Nuclear math doesn’t add up

Nuclear proponents in the US also ignore the history of cost overruns, construction mismanagement and profiteering in nuclear plant construction. The Washington Public Power Supply System’s (WPPSS) attempt to build two plants resulted in one of the largest municipal bond defaults in US history, $2.25 billion. A reporter friend of mine covered the mess, documenting the use of unapproved substandard steel, welders who couldn’t weld and walls built, torn down and rebuilt repeatedly as they failed inspections. Not exactly confidence building. And we still don’t have a solution for long-term storage of radioactive waste. Amusingly, the abandoned plants in Elma, WA were used as sets for some shots in The Hunger Games.

I think that with the global potential of renewable energy generation nuclear energy doesn’t make sense, not only with the economic costs, but also with the environmental problems of the waste.

There are discussions in Australia about a nuclear waste facility in South Australia, which after a Royal Commission the S.A. government has said “400 people [will] be selected to sit on citizens’ juries to consider the state’s approach to its nuclear future.”
http://www.news.com.au/technology/environment/australians-face-big-decision-on-nuclear-waste-dump/news-story/dac386f5403f21a105642d5bfa7f43c2

Earlier this year China outlined plans for a global solar and wind renewable energy grid to be built by 2050, so countries can share energy across borders. I think this really means that new nuclear power plants are not going to be needed in the future, when there is safer and cleaner energy available from solar and wind RET.
http://futurism.com/building-big-forget-great-wall-china-wants-build-50-trillion-global-power-grid-2050/

The positive news about the the Democratic Party platform is that Russell Greene sponsored an amendment to the platform which was inserted and it includes language about a WW2 style climate mobilisation. The amendment says:

“Democrats believe it would be a grave mistake for the United States to wait for another nation to lead the world in combating the global climate emergency. In fact, we must move first in launching a green industrial revolution, because that is the key to getting others to follow; and because it is in our own national interest to do so.

Just as America’s greatest generation led the effort to defeat the Axis Powers during World War II, so must our generation now lead a World War II-type national mobilization to save civilization from catastrophic consequences.

We must think beyond Paris. In the first 100 days of the next administration, the President will convene a summit of the world’s best engineers, climate scientists, climate experts, policy experts, activists and indigenous communities to chart a course toward the healthy future we all want for our families and communities.”

Russell Greene introduced the amendment by saying:

“Global warming threatens to destabilize the climate system for tens of thousands of years, sending the Earth into a hot state inhospitable to humanity. If business as usual continues for much longer, a chain reaction of proliferating droughts, famines, and subsequent state failures will cause the unraveling of an organized human community. Furthermore, the global economy’s overshoot of the planet’s limits has initiated the 6th mass extinction of species, which threatens to wipe out much of life on Earth within this century. If allowed to unfold, it could take some 10 million years for life to fully recover…

We must call this out explicitly. It is not enough that this is the most progressive platform in Democratic history. What matters is that this matches the moment. What will be remembered from our work here this weekend is not only what we have put in, but what we have left out. Prioritizing a climate mobilization provides the most immediate path to social, economic, and environmental justice. As with the WWII home front mobilization, the climate mobilization will create full employment and drastically reduce inequality, with the principle of fair and shared sacrifice embraced by all Americans.”

http://www.theclimatemobilization.org/dnc_ww2

In news from Australia, the State government of Victoria, where I live, has issued its first Green Bond of $300 million which was fully subscribed in a bit over 24 hours. I think this is the first Green Bond scheme in the world.

http://www.premier.vic.gov.au/victorian-green-bonds-an-australian-and-world-first/

There’s a chance we will get some sensible public economic analysis of the European nuclear power sector with the EU commission investigation of the Areva restructuring, which in turn is closely tied to UK nuclear power expansion. Of course an unstated case for Brexit may have been the freedom to build more nuclear power plants without having to consider market distortion.

http://europa.eu/rapid/press-release_IP-16-2587_en.htm?locale=en.

I guess the small scale reactors being developed offer a bit of hope to Nuclear proponents. Supposedly these are cheaper/safer. I am not enthusiastic. I’m sure a few will be built, but it would take a lot to make a serious dent.

@6:
Again, it not really cost, but practicality of monetisation.
Consumer renewable are sold as a consumer product; the manufacturer gets the money straight away, instead of waiting to make it back by selling electricity.

For utility-scale renewables, the main investment is in land ownership. So long as that land has other uses, there is a credible threat to walk away and use the land for something else. So prices can be kept high enough to make a profit (and if the market does collapse, you have still got the land).

Which is all fair enough as long as you are relying on it being the case that current market conditions and property ownership regimes are naturally going to lead to adequate decarbonisation; nothing needs to change.

However, if you are talking about political action, then that needs to be assessed in appropriate terms, not self-referential ones.

@Richard M #9
In the case of a contaminated nuclear site I’d say there is little credible threat that you’re going to use the land for anything else. If I follow your argument, this would be advantage renewables (at least for commercial management)

@Matt
I think he’s talking about the opportunity cost of the land. The possibility that the owner could not produce power and thus bargain for higher electricity prices. If the producer can’t use the land for anything else, then the buyer knows that the producer will sell the power for any price above their marginal cost of production. (even if it is below their average cost, and thus an unprofitable investment, overall)

The real economic problem with nuclear is that it produces energy too cheaply. All the costs are up front, so the owner has no leverage to control prices.

This is not true either. If nuclear power were the lowest cost producer around, like large scale hydroelectric generators, it would be the last producer left standing in an environment of low wholesale prices. Instead it’s getting squeezed out by low priced natural gas and (to a lesser extent) wind power.

Nuclear generation has by far the highest O&M costs per unit of capacity — $198 per nameplate kilowatt per year on average. Or $198 million per year for a 1000 megawatt plant. If that plant runs at 91.3% capacity factor, the O&M cost per megawatt hour is:

198000 / (24 * 365.25 * 0.913) = $24.74

For comparison, solar is $25 per nameplate kilowatt per year, and normalized by its lower capacity factor that comes to:

25000 / (24 * 365.25 * 0.2725) = $10.47 per megawatt hour.

I suspect that a large part of these costs comes from nuclear labor. The two new AP1000 units at Plant Vogtle are supposed to provide 800 permanent jobs once construction completes. Each unit produces 1117 MWe nominal. At a 91.3% capacity factor that’s:

(1117 * 2 * 0.913) / 800 = 2.55 annualized real megawatts per worker.

The 550 MW Topaz Solar Farm provides 12 permanent jobs in the operational phase. Last year it generated 1301337 MWh, for an annualized power of 148.45 MW, meaning the power-to-labor ratio was

148.45 / 12 = 12.37 annualized real megawatts per worker.

That’s nearly 4.9 times as much annual electricity output per full time employee. For nuclear to lower its operational labor costs below those of utility scale solar, it would have to pay workers a lot less than they would earn in solar operation. But I don’t think that is possible.

@Matt

>That’s nearly 4.9 times as much annual electricity output per full time employee. For nuclear to lower its operational labor costs below those of utility scale solar, it would have to pay workers a lot less than they would earn in solar operation. But I don’t think that is possible.

The trouble with this comparison is that there’s a lot of work that goes into a power plant that isn’t directly onsite in terms of things like spare parts, mining and transporting fuel, etc. A solar farm doesn’t need a lot of people onsite, but it does need a factory somewhere cranking out replacement panels. I can’t prove it, but I’d bet a bundle that nuclear would be on the low end for offsite labor because of the robustness of the initial construction and the low demand (volume wise) for fuel.

The trouble with this comparison is that there’s a lot of work that goes into a power plant that isn’t directly onsite in terms of things like spare parts, mining and transporting fuel, etc. A solar farm doesn’t need a lot of people onsite, but it does need a factory somewhere cranking out replacement panels. I can’t prove it, but I’d bet a bundle that nuclear would be on the low end for offsite labor because of the robustness of the initial construction and the low demand (volume wise) for fuel.

Spare parts and fuel are included in O&M costs. As shown above, nuclear O&M runs to $24.74 per megawatt hour and utility scale solar PV O&M runs to $10.47 per megawatt hour. The lower advantage ratio here — 2.4 to 1 in favor of solar, rather than the 4.9 to 1 ratio in favor of solar for onsite labor efficiency — does indicate that relatively more of solar’s O&M expenses are offsite. But it’s not more expensive on an absolute basis.

Are those overruns worse than overruns for non-nuclear projects, or do they just get more attention because of the large sums (nuclear power being paid for up front) and the degree of government involvement?

Construction Cost Overruns and Electricity Infrastructure: An Unavoidable Risk? reviewed utility scale generation cost and schedule overruns. Large scale hydroelectric projects were the worst, then nuclear, then thermal combustion generation, and solar PV was by far the best at sticking to budget and schedule. The authors offer several explanations for the different patterns found. The simplest correlation I notice is that larger projects are more likely to encounter problems.

There is no such thing (yet) as a commercial nuclear reactor that you can ship assembled from the factory, like an aeroderivative gas turbine, and use to generate a few tens of megawatts, like an aeroderivative gas turbine. If you want to build a nuclear reactor in the US or Europe you’re currently committing to more than 1000 megawatts at a time. Nuclear power is only available in “extra chunky” style at present. Making reactors bigger was supposed to lower generating costs relative to the older Gen II designs that could be built in somewhat smaller increments, but that only works if you can get repeat business and learn by doing. The early experiences building these larger, newer reactors are so discouraging that repeat business looks like it may be hard to come by.

I think that you could probably get costs under control again if there were a long term commitment to a sustained nuclear build, like France did in the 1970s and 1980s. But that was achieved while France had a single state owned national monopoly supplier for electricity. France has completed only 1 reactor since the electricity monopoly ended, and their costs/schedules are going terribly for newer reactors. There’s something for everyone to hate in the one big successful switch to nuclear power: a substantial part of the environmental-left is wary of nuclear power as a technology and the freemarket-right is wary of the heavy state involvement that is (apparently) needed for success.

Fuel supply isn’t really an issue.

Firstly the known resources of uranium relative to current consumption are large enough that there isn’t any real point in looking for more. There are a fair number of locations which would be likely to have usable resources, if we had reason to look we would find more.

Secondly even if the mineable ores ran out we know how to extract uranium from seawater. The cost is about ten times the current market price. About $250 per kg. The concentration is low but the sheer quantity of uranium in the oceans is vast. At that price supply is effectively unlimited.

Fuel is a relatively small part of the operating cost of a nuclear power station.

The main problem is an exaggerated fear of radiation. The aftermath of Chernobyl failed to show any detectable excess of cancer cases in areas within the plume compared to areas outside the plume. With the exception of 2,000 cases of one very rare extremely low mortality (#0.5%) childhood thyroid cancer which is almost exclusively linked to exposure to radioactive iodine. In all other cases any signal was overwhelmed by the noise of the background rate of the cancer. The linear no-threshold model appears to seriously overstate the risk of low level exposure. With more evidence based safety standards nuclear would be lot more cost effective.

A properly functioning emission permit market or a carbon tax would also help by forcing the fossil fuel generators to price in the externalities caused by their emissions.

In fact per unit generated nuclear is the safest form of power generation, solar is second (installation accidents, such as falls are the main risk there).

high safety standards lead to overengineering, which are not ‘natural’ engineering solutions in that they require often novel changes to routine systems and complex systems interactions may lead to unintended effects resulting from modifications. basically modifications made without taking into account their effects, either due to lack of awareness of potential implications or simply nature of designing new technology, introduces a lot of cost uncertainty and overrun potential.

“That’s enough to check Bryce’s claim. The gap between nuclear and renewables is currently 552 TW/year. The current rate of addition of capacity for solar and wind combined amounts to additional generation 25.7 TW/h per year. Doubling that gives 51.4. So, assuming a doubling of the current installation rate, which seems plausible even without any new policies, it will take 10.5 years for renewable generation (including hydro) to surpass nuclear. Excluding hydro it will take 15 years. So, if the goal of policy were to replace nuclear with renewables, the answer is that it could be achieved in the fairly near future.”

John, no, this is all mixed up.

Bryce was talking about “replacing” nuclear power–that is, growing renewables enough so that the *additional increment* over present production is enough to equal current nuclear production. (The issue is how soon could growth in renewables make up for the loss of low-carbon electricity if the entire nuclear fleet is closed, as Bernie Sanders wants to do.) But then you did a calculation for how soon renewable generation, current plus additional growth, would “surpass” nuclear. That’s not what Bryce was talking about. Then, you switched back to the word “replace,” which is what Bryce was talking about, but your calculation is wrong for “replacement” of nuclear power.

At your assumed growth rate of 51.4 twh per year, it would take about 16 years to *replace* nuclear.

Also, “the gap between nuclear and renewables is currently 552 TW/year,” no. (TWh/year.) 2015 net nuclear generation was 797 TWh while total RE generation was 562 TWh, for a “gap” of 245 TWh. But closing the “gap” is meaningless as a decarbonization goal, since if we then lose the nuclear sector we’re still left with dramatically less carbon-free generation than we have now.

Anyway, your assumption of 51 TWh/year growth in RE generation is unrealistic. From 2006 to 2015, growth in total RE generation was 176 TWh, a growth rate of 20 TWh per year. At that rate, 40 years for RE to replace nuclear; 20 years if growth rates double.

But in recent years RE growth rates have slowed. From 2011 to 2015, total RE generation grew 48.3 TWh; that’s 66 years to replace nuclear, 33 years if growth doubles. From 2014 to 2015 RE grew 13.8 TWh; 58 years to replace nuclear, 29 years if that growth rate doubles. (Replacement times assume no attrition or productivity decline in wind and solar facilities, which would lower the net growth rates.)

Growth rates are very unlikely to double. The 2015 energy bill phases out federal wind and solar subsidies over the next decade. In 2013, when federal wind subsidies briefly lapsed, just 1 GW of wind was installed.

(See https://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_1_1 for US electricity generation data.)

Bottom line: RE is growing very slowly (much slower than historical rates of nuclear buildout) and wind and solar are still dependent on subsidies. If nuclear reactors close prematurely it will take decades for RE to make up the loss, with nothing left over in the meantime to displace fossil fuels from the grid.

Any thoughts regarding Transatomic Power’s MSR or TVA’s Clinch River Early Site Permit?

From:
http://www.transatomicpower.com/wp-content/uploads/2015/04/transatomic-white-paper.pdf

“The overnight cost for an nth-of-a-kind 520 MWe size, including on-line fission product removal and storage, is estimated at $2.0 billion with a three year construction schedule.”

“The $2 billion price point can greatly expand the demand for nuclear energy, because it has a lower entry cost than do large-sized nuclear power plants. The Vogtle 3 and 4 plants, each 1100 MWe and built in parallel, have a combined project cost of over $19 billion for approximately four times higher output [18]. Similarly, the Flamanville EPR is estimated to cost $11.7 billion for 1630 MWe, or just over three times higher output [19]. Even if the cost per watt were the same, a lower price for a smaller unit will still expand the number of utilities that can afford to buy nuclear reactors, better match slow changes in demand, allow greater site feasibility, and reach profitability faster. The speed of construction and faster payback also reduce financing costs.”

From:
http://www.knoxnews.com/opinion/editorials/tva-sensible-to-move-forward-on-modular-reactors-330d8663-2497-4069-e053-0100007fd1e3-380001651.html

“TVA is the first in the industry to submit any type of application related to SMRs (small modular reactor) to the NRC.”

Environmentalist, definition: someone who thinks that now we have ‘every chance’ of being saved from global warming by exponentially cheap solar, nothing needs to be done.

I think it would be preferable if the grid was built by international efforts rather than just China though, since having one country in control of the whole world’s shared energy grid seems kind of problematic to me.

The USA might come on board with it, given the positive news from the Democratic Party Convention, where the platform was changed by Russell Greene sponsoring an amendment including language about a WW2 style climate mobilisation.

The amendment says:

“Democrats believe it would be a grave mistake for the United States to wait for another nation to lead the world in combating the global climate emergency. In fact, we must move first in launching a green industrial revolution, because that is the key to getting others to follow; and because it is in our own national interest to do so.

Just as America’s greatest generation led the effort to defeat the Axis Powers during World War II, so must our generation now lead a World War II-type national mobilization to save civilization from catastrophic consequences.

We must think beyond Paris. In the first 100 days of the next administration, the President will convene a summit of the world’s best engineers, climate scientists, climate experts, policy experts, activists and indigenous communities to chart a course toward the healthy future we all want for our families and communities.”

Russell Greene introduced the amendment by saying:

“Global warming threatens to destabilize the climate system for tens of thousands of years, sending the Earth into a hot state inhospitable to humanity. If business as usual continues for much longer, a chain reaction of proliferating droughts, famines, and subsequent state failures will cause the unraveling of an organized human community. Furthermore, the global economy’s overshoot of the planet’s limits has initiated the 6th mass extinction of species, which threatens to wipe out much of life on Earth within this century. If allowed to unfold, it could take some 10 million years for life to fully recover…

We must call this out explicitly. It is not enough that this is the most progressive platform in Democratic history. What matters is that this matches the moment. What will be remembered from our work here this weekend is not only what we have put in, but what we have left out. Prioritizing a climate mobilization provides the most immediate path to social, economic, and environmental justice. As with the WWII home front mobilization, the climate mobilization will create full employment and drastically reduce inequality, with the principle of fair and shared sacrifice embraced by all Americans.”

http://www.theclimatemobilization.org/dnc_ww2

I think that is really great news. Definitely another reason to vote for the Democratic Party in the elections.

Even for people who say that Trump would be better on jobs — if the Democrats are in government and start making policy for a climate mobilisation this will create so many jobs for Americas, in a way that looks to the future, rather than backwards to isolationism and walls and protectionism.

In good news from Australia, the State government of Victoria, where I live, has issued its first Green Bond of $300 million which was fully subscribed in a bit over 24 hours. I think this is the first Green Bond scheme in the world.

http://www.premier.vic.gov.au/victorian-green-bonds-an-australian-and-world-first/

Soru & Rich Puchalsky: “saved from global warming” != “irrelevance of nuclear” != “nothing needs to be done”.

“No, wind and solar growth rates have definitely not been exponential over recent years.”

It sounds like you’d be game for an agreement involving chessboards and grains of rice.

By your numbers, the compound annual growth rate from 2007 to 2015 of solar output was 55%, that is, the production in 2015 was 1.55^8 times that of 2007. Continued at that rate for another 25 years, production would “only” be 1600 terawatts (14,501,501 terawatt-hours for the year). Current consumption is about 12 terawatts, or let’s say in the tens of terawatts. So I think those paltry growth rates of solar from 2007 to 2015, if continued for awhile, just might manage to keep up with world energy needs.

Even the relatively “weak” 32% growth you quote for 2015, if continued for another 25 years, would result in production of about 30 terawatts, or about three times current world consumption.

And that’s just solar.

Solar and wind are now competitive without subsidies in many places, and still getting cheaper and more competitive in more places. So subsidies are less and less relevant to growth.

So yeah, nuclear is irrelevant.

Kiwanda @ 36,

No, you need to think a bit harder about what you’re saying here and try to understand what exponential growth really is and why it doesn’t occur for very long in the real world.

Renewables growth won’t “compound” at 55 percent or 32 percent. The relative growth rate will keep dropping, just like it has in preceding years. It won’t hold up at 32 percent for 25 years, just like it didn’t hold up earlier at 92 percent or 59 percent; it just keeps dropping. The data show this clearly; no use trying to deny them.

“It sounds like you’d be game for an agreement involving chessboards and grains of rice.”

Uh, Kiwanda, think harder about how that “agreement” actually plays out. Do you think it ends by putting quadrillions of rice grains onto the chessboard, (or building 30 terrawatts of solar)? No. In the real world, material and logistic constraints always snuff out exponential growth rates. Imagining that wind and solar are somehow immune to this fact is just day-dreaming, and it’s clearly disproven by the stats that I cited.

Again, it’s really irresponsible to talk about wind and solar growing exponentially, because that vastly overestimates how fast they will actually grow, and therefore encourages people to not take the necessary policy steps to ensure that other low-carbon energy sources are brought onto the grid.

I agree with Will that exponential curves don’t last long in the real world. But I would disagree that solar growth is being driven more by subsidies than by falling prices. The US federal government has offered the same 30% investment tax credit for solar power projects since 2005 — and it was extended through 2019 at the end of last year — but capacity additions have been going up each year even as the tax credit remains the same. (Though expect 2017 to be lower than 2016, since extra projects were scheduled for 2016 in anticipation of the expiration of the now-extended tax credit). Globally, money spent on solar projects last year declined even as the sector set a new record for capacity additions. That’s because of rapidly falling prices.

Most of the world’s population lives in places significantly sunnier and less densely populated than Germany, significantly less sunny and more densely populated than the Atacama desert. I think that the continental United States is less of an outlier than either Germany or Chile in this regard.

@Will if you have a citation for Georgia state regulators reckon the lifetime cost of power from the new Vogtle AP1000 reactors at about 6.2 cents per kwh, I’d love to read it. I have a hard time figuring out how one should convert capital-intensive, long-life energy project costs into per-unit output costs.

“I agree with Will that exponential curves don’t last long in the real world. ”

I thought Will’s argument was that because exponential growth can’t last forever, it doesn’t exist: eight years of growth better than Moore’s law (actually, longer than eight), with a virtuous circle where higher adoption leads to lower prices and vice versa, must surely be ending because last year the growth was “only” 32%.

Sure, ultimately we’ll see something like a logistic curve, as renewables saturate the market. So also Moore’s law must end, the world’s population cannot increase without bound, and disease epidemics cannot kill more hosts than exist. What matters is how long the early, exponential part of the logistic curve continues. With the current and improving competitiveness of wind and solar, it won’t end for a good long time.

“Globally, money spent on solar projects last year declined even as the sector set a new record for capacity additions. That’s because of rapidly falling prices.”

It’s that virtuous circle continuing.

While 6.2 cents/kwh, was estimated by Georgia regulators, a solar contract for 3 cents/kwh was signed in Dubai, and for 3.6cents/kwh in Mexico.

@ Matt 40,

“The US federal government has offered the same 30% investment tax credit for solar power projects since 2005 — and it was extended through 2019 at the end of last year — but capacity additions have been going up each year even as the tax credit remains the same. ”

That’s just one of many subsidies solar gets. Lots of states have their own renewable tax credits and grants. State RPSs have separate renewables quotas that ratchet up as time passes, driving utilities to buy more solar; that creates a market in SRECs which is a whole nother revenue stream for solar plants. A lot of the solar growth has been rooftop, which gets state and local rebates in addition to federal subsidies. Then there’s net metering, which is a huge subsidy; it essentially makes utilities pay rooftop PV retail rates (average 12-13 cents per kwh) for electricity that they could buy on wholesale markets for 4-5 cents per kwh. Solar gets a lot of different subsidies, and the industry has been very active and effective lobbying for accelerated subsidies in recent years.

Yes, solar has also accelerated because of cost reductions, but without the subsidies it would still be dead. Again, take Germany. Subsidized market premium rates are now about 8 eurocents per kwh for utility scale, with rooftop drawing FITs of 11-12 eurocents. But wholesale prices have plunged to maybe 2-4 eucents per kwh because of RE overcapacity. Without those subsidies there’s simply no way any new solar in Germany could compete against wholesale markets and none would get built. As it is, even with the subsidies, German solar is almost dead.

Will Boisvert 48: don’t know about subsidies, implicit or otherwise, for the Dubai or Mexico projects; there are claimed to be none. Although it’s true, the estimates of regulators are surely more accurate than signed contracts.

Speaking of implicit subsidies, solar doesn’t see a lot of taxpayer coverage of cost overruns, abandoned plants, and waste storage, unlike nuclear. But subsidies for solar will be ending, now that this emerging technology has taken off. Meanwhile, new ideas for nuclear are just around the corner, the same corner they’ve been around for 50 years.

True, Georgia doesn’t have the vast solar resources of California, Texas, Arizona, or Nevada, or the vast wind resources of Texas, Kansas, Nebraska, or the Dakotas. (Or even Iowa, which will be getting about 40% of its power from wind.) This is why a better grid for load and generation balancing, like the Plains and Eastern project allowing Georgia to buy cheap wind power, is so crucial.

@ Kiwanda 49,

“Speaking of implicit subsidies, solar doesn’t see a lot of taxpayer coverage of cost overruns, abandoned plants, and waste storage, unlike nuclear.”

Huh? federal taxpayers pay 30 percent of the cost of every solar plant and rooftop rig that gets installed, with state and local taxpayers paying more. Nuclear overruns and abandoned plants aren’t paid for by taxpayers, they’re paid for by rate-payers—the same rate-payers who pay extra charges for net metering for solar homeowners and above-market rates for solar plants required by renewable portfolio standards. All storage of waste from nuclear plants is paid for by nuclear utilities, not by taxpayers. In fact, nuclear plants paid about $25 billion in fees to the federal government to fund Yucca Mountain, which was cancelled—which means the government owes nuclear plants about $25 billion.

“But subsidies for solar will be ending, now that this emerging technology has taken off.”

Not hardly. The Federal Investment Tax Credit will taper from 30 percent to 10 percent over the next decade, then stay there. But state and local subsidies for solar—tax credits, rebates, net metering, SREC income and above-market PPAs mandated by renewable portfolio standards—are huge and are, if anything, growing.

“Meanwhile, new ideas for nuclear are just around the corner, the same corner they’ve been around for 50 years.”

Yeah, nuclear, always the same old tiresome ideas—generating 60 percent of America’s low-carbon electricity, 24/7/365. Why don’t they try something new, something that needs bigger subsidies and conks out when the sun sets or the wind dies down?

It’s because the “economic critique” is just a nonsensical cover for the irrational ideological animus greens hold against nuclear power, all too evident in this discussion.

Do you forget history so quickly? Simple cost comparisons with no caveats or attached spreadsheets were by far the most straightforward and common argument against solar power a decade ago (deployed against solar by advocates for nuclear power and fossil fuels alike). I found those arguments convincing! I used to have a “nuclear green”-type perspective on the matter. Solar power was fine for millionaires, but nuclear power was clearly the everyman’s route to decarbonizing electricity. All you had to do was look at two numbers: capital costs and capacity factor. Divide A by B and the lower number wins. Nuclear power was easily the winner. Solar advocates who wanted to make lengthy arguments about the hidden benefits of solar and the hidden costs of nuclear appeared to be motivated reasoners who had an immovable position and were only good at identifying additional issues in service of their fixed conclusions.

But solar PV subsequently cut costs and increased scale beyond my most optimistic expectations. And the “nuclear renaissance” I expected stalled, with costs and schedules slipping worse than I expected. Nowadays if I divide A by B, there are a lot of places that nuclear power can’t win that simple numbers game. We probably shouldn’t have used such simple metrics in the first place, but I assure you that they were wielded in favor of nuclear power in the recent past. Turnabout is fair play.

FWIW I still agree with the “nuclear greens” that nuclear fuel is abundant and that electricity from nuclear power is safer than fossil combustion. It’s just the changing numbers that have made me more enthusiastic about solar. I agree that solar probably can’t serve 100% of demand even in sunny California without extraordinary costs. But the same is true of 100% nuclear without storage. (Are you going to build an extra 10 gigawatts of reactors needed only a few days each year for peak summer cooling demand?) Criticizing near-term projects that take solar up to maybe 10% of supply, on the basis that they’d be financial disasters at 90% penetration, makes no more sense than the parallel argument against near-term nuclear projects (“OMG all that unneeded electricity while everyone sleeps” vs “OMG all that unneeded electricity in the middle of the day”).

PS to mods I have a previous comment stuck in moderation. Please un-stick it at your convenience.

@ Matt 52,

“We probably shouldn’t have used such simple metrics in the first place, but I assure you that they were wielded in favor of nuclear power in the recent past. Turnabout is fair play.”

Huh? Bad, misleading energy-economics analysis is now “fair play?”

And Matt, the historical equivalence you’re trying to draw between pr0-renewables and pro-nuclear sides of the debate is bogus. When have pro-nukes called for banning wind and solar? When have they called even for banning *subsidies* for wind and solar? They haven’t and they don’t.

While pro-nukes have consistently argued that nuclear power is more economic than wind and solar, their consensus policy is simply that nuclear should get subsidies equivalent to those of wind and solar. That’s a stark contrast to leading renewables advocates—Greenpeace, FOE, NRDC, Sierra, Green parties in European governing coalitions, Bernie Sanders, Ralph Nader— who have made a legal ban on nuclear a principle demand since the 1970s.

–And actually, the simplistic A over B analysis is still in general very favorable to nuclear power, without any caveats about system costs.

For example, in 2015, 10 new reactors came on line globally with 9.4 GW of capacity. Their total capital cost, financing included, was about $31 billion. Last year’s crop of new solar installations was 59 GW, costing about $92 billion. Assuming the nuclear has a cf of 86 percent and lasts 60 years and the solar has a cf of 15 percent (that’s the global average) and lasts 40 years, then the nuclear capacity will produce 4,260 TWh over its lifetime and the solar will produce 3,120 TWh. Last year, A / B = $3,720 per kw for nuclear, $10,359 for solar.

In other words, last year the world got new nuclear capacity that will produce 36 percent more clean energy over its service life than will last year’s crop of solar capacity, and got it for *one third* the cost of the solar capacity. Solar operating costs will be about 1 cent per kwh lower than those of nuclear, but that’s greatly overweighed by the system costs of transmission, storage, and gas backup that solar imposes.

The nuclear capacity was heavily weighted towards Chinese reactors (80 percent), but the comparison still holds up. In China, solar LCOE’s and feed-in tariffs are twice as high as those for Chinese nuclear, the best comparison of costs given comparably supportive policies for both technologies. Also, South Korea’s two latest reactors, coming on line last year and in January (its Gen III APR1400), cost the same as the Chinese reactors, about $3,200 per kw with financing costs. South Korea is a developed country with high wages, so there’s no reason its success can’t be replicated everywhere in the West. Indeed, cheap builds are historically the rule, not the exception; France’s fleet, for example, was built for an average cost of about 1,600 euros per kw in 2010 euros.

So there’s no doubt that new nuclear is still far more economical than wind and solar, even leaving very high system integration costs aside and without looking ahead to the wall of diminishing returns at even moderate intermittent penetrations. With supportive industrial policy that performance could be achieved everywhere and hugely accelerate the decarbonization of the energy supply.

I don’t argue that solar and wind should not be subsidized and deployed. The reason I point out their severe shortcomings is to challenge the green group-think that intermittent renewables are “competitive” while nuclear is “hopelessly uneconomic.” That false meme grossly distorts energy policy and impedes effective action on climate change, and I hope the people who put it out will think a little harder about their responsibilities as public intellectuals.

While pro-nukes have consistently argued that nuclear power is more economic than wind and solar, their consensus policy is simply that nuclear should get subsidies equivalent to those of wind and solar.

This at least I can fully agree with. Since production tax credits can drive negative pricing, I’d prefer an investment tax credit offered at the same rate across the board for nuclear, solar, wind, storage, and transmission serving any of the above. (Or better yet a carbon tax, but that seems like it won’t ever happen.)

I also agree that it’s horrible to shut down nuclear plants that remain technically operational while there are still fossils being burned for electricity. If states want to keep increasing renewables and existing nuclear gets in the way, require storage to go with new renewables. Or use real-time pricing to encourage demand to chase production peaks. Or build interconnections over longer distance to enable more flexibility. Anything that shuts down fossils before emissions-free sources instead of pitting one emissions-free source against another.

“Huh? federal taxpayers pay 30 percent of the cost of every solar plant and rooftop rig that gets installed, with state and local taxpayers paying more. ”

In response to the observation that nuclear gets subsidies, you note that solar gets subsidies, a point not in dispute.

“Nuclear overruns and abandoned plants aren’t paid for by taxpayers, they’re paid for by rate-payers”

Oh, sorry, rate-payers. If the difference significant?

“But state and local subsidies for solar…are huge and are, if anything, growing”

I assume you mean “rate of subsidy”, because otherwise this is merely agreement that solar is growing rapidly. Why would a state want to encourage the growth of a power source as clean, reliable, and low-water-use as solar? Hard to figure.

“And the cost of those grid improvements would be another large implicit subsidy that should be added to the tab for renewables. ”

“Your suggestion to transmit renewable power thousands of miles to Georgia from Arizona or the Dakotas would add even more to the real price of the kwhs”

I did not suggest that. Besides, as Matt notes, Georgia has pretty good solar resources, and some offshore wind potential. That Plains and Eastern project I mentioned, for transmission capacity from OK and TX to the southeast, looks pretty serious, and is not government funded: it’s a business venture, to build and sell transmission capacity.

“That’s why simplistic power-purchase prices like the ones you quoted for solar are a very bad measure of the comparative cost of renewables—they don’t capture the enormous system costs of storage, gas backup, demand response and, yes, transmission that weather-dependent RE imposes on the grid.”

The solar *contracts* are 1/2 or 2/3 of the *estimate* for a nuclear project that hasn’t completed, but somehow they’re not believable, because of some hypothesized costs: well, that’s a pretty tight argument.

Yet we hear this argument against nuclear all the time. “So what if Vogtle is 6.2 cents per kwh, Dubai solar is 3 cents per kwh! Nuclear is therefore hopelessly uneconomic and irrelevant!”

” It’s because the “economic critique” is just a nonsensical cover for the irrational ideological animus greens hold against nuclear power, all too evident in this discussion.”

Matt has answered this last at length above, but: you *were* making a cost argument, that wind and solar are hopelessly uneconomic. But wind and solar have been dropping rapidly in price, in tandem with wider adoption. In favorable settings, they have crossed an important threshold: they are cheaper than other sources. And the price trend shows no sign of stopping. Nuclear, meanwhile, has no such prospects. So in the face of unfavorable economic trends, you’re left with blowing smoke and making accusations of bad faith.

“Seriously, do you think it’s possible to shift a large portion of the economy to a chaotic surge-and-slump production mode to match surge-and-slump intermittent power”

“Solar operating costs will be about 1 cent per kwh lower than those of nuclear, but that’s greatly overweighed by the system costs of transmission, storage, and gas backup that solar imposes.”

Maybe you could have phrased that as “0.001 cents per watt-hour” advantage. Just as a rough comparison, regarding that 1c/kWh as saying that nuclear operating costs are 1c/kWh hour and solar operating costs are zero, the cost per megawatt-hour for that nuclear capacity over its lifetime is $17, while that for solar is about $10, using your numbers. Putting in some x cents/kWh for solar and x+1 cents/kWh for nuclear would decrease that ratio, but solar would still have the cost advantage. But that’s not accounting for the cost of transmission for solar, which is often from the roof to inside the house, so maybe that’s not fair.

And solar prices are still decreasing rapidly.

“Storage is still colossally expensive, ”

Lithium-ion battery packs are about $190/kWh: not economical for all applications, but not what I’d call “colossal”. Several manufacturers now offer home energy storage systems, which can be based on cheaper technologies than lithium-ion; these can help with load balancing, especially combined with using all-electric vehicles for demand response.

@ Kiwanda 62,

“I assume you mean “rate of subsidy”, because otherwise this is merely agreement that solar is growing rapidly.”

Right Kiwanda, the rate of subsidy. But total value of subsidies matters too, because it reliably causes governments to slash subsidy rates when the total grows insupportable in order to stall installations. That’s what has happened in Germany, and Britain recently.

“Why would a state want to encourage the growth of a power source as clean, reliable, and low-water-use as solar? Hard to figure.”

Solar reliable, Kiwanda? Capacity factors for the best solar plants run about 33 percent, the global average is 15 percent, in Germany and Britain they are about 11 percent, compared to nuclear’s US fleetwide average of 90 percent. In Germany the entire solar sector goes dark for weeks on end in the winter. The only thing that you can count on with solar is that it will produce no power during most hours of the year. Calling solar “reliable” is positively Orwellian.

@ Kiwanda 46, 50, 62,
“While 6.2 cents/kwh, was estimated by Georgia regulators, a solar contract for 3 cents/kwh was signed in Dubai, and for 3.6cents/kwh in Mexico.”

“Will Boisvert 48: don’t know about subsidies, implicit or otherwise, for the Dubai or Mexico projects; there are claimed to be none. Although it’s true, the estimates of regulators are surely more accurate than signed contracts.”

“The solar *contracts* are 1/2 or 2/3 of the *estimate* for a nuclear project that hasn’t completed, but somehow they’re not believable, because of some hypothesized costs: well, that’s a pretty tight argument.”

So I looked up the most recent Mexican capacity auction. http://www.greentechmedia.com/articles/read/Solar-Stuns-in-Mexicos-First-Clean-Energy-Auction-1860-MW-Won-at-50.7-P I find that the average contract price for solar was $50.7 per mwh, range $35-67, and then look at this: “Additionally, all 11 projects have won contracts for a combined 4 million Clean Energy Certificates.” [A number equal to the projects’ combined yearly mwh output.]You know what that means, Kiwanda? The solar plants are selling those Clean Energy Certificates at home or abroad on the international carbon offset market for a whole nother income stream in addition to the contract price for the power.

So you were flat wrong when you claimed that the Mexican solar contract prices were unsubsidized. In fact, they get big, honking government subsidies, just as I surmised; that’s why contract prices are so low.

And of course, as the Plains and Eastern transmission project you tout show, the cost of actually shipping that Mexican solar (1000 miles? 2000 miles?) to Georgia would double the contract price and more. And we haven’t yet accounted for the system costs and carbon costs of storage and gas backup for unreliable solar.

That’s why it’s utterly meaningless, indeed actively misleading, to cite the contract price of Mexican solar in a discussion of whether the Vogtle plant makes economic sense, as you did.

@ Ogden Wernstrom 63,

“I’ve lost track – is that about nuclear wastes, or solar-generated energy?
Although it makes little difference in who ends up paying, does the cost of nuclear waste treatment and storage fall to ratepayers, taxpayers, future generations, or someone else I have failed to consider. (Oh, yes – shareholders! Not likely, is it?)”

I meant storing wind and solar power.

Storage costs for nuclear waste are paid out of the plant’s revenue. The US DOE assessed a fee for a permanent waste depository at Yucca Mountain (now canceled because of political opposition), to fund construction and an endowment for maintenance in perpetuity. The fee was 0.1 cents per kwh, to be paid until the repository was full and closed. (Then the fee would have been earmarked for another depository for subsequent waste.)