How America’s Big Build fell short
The United States once pledged to build 1,000 reactors by the year 2000. What went wrong? And what are the lessons learned? And what will it take for American nuclear to return to its glory days?
This edition of the Decouple Dispatch is sourced from an episode of the Decouple Podcast featuring physicist and pro-nuclear advocate James Krellenstein. You can watch to the full episode here or listen on your favorite podcast platform. Support the work of Decouple Media by signing up as a patron on Patreon.
In the “thumbnail sketch” of history embedded in our collective consciousness, we remember Three Mile Island incident as the turning point in history when the United States turned away from nuclear power. But as we look ahead to another nuclear renaissance, it’s important to look back at the past and ask: “What really happened?”
Where it all began
The United States of America is where practically all of the world’s current commercial nuclear technology originated. Despite talking a good game about governments taking a back seat to markets in the land of the free, both the Pressurized Water Reactor (PWR) and Boiling Water Reactor (BWR) family of reactors have their roots in big US government programs: PWRs out of the Navy under the leadership of Admiral Hyman Rickover, and the BWR coming later from the Argonne National Labs.
This was a public-private partnership with a distinctly muscular post-WWII flavor. The US government built the prototype plant, then identified a commercial partner: Westinghouse for the PWR, General Electric for the BWR. “Almost all of the global nuclear power generation capacity we have today had its genesis from these two programs,” said Krellenstein.
After that, American nuclear development went from a “Set Menu” to a “Buffet Lunch.” Unlike most countries, the US didn’t have a national utility, but a patchwork of power providers…some big, some small, some public, some private. This allowed each utility to choose their own favored reactor technology with little federal coordination. As a result, the nuclear fleet in the US became “maybe the most non-standardized global fleet anywhere,” as well as the largest.
Nuclear Mania
Fast forward to the year 1973. Tony Orland’s folksy “Tie a Yellow Ribbon” jostled with Roberta Flack’s “Killing Me Softly With His Song” at the top of the charts. And Americans were lining up around the block anxiously at gas stations. Nuclear energy has come of age just as the Oil Embargo of 1973 was biting. Americans were also increasingly convinced by legendary geophysicist M. King Hubbard that we were going to run out of oil.
Is it any wonder that utilities all over America rushed to order new nuclear power plants! This was an opportunity to generate power in any location, free from the need for oil from the Middle East. The peak number of orders was more than 100 in 1973! This was also the year when Nixon announced “Project Independence.” With 1,000 reactors planned for the year 2000, it would have been the American Messmer plan.
From Pots of Gold to Utility Killers
The manic optimism of ‘73 almost immediately began to sour. The same energy crisis that had utilities seeing dollar signs dancing around nuclear power plants also brought forth real inflation and labor shortages, making plants more difficult to build.
Even worse was the Federal Reserve raising interest rates very significantly to tamp down on inflation. This raised the capital costs of US plants, which were privately financed on the capital markets. Even if the “overnight” cost of the plants remained constant, the projects would have to pay so much more in interest it would make them much more expensive. To the point where they can put the entire company at risk.
With a captive customer base and a product that’s essential to modern life, utilities rarely go bankrupt…but a nuclear project with escalating capital costs can do it, and did on several occasions. This was how nuclear plants gained a reputation for being “utility killers.”
The problem with the “Buffet Lunch” approach to reactor selection also began to raise its ugly head. For every plant of a different design — and they were pretty much ALL plants of different designs — you had to have different valves and different pumps, snarling up the qualified supply chains that were already straining under inflation-related difficulties.
“This contributed to the overall crunch that started happening post-’75,” said Krellenstein, “as a result, the completion rate of nuclear power plants dropped below 1% per month.”
The utilities of the 60s and 70s were not totally irrational in ordering such an assortment of reactors.
“They thought they were prototyping which reactor worked best, then the second order would all be of that design,” explained Krellenstein.
That second order never came.
The road to Three Mile Island
In the popular imagination, Three Mile Island happened, then came the film “The China Syndrome,” then we stopped building new nuclear plants. That’s actually The opposite order to how things happened.
Nuclear orders were drying up BEFORE the film fanning anti-nuclear fears came out, which was just 12 days BEFORE the Three Mile Island incident itself. But somehow it all came to a head in 1979.
Dare we imagine an alternative scenario where the US nuclear industry coalesced at an earlier stage around a standardized reactor and started to aggressively cost-down? In a world where more projects survived the inflation crunch and started making money for their utilities rather than dragging them down, would we have remembered Three Mile Island more as an industrial footnote than a turning point?
The Good-enough Reactor
The US paid a heavy price for its experimental spree with reactors. So what’s the result? Which one is the best? Who won?
What if I told you if none of it even mattered?
“What’s fascinating is that you can’t really find a clear signal,” said Krellenstein, “the management and operations turned out to be more important to the performance of the plants than the reactor design.”
When the US nuclear industry stopped building nuclear power plants, they became the world’s best operator of the nuclear power plants they did build. The same fleet that were struggling to get above 65% capacity factor in the 70s are now operating at 90%+ capacity factor. In the process of that improvement, any initial difference between the reactor types were blown away.
“Even if you try to compete between PWRs and BWRs, they come down within a couple of tenths of one percent on the capacity factor.”
Looking back, the mistakes of the US nuclear industry was obvious. Every single reactor type was actually good enough, but they needed to build 100 of one type in order to get on that learning curve.
The bullish case for US nuclear
In Krellenstein’s view, the obvious choice for the “good-enough reactor” of 2023 is without a doubt Westinghouse’s AP1000. Perhaps you might have heard that Vogtle Unit 3 just went into commercial operation.
Krellenstein does not deny that the plant’s build — the first in the US for decades from scratch — was beset with cost and schedule overruns. Counterintuitively, that’s exactly the reason why we need to turn around and build another AP1000.
“We finally got it to work. It’s built and finalized in the real world. That’s a killer asset,” said Krellenstein, “we have to go back and do it again. We need to figure out a way that we can actually repeat the same lessons, the same designs over and over again.”
The root cause of Vogtle’s cost-overruns were an incomplete design, non-existent supply chains and lack of a workforce familiar with the plant, reasons Krellenstein. By turning away from the AP1000 to new designs that seems to have less baggage, the industry is ironically choosing to expose itself to the same risks that just got rectified with the AP1000 after Vogtle 3’s completion.
“As an industry and an advocacy community, we haven’t talked enough about what went wrong at Vogtle. Instead, we’re hand waving around it. We’re saying we have these marvelous new small modular reactors that are going to solve all those issues.”
Krellenstein warns that if we don’t face up to the hard lesson of Vogtle, we are at risk of repeating history while diluting away the hard-won learnings gained by Vogtle’s completion.
“It’s not about how fancy or cool the technology is. It’s about the familiarity of the workforce. The maturity of the design. The completeness of the supply chains. The predictability of the regulatory process.”
Angelica’s Take
2023 is a funny time for the nuclear industry. We appear to be undergoing a milder form of both the nuclear mania of 1973 simultaneously with the harrowing nuclear crash of 1979. The industry is understandably traumatized by the painful cost overruns at Vogtle, while simultaneously high on the narrative of SMRs. We are seeing a great proliferation of technologies too, ignoring the fact that history is screaming at us to keep things simple and learn by doing the exact same thing over and over again.
Let’s not forget that as botched and misbegotten as the American nuclear buildout might appear to be in hindsight, the industry took a brand new technology and created mighty machines that pumps out massive amount of power and brought it to peak performance. Less than 100 of those machines in the United States provide 20% of the electricity on the grid.
How can we build some more?
We listened to both episodes with James. Really well done. His recall of the history and specifics of various units, technologies is amazing.
Incredible what you and C4NE have accomplished. Gives us great hope for the U.S.
We agree with James. Much learned on AP-1000s from Vogtle. Glad you or he referenced the MIT study by Shirvan. For those interested, Shirvan concludes cost of 10th AP-1000 approx. 1/3 of Vogtle Units 3/4. https://web.mit.edu/kshirvan/www/research/ANP193%20TR%20CANES.pdf
One suggestion I have seen from none other than Tim Echols who is a prominent pro nuclear member of the Georgia Public Service Commission(that has been extensively involved in the construction of Vogtle) is that Congress and the US Federal govt should essentially give the Tennessee Valley Authority some type of catastrophic risk guarantee to build the next two AP1000s. Basically the US Federal govt would guarantee that TVA ratepayers would not pay for any costs of building a new AP1000 unit above a certain unlikely to exceed price ceiling. I personally think there is a lot to be said for this idea. Echols also suggests this type of guarantee could be used with a privately owned utility as well but Congress is for political reasons going to be more likely to give this type of guarantee to the TVA as a Federal agency than a private or US state/municipal controlled agency, a sentiment I agree with.
The other thing I will note is not all in the pro nuclear community actually agree with James Krellenstein. Many think the NRC has to be "blown up" to the point that SMR's become as cheap as combined cycle natural gas. Thinking of someone like James Hopf having this later view.