the recipe for the Western cost curse.

  The encouraging part is that India has just started moving in the right direction:

• A roadmap to 100 GW by 2047, with indigenous small modular reactors (the Bharat Small Reactor) as a centrepiece.¹⁴

• A push to open the sector to private capital and amend the old liability and atomic-energy laws that scared investors off.¹⁴

• Around 7 to 8 GW already under construction, including the Russian-built Kudankulam units.³

   

  
  
   

  The negatives India has to stare at honestly:

• India cannot simply buy China's cheap reactors. The dual-use, opacity and border realities make that a non-starter, so India has to build the cost curve itself.

• India's uranium position is weak, and China has spent a decade locking up mines from Namibia to Kazakhstan.² That is the supply-chain lesson underneath the reactor lesson.

• Pace is a mindset. China's advantage is not a secret technology. It is the discipline of never stopping, and that is the hardest thing of all to import.

What to watch

Concrete markers over the coming months:

• The first concrete pour on the Hualong One 2.0 units at Jinqimen and Taipingling, the real test of whether Version 2.0 cuts build time as promised.⁷

• India's SMR tenders and the promised amendments to the Atomic Energy Act and the nuclear liability law. If private money actually flows, the 100 GW target stops being a slide.¹⁴

• The second CFR-600 unit reaching full operation, and whether the IAEA gets any new transparency in return.¹⁰

• Whether the US, under the ADVANCE Act, can bring a single follow-on AP1000 in near MIT's hoped-for $4,300/kW. That number is the West's referendum on itself.⁶

The one line

China's real nuclear breakthrough is not a reactor, it is a production line, and the country that learns to copy the boring discipline of building the same thing over and over will own the next century of clean baseload.

And well that is it for today's edition. That said, do check out my core WhatsApp community Biz News+ where I share 4-5 deepdives from the world of business, economics & public economics daily: https://t.ly/h2jq1