Senior decision-makers come together to connect around strategies and business trends affecting utilities.


If You Want To Get Climate Serious, Get Nuclear Serious

image credit: Wikimedia Commons
Llewellyn King's picture
Executive Producer and Host White House Media, LLC

Llewellyn King is the creator, executive producer and host of “White House Chronicle,” a weekly news and public affairs program, airing nationwide on PBS and public, educational and government...

  • Member since 2018
  • 75 items added with 78,907 views
  • Jul 16, 2022

If you have gasped in Dallas, sweltered in London or baked in Tokyo this summer, you will likely believe that global warming is real.

You are also likely to believe that governments — at least the caring ones — are desperate to cut the amount of carbon released into the atmosphere from power plants and vehicles.

More electricity is needed to cut the greenhouse-gas emissions from cars, trucks, buses, trains and, eventually, aircraft. The government figures that U.S. electricity demand will double by 2050, even as the fuels producing it change.

There are three technologies for producing new large quantities of electricity without producing greenhouse gases: wind, solar and nuclear.

Leading global energy institutions, including the International Energy Agency, are adamant that nuclear must be part of the future energy mix. Nuclear is desirable in many ways:

—It isn’t dependent on foreign supply except for some heavy components, like the castings for large pressure vessels. If needed in newer reactors, these can be acquired from reliable allies, including South Korea and Japan. Alternatively, we could reinvigorate our large component industry.

—When it comes to the supply chain, nuclear component manufacture can be brought on shore. It doesn’t have a Chinese component. Wind is dependent on rare earths — they are a multiplier in wind turbines, increasing output up to five times. More than 90 percent of rare earths are processed in China, even if they are mined elsewhere. It will take precious decades to replicate the Chinese rare earths infrastructure. Also, China dominates the manufacture of cheap solar cells.

—Nuclear offers long-term planning: The design life of a plant can be as long as 100 years. These plants are clean, safe — and getting safer. They have a high energy density and low land use, which contrast with solar and wind.

Incredibly, the world, outside of China and Russia, seems to have lost the ability to build nuclear plants. It is as though talent and institutional knowledge have disappeared. Those under construction are running many times over their projected costs and a decade or more behind schedule. They represent a systemic industrial failure, whether it is Plant Vogtle in Georgia, Flamanville-3 in France (which gets 70 percent of its electricity from nuclear; we get 19 percent), or Olkiluoto 3 in Finland.

At the heart of these failures — complex and far-reaching — is a failure of welds and a shortage of welders.

As a first step, the United States, in conjunction with the nuclear manufacturing industry, needs to find out what it is that we have lost in expertise and how to recapture it. We built more than 100 reactors in the 1960s and 1970s. There were some delays back then, but they were nothing like the catastrophic ones of today. Particularly, examining what has gone wrong with nuclear building needs to concentrate on welding. Is this an old trade that needs updating? Can we fix some of the human error that has plagued big industrial welding, from nuclear plants to new ships, through automation and AI?

Not since the 1960s, I am told by nuclear lobbyists, has the public policy apparatus been so aligned to favor nuclear. One of these lobbyists said, “Both houses of Congress are on board, the administration is on board, the regulatory agencies are on board, and public acceptance is greater than it has been in years. But the industry is on its back.”

The issue, to my mind, is not whether we can relearn how to do what we used to do but that there is no mechanism for the utilities to buy and build nuclear plants, whether they are the new generation of small modular reactors now under development or updated, large (about 1,000 megawatts), more traditional light water reactors. No utility can take the risk in the deregulated world. It is too much to ask.

The nation needs a coherent plan whereby a new generation of nuclear power can be built quickly. It has been done in the past, and it can be done again.

I would suggest — as I have suggested over many years — that nuclear needs government safety oversight, proliferation safeguards and approval that a tranche of reactors be built on government sites, financed by the government and sold to commercial consortia to operate. These needn’t necessarily be utility companies. Wind and solar are being developed by merchant companies in many cases.

There is a national climate crisis, and a national electricity crisis is building. Utilities are having to produce more electricity while giving up coal and gas to do it. Nuclear is the strong third leg of the future electricity stool.

Llewellyn King's picture
Thank Llewellyn for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member
Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Matt Chester's picture
Matt Chester on Jul 18, 2022

And because many nuclear solutions take so long to plan, build, and connect to the grid, time is imperative. Every nuclear plant that closes is setting us back on our net zero goals, and every delay in getting new ones online is time we'll likely regret in pursuit of those clean energy goals. Thanks for sharing this important piece, Llewellyn

Michael Keller's picture
Michael Keller on Jul 25, 2022

Exceptionally high nuclear construction costs are caused by significant overregulation by the Nuclear Regulatory Commission. The bureaucrats prescriptively dictate all manner of activities which significantly balloons the cost of all activities. There is virtually no doubt that this occurs, as demonstrated by contrasting the unit production costs between nuclear and fossil plants. In my 50+ year career, I have been heavily involved with both types of power plants and the contrast between nuclear and fossil are both stark and deeply disheartening.
In effect, the NRC staff epitomizes a bureaucracy decoupled from accountability, with the regulations they have more or less unilaterally created stunningly complicated and counterproductive. Vast amounts of effort are spent on activities well removed from protecting the public from hazardous radiation and that decreases public safety.

The NRC staff is currently on a crusade to revamp regulations for advanced reactors in response to very simple new law to modernize regulations. The proposed new regulations are much larger (about 60%) than existing regulations and are fiendishly complex. So much for “modernization” (AKA simplification). Classic example of bureaucrats gone wild in utter defiance of the law. Hundreds of pages of comments by the public and industry have been submitted to the NRC. The NRC staff has virtually ignored all the comments.

While I agree nuclear power is essential to efforts to reduce CO2 emissions, the practical reality is that the needlessly inflicted overregulation will doom the effort to deploy the passively fail-safe next generation of advanced reactors. The financial risk in building these power plants is “off-scale high”; no idea of the cost and construction duration. 

Richard McCann's picture
Richard McCann on Jul 25, 2022

Costs are rising everywhere globally. This isn't a single nation issue--it appears to be inherent in the technology. 

Michael Keller's picture
Michael Keller on Jul 25, 2022

The massive cost overrun problem is endemic in regions with extensive government overregulation, e.g. US, and Europe. Regions with more of a reasonable approach are not particularly affected by heavily bloated costs, e.g. China, Middle East.

Just recently, state report on delays and overruns at Vogtle plant in Georgia cited regulatory driven paperwork issues.

The technology needed to build the plants is quite straightforward. Dealing with overregulation is anything but straightforward.

Richard McCann's picture
Richard McCann on Aug 1, 2022

Not true--costs are up everywhere, even China. Whether the costs are transparent elsewhere is 

Of course Georgia regulators are trying to cover themselves from blame for Southern Company's mismanagement (which appears to be endemic to the technology.) You don't get to $30B solely through paperwork issues:

Even MIT notes that the rise in costs is not related to regulation or lack of a trained workforce:

Other studies have dug into the reasons and trends for the rise in costs.


Michael Keller's picture
Michael Keller on Aug 2, 2022

The cost increases in China do not remotely resemble those in the US and Europe. 
I’ve been in the energy business for over 50 years and have seen 1st hand the impacts of massive overregulation by the NRC. All activities are significantly ballooned in terms of material costs, fabrication costs, construction costs, and operations cost. Everything take a hell of a lot more time because of paperwork and watchers-watching-the-watchers. There is no parallel in the fossil plants, particularly the gas turbine combine-cycle units.

The academics at MIT has never designed, built, and operated a power plant. I have, both nuclear and fossil. I also hold several patents for advanced nuclear power plants.

Joe Deely's picture
Joe Deely on Aug 2, 2022

Nuclear in China is actually much more similar to the rest of the world than most folks think.

Of course with government dictating everything... things are a little cheaper and a little faster. However, the same goes for renewables and like everywhere else in the world renewables are doing WAY better than nuclear in China.

Even in China it takes a couple of years to plan nuclear construction and then it takes on average almost 6 years to finish construction on a reactor. Obviously better than Vogtle but not really competitive.

Renewables lead over nuclear in China is going to grow dramatically bigger over the rest of this decade.

Robert Borlick's picture
Robert Borlick on Aug 1, 2022

"Costs are rising everywhere globally...."  Really Richard?  Not in China, which builds nuclear plants for far less than the West does.  

Robert Borlick's picture
Robert Borlick on Aug 1, 2022

I disagree.  Analyses done in the 1980s suggest that NRC regulations added about 20 percent to the overnight costs of the typical nuclear plant.  That is not an unacceptable price to pay for the added safety produced by government oversight.  Some inefficiency and overkill is still preferred to an increased risk of an accident that causes a radioactive release.  

There will always be people who rail against environmental regulation but the country has greatly benefitted from cleaner air and water that it has produced since EPA was founded 50 years ago.

Michael Keller's picture
Michael Keller on Aug 2, 2022

The regulations in the 1970’s were not that onerous - I was part of industry at the time. Those in the 1980’s increased quite a bit because of the Three Mile Island accident. Thereafter, the regulations increased at a stupefying rate. 

Today’s nuclear regulations are stunningly complex, with the vast majority being “regulatory guides”, branch technical positions, and allied NRC staff guidance. These things are more-or-less unilaterally developed by the NRC staff. In the 1970’s these types of documents number in the dozens, now they number in the hundreds and hundreds. 

The fundamental problem with the NRC is bureaucrats prescriptively dictating how to design, build, and operate nuclear plants. That actually decreases safety because vast resources are directed at activities well removed from protecting the public from hazardous radiation.

The key regulations governing nuclear power (10CFR50/52) are actually relatively high-level in nature. The NRC staff has chosen to create an armada of lower tier complicated and highly prescriptive regulations.

The all-in cost to license the NUSCALE reactor was well over a half billion dollars. Relative to earlier licensing efforts, the incremental improvement in public safety is insignificant.

The NRC is no different than any other Federal regulatory agency striving to constantly increase their power; that’s what bureaucrats do. However, as we have seen in the recent Supreme Court ruling against the EPA, these abuses are running afoul of the law.

Richard McCann's picture
Richard McCann on Jul 25, 2022

King asserts "the world...seems to have lost the ability to build nuclear plants." He assigns the current woes to "a failure of welds and shortage of welders."

This rosy view of the past is a myth. On the West Coast at least, the nuclear industry has a trail of failures and boondoggles. Diablo Canyon costs ran to ten-fold the original estimate and San Onofre has similar overruns. Both Trojan and Rancho Seco ran so poorly that they were shutdown prematurely. The entire Pacific Northwest economy was stymied in the early 1980s by the WPPSS bond default to build five plants, of which only one unit was built. Yes, Columbia and Palo Verde has run well and cheaply, but they are the exception, not the rule. The debacles that King lists are not surprising nor exceptional.

New technologies such as small modular reactors (SMRs) may deliver, but the assumptions used in the studies to justify the economics are quite optomistic.

Robert Borlick's picture
Robert Borlick on Aug 8, 2022


You make some valid points and some not so much,  

It is true that the completed cost of virtually every nuclear plant was about tenfold higher than the original estimate.  This was certainly the case for the South Texas Nuclear Plant, which I studied in detail for the Public Utility Commission of Texas.  What I was surprised to discover was that the engineers that initially estimated the plant's overnight costs lacked a detailed design and grossly underestimated the quantities of the inputs, e.g., concrete, rebar, wires, and man-hours of labor.  As construction progressed the engineers would produce the detailed blueprints just months ahead of the construction crew.  And each year the quantities of the inputs increased.  Also, the construction period kept getting extended, which increased the amount of interest during construction (referred to as AFUDC).  

While cost overruns and completion delays caused some partially completed plants to be cancelled, the most important factor was that electricity demand growth dropped off substantially through the 1970s when most of the cancelled plants were initially undertaken.  That was particularly true in the Pacific Northwest.  

WPPSS undertook that ambitious nuclear construction program because the head of Bonneville Power Administration convinced them that those plants would be needed due to BPA's forecast of demand growth.  You can't blame the nuclear industry for BPA's bad planning.  WPPSS depended on BPA resource planning process as it always had in the past when BPA's hydropower was the predominate source of electricity for all of the publicly-owned utilities in the Pacific Northwest.  

I haven't studied the California nuclear plant cost overruns.  However, I vaguely recall that Rancho Seco had a troubled history that was directly related to design flaws and construction mistakes.  That was specific to the competence of the Engineering firm and Construction contractors, not nuclear in general.  

The new SMR designs avoid many of the shortcomings of the large LWRs because their designs are standardized and they can be manufactured on assembly lines where both costs and quality are easier to control.  But I agree, until we see the first SMRs go into service we won't know whether they will deliver on the promise.  

Audra Drazga's picture
Audra Drazga on Jul 26, 2022

Curious about ways that utilities that have already invested or are investing in Wind and Solar - can pivot to nuclear? 

Joe Deely's picture
Joe Deely on Jul 26, 2022

Meanwhile... here is what is actually happening to generation around the world.

Peter Farley's picture
Peter Farley on Jul 26, 2022

The good old days of nuclear are a myth Vogtle 1 &2 $660m estimate $8.3bn actual. Barrakah five years estimate actual 10 years.

 Space efficiency, myth. Plant Vogtle is 5 square km generating about 30TWh per year when all four reactors are running. Using as much water as a city of 500,000 people. 3,300 5MW class wind turbines at 300 square meters each US average CF use 3,300 x 300/1,000,000 = 1 square km to produce 60-65 TWh.

 Cost is a myth: Vogtle 2&3 $25bn for 17TWh/y. 5 GW of wind for 19 GWh/y $8 bn. Throw in 2.4 GW of storage to match nuclear reliability and exceed peak summer output total cost $11 bn.

Operating cost for wind + Batteries $20/MWh Nuclear $40.

With all China's nuclear prowess it is connecting 3-5GW of nuclear per year but commissioning 50-100GW of solar and 50-70GW of wind. If the US connects an average of 70 GW each of solar and wind per year for the next 15 years before the next nuclear plant could come on line, the added generation would be 5,800 TWh/y or almost double combined nuclear and FF output this year. Total generation would from all sources with hydro 80% of existing nuclear and other existing renewables 7,200 TWh/y or about 800MWh/square km. By 2030 Germany expects to be generating 1,800MWh/square km from renewables with far less hydro, lower wind speeds and less solar and about 1/4 of the open space per resident


Michael Keller's picture
Michael Keller on Jul 26, 2022

Your land area calculation for wind turbines is not correct. The spacing is based on optimizing air flow, not blade diameter. Additionally, an uninhabitable area is required based on blades and objects (e.g. ice, parts of blades) thrown from the massive rotating machines. Generally, the zone is about 3 times the tower height. Further, in order to match the output of a nuclear reactor, about 3 times the reactor’s output is required because of the low capacity factor for wind turbines and unreliability of the wind
A nuclear plant the size of Vogtle typically sits on about 750 acres. However the actual plant area is considerably less.

The cost of a nuclear plant is an obvious issue, but as I pointed out earlier, that is largely driven by excessive overregulation.

So how is Germany doing today with all their wind energy and Russian gas? They actually are in an energy crisis. That is the direct result of Germany’s mindless embrace of renewable energy.

Peter Farley's picture
Peter Farley on Aug 9, 2022

You are right the land area calculation is incorrect because putting wind turbines further apart does not affect how much land they use, crops and forest still grow on the land between them and cows still graze right up to the base. Actually in the US with the latest generation turbines averaging 44% and nuclear 90% you need about 2.1 GW of wind for every 1GW of nuclear. I was being generous 3,300 x 5 = 16,500 MW. Plant Vogtle I-IV total 4,640 MW. i.e. I allowed for almost twice as much wind power as I should have

As for land area of plant Vogtle go on Google maps and measure it. If we want to talk about the footprint of the actual wind generator is less than 100 square meters and for a strict output comparison using GE Cypress 6.0-164 units you would need 1,500 turbines or 150,000 square meters . Throw in a battery three times the size of the planned Moss Landing battery (3 x (1,600MW/4,800 MWh) at three 200m x 300m = 180,000 square meters for batteries 150,000 square meters of generators 330,000 square meters or 81 acres

Germany's problems have little to do with renewables, everything to do with nuclear and gas. Renewables are up from 45% to 51% of supply. Instead of importing a net 1-2% of its electricity from France, France's nuclear output is so low that it is importing 4% of demand from Germany and burning 50% more gas. France's trade balance with Britain and Spain has also reversed. 

Robert Borlick's picture
Robert Borlick on Aug 8, 2022

I haven't examined the veracity of your numbers but what you appear to have missed is that renewables have to be substantially overbuilt to account for the seasonal variation in output.  At latitudes similar to Southern California wind and solar energy production in December is less than half of what it is in July.  Trying to compensate for this seasonal variation with storage is prohibitively expensive. In California the required overbuilding effectively doubles the cost of the energy the renewables produce.  In addition there is still a need for storage to accommodate the diurnal cycle for solar and the fact that wind production typically drops to near zero for several consecutive days.  As we move to more northern latitudes the overbuilding requirement gets even worse.  

Several studies (e.g., Jenkins at Princeton) have shown that adding a reliable caseload resource, such as nuclear, can reduce the cost of reliably serving demand can be cut in half.  

Lastly, your assumption that new nuclear will not be available until 2037 is really pessimistic.  NuScale is scheduled to start up its first reactor in 2029 and other SMR technologies are not far behind.  Once the first few SMRs demonstrate their economic viability they can be produced on a cookie cutter basis.  

Peter Farley's picture
Peter Farley on Aug 9, 2022

I understand that you have to overbuild renewables and that is why I suggested 3,300 wind turbines instead of the strictly required 1,500-2,400 depending on size. You are exactly correct that relying on storage for seasonal variation is complete folly. It varies region by region but there are three rules of thumb,

a) have enough wind and solar at normal capacity factors to generate 30% more energy per year than you need. Ensure that most of your solar has a west facing bias or east west with as little as possible directly south facing and that your wind turbines are as tall as possible with very high swept area/power ratios. 

b) have a rough balance between annual output of wind and solar. winds are stronger in winter and at night, solar during the day and in summer.

c) For the next 20 years at least, expect 5-10% of your annual demand and up to 35% of instantaneous demand to come from thermal sources

California violates almost all the above advice so it will need much more backup than it should

In 1987 Scientific American promised SMRs in 4 years, in 2014 Nuscale promised an operating reactor in 2025. Last year it announced an "order" for a 840MW plant in Idaho. This year it went public and raised enough cash to keep going for 18 months and the plant is now half the size.  Forgive me if I am skeptical


Robert Borlick's picture
Robert Borlick on Sep 6, 2022

Vogtle 1-4:  4640 MW at 90% CF produces 36581760 MWh per yr. or 3048480 MWh per month year-round.

Onshore wind and solar produce about half as much energy in January as in June at average capacity factors of 41% and 29%, respectively, (for utility solar installations in the southern-most latitudes of the US). 

So to satisfy the same winter demand as Vogtle can:

     Utility Solar installed capacity = 3048480/(720)*.29 = 14,600 MW


     Onshore Wind installed capacity = 3048480/(720)*.41 = 10,327 MW

In addition, both resources also requires enough storage to supply load when the resource is not producing enough.

For solar that amounts to total load served from about 7 PM to the following 7 AM, since the sun doesn’t shine at night

For Wind, the amount of storage has to be sufficient to serve the total load for at least 2 consecutive days, because calm conditions have been observed to occur for such lengthy intervals, and even longer.

Based on these rather straightforward calculations, it appears that you have grossly underestimated the amount of renewable resources plus storage needed to replicate the performance of the Vogtle plant.  


Peter Farley's picture
Peter Farley on Sep 12, 2022


The whole point of a renewable system is to have a balance between wind and solar in annual generation terms so comparing nuclear with wind alone or solar alone is beside the point. solar is obviously higher in summer and lower in winter and wind is roughly opposite. Wind is also generally stronger at night while demand is lower so the shortfall at night is not demand from 7PM to 7AM

The combined monthly share only varies by a few percent so if you have enough wind and solar to generate 130% of annual demand the monthly potential output will vary from about 110% to 150%. Some of that will be diverted to storage, some will be used to reduce hydro output to allow reservoirs to refil and some will just be curtailed.

You definitely have to plan for low wind periods of more than a few days and low solar for that matter, so there is no doubt storage is needed and in my view for a long time some thermal capacity will be retained. 

But you also have to have backup for nuclear and in fact the French experience this summer has shown even more storage duration is required. This year from April 1 to now France's trade has fallen from a surplus of 22,000 GWh to a deficit of 11,000 GWh and gas generation has increased by 12,000 GWh. In effect France needed 52,000 GWh of new generation to replace declining nuclear. Last year wind and solar generation in Germany declined from 176 TWh in 2020 to 162 TWh in 2021. Allowing for increased capacity, about 17,000 GWh extra backup over 12 months. In other words, a bad nuclear period in France required three times as much backup as a bad renewable year in Germany. Now the comparison is not that simple because in a normal year France's nuclear output is more than double Germany's wind and solar output but on the other hand EDF is expecting nuclear output this year to be about 290 TWh vs 2019 output of 377 TWh and 2015 output of 415 TWh so even if Germany increased potential wind and solar output to 415 TWh and a bad year came along it would be unlikely to fall below 370 TWh, less than half the decline in nuclear. 


Michael Keller's picture
Michael Keller on Sep 12, 2022

I believe you will find the French are returning the majority of the nuclear fleet to service for the upcoming winter.

Energy costs in most of Europe are skyrocketing, compliments of just plan dumb policy. Namely over reliance on Russian gas, shutting down coal plants as well as nuclear plants. This is the direct result of hysterical overreaction to CO2 in the air and the irrational claim that CO2 controls the climate. It does not. Energy from the sun and the complex transfer of energy around the globe drive the climate. CO2 is a bit player, at best.

Robert Borlick's picture
Robert Borlick on Aug 1, 2022

I totally agree with my friend, Llewellyn.  There is no doubt that Vogtle 3 & 4 demonstrated that we have lost the knowhow for building large nuclear plants, although I think the skills shortfall extends beyond welding.  The forthcoming small modular reactors may solve this problem as they will be fabricated on assembly lines that can control quality more closely and will benefit from the experience curve.  

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »