This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Nuclear Agency: Nuclear Power Will Play Only A Modest Role In Stopping Climate Change

Joseph Romm's picture
, American Progress
  • Member since 2018
  • 1,651 items added with 368,059 views
  • May 20, 2015

A journalist checks radiation levels at stricken Fukushima Dai-ichi nuclear power plant. The 2011 disaster, which will cost Japan over $100 billion, set back the nuclear industry worldwide. CREDIT: AP Photo

Nuclear power can play a modest, but important, role in avoiding catastrophic global warming — if it can solve its various problems including high construction cost without sacrificing safety.

That is the conclusion of a comprehensive 2015 “Technology Roadmap” from the International Energy Agency (IEA) and Nuclear Energy Agency (NEA). It is also what I’ve been arguing on Climate Progress for a long, long time.

The IEA is the global body responsible for energy analysis, and one of the few independent agencies in the world with a sophisticated enough energy and economic model to credibly examine in detail the role of various low carbon technologies in a 2°C scenario (2DS) aimed at averting catastrophic global warming. The NEA was set up by the industrialized OECD countries “To assist its member countries in maintaining and further developing, through international co-operation, the scientific, technological and legal bases required for a safe, environmentally friendly and economical use of nuclear energy for peaceful purposes.”

Here is what the IEA and NEA project is a plausible though “challenging” pathway for the nuclear energy industry in a carbon constrained 2DS world — if it can get its act together:


The IEA and NEA project nuclear could rise from its current 11% of world electricity capacity to 17% in 2050 in a carbon constrained world. (Click to enlarge.)

Because it is a low-carbon source of around-the-clock (baseload) power, a number of scientists and others have called for a reexamination of nuclear policy. The Chinese in particular have been building nuclear power plants at a steady pace. Yet very few new plants have been ordered and built in the past two decades in countries with market economies, such as the United States, which derives a fifth of its power from nuclear. That is primarily because new nuclear plants are so costly, but also because dealing with the radioactive nuclear waste remains problematic and the costs of an accident are so enormous.

In particular, the 2011 Fukushima nuclear disaster in Japan slowed the rate of new plant construction starts. In 2014 there were only three new plants put under construction.


At the same time, Fukushima caused a number of countries, including Japan and Germany, to reconsider their dependence on nuclear power. Germany is phasing out all its nuclear power plants on an accelerated schedule. In 2014, Japanese college professors calculated that the accident “will cost 11.08 trillion yen ($105 billion), twice as much as Japanese authorities predicted at the end of 2011.” That includes both radiation clean up and compensation paid to the victims.

The IEA and NEA note, “France, which today generates 75% of all its electricity from nuclear, still plans to reduce this share to 50% by 2025 while proposing to maintain nuclear capacity at its present level.”

The costs of new nuclear reactors have been rising for decades, and they are now extremely expensive, costing up to $10 billion dollars apiece. While solar power and wind power continue to march down the experience curve to ever lower costs — solar panels have seen a staggering 99 percent drop in cost since 1977 — nuclear power has been heading in the opposite direction.

Nuclear power appears to have a negative learning curve:

Average and min/max reactor construction costs.

Average and min/max reactor construction costs per year of completion date for US and France versus cumulative capacity completed.

In the past several years, utilities have told state regulators that the cost of new nuclear plants is in the $5,500 to $8,100 per kilowatt range (see Nuclear power: The price is not right and Exclusive analysis: The staggering cost of new nuclear power).

A key reason new reactors are inherently so expensive is that they must be designed to survive almost any imaginable risk, including major disasters and human error. Even the most unlikely threats must be planned for and eliminated when the possible result of a disaster is the poisoning of thousands of people, the long-term contamination of large areas of land, and $100 billion in damages.

In 2014, just 5 gigawatts of capacity were added. In their “Nuclear Roadmap,” the IEA and NEA explain what level of capacity additions would be required in its 2 degrees Celsius scenario: “In order for nuclear to reach its deployment targets under the 2D scenario, annual connection rates should increase from 5 GW in 2014 to well over 20 GW during the coming decade.” That means returning to a nuclear build rate previously achieved for only one decade — 20 gigawatts per year during the 1980s. That target has many challenges in a post-Fukushima world.


The IEA and NEA themselves note that “such rapid growth will only be possible” if several actions take place including: “vendors must demonstrate the ability to build on time and to budget, and to reduce the costs of new designs.” Also, both governments and the industry need to maintain and improve safety. If such advances do occur, then new nuclear plants could provide a moderate amount of the needed new carbon-free power for the 2°C scenario.

For the medium term, the Department of Energy and others have been working to develop small modular reactors that could start to ramp up production in 2030 and beyond. Constructed in factories, these reactors would cost $3-5 billion each. Ideally, they would be much safer than the large reactors. But because they are smaller and generate much less electricity, it’s not clear that their cost per kilowatt hour of delivering electricity would be much lower than current nuclear plants.

The final wild-card is the possibility of one or more major nuclear disasters in the coming years, which might once again impact existing and planned nuclear reactors. An April 2015 study “compiled the most comprehensive list of nuclear accidents ever created and used it to calculate the likelihood of other accidents in future.” Researchers concluded “In dollar losses we compute a 50% chance that (i) a Fukushima event (or larger) occurs in the next 50 years, (ii) a Chernobyl event (or larger) occurs in the next 27 years and (iii) a TMI event (or larger) occurs in the next 10 years.”

Bottom Line: If the world is able to put itself on the 2°C path in the coming years, and if the nuclear industry can resolve a variety of issues and avoid a major disaster, then nuclear power can make a modest but important contribution. At the same time, the IEA and many others have concluded that new renewable energy will play a far bigger role in the transition.

The post Nuclear Power Will Play Only A Modest Role In Stopping Climate Change, Nuclear Agency Says appeared first on ThinkProgress.

Bob Meinetz's picture
Bob Meinetz on May 20, 2015

Joe, a big LOL for your claim that

A key reason new reactors are inherently so expensive is that they must be designed to survive almost any imaginable risk, including major disasters and human error.

I was under the impression nuclear’s expense was the result of calculated risk – but instead we’ve been relying on the rock-solid antinuclear imagination as a reference.

I don’t know whether to be relieved or profoundly disappointed, but I would guess you’re not the best person to ask.

Joris van Dorp's picture
Joris van Dorp on May 20, 2015

I always love the ‘negative learning curve’ gobbledygook which Joe Romm has rarely failed to push into his readers faces whenever nuclear comes up.

Any sane person would look at those cost trends and conclude there is something very strange going on with nuclear power plant costs in those countries. After all, no other utility product of human ingenuity in the history of the industrialised world has ever become nominally more expensive over time.  Only nuclear, in some countries. A sane person would want to know why that is, not merely that it is.

A sane person would also be cognizant that nuclear power has had a devastating effect on the market for oil-fired electricity generation in the last century, all but eliminating the use of oil for electricity in all countries which adopted nuclear, and severely reducing coal and gas as well in many nuclear countries. In fact, nuclear is the only non-hydro, non-biomass technology which has taken on fossil fuels on a level playing field, at scale, and won big time! A sane person would not overlook that striking fact. Not in these times of spiralling oil-related conflict and the monstrous ongoing global failure at reigning in rampant greenhouse gas emissions.

Finally, a sane person would be aware of the fact that nuclear power is also the cleanest and safest source of electricity ever developed, on a straight per kWh basis, even despite accidents like Chernobyl, TMI and Fukushima. Such a sane person would note that fact very seriously indeed.

But not Joe Romm. Never Joe Romm. Until today. In this article, it is the first time that Mr. Romm has had something positive to say about nuclear power, although he was quick to also regurgitate his standard anti-nuclear fare as if to cleanse himself from what was probably hard for him to do.

Yet this is progress in my opinion. I would not be very surprised if Mr. Romm becomes one of the most influential born-again pro-nukes in a few years. Mr. Romm may not be quite sane (in my humble opinion) but he is obviously not stupid, and he is a prolific blogger, so if he allows his anti-nuclear bias to fray around the edges like he seems to be doing in this article there is no telling what might happen in future.

I venture that the impending implosion of the German ‘Energiewende’ could be the tipping point for Romm, and presumably some of the other high-profile anti-nukes. That implosion could happen any time between now and the next few years, I suppose. During the same period, an increasing number of Chinese nukes will be coming online, and the data coming from the Chinese nuclear program will make it increasingly difficult for Romm to keep peddling his ‘negative learning curve’ argument.

Time will tell.




Bob Meinetz's picture
Bob Meinetz on May 20, 2015

Joris, neither the title of his piece:

Nuclear Agency: Nuclear Power Will Play Only A Modest Role In Stopping Climate Change

nor the softened first line:

Nuclear power can play a modest, but important, role in avoiding catastrophic global warming.

are found anywhere in IEA’s report, but are fictions Joe has invented en route to rewriting his personal history of antinuclearism. For those wondering what on earth IEA did say, here’s an exact quote of the first Key Finding in the report (p5):

Nuclear power is the largest source of low-carbon electricity in OECD countries, with an 18% overall share of electricity production in 2013 and second at global levels with an 11% share. The updated vision for the 2014 Nuclear Roadmap – based on the 2 degrees Celsius (°C) scenario (2DS) of Energy Technology Perspectives: Scenarios and Strategies to 2050 (IEA, forthcoming 2015) – sees nuclear continuing to play a major role in lowering emissions from the power sector, while improving security of energy supply, supporting fuel diversity and providing large-scale electricity at stable production costs.

Sadly, Joe seems to have convinced himself that some interpretation of the Roadmap is

…what I’ve been arguing on Climate Progress for a long, long time.

Hmm. Anyone who’s been reading Joe’s column for the last decade recognizes this egregious deception for what it is, but hey – I agree Joris, in the grand scheme it represents progress. Now is not the time to beat on this influential voice for climate, but wish him luck and godspeed to his journey’s inevitable endpoint – and that he might redeem himself by blazing a trail for Helen Caldicott, Arnie Gunderson and others to follow.

Nathan Wilson's picture
Nathan Wilson on May 20, 2015

Regarding the notion of negative learning curves, I think the results of the solar industry can be instructive.  Way back in 2008, Romm wrote this article in Salon, in which he argued that concentrating solar thermal electric power (CSP) plants would eventually eclipse PV and wind to become “the most important form of carbon-free power in the 21st century“.  The prediction was based primarily on the relative ease of storing thermal energy, and the use of plentiful construction ingredients (steel and glass).  This obviously hasn’t happened yet, in fact CSP appears to have a flat or negative learning curve, compared to PV which has come down in cost admirably.

So why has CSP failed so badly compared to PV?

  • PV can be located in any region (like nuclear), whereas CSP is best suited to deserts.
  • PV was pushed very hard by a major government (Germany) but CSP was not.
  • The PV deployment rate steadily grew, whereas the CSP deployment rate was inconsistent (like nuclear).
  • PV has grass roots advocacy and even though most PV is utility scale, there is still a residential component that helps promote awareness.
  • Note that the toxic and exotic chemicals used in PV (and battery) manufacturing have not held back PV, even though CSP uses wholesome steel and glass.
  • Note that the extremely esoteric physics of the photo-electric effect, valence versus conduction band electrons, electron-hole recombination, and the complicated way that PV cells actually make electricity has not harmed PV’s image either (even though the sun warms our skin the same way it does the steam in a CSP systems).
  • The historic cost reductions for PV are exaggerated by starting the cost curve before the first 10 MWatts (i.e. satellite power), which greatly inflates the early costs.  If we included data for other energy sources from this low volume period, they too would show 99% cost reductions.

I therefore conclude that the primary factor needed for nuclear to resume a normal learning curve is a critical mass of positive support from the environment community, which will lead to support from government.  I think in colder regions, the introduction of district heat networks supplied by nuclear combined heat and power plants will help too; the active choice of cancelling the oil or gas service and converting to safe and clean district heat will make people feel involved.

Another major issue is the problem of the anti-nukes not wanting to admit that they were wrong (and therefore helped prolong the fossil fuel era by 40 years).  A good solution to that would be advocacy of modern SMRs and Gen III technology which can boast of many safety improvements (e.g. no Chernobyl-style positive feedback explosions and no Fukushima-style melt-downs in response to multi-hour-long station black-outs).  An even bigger boost will come with nuclear plants that use TRISO fuel, such as the 105 MWe HTR Pebble-bed Modular reactor now under construction at the Shidaowan site in China, as well as the salt-cooled FHR reactors under study in China and the US.  This remarkably robust fuel form makes it easy to explain why the resulting reactor is melt-down-proof, and that the spent fuel is not a threat to safety.  It will decrease the cost penalty for dry-cooling.  It might be used with thermal energy storage, to help compliment solar PV.  It may even allow a stream-lining of the onerous regulation and liability concerns that hamper nuclear plant construction.

Joris van Dorp's picture
Joris van Dorp on May 21, 2015


I had to chuckle at “wholesome steel and glass.” Very apt! Love it.

I also like TRISO fuel, and I happen to believe TRISO fuel based nuclear is the way to go, at least in the short-term, and I am trying (with about as much success as others before me, meaning very little) to (re)awaken interest in this brilliant technology in my country. I recognise that this technology has a number of unique benefits compared to the other (also very good) nuclear technology options.

Concerning CSP, I spent some time a decade ago actually working with a group of people advocating this technology, because I too believed it was superior to PV. Essentially, CSP technology allows the addition of (thermal) storage in a relatively simple way as you know, while storing the electricity from PV is much harder and more expensive.

So I suppose I would add one more reason to your list of why CSP has not been as successfull as some people – including myself – expected at one point, and that is that PV has (up to now) somewhat surprisingly never had to internalise the high external costs of it’s horrific intermittency and variability performance. If PV and CSP would have had to compete on a level playing field on this account, for example by demanding that they each supply at least a minimal amount of (quasi) firm capacity, then CSP would have been probably been seen to be superior from the start and perhaps have recieved more support and hence done better on the learning curve.

For what it’s worth, CSP in the desert does have it’s share of challenges, most obviously the fact that it’s in the desert. People who have never been to a desert tend to underestimate the harshness of desert conditions. There is a good reason that deserts consist largely of dust. It is because that’s what deserts do: turn things put into them into dust. Including CSP installations.

Joris van Dorp's picture
Joris van Dorp on May 21, 2015

I agree.

As an aside, I watched a rather discomforting documentary about “Scientology” a few days ago, and the take home point for me was the extraordinary effort human beings tend to expend on rationalising and denying facts, once they have commited themselves to a particular cause. I suppose outspoken opponents of nuclear power suffer from a similar kind of self-inflicted mental incarceration, which only gets worse every time the act on their conviction.

I can understand this. In my time, I have also wrestled with (relatively minor, but still painfull to renounce) prejudices, and I suppose most people have to deal with such things at one point in their lives. I am still concerned about the ongoing risk of developing attachment to false convictions even today, because it is so easy to fall into that trap. It requires an almost obsessive level of self-denial to be one’s own worst critic at all times, but that is the only way to protect oneself, I believe.

Influential anti-nukes like Romm, Gunderson and Caldicott face the prospect of recognising the damage they have done, if they ever decide to entertain even the possibility that they may have been wrong about nuclear power. It would take almost superhuman psychological resillience to honestly admit that you inadvertently:

– caused people to develop morbid radiophobia leading to economic and social hardship, depression, waves of unnecessary abortions and even suicide.

– helped deepen one of the most important causes of ongoing resource wars in the Middle East, namely the persistent non-existence of viable alternatives to natural hydrocarbons as an energy source, while nuclear is being suppressed.

– helped sully the reputations of countless scientists and engineers in the nuclear sector who are merely trying to (actually) protect the environment and improve the human condition.

– helped undermine the public debate concerning energy and climate by endorsing myriad biased ‘studies’ created by dedicated anti-nuclear propaganda outfits.

– helped contribute to relentlessly increasing global greenhouse gas emissions

– been blind to the countless occasions when others have pointed out fatal flaws in your reasoning

– prevented millions of people from gaining access to affordable, reliable and clean energy

Hops Gegangen's picture
Hops Gegangen on May 21, 2015


Having done some safety analysis, I’m sure that some very practical and quantitative engineers are at some point in the process creating a list of hazards based on what they can imagine. You don’t have to be anti-nuclear to do that.

The trouble is, the tolerable hazard rate is very low, so you have to mitigate even rare events like the once in 1000 years magnitude 8 earthquake.




Joris van Dorp's picture
Joris van Dorp on May 21, 2015

By the way, it is interesting to see in the graph of nuclear construction starts in Mr. Romm’s article that the decline of nuclear power construction started years before TMI happened. This is a point noted by several chroniclers of the history of nuclear power, who explain that the original decline of nuclear power had little or nothing to do with the occurance of accidents. I remember Bernard Cohen had an interesting chapter in his book “The Nuclear Energy Option” about what really happened. And Rod Adams has a long list of articles about the apparant role of various fossil fuel interests in causing and exacerbating nuclear’s troubled history.

The recent slow-down of construction starts caused by Fukushima is obvious from the graph, but it can be seen that TMI and later Chernobyl merely hastened a decline which had already set in years earlier. This should invite those interested in the history of nuclear power to find out why this decline set-in like it did, and what coiuld be done about it (something which Bernard Cohen also discusses in his book, linked to above).

Bob Meinetz's picture
Bob Meinetz on May 21, 2015

Hops, competent engineers do not consider the “survival” of a reactor, “imaginable” risks, or “major” disasters (epidemics? revolutions?). These generic legacies of The China Syndrome make great copy and draw banner advertising to Climate Progress, but would be laughable were it not for the deplorable state of education about everything nuclear.

Paxus Calta-Star's picture
Paxus Calta-Star on May 22, 2015

The authors of the study write:

 “vendors must demonstrate the ability to build on time and to budget, and to reduce the costs of new designs.” 

And here is the rub.  Not only have vendors not demonstrated this, every reactor under construction in the US and western Europe is late and over budget.  Olkiluoto-3 Finland, 9 years late 270% over budget, Flamanville-3 France, 6 years late at least 250% over budget many have to replace pressure vessel at huge cost and delays.  Watts Bar 2, US – decades late, over 400% over budget, Vogtle 3&4 at least a year late (and construction just started) and at least a $billion over budget.  Summer 2&3 at least a year late and by some estiamtes $4 billion over budget already.

With every single example failing in this region which has more than half of the worlds reactors Wall Street remains signulrly uninterested in this failed technology.

Nathan Wilson's picture
Nathan Wilson on May 22, 2015

Yep, while Wall Street focuses all their efforts on the fossil fuel industry, the rest of the world has 65 nuclear reactors under construction, worth 68 GWatts when they come on-line.  While we stand on the side-lines, China is trying to become the Saudi Arabia of nuclear power:

The Chinese are on schedule to hand a humiliating defeat to the European nuclear industry, by being the first nation to get an EPR (European Pressurized Reactor) on-line in 2016 (ahead of  Olkiluoto-3 and likely ahead of Flamanville).

The Chinese should also have the world’s first AP1000 (US designed) reactors on-line next year; no humiliation here, due to the fact that they are planning to honor the US by building up 24 more of the design.  Our glory will be short-lived however, as they have already broken ground on their new Hualong-One design, a “Gen III” design with modern safety features, and full Chinese ownership of all intellectual property (this reactor is likely to be a huge seller, as 6 of them are already being planned for China, plus exports to Pakistan and Argentina, and they’ve started paperwork requesting approval in the UK).

Paxus Calta-Star's picture
Paxus Calta-Star on May 22, 2015

And you are fully satisfied with the Chinese nuclear regulator agencies and are confident these will instill a great safety culture at all these reactors?

Nathan Wilson's picture
Nathan Wilson on May 22, 2015

I believe that particularly with modern Gen III reactors, government regulations do not play the biggest role in insuring safety. The regulations do make it easier for companies to maintain competence, but regardless of how a company feels about safety culture, every company is profit motivated, and the accident at Fukushima and the nuclear reactor destroyed by flawed maintenace at Crystal River provide extremely strong incentives for companies to operate their reactor safely. 

But in any case, the accidents at Chernobyl and Fukushima have taught us that nuclear accidents are simply not that dangerous (i.e. zero fatalities from Fukushima radiation).  There is simply no question that nuclear power is better for the environment and better for human health than fossil fuel use (this includes realistic combinations of renewables with fossil backup).

Paxus Calta-Star's picture
Paxus Calta-Star on May 23, 2015

I wonder if the 4000 premature dead from Chernobyl would agree with you?  This is the lowest estimate byfar form WHO and the IAEA. Most other estimates, including those using the same data, start with fatality levels at 27K.  See 


Similarly the 80K people still displaced because of the Fukushima triple meltdowns are not rushing to agree with you that Japans environment is improved by operating reactors.  

The idea that profit motivated companies need not be regulated because they have learned from Fukushima and Chernobyl, flies in the face of the fact that they did not learn from TMI or WIndscale.  The best profit maximizing strategy is clearly buy off politicans and capture regulators and in the event of catastrophic accident push the costs off onto the state through liability scams like Price Anderson.

Nathan Wilson's picture
Nathan Wilson on May 23, 2015

A few thousand dead from Chernobyl, over many decades?  Are you not aware that pollution from coal combustion kills 10,000 American and 22,000 Europeans each year? (and millions more globally).  This report finds that the proposed Swedish nuclear phase-out would result in 50,000 additional energy-related deaths (due to pre-mature closure of the existing fleet).

So given the proven advances in nuclear safety (Gen III reactors are projected to have serious accidents many orders of magnitude less often than the RBMKs like Chernobyl), I’d say you’re barking up the wrong safety tree.

Paxus Calta-Star's picture
Paxus Calta-Star on May 23, 2015

No one is proposing we replace nuclear with coal, please update your rhetoric.  The Japanese and the Germans are doing quite a fine job of fast phase out of nuclear with efficiency and real renewables.  Or perhaps you have not noticed the tremendous drop in renewables prices and the break through battery technology coming on line.  With new nuclear costs seriously uncontrolled in the west, banks will look to renewables, distributed generation and natural gas for new generation investment in the coming time.

Paxus Calta-Star's picture
Paxus Calta-Star on May 23, 2015

And the idea the untested Gen 3 reactors are super safe is also fanciful.  It was computer safety models by the same utilties which were used to justify previous generations of reactors, including all those which failed.  Saying “our new computer models prove these designs are safer” is as compelling as the endless promises that new reactors will be cheaper than other energy sources.

The tide is against nuclear and fossil sources as extraction techniques with increasing fuel costs compared with real renwables which are harvest techniques and get cheaper and more efficient with time using free fuel and not having high waste handling costs.

Nathan Wilson's picture
Nathan Wilson on May 23, 2015

The Japanese and the Germans are doing quite a fine job of fast phase out of nuclear with efficiency and real renewables.”

Acutally, the Japanese have increased their use of imported fossil fuels (including coal) by about $30 billion per year, and the Germans are keeping their coal consumption about the same.

The fundamental assumption that reducing the use of nuclear power is more urgent than reducing the use of fossil fuel (which is the basis of the German and Japanese policies) is refuted by modern science; use of fossil fuels is enormously more harmful to human health and the environment.  The fact that much of the public holds this assumption is a very bad thing.

I looked at the article (on your website) that you linked, which claimed that the Japanese would be better off not re-starting their reactors.  It was an unconvincing and unsupported rant.  The economic cost of the shutdown run in the tens of billions per year, the CO2 emissions have skyrocketed, and there is zero benefit.  It is a great shame that you waste your talents advocating such destructive policies.

Alistair Newbould's picture
Alistair Newbould on May 23, 2015

In 1981 I was part of a sit in protest against the building of Torness nuclear power station. I was young, ignorant, opinionated and wrong. I apologise.


I am trying to make amends rather late in life, by minimising my personal carbon footprint and advocating for real progress on this issue, using whatever (moral) means we have available.


Alistair Newbould

Clayton Handleman's picture
Clayton Handleman on May 25, 2015

Romm’s predictions were reasonable in 2008.  There were contracts for a number of CSP projects at the GW scale that would have offered us the opportunity for a consistent build out and to see how scaling affected the learning curve for CSP.  These included nearly a GW of Dish Stirling engines.  The extent of the Chinese commitment to PV was just beginning to become evident  in 2008.  When PV prices collapsed due to the extraordinary Chinese efforts, there was a pullback from solar thermal.  It is absurd to point a finger at Romm on this, none of the mainstream thinking, at that time, was that PV was going to take an unquestioned price lead over CSP in under a decade. 

Your comment about the PV industry going back to satelite days sinks to the level of propaganda.  You are correct that the curves for PV go back quite far.  People started keeping records carefully starting in the 70’s when a concerted effort was begun to develop PV for utility scale power production.  However it is not some nefarious plot to create a distorted picture of costs.   It was because the industry (starting with Paul Maycock in the Carter years and working meticulously into the 90’s) wisely, kept records back that far and they are useful at showing an ongoing, reliable trend. I collaborated with Paul, who worked tirelessly to keep records, calling and visiting PV companies to establish as accurate a dataset on production as possible as the industry ramped so that we could understand, with confidence, the learning curve for PV.  His integrity was unquestionable and he earned the trust of the majority of PV manufacturers.  Despite the highly proprietary nature of their production costs and process information they shared with him the information he needed to profile the industry costs in his annual updates and projections. 

But if you don’t want to start at 10 MW you can jump ahead to 100 MW and still the conclusion is the same and again to 1GW – same, the experience curve is 17% cost reduction for every cumulative doubling of capacity.  And it has been 17% for 40 years.

You, on the other hand, are making stuff up.  You have conjectured something about other sources following similar curves.  But you have provided no data.  Not even poor data.  And, since the cost of FF electricity has a significant fuel component the up front capital is less important anyway. 

The primary reason for the premature slowdown in solar thermal was due to the surprise that PV dropped in price as quickly as it did. 

As is illustrated clearly below the learning curve is very steady pretty much wherever you choose to start.  The price of PV stalled at $4:00 per watt as the industry shifted from using the leftover silicon from the electronics industry to building dedicated foundaries to produce silicon for the PV market.  As such there was a shortage of silicon for several years which kept costs high.  During that time the Chinese were declaring economic war on the Western PV industry by pouring billions into massive production capacity.  The silicon crisis eased with improved, lower cost silicon feeding the new scaled up PV fabs to create a perfect storm and collapse the price of PV.

Here is an overview of how experience curves offer a simple and reliable way to predict cost reductions in manufactured products.  This link offers experience curves for a number of renewable technologies.



Bob Meinetz's picture
Bob Meinetz on May 25, 2015

Paxus, if you’re using “late and over budget” as your criteria for failure, you will find that something north of 90% of all construction projects, including wind and solar, are abject failures.

Nuclear energy is an extremely potent source of energy, and because of that is an extremely powerful tool for combating global warming. If we only rely on benign, ineffectual sources of energy like wind, solar, small hydro, etc. we run very little risk of localized accidents resulting in injury or fatalities. We also make combating global warming exponentially more difficult, and for that reason it’s critical we don’t let irrational fears guide energy policy.

Clayton Handleman's picture
Clayton Handleman on May 25, 2015

More sillyness from a fact starved anonymous Internet poster. 

Of course the difference between renewables and nuclear is that the ratepayers and taxpayers are forced to pay for nuclear gone bad.  The track record for wind and solar is very good in terms of project timelines and most importantly project investors are the ones who eat the penalties when projects go over budget or blow their timelines.  With 40 years of commercialization it is time that the risk for nuclear be passed forward to investors who buy in voluntarily. 







Nathan Wilson's picture
Nathan Wilson on May 25, 2015

“… the experience curve is 17% cost reduction for every cumulative doubling of capacity … You have conjectured something about other sources following similar curves.”

Yes, it is a generally accepted fact that the learning effect is universal, until some limit is reached.  I have seen no compelling arguments that either PV or nuclear are unique; the only exceptions I know of are software and micro-electronics.  PV is not software, and does not share the underlying mechanism of Moore’s law (shrinking components via lithography).  Your comments on the early PV industry support my claim that it is the reporting of PV’s learning effect that is the only unique aspect, not the effect itself.

To summarize: my hypothesis is that the claimed negative learning curve for nuclear is the result of confounding factors.  I have cited the example of CSP as another technology which has failed to meet the expected learning curve, and claim that both CSP and nuclear could display a normal learning curve under the right circumstances (a sustained push by a competitive industry along with government and public support).  However nuclear could beat CSP, for some of the same reasons that PV has beaten CSP.

To support my claim that nuclear has not reached a limit, I would offer this 2005 report from UC Berkeley, which shows that the steel and concrete material inputs used for nuclear plants is about 50% of that required for coal power, and about 10% of that for wind power.

The “negative learning curve” is like cold fusion; it is possible, but it is so far removed from everything else we know that there is a very high burden of proof. 

Clayton Handleman's picture
Clayton Handleman on May 25, 2015

“To summarize: my hypothesis is that the claimed negative learning curve for nuclear is the result of confounding factors.  I have cited the example of CSP as another technology which has failed to meet the expected learning curve, and claim that both CSP and nuclear could display a normal learning curve under the right circumstances (a sustained push by a competitive industry along with government and public support).”

I think you are right, eventually nuclear could reverse its learning curve and move in a favorable direction.  However what is consistently avoided or denied is that the negative learning curve effect is not an irrational reaction to nuclear power.  It is in fact a response to a highly complex and dangerous method of generating power.  The initial commercialization made reasonable efforts at safety but there was much to be learned after deployment.  Compounding that was a nuclear regulatory environment that was not transparent and that was captive to the industry leading to poor safety regulation.  As the curtain was lifted there were many problems uncovered.  That is what is at cause and the industry that grew to 1/5th of the US power producing fleet cannot claim to be an immature industry at this point.  Yet even the most recent efforts are fraught with difficulties and it is unclear whether the negative experience curve has played out yet.

Solar thermal is a different animal.  It is relatively immature and to the degree that it had a chance at roll out it demonstrated that it would ramp down the experience curve.  This was done by Luz in the 1980’s and the SEGS plants are still running today.  They demonstrated rapid cost improvements with scale.  I think we would be well served to see an orderly roll out of on the order of 10 successive projects to assess the experience curve for that technology. 

This would constitute a tiny fraction of what the US spends on, for example, the F-35 bomber program and probably yeild much better results in terms of national defense since solar in conjunction with EVs reduces our reliance on over seas oil.  So arguments that it is money poorly spent really are not very meaningful.

Bob Meinetz's picture
Bob Meinetz on May 25, 2015

Clayton, in another thread on TEC there’s a discussion of solar power in North Carolina, where taxpayers have paid $600 million (30% of $2 billion) for six-tenths of one percent of NC electricity. As a proportion of actual generation, solar’s horrendously poor performance in most of the U.S. makes nuclear cost overruns pale by comparison.

Yes, I’m an anonymous internet poster, but when I make mistakes I’m usually pretty good about coming clean on them. I like facts – the truth is good. Which raises the issue of your claims to be a physicist and an engineer – you’ve persistently avoided providing any accounting of your training in those fields. It’s a valid question, Clayton – what’s your training, or did you bestow those titles on yourself?

Clayton Handleman's picture
Clayton Handleman on May 25, 2015

Your comments on the early PV industry support my claim that it is the reporting of PV’s learning effect that is the only unique aspect, not the effect itself.”

No, your claim was that the reporting of PV’s learning curve somehow exaggerated it.  My numbers showed that the industry has reported it accurately and based the reporting on carefully gathered data.  It is not a black mark on Solar that the industry does a good job of this it is a problem with other industries if they don’t. 

There also has been considerable data gathered on the Nuclear industry showing that the costs continue to escalate.  Most who support nuclear try to wish that away.  It is fact, the costs have risen dramatically.  And if people want to see nuclear happen it would make much more sense to be looking at why, even today, even with over 4 decades of commercial experience, the nuclear industry continues to sell product with poor quality.

A valuable discussion would be to look at how the nuclear industry could clean up its act.  What is systemically wrong with it.  I suggest moving to an investor driven model instead of the old utility model.  But on this board, the effort is instead focussed on diverting attention to what is wrong with renewables and suggesting some conspiracy against Nuclear.

I think one of the reasons renewable energy is succeeding in the face of what might be argued is a harder set of problems, is that the renewables industry acknowledges the problems and embraces the discussion.  Yes intermittency is a problem.  Now how can we solve it, yes storage is a problem, how can we solve it etc.:

Nuclear, when a problem is presented what we see is deflection, denial, and claims that the other approaches are foolish.  I would love to see you come on and say, gee, there have been a lot of safety issues identified, here are some real ones, X,Y and Z have been solved but A, B and C are still problematic and here are ways that they might be addressed.  Now that would be an interesting and engaging and useful discussion.

Clayton Handleman's picture
Clayton Handleman on May 25, 2015

Each poster and commenter on TEC has a profile.  You can review that profile to find any personal information that they choose to provide.  My credentials are readily verified in my profile or through simple google searches on my name.  I look forward to your apology. 

As to admitting you are wrong, I look forward to seeing examples of that in the comment thread, to date what I have seen is that when confronted with facts you resort to insults or changing the subject. 






Bob Meinetz's picture
Bob Meinetz on May 25, 2015

Clayton, I can understand why as a renewable energy supporter you might like to hear a real engineer like Nathan say, “Gee, there have been a lot of safety issues identified with nuclear energy.” There are a couple of reasons why you’ll probably be disappointed:

  • It’s not true. On a per unit of energy basis, nuclear is the safest form of electricity generation.
  • Real engineers seldom use the expression “Gee…”

Nathan Wilson's picture
Nathan Wilson on May 25, 2015

“… if people want to see nuclear happen it would make much more sense to be looking at why … the nuclear industry continues to sell product with poor quality.  A valuable discussion would be to look at how the nuclear industry could clean up its act.”

Ok, this is an area where we disagree pretty deeply.  My understanding of the existing nuclear power industry is that if we used only today’s nuclear technology, quality, and cost (assuming no learning, which I consider improbable), and grew the current US nuclear fleet to displace 100% of coal and 50% of gas use in electricity generation:

  • based on EIA cost estimates and a gradual (40 year) transition, any increase in the cost of electricity would be small or negative, and over-whelmed by saving of the externalized cost of fossil fuel combustion.
  • the effects of any safety deficiencies and accidents would cause enormously less harm to human health and the environment than continued fossil fuel use.
  • the fuel sustainability of the needed uranium would be better than the fuel sustainability of our fossil fuel system.

Given the environmental importance of substantially reducing emissions from fossil fuel use, and the fact that nuclear is the only scalable non-fossil energy source which is proven at high penetration, I find the notion that the nuclear industry must make X or Y improvements before gaining the endorsement of the environment community to be completely irrational and self-defeating.

The hope that a non-nuclear portfolio based on renewables might some day be able to offer the same benefits as one rich in nuclear, but at a lower cost (but higher environmental footprint) is in my mind, not worth the risk of fossil fuel lock-in.  The implied benefit of non-nuclear paths that is most often hinted-at is the avoidance of some unstated catastrophe; I’ve seen no evidence of a plausible nuclear power catostrophe, in spite of numerous historic nuclear accidents which all turned out to be uneventful, except for the over-zealous political responses. 

Bob Meinetz's picture
Bob Meinetz on May 25, 2015

Clayton, your bio on your website shows no evidence of any training whatsoever:

which leads me to believe calling yourself a “physicist” and an “engineer” is a gross misrepresentation of your qualifications. Absent any reference thereto, I think it’s your readers who deserve the apology.

Clayton Handleman's picture
Clayton Handleman on May 25, 2015

“Clayton, your bio on your website . . . “

Last time I checked this was TEC not my blog.  But by all means, if you have nothing better to do, please feel free to go through my blog and suggest changes.  I am employed so my time is limited and an editor would be very helpful.  Seems like you have lots of time on your hands, lets exchange contact info so you can send along your recommended edits – . . .  oh that won’t work will it?  You are hiding, anonymous and all that, never mind.  But please don’t keep cluttering TEC comment section with your drivel.  Of course you are aware, there is a thorough bio of me on TEC.  But you have gone fishing in my blog, one wonders why. 

Perhaps because you cannot debate me effectively so all you have left is to attack my credentials while hiding pathetically behind the anonymity of the Internet. 

There are huge monied interests that would like nothing more than to poison the discussion threads on reputable blogs such as TEC.  No doubt there are people whom they pay to subvert the threads.  There are many tactics that trolls can use effectively.  Anonymity is one, another is to divert threads to irrelevant red herring discussions.  Another is to attempt to discredit contributors.  If you keep this up Bob, people may start getting the wrong idea about you, or, is it the right idea? 

You have made the point that credentials matter, why don’t you set a good example and provide yours?  Do you have any?  Or let us know what you are hiding from so we won’t come to the wrong conclusion. 










Bob Meinetz's picture
Bob Meinetz on May 25, 2015

Clayton, I wouldn’t dare assess my own debating credentials. I do the best I can, then leave that up to other readers to judge.

I’ll be the first to tell you I’m not an engineer, and I’m not a physicist. But the wonderful thing about internet debate is that because misrepresentation is rampant, claimed credentials don’t stand for much –  it’s up to everyone to create a compelling argument based on facts and analytical skill. Of that I have a fair amount, as well as a fairly accomplished eye for spotting frauds.

So when you label yourself an “engineer” and a “physicist” then are incapable of providing any details of your education, that tells me a lot more about your intellectual honesty than technical proficiency. By doing so, you disingenuously attempt to put yourself in company with people on this board who have studied years to earn those titles. Instead, you prove you’re not above fibbing when it might advance your agenda. I’m not sure why you’d be flattered – it doesn’t do you or your agenda any favors.

Clayton Handleman's picture
Clayton Handleman on May 25, 2015


“then are incapable of providing any details of your education, that tells me a lot more about your intellectual honesty than technical proficiency.”


Your ongoing missrepresentation of my biographical information is unfortunate and I hope TEC moderators will take notice.  In any event for anyone who has followed this thread, my bio can be found by clicking on my name or HERE.  My college credentials are listed at the bottom of the bio and have been for some time. 

 Here is An interesting piece on trolling.



Joris van Dorp's picture
Joris van Dorp on May 26, 2015

Paxus Calta-Star wrote:

“No one is proposing we replace nuclear with coal, please update your rhetoric.”

Sadly, you are absolutely wrong.

In my country, The Netherlands, one of our largest and most prominent NGO’s for sustainability (called “Natuur en Milieu” or “Nature and Environment” in English) *DOES* in fact state that coal power is better for humanity than nuclear power. They stated this in a personal email to me, in response to a question of mine about what their opinion of nuclear power is.

I suppose you’ve just displayed the worst flaw of popular anti-nuclearism. It is the ridiculous *ASSUMPTION* that fossil fuels will replaced by bona fide renewable energy *IF ONLY* nuclear power is destroyed (by spreading lies about nuclear technology and radiation health effects).

Joris van Dorp's picture
Joris van Dorp on May 26, 2015

“With 40 years of commercialization it is time that the risk for nuclear be passed forward to investors who buy in voluntarily. “

Nonsense. Not all risks should be passed onto investors. Some examples of risks which need to be born by the public:

– the risk of costly changes in nuclear power regulations occurring after a project has been given the go-ahead

– the risk of sabotage by the anti-nuclear movement, for example by delaying construction using blockades, spurious legal challenges, or bombing construction sites or the assassination of project executives.

– the risk of policy changes impacting the overall economics of the project, after a project has recieved approval, including:

a. policy changes which artifically reduce the price of non-nuclear electricity, such as market-distorting subsidisation of certain sources of electricity.

b. policy changes which artificially increase the cost of nuclear electricity, such as arbitrary taxes on nuclear energy.

It is these risks which spook “Wallstreet”. And understandably so. The easiest way to solve these risk is probably by loan guarantees. Such loan guarantees have a double benefit:

1. They reduce the cost of capital, thereby reducing the price paid for nuclear energy. In effect, more of the benefit of nuclear power goes to ratepayers, as opposed to the financial markets.

2. They cause the public to have a stake in the project, reducing the chance that the public will approve of blatant sabotage perpetrated by the anti-nuclear movement.

Loan guarantees don’t protect against the risk of electricity market distortion. To protect against that, Power Purchase Agreements could be used. However, as the situation in Germany shows, politically inspired market distortion is a wrecking ball which hurts everybody, and protecting against that risk is perhaps impossible. It takes an engaged and knowledgeable public to prevent such market distortion. Nuclear power plants especially, but also renewable energies – due to their high up-front costs – suffer most from political market distortion. Fossil fuel power generation options, where most of the lifetime cost is for fuel and O&M, have a relative advantage for investors in markets wracked by political distortion. It is no coincidence that fossil fuel power generation is doing relatively well in such distorted markets.

Joris van Dorp's picture
Joris van Dorp on Jun 10, 2015

Why do you suppose I haven’t read it?

What makes you think I think they changed their policy?

Why are you not banned from TEC yet, for continuing your campaign of promoting fear-mongering lies about nuclear power?

Paxus Calta-Star's picture
Paxus Calta-Star on Jul 28, 2015

The problem is not massive earthquakes in Japan, meltdowns and near meltdowns can and have happened in the US, by simply bad planning on the part of nuclear utilties.  As we saw in 2011 at the Fort Calhoun reactor in Nebraska.  The only thing which prevented a meltdown here was that it stopped raining.  Hardly reassuring.

There are several commenters who seem worried about China leading the world in nuclear power.  These folks seem oblivious to the fact that in 2014 spent about $9 billion on nuclear, while it spent $83 billion on wind and solar.  China has more real renewables operating already and at this rate obviously this is not going to change.  See

Bob Meinetz's picture
Bob Meinetz on Jul 28, 2015

Paxus, that’s an inaccurate account of what happened at Fort Calhoun. As an example of excellent planning on the part of a nuclear utility, Omaha Public Power District intentionally shut the reactor down nine days in advance in anticipation of flooding:

It was reported on June 17, 2011 that the plant was in “safe cold shutdown” mode for refueling and the anticipation of flooding,and that four weeks’ worth of additional fuel had been brought in to power backup generators, should they be needed.

Not as scary and ominous as your version, but the truth seldom supports the antinuclear script.

Paxus Calta-Star's picture
Paxus Calta-Star on Jul 28, 2015

Bob do you know how reactors work?  Flood of the building is not a good thing.  The reactor being shut down would be no insurance of public safety in the event that the entire structure was innudated with water.  You can’t simply turn them of like a light switch and then run tons of water through the building.  Cold shutdown still requires active systems to maintain public safety.

Omaha got lucky.

Joseph Romm's picture
Thank Joseph 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

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 »