I composed this essay over much of 2026, as I was researching Beyond Vogtle — my new study (co-authored with James Boucher) of U.S. nuclear costs from the 1980s post-Three Mile Island reactors to the Vogtle 3 and 4 units completed earlier in this decade. It traces my journey from nuclear power doubter to proponent.
The story is both general and personal: what was changing in the wide world of nuclear power, energy policy and climate; and how I regarded those changes. I hope this mode of story-telling resonates with a wide audience and also provide a backdrop to “Beyond Vogtle.”
I’ve divided it into three parts on account of its length.

Detail from 1979 rally poster. Full poster appears below.
This installment begins at the big Washington rally called to mark the Three Mile Island reactor accident, and covers the subsequent withering of nuclear dread, the advent of virtually always-on reactor operation, and nuclear power’s stature as the least carbon-emitting energy supply technology.
Installment #2 begins with the failure (to date) of carbon tax advocacy — “A Climate Cure No One Wanted” — and continues with the closure of Indian Point and the concurrent slow dissolve of my dream that renewable energy could do it all.
The third and final installment begins with a slight detour in which I contrast the appalling human damage from automobility with nuclear power’s increasingly remote dangers, and then return to energy policy with a fresh way to regard nuclear power’s potential contribution to decarbonizing U.S. grids.
Thank for reading. Also, be sure to download and read my new, richly detailed report, Beyond Vogtle: What History Tells Us About the Cost of New Nuclear, co-authored with James Boucher. It’s a fresh, empirically based way of looking at the cost to build nuclear plants in the U.S. Journalist Alexander Kaufman’s reviewed it earlier today for Heatmap in a scintillating story, Nuclear Is Cheaper Than We Think It is.
— C.K.
* * * * * * * * *
I’m pro-nuclear power. Big time.
I don’t just want the U.S. to keep running its 95 extant reactors. I want us to build more. Lots more. Hundreds.
I’m not alone. Public opinion is shifting toward nuclear power. But I came to that conclusion after spending years arguing the other side. From the mid-1970s to the late 1980s, I published a raft of critical analyses that got a good deal of public attention ― and even some traction within the nuclear industry itself. My core argument was that the cost of building nuclear plants was rising much faster than the costs of competing sources of energy. The main driver, I said, was a cascade of safety-requirement changes triggered by widespread fears of reactor accidents.

Here’s the full poster.
That work earned me a speaking spot at the massive May 6, 1979 rally in Washington demanding a halt to reactor construction in the wake of the March 28 meltdown at Three Mile Island in Pennsylvania. In the weeks before the rally, my research was cited in hundreds of news stories. At the Capitol, I stood alongside consumer advocate Ralph Nader, the leading figure in the U.S. antinuke movement; actress Jane Fonda, whose new film, “The China Syndrome,” had eerily foretold the Three Mile Island meltdown; and folk-rock icon Jackson Browne. I told the cheering throng that nuclear power was finished.
Cost overruns and canceled reactor projects were putting billions of dollars on the line, and I spent much of the next decade in courtrooms all over the country, explaining patiently to utility regulators why investors, not customers, should bear those losses. I eventually moved on to other public policy work, sparking big-city bicycling and helping bring congestion pricing to New York. But I kept watching nuclear power and the broader energy landscape. Over time, what I saw led me to turn from skeptic to supporter. Here’s why, in eight parts.
1. Fear and Dread Recede
Ask 100 random people today what “TMI” means, and at least 90 will say “Too Much Information.” Maybe one will mention the 1979 accident at Three Mile Island.
That’s a real shift.
The 1970s were commercial nuclear power’s first decade, and the Three Mile Island accident brought it to a terrifying close. Sixty hours into the slow-unfolding crisis, as fears grew that a “hydrogen bubble” in the reactor might explode and rupture the containment dome, CBS Evening News anchor Walter Cronkite captured the nation’s dread: “The world has never known a day quite like today. It faced the considerable uncertainties and dangers of the worst nuclear power plant accident of the atomic age. And the horror tonight is that it could get much worse.”[1]
The stricken reactor was eventually stabilized. But a combination of soaring costs and public apprehension brought an end to nuclear power’s rapid expansion in the United States.
Cronkite’s warning was about Three Mile Island itself, but it seemed to foreshadow more disasters to come. Yet since 1979, the U.S. nuclear industry has accumulated nearly 20 times as much reactor operating experience as it had built up in its entire history before the meltdown, without a sequel.[2] Like the dog that didn’t bark, this quiet fadeout of reactor calamities is an overlooked shift ― and it is prompting a rethink of long-held certitudes about nuclear power.
With each passing year there are fewer people who grew up with dread from Cold War-era A-bomb drills and weapons test fallout that got bundled into nuclear power. There are more young people like Zeke, a Brooklyn high-schooler who in 2019 went to Battery Park City to greet climate activist Greta Thunberg after her sail-powered trans-Atlantic voyage, but who this past May called fears about nuclear power “over-exaggerated.”[3] (Thunberg herself has said that Germany made a mistake by phasing out its nuclear plants, since it led to a sharp rise in coal-fired electricity.)
I’ve heard the same sentiment in hundreds of casual conversations over the past decade ― at climate rallies and on Trader Joe’s checkout lines, in California and in New York. These conversations suggest that the oppositional currents that once compelled federal nuclear regulators to keep piling on costly new safety requirements are losing force.
2. Permanent Peak Performance
Even before construction costs began to soar in the 1970s, nuclear power had another Achilles heel: spotty operating performance.
Throughout the 1970s and 1980s, the U.S. nuclear power sector struggled to maintain even a 60 percent “capacity factor” ― a measure of how much of a plant’s potential output it generates. That’s a dismal rate for equipment that’s expensive to build. I know this because I researched and wrote the first full-length study of shortfalls in U.S. nuclear plant performance, in 1976.[4]
That started to change in the mid-1980s. Plants finished safety upgrades required after Three Mile Island. The industry began sharing best practices ― and mistakes to avoid. Economic incentives helped too, as utility earnings became tied to how often plants actually ran.

A remarkable turnaround, though seldom credited in climate and nuclear discourse.
The turnaround has been dramatic. Since 2000, U.S. nuclear plants have averaged 90 percent capacity factor ― a huge leap from the earlier 60 percent. In effect, downtime has dropped four-fold, from 40 percent of the time to just 10 percent. Repair jobs and retrofits that used to drag on are now precision-scheduled like the train heist in “Breaking Bad.”
The higher reliability brings a huge symbolic benefit. In the 1970s, U.S. reactors seemed to stumble from one fiasco to the next. In Alabama, a technician using a lit candle to locate an air leak started a fire that burned through a thousand cables and knocked two brand-new reactors offline for 19 months. At some ocean-cooled plants, saltwater corroded delicate heat-transfer tubes, forcing protracted repairs. Profits and industry morale took a beating, and nuclear power became a punchline on “The Simpsons.”
Those days are long past. Nuclear power, uniquely, has blossomed into both grid bulwark and climate hero. Thanks to those higher ― much higher! ― capacity factors, each nuclear plant now displaces 50 percent more carbon-emitting power generation than it used to.[5] In fact, at a 90 percent capacity factor, a kilowatt of nuclear power delivers double or triple the climate benefit of a kilowatt of wind power (which averages 30 to 40 percent capacity factor) and roughly five times that of solar (15 to 20 percent) ― a crucial distinction that’s often missing from gushing coverage of renewable energy.
3. Climate to the Fore
Nuclear power’s newfound operational mastery would matter much less but for the urgency of the climate crisis and the persistence of U.S. and global carbon emissions. Like wind and solar, nuclear power generates electricity without burning carbon.

Squint to see that nuclear lifecycle greenhouse gas emissions range from 5.1 to 6.4 (in g CO2 equivalent per kWh); analogous range for solar-PV is 7.4 – 83.0; wind, 7.8 – 23.0. Source, Dinon et al., in report linked in this section’s second paragraph.
It is true that fossil fuels are implicated in nuclear power’s supply chain. Uranium mining uses petroleum, and enriching nuclear fuel requires electricity. But even counting upstream carbon, nuclear power’s climate footprint is smaller than that of wind, solar or hydropower, according to an authoritative 2022 analysis by a multinational team for the UN Economic Commission for Europe. (Their finding matched that of the similarly comprehensive 2018 report by the UN-chartered Intergovernmental Panel on Climate Change.[6])
That’s another big change. In nuclear power’s early years, uranium fuel enrichment was so energy-intensive that the three U.S. “gaseous diffusion” plants were said to consume 10 percent of all electricity used by American factories. But diffusion enrichment has given way to gas centrifuges and, more recently, laser isotopic separation ― methods that use 20 times less energy to isolate fissile U-235 from U-238.
The bottom line: kilowatt-hour for kilowatt-hour, nuclear-generated electricity is at least as effective as solar and wind at cutting climate pollution. Meanwhile, Hurricane Katrina, Superstorm Sandy, “heat domes” and wildfires, and, in August, the first Himalayan glacial collapse, have made once-hypothetical climate death and disruption a daily reality. More than rising electricity demand from A.I., it’s the climate crisis that’s driving renewed interest in nuclear power.
Click here for the second installment, Why I’m Pro-Nuke Now: Centerpiece.
[1] Quoted passage is from a local (PA) news site, though the an archived Channel 2 broadcast it cited is no longer on line.
[2] The sole “near-miss,” and a major one, was the 2002 discovery by operators at the Davis-Besse nuclear plant near Toledo, OH of extensive corrosion of the reactor vessel head — a vital barrier against loss of coolant and release of radiation. The U.S. General Accounting Office sternly rebuked the Nuclear Regulatory Commission for failing to identify and prevent the corrosion. See GAO, Nuclear Regulation: NRC Needs to More Aggressively and Comprehensively Resolve Issues Related to the Davis-Besse Nuclear Power Plant’s Shutdown, GAO-04-415, May 2004.
[3] In-person conversation at People’s Policy Conference at the New School for Social Research in New York, May 2, 2026.
[4] C. Komanoff, Power Plant Performance: Nuclear and Coal Capacity Factors and Economics (15 MB pdf), Council on Economic Priorities, 1976.
[5] Dividing today’s 90% uptime by the former 60% yields 1.50, indicating 50 percent more kilowatt-hours per kW.
[6] IPCC Annex III report, Technology-Specific Cost and Performance Parameters, 2018. See table on p. 1333.

I vaguely knew that a 2017 deal ― pushed by the self-proclaimed environmental group Riverkeeper and brokered by then-Gov. Andrew Cuomo ― was about to shut down the Indian Point nuclear plant, located on the Hudson River 35 miles north of midtown Manhattan. The older of its two reactors unit would (literally) get the chop within a week; its twin would follow in a year. Both reactor vessels would be cut to pieces and their radioactive components chemically dissolved. Once that process began, there’d be no turning back.
By that gauge, wind and solar look mediocre. Take those 42-inch square “balcony solar” arrays that Germans are buying like hotcakes ― they’re a neat idea, but it would take 50 million of them to match the carbon reduction Indian Point provided, as 











An astonishing $33 billion had gone down the drain. But why? Klein and Thompson describe how the project immediately bogged down in innumerable negotiations over property rights. “Negotiating with courts, with funders, with business owners, with homeowners, with farm owners,” they write. “Those negotiations cost time, which costs money. Those negotiations lead to changes in the route or the build or the design, which costs money.” State officials relied on expensive consultants who made matters worse: “It was one of these consultants – WSP – that estimated the system would cost only $33 billion and take only twelve years to build. But WSP was joined by Project Finance Advisory, Cambridge Systematics, Arup, TYLin, HNTB, PGH Wong Engineering, Harris & Associates, Arcadis, STV, Sener, and Parsons Corporation.” The Los Angeles Times concluded that outsourcing “proved to be a foundational error in the project’s execution – a miscalculation that has resulted in the California High-Speed Rail Authority being overly reliant on a network of high-cost consultants who have consistently underestimated the difficulty of the task.”
Abundance reflects this growing pessimism, but never asks why. It quotes a University of Chicago economist named Chad Syverson on a construction-industry regulatory process that is complicated beyond reason. “There are a million veto points,” he complains. Yet there’s no sense that the problem involves not just economic regulation, but a legislative process that also contains a million veto points thanks to hundreds of congressional committees and subcommittees, outrageous filibuster rules, a “hold” system that allows individual senators to tie up legislation for months, and so forth. If Capitol Hill is in chaos, is it any surprise that the regulatory structure it gave rise to is in chaos too?
From a climate standpoint, the true Indian Point disaster is the plant’s closure and dismantlement. Both reactors are now kaput, their reactor cores chopped up. Unsurprisingly, the effort to decarbonize the state power grid — New York’s lowest-hanging climate fruit — is in reverse. Emissions are mounting, and in New York City and other downstate areas formerly supplied by Indian Point, electricity is getting costlier and less dependable.







True, what was on the ballot in Washington last week — making gasoline and other fossil fuels more costly to elevate lower-emission substitutes from smaller cars to electric cars and, best of all, less driving — isn’t the same as congestion pricing. But it’s a close cousin. What stands out is that a policy rejected by nearly 60 percent of voters in 2016 and again in 2018, won with around 60 percent in 2024. Attitudes can shift when facts warrant.