If you’re trying to figure out whether a nuclear phase-out helps or hurts a climate plan, Germany already ran the experiment. Coal came back before renewables did, and emissions fell anyway once they caught up.

What filled the hole nuclear left behind

Nuclear supplied more than a quarter of German electricity as recently as 2010, when the country’s 17 reactors generated 133 terawatt-hours. After 8 of those 17 reactors went offline in 2011, cutting nuclear output roughly in half, the lost generation didn’t shift to wind turbines. Reporting on the years since Fukushima shows reductions in nuclear generation were offset primarily by increases in coal, not by the renewable buildout that was supposed to be the long-term answer (CNN).

Renewables kept growing through the 2010s, but not fast enough to close the gap outright. By the time the last three reactors closed in April 2023, nuclear was still supplying about 6 percent of German electricity, and coal had climbed to more than 30 percent.

When a reactor goes offline, the grid needs something to answer the gap that same hour. A coal plant already sits on standby, its stacks built and permitted, while a solar field a few miles away still needs two or three years of construction before it can connect to the grid. Wind and solar add capacity gradually. Without enough storage or transmission to move their output where it’s needed, they can’t step in for a retired reactor on short notice.

The 2022 energy crisis pushed the same lever harder. If you heated a German home that winter, you felt it directly. Gas covers almost half of German home heating, and Russia had supplied 55 percent of the country’s gas imports in 2021.

When Russian gas deliveries collapsed after the invasion of Ukraine, Berlin passed the Substitute Power Plants Maintenance Act to bring mothballed coal capacity back into the grid. It moved roughly 2.7 gigawatts of lignite units and several gigawatts of hard coal plants out of cold reserve between mid-2022 and early 2023, to save gas for winter heating. Delaying the nuclear exit and reviving coal plants were two sides of the same emergency response.

Climate scientists don’t agree on whether losing that last 6 percent of nuclear was worth it, or how safe those reactors were to begin with. Hans von Storch, a climate researcher at Germany’s Institute for Coastal Research, called the decision “incomprehensible” from a climate standpoint. Niklas Höhne, who studies climate policy at Wageningen University, took the opposite position. Phasing out nuclear, he argued, gave renewables the certainty to scale “much faster, cheaper and safer” than nuclear power could.

What economists found it cost

An economics team, Stephen Jarvis, Olivier Deschenes and Akshaya Jha, tried to put a number on that substitution. Their study traced what happened after Germany shut down more than half its nuclear capacity in the wake of Fukushima. It circulated first as an NBER working paper in 2019 and was published in 2022 in the Journal of the European Economic Association.

Their finding matches the pattern above. The lost nuclear output was replaced primarily by coal-fired production and net electricity imports.

The published version puts the added social cost of that substitution at €3 billion to €8 billion a year, mostly from the mortality risk that local air pollution from burning coal creates. If you live near one of the plants that picked up nuclear’s old load, that mortality risk is the bill you paid instead of a bigger charge on a power statement. Most of the total has nothing to do with carbon.

An earlier working-paper version of the same research had put the annual toll at about $12 billion, with more than 70 percent of it coming from that same air-pollution mortality channel. The estimates moved between drafts. The conclusion didn’t. The accident and waste-disposal risks that motivated the phase-out were real, and by the authors’ own accounting, still smaller than the health and climate costs of running coal harder in nuclear’s place.

Where German power emissions stand now

The renewable energy buildout that was meant to replace nuclear has largely caught up, years after the fact. Renewables supplied more than half of Germany’s gross electricity consumption for the first time in 2023, the same year power-sector CO2 emissions fell 21 percent to 177 million tons. In 2024, renewables reached 55 percent of consumption, and coal-fired generation dropped another 16 percent year on year.

Germany’s federal environment agency put the energy sector’s 2024 emissions decline at more than 15 million tons, a 7.4 percent year-on-year drop, driven mainly by lower coal and lignite burning. The carbon intensity of German grid power fell alongside it, from 433 grams of CO2 per kilowatt-hour in 2022 to 386 grams in 2023 and 363 grams in 2024. That’s a cleaner electron for anything you plug in, a dishwasher or an EV charger, than the German grid delivered two years earlier.

What 2025 looked like without nuclear at all

The full 2025 numbers show the same trend continuing, with no nuclear anywhere in the mix. Wind became Germany’s largest electricity source for the year at 132 terawatt-hours, and solar generation rose 21 percent to overtake lignite for the first time on record. That’s what solar can do without nuclear left to lean on, and it’s real context if you’re weighing panels for your own roof. Renewables held at 55.9 percent of net public electricity generation, per Fraunhofer ISE.

Coal hasn’t gone away. It still covers the gaps on windless, sunless days, the moment you’d notice it most if you lived on that grid, and it came back briefly during the 2022 gas crisis. Every year since the phase-out finished, though, coal has lost ground to renewables and imports.

How the phase-out happened

Germany’s exit from nuclear power didn’t start with Fukushima. It started nine years earlier, when Chancellor Gerhard Schröder’s SPD-Green government struck what’s known in Germany as the nuclear consensus, a 2000 deal with the country’s utilities to cap reactor lifespans at roughly 32 years and wind the fleet down by around 2022, written into law in 2002 as an amendment to the Atomic Energy Act.

The center-right government elected in 2009 tried to undo it. Angela Merkel’s coalition with the FDP extended reactor operating licenses by eight to fourteen years in 2010, which drew large protests. Roughly 40,000 people marched in Berlin that autumn.

Then came Fukushima. Merkel imposed a three-month moratorium within days of the disaster and took 8 of the country’s 17 operating reactors offline for review. By June, she had reversed course for good, keeping those reactors shut permanently and setting the remaining nine to close by 2022 with support from more than 80 percent of Bundestag members.

The last three plants, Emsland, Isar 2 and Neckarwestheim, were due to close at the end of 2022. Russia’s invasion of Ukraine and the resulting gas squeeze pushed Berlin to delay that by three and a half months. Parliament approved a one-time extension to April 15, 2023, with no further extensions on offer. That’s the date nuclear power in Germany ended, and the last day it showed up on a German electricity bill in any form.

Source: Clean Energy Wire, U.S. Energy Information Administration (EIA)

Fusion and small reactors are the only nuclear left on the table

Katherina Reiche, Germany’s energy minister, said in May 2025 that the phase-out is complete, and that restarting conventional reactors would need investor money and trust the country doesn’t have right now. “I don’t see companies investing on their own initiative,” she said. The coalition agreement she serves under still funds a first nuclear fusion reactor in Germany, and Reiche called small modular reactors “another matter” worth a separate look.

Everything else in her ministry’s actual plan, wind, solar, storage and transmission, is the technology already running the German grid without a single reactor. Conventional nuclear in Germany is closed for good. The fusion reactor and the small modular reactors Reiche mentioned are still lines in a coalition agreement. Neither one has produced a kilowatt-hour yet.

The quake, then the wave

The earthquake hit at 2:46 p.m. local time. Fukushima Daiichi’s three operating reactors, units 1 through 3, shut down automatically the instant the shaking started, exactly what a reactor is built to do. Units 4 through 6 were already offline for maintenance. The quake alone didn’t cause a meltdown.

What came next did.

The first tsunami wave reached the plant about 41 minutes later, at 3:42 p.m.. It crested at 14 to 15 meters, more than double the 6.1-meter seawall TEPCO had built by 2009. That wall was itself a jump from the plant’s original 3.1-meter design from the 1960s.

The water went over the wall, drowning 12 of the plant’s 13 backup generators along with the seawater pumps that cooled them. All six of the plant’s off-site power lines were already dead from the earthquake itself, so units 1 through 3 lost every source of power for their cooling systems at once. No grid power and no working generators is called a station blackout, the condition that turned an earthquake and a flood into a nuclear accident. It’s the failure you might have during a bad storm at home, except here the backup pumps were what stood between the reactor cores and overheating.

Why a shut-down reactor still needs cooling

A reactor that’s shut down isn’t a reactor that’s cold. The moment the control rods drop and the chain reaction stops, the fuel still holds radioactive fission byproducts such as cesium and iodine. They keep decaying on their own, and that decay throws off heat. Decay heat runs about 7% of full power at shutdown, falling to roughly 2% within the first hour and about 1% after a day, according to DOE reactor operator training material.

You don’t need reactor physics to feel why one percent still isn’t nothing, since a pot pulled off a hot burner stays too hot to touch long after the flame’s out. Unit 1 was rated at 460 megawatts of electrical output, so even a sliver of its full power is real heat with nowhere to go unless water keeps moving through the core. Cut the pumps, and the water already in the vessel boils away on its own. Once the fuel rods sit exposed to steam instead of water, the zirconium cladding around them reacts with that steam, throws off hydrogen gas, and the fuel starts to melt.

How three reactors melted down

With no power for the pumps, each of the plant’s three running reactors melted on its own timeline, as the hours below show.

Source: World Nuclear Association

Plant operators tried venting steam and pumping in seawater, improvised responses to a situation the plant wasn’t built for. It’s the kind of call you’d dread making at work with half your instruments dead, except here the stakes were three reactor cores. Neither stopped the melting, and the fuel in all three reactors slumped, in places burning through the steel of the vessels themselves.

The hydrogen building up inside the damaged reactors found an ignition source in three separate buildings. Unit 1’s reactor building exploded on the afternoon of March 12, blowing the roof and upper walls apart, and Unit 3’s building went the same way on the morning of March 14. Unit 4 had no fuel loaded in its own reactor, but it exploded anyway, after hydrogen vented from Unit 3 backed up through shared ducts and ignited early on March 15.

The blasts scattered debris and made it much harder for workers to get close enough to restore cooling.

Evacuating a quarter million people, and the deaths that followed

Japan widened its evacuation zone in stages as the accident worsened, starting with a 2-kilometer radius on the evening of March 11. By the next evening, the radius had grown to 3, then 10, then 20 kilometers.

At its peak, in May 2012, about 160,000 people had left their homes, roughly 100,000 of them still inside Fukushima Prefecture and the other 60,000 scattered elsewhere in Japan. As of May 1, 2026, the prefecture still counted 22,625 people as evacuees. If you had to leave your own town and still couldn’t move back nearly fifteen years later, that’s what the number means, since many of those evacuees come from towns where the return order has never fully lifted.

The evacuation itself had a body count. A peer-reviewed study covering March 2011 through July 2021 found 2,326 certified disaster-related deaths in Fukushima Prefecture, deaths from the physical and medical strain of the evacuation itself. The study separates these from deaths the earthquake and tsunami caused directly, and from any radiation-linked deaths. Elderly hospital and nursing-home patients were hit hardest, moved long distances under chaotic conditions with medical care interrupted.

That toll, 2,326 disaster-related deaths, was well above the 1,610 people Fukushima Prefecture lost directly to the earthquake and tsunami.

What radiation did to people’s health

The United Nations Scientific Committee on the Effects of Atomic Radiation, known as UNSCEAR, studied Fukushima’s radiation exposures for a decade, publishing its fullest assessment in 2020 and 2021. It found no adverse health effects among residents that could be attributed to radiation from the accident. It also expects future radiation-linked health effects to stay undetectable because the doses were so low.

The thyroid cancers found in screened children weren’t caused by radiation either. UNSCEAR attributes them to an unusually sensitive ultrasound program that caught thyroid abnormalities doctors wouldn’t otherwise have found at that age.

Emergency workers got a heavier dose than the public, as you’d expect from people running hoses inside a damaged plant. They averaged about 13 millisieverts between March 2011 and March 2012, and 36% of them received more than 10 millisieverts. UNSCEAR still didn’t attribute any deaths or any likely detectable rise in cancer to that exposure. That finding runs against the popular fear surrounding the disaster, worth keeping in mind whenever people debate whether nuclear power is safe.

Treated water out, fuel debris still stuck inside

Cooling the melted cores never stopped, and the water used has to go somewhere once it’s contaminated. TEPCO built the Advanced Liquid Processing System, ALPS, to filter out 62 of the water’s radioactive elements before releasing it through an offshore pipe starting August 24, 2023, keeping everything but tritium, which is chemically bound into the water and can’t be filtered out. If you’ve heard concerns about seafood near Fukushima, this is the process built to answer them, stripping out everything the filters can catch before any of it reaches the ocean.

By the end of 2025, TEPCO had discharged roughly 133,000 cubic meters across 17 batches, each diluted below regulatory tritium limits before release. The IAEA’s most recent review found the discharge meeting international safety standards in its September 2025 report.

The parts of the plant that could be defueled the normal way mostly have been. 1,331 spent fuel assemblies came out of Unit 4’s storage pool by December 2014, and Unit 3’s pool was emptied by February 2021. The harder problem is the roughly 880 tonnes of melted fuel debris fused with structural metal and concrete inside units 1 through 3, in ways nobody has fully mapped yet.

Progress there is slow on purpose. Workers pulled the first trial sample of fuel debris from Unit 2, about 0.7 grams, in November 2024, gripped by a tool built to reach through a small opening in the damaged container. TEPCO has since concluded that preparing for full-scale removal alone needs 12 to 15 years, pushing the start date to 2037 at the earliest, well past the early-2030s target officials once set.

A 2017 government estimate put the total decommissioning cost near 22 trillion yen, about $190 billion at the time, double what officials had projected a few years earlier.

What the nuclear industry changed afterward

Fukushima happened because a plant lost every source of power at once with no way to cool its cores. Regulators everywhere rewrote their rules around that failure mode. In the United States, the NRC ordered reactors to keep portable pumps and generators on hand, along with battery packs and chargers, starting in March 2012, all capable of cooling the reactor and its spent fuel pool through an event beyond the plant’s original design.

The industry built this out under the name FLEX, backed by two national response centers in Memphis and Phoenix that can ship heavier equipment to any reactor site within 24 hours. If a reactor near you shares Fukushima’s General Electric containment design, it already has the hardened vents and spent fuel pool instrumentation the NRC ordered for the 31 U.S. reactors built that way. The industry finished installing all of it at every U.S. plant, at a combined cost topping $4 billion, according to the Nuclear Energy Institute.

Spending that much money is what it costs to guard against an accident nobody predicted, not the one the plants were designed for. Not every country chose to keep paying it. Germany shut its nuclear fleet down entirely rather than retrofit it, a decision that still shapes the nuclear power outlook for the reactors still running today.

Nuclear power supplied 19% of the electricity generated in the US in 2024, and American plants are doing something the industry has never done before. Reactors that fully shut down are switching back on instead of staying closed for good. If you’re on the US grid, roughly one in five of the electrons reaching your outlet already came from one of those 94 reactors at 54 plants, which together produce about 97 gigawatts.

Worldwide, reactor counts are flat too, but roughly half of the new construction is happening in China. Nuclear’s share of the world’s electricity has slid to 9% in 2025, down from a peak of 17.5% in 1996, per the World Nuclear Industry Status Report.

How many US reactors are operating

That fleet isn’t the whole clean-power picture where you live. Nuclear and renewables together supplied about 40% of US electricity generation in the first ten months of 2023, with nuclear responsible for 18 of those percentage points and renewable energy sources the other 22. If you pay a power bill anywhere in the US, roughly two of every five kilowatt-hours behind it already came from a source that doesn’t burn fuel.

Worldwide, the fleet is bigger and holding roughly steady too. The World Nuclear Association counts about 440 reactors running in 31 countries. Combined, those reactors produce 400 to 404 gigawatts of capacity.

Retirements happen even while that total holds steady. In 2025, the IEA counted 3 gigawatts of nuclear capacity retired worldwide, two-thirds of it in Belgium, offset by 3 gigawatts of new capacity in China, India and Russia. What’s shrinking is nuclear’s slice of a power market that keeps getting bigger.

American plants are coming back online

Palisades, an 805-megawatt plant in Michigan, shut down in May 2022 and began reloading fuel in late August 2026, aiming to become one of the first US commercial reactors to fully restart after closing for good. Then it hit a real snag.

The NRC confirmed on September 23, 2026 that a fuel assembly had tipped during loading, with no radiological release, and the plant’s final restart date is on hold while regulators sort out next steps.

Pennsylvania’s Three Mile Island Unit 1, rebranded the Crane Clean Energy Center, is targeting a 2027 commercial restart at 819 megawatts. Iowa’s Duane Arnold, at 601 megawatts, is aiming for an early 2029 return.

California’s Diablo Canyon, once scheduled to close around 2025, got a 20-year license renewal from the NRC in April 2026 and remains the state’s only operating nuclear plant, so every nuclear watt reaching a California home or business still comes from that one site. For how nuclear’s safety compares with other power sources, see how safe nuclear power is.

AI’s power demand is paying for the restarts

The restarts aren’t happening out of nostalgia. Microsoft signed a 20-year power purchase agreement worth roughly $16 billion, announced in September 2024, to buy electricity from the Crane Clean Energy Center once it’s running. The US Department of Energy backed the project with a $1 billion loan in November 2025.

Other companies are signing similar deals. By July 2026, tech companies had committed to roughly 9.8 gigawatts across 13 separate deals for nuclear power.

Meta’s agreements add up to as much as 5.2 gigawatts, including 4 gigawatts of TerraPower’s Natrium reactor for delivery between 2032 and 2035, plus 1.2 gigawatts from Oklo’s Aurora reactor.

Amazon has committed to up to 960 megawatts through a dozen X-energy Xe-100 reactors, backed by a $700 million investment. Google struck its own deal with Kairos Power for a 500-megawatt fleet of small reactors aimed at 2030.

None of this is free for everyone else on the grid. PJM, the grid operator that covers Washington DC and states from Illinois to New Jersey, expects data center demand to push some customers’ bills up by as much as 5% in the year starting June 2026.

China is building about half of the world’s new reactors

China’s fleet stands at 64 reactors totaling 63,985 megawatts, and the country isn’t slowing down. Another 38 reactors are under construction there, adding up to 39,694 megawatts, which is close to half of the roughly 80 reactors the World Nuclear Association counts under construction worldwide. In March 2026, China’s 15th Five-Year Plan set a target of 110 gigawatts of nuclear capacity by 2030.

Outside China, growth is thin. Of the 10 reactor construction starts worldwide in 2025, nine were in China and one was in Russia, according to the International Energy Agency. If you were hoping one of these projects would break ground near you and bring construction jobs with it, the odds outside those two countries were close to zero.

Small reactors are closer to real, years behind schedule

NuScale was supposed to prove small modular reactors could work in the US, and it didn’t. Its VOYGR design became the first SMR certified by the NRC in January 2023. Its flagship Utah project collapsed in November 2023, after costs rose from $5.3 billion to $9.2 billion. That overrun pushed the target price from $58 to $89 per megawatt-hour, the rate the towns and utilities that had signed on would have paid for every megawatt-hour NuScale delivered.

For more on how nuclear’s costs stack up against other power sources, see how solar and nuclear costs compare. NuScale’s lead project shifted to Romania, a 462-megawatt plant with six modules that reached a final investment decision, though construction hadn’t started there as of September 2026.

Back in the US, the Tennessee Valley Authority’s BWRX-300 reactor at Clinch River got an NRC staff recommendation for a construction permit in June 2026, with a mandatory hearing held August 13, 2026, and commissioners’ final decision still pending.

Kairos Power broke ground in April 2026 on its Hermes 2 demonstration reactor in Oak Ridge, Tennessee, its first design meant to generate power, separate from the non-power Hermes 1 test unit it’s been building since 2024. And TerraPower’s Natrium reactor in Kemmerer, Wyoming, became the first full-scale advanced reactor design to win an NRC construction permit in March 2026, a 345-megawatt design that stores heat in molten salt.

Nuclear’s slice of the world’s power keeps shrinking

Nuclear’s share of global electricity fell to 9% in 2025, down from a peak of 17.5% in 1996, according to the World Nuclear Industry Status Report, a group that’s generally critical of the nuclear industry but whose numbers get cited widely anyway. World electricity demand has grown faster than new reactors have, even as a decade of debate over climate goals argued nuclear needs to be part of the answer.

For anyone whose power already leans on nuclear, that slide doesn’t cut watts today. Grid planners will have to find the next decade’s clean-power growth from sources other than nuclear instead, and both the US restart wave and China’s buildout are still too small to close that gap by themselves.

The restart wave hasn’t reversed nuclear’s shrinking share yet

A handful of decisions still pending this decade will say whether that changes. The NRC’s commissioners still have to rule on the Tennessee Valley Authority’s Clinch River permit after the August 13, 2026 hearing, a ruling that will decide whether TVA’s BWRX-300 becomes the second advanced reactor design under construction in the US alongside TerraPower’s Natrium.

Palisades’ final restart timeline is on hold while regulators finish reviewing the fuel-assembly incident, and Duane Arnold is aiming for an early 2029 return. China, meanwhile, still has to close most of the distance to its 110-gigawatt target for 2030 from a base of roughly 64 gigawatts operating and 40 gigawatts under construction.

A federal regulator, a Michigan repair crew, an Iowa utility’s schedule and Chinese state planners each hold one piece of whether the 2030s look different from the 2020s for anyone counting on nuclear power to help run their grid.

Nuclear waste still has to go somewhere. A bad accident costs far more than the death count alone shows.

What deaths per terawatt-hour measures

One terawatt-hour is roughly what a mid-sized country’s grid delivers in a few hours, or what a single large power plant produces running for a year or two. The comparison below combines a 2007 life-cycle study of electricity and health published in The Lancet with a database of 279 major energy accidents.

By that tally, coal kills about 24.6 people per terawatt-hour and oil about 18.4, almost entirely from the sulfur dioxide and fine particulates that combustion pushes into the air year after year. Gas comes in at 2.8, and nuclear is lower still. If your electricity comes from a coal-heavy grid, that pollution does not stay at the plant. It moves into the same air your town breathes, which is what a death rate like this is counting.

Source: Our World in Data, “What are the safest and cleanest sources of energy?”, combining Markandya and Wilkinson (2007) with a database of historical energy accidents.

Hydropower’s low average hides a single outlier. Nearly all of that death toll traces to one event, the Banqiao Dam failure in China in 1975. The same analysis puts the toll at an estimated 171,000 people.

Estimates for that single event vary widely. Chinese authorities have cited 26,000 confirmed dead, versus a Human Rights Watch estimate of 230,000 once deaths from the disease and famine that followed the flood get counted in.

Chernobyl’s death toll

The Chernobyl reactor exploded in April 1986, in what is now Ukraine. By 2005, fewer than 50 deaths had been directly attributed to radiation exposure, almost all of them rescue workers who took the highest doses in the accident’s first hours and months, according to the World Health Organization.

The WHO’s 2005 assessment also projected up to 4,000 eventual deaths from radiation-induced cancer among the roughly 600,000 people most exposed. The agency put that at an increase of about 3% over the cancer rate that group would see anyway, too small a rise to observe directly against the background rate. If you had been among that group, the added risk would have been small enough to hide inside your normal odds, invisible on any chart you would see.

Radioactive iodine released by the fire also drove up thyroid cancer among people who were children in the contaminated region at the time. By 2015, close to 20,000 thyroid cancer cases had been registered in the affected areas, and the United Nations’ scientific committee on radiation attributed about one in four of them to the accident.

The reactor involved was an RBMK design that lacked the containment structure standard on Western reactors and had a control-rod flaw that, on an emergency shutdown, briefly added reactivity to the bottom of the core instead of cutting it. The World Nuclear Association treats that flaw as central to why the accident escalated the way it did.

Fukushima’s death toll

The Fukushima Daiichi meltdown, in March 2011, didn’t kill anyone through direct radiation exposure. One plant worker died in 2018 of lung cancer that Japan’s government attributed to his radiation exposure on the job, the only radiation-linked death the accident has produced.

More than 100,000 people were moved out of the area as a precaution, on top of about 19,500 people killed by the earthquake and tsunami itself. Japan has certified 2,313 deaths as disaster-related among Fukushima evacuees, most from disrupted hospital care and the strain of relocating elderly and hospitalized patients during the evacuation. If a meltdown forced an evacuation where you live, the disruption of packing up your home on short notice would likely affect you more than radiation itself.

About 27,000 people from Fukushima were still living as evacuees as of November 2025, 14 years on.

Nuclear waste, in volumes

A reactor generating 1,000 megawatts produces about 3 cubic meters of high-level waste a year when its fuel is reprocessed, roughly the size of a shed you might keep in your backyard, against roughly 300,000 tonnes of ash from a coal plant generating a similar amount of power.

High-level waste and used fuel together make up less than 3% of nuclear waste by volume but hold 95% of its radioactivity, the association says. Intermediate-level waste adds another 7%, and the remaining 90% is only lightly contaminated tools and clothing.

About 97% of used fuel can be recycled into new fuel, most of it reusable uranium. What can’t be recycled is vitrified, mixed into glass and sealed inside metal canisters, then buried in deep geological repositories, an approach France, Japan, Belgium and Russia already use for permanent disposal.

Radiation risk fades with time

Radioactivity isn’t a fixed hazard. It decays on a known schedule. Used fuel’s radioactivity falls to about a thousandth of its initial level after 40 years, the World Nuclear Association says.

The same group states that nuclear waste has never caused harm to people since the civil industry began, a claim that holds up against decades of operating records. That’s a claim about the industry’s history, though, not a guarantee about storage arrangements holding for the thousands of years some of this material stays hazardous.

Four decades is close to a full career. If nuclear expansion means new jobs in your county, guarding and monitoring that waste is the part of the job that outlasts whoever signs up for it.

What the death rate doesn’t put a price on

Deaths per terawatt-hour answers one question well. Historically, which sources of electricity have killed the fewest people relative to how much power they’ve produced? On that question, nuclear does about as well as wind and solar, and far better than any fossil fuel.

It answers a narrower question than people often assume. It’s a global historical average built mostly from twentieth-century reactors and two accidents. It doesn’t forecast what a reactor built under current safety regulation would do. It also treats a death from evacuation stress the same as a death from radiation exposure, which is fair for counting purposes but blurs how differently those two causes of death happen.

The number says nothing about what a bad accident costs. Estimates from the 1990s put Chernobyl’s cleanup and economic costs at hundreds of billions of dollars over two decades, the World Health Organization says. Fukushima evacuees alone had received about 9.7 trillion yen in compensation as of January 2021. None of that shows up on your electricity bill today, but it is the bill someone eventually pays when a reactor fails, and it runs far higher for nuclear than for the low-mortality alternatives it gets grouped with.

None of that changes the mortality data. Being statistically safe and having no downsides worth discussing are two different claims, and the costs above show which one survives. In 2024, nuclear supplied about 9% of the world’s electricity, or 2,667 terawatt-hours.

Germany read those same safety numbers and still phased out nuclear power entirely, leaning on coal to fill the gap, a decision about cost and politics more than accident risk. China read them the other way. 28 new reactors were under construction there as of mid-2025, the fastest pace of any country. On the death-rate numbers alone, nuclear is about as safe as wind and solar, and far safer than any fuel that burns.