What is the cheapest form of green power? (let’s ask Petr Beckmann)

From Free Exchange:

…levelised costs do not take account of the costs of intermittency…

Seven solar plants or four wind farms would thus be needed to produce the same amount of electricity over time as a similar-sized coal-fired plant. And all that extra solar and wind capacity is expensive.

If all the costs and benefits are totted up using Mr Frank’s calculation, solar power is by far the most expensive way of reducing carbon emissions. It costs $189,000 to replace 1MW per year of power from coal. Wind is the next most expensive. Hydropower provides a modest net benefit. But the most cost-effective zero-emission technology is nuclear power. The pattern is similar if 1MW of gas-fired capacity is displaced instead of coal. And all this assumes a carbon price of $50 a tonne. Using actual carbon prices (below $10 in Europe) makes solar and wind look even worse. The carbon price would have to rise to $185 a tonne before solar power shows a net benefit.

There is more here.  The relevant cited studies you can find here.


I wonder for how long this sort of thing will be true. Once, solar, wind etc more expensive than existing technologies even if you ignored intermittency, now they are much cheaper thanks to improved technology. Nuclear fusion is old technology that is not going to get any cheaper. Maybe windmills are old technology too. But what about photovoltaic cells?


I wish nuclear fusion was old technology.

Windmills are way way older than nuclear power. Solar cells are probably about as old as nuclear power. What bearing does their age have on anything?

Of course age has a bearing: Has the cost / MW of solar / wind / coal power gone down equally rapidly over the last 50 years?

My guess says solar & wind prices have fallen a lot more rapidly than coal energy. Technology maturation & economies of scale cause the price of a less mature tech. to fall more rapidly than a mature one. Note this is only rates & not saying anything about absolute levels.

So to restate @Adrian's question: Where's the price asymptote for wind / coal etc? I'm not sure that I know.

Solar has dropped because of manufacturing improvements common to electronic devices. It faces the hard wall of the area required to capture sun energy. They can't make their chips smaller. Has wind installations dropped in price? I would like to see some numbers on that.

If you price wind and solar by mwh of demand satisfied, the numbers look far different.

Sure - 'Declining Cost of Wind Energy Over Time

Advanced technology, improved siting techniques, and learning across all sectors as the industry scales up have all influenced the cost of wind energy over time. The Department of Energy, below, depicts the cost reduction in wind energy alongside U.S. wind energy deployment, showing a decrease in cost of more than 90% since the early 1980's.' http://www.awea.org/Resources/Content.aspx?ItemNumber=5547

There is also a DOE graphic at the link.

You believe what the Wind Association says?

I'm looking for cost per mwh load satisfied.

You aksed - 'Has wind installations dropped in price?'

The answer, according to the DOE (Revolution Now The Future Arrives for Four Clean Energy Technologies. September, 2013) is yes.

Of course, you are welcome to disregard any data you wish. Here is some in German, comparing the current cost for building onshore windmills in the best locations compared to the higher cost of building coal or gas plants - 'An sehr guten Onshore-Windstandorten produzieren WEA Strom bereits heute zu geringeren Kosten als neue Steinkohle- oder GuD-Kraftwerke. Die Stromgestehungskosten von Onshore-WEA (spez. Invest zw. 1000 und 1800 Euro/kW) liegen heute zwischen 0,045 und 0,107 Euro/kWh.' http://www.ise.fraunhofer.de/de/veroeffentlichungen/veroeffentlichungen-pdf-dateien/studien-und-konzeptpapiere/studie-stromgestehungskosten-erneuerbare-energien.pdf

Solar has dropped because of massive overbuilding of manufacturing capacity in China.


There is a hard wall, but it is at a generous 1kW/m^2. So you cordon off a very small sheep farm and you get a gigawatt of power -- at the hard limit. The technological question is how far below that hard limit do we have to be?

The age of the technology seems irrelevant.

Economic growth requires energy, it was the tapping of the petroleum that fueled growth over the past 100+ years. If you were long term planning and trying to figure out what power source can deliver an exponential increase in energy output going forward, you're not even going to consider wind, and only solar if you're thinking of space based power generation. But what if the citizens of the world still want to keep growing and current output is only 10% of global energy needs in 2100 or 2150? Nuclear.

Beckmann died over 20 years ago, so his comments should be revisited in light of current pricing. Natural gas actually beats coal by a large margin these days, and the cost of solar photovoltaic cells has greatly declined.

Like everything in economics, it is true until it isn't.

Microeconomics tells us that people respond to RELATIVE price changes. Consequently, we will NEVER run out of oil. Long before we run out, alternatives will become relatively cheaper to the point that the technology is replaced.

In California, it's not uncommon to pump water uphill during off hours so it can be released back down to spin hydropower turbines during the afternoon when air conditioning is most needed.

Yes, that's common in UK too.

Pumped hydro is the only currently affordable way to store energy in significant volumes for later use. However, all feasible hydro sites are pretty much developed now, and it will never be more than a few % of total generation for large countries.

Yeah, but a little bit of storage can go a long way to smooth out short-term intermittency. If it's a matter of delaying the consumption of solar power until the evening, even a modest reservoir can help a lot. I know that winter solar is weaker than summer solar, and reservoirs will not help with that kind of supply/demand variance. To balance this, we probably will need to burn fossil fuels. But the point is, the rest of the year, we *don't* need to be burning them. Even when those gas/coal plants can be idling for a few months, I count that as a partial win.

[reality check needed]:

You also disregarded whole bussiness model of pumped-storage and silently assumed something along the lines of authoritarian government that both simply ordains operators to run it your way AND kowtows to the environmental pressure groups' notions of what should be done. Same goes for the image of somebody building and operating those idling gas/coal plants without profit. No, idling doesn't earn you money, unless you get hefty subsidy, google "Irsching" and "capacity market."
Little bit of storage can go a long way, indeed, but when your little bits are of values close to Planck's constant and the way necessary to undergo to keep grid together is in meters, rarely the twain shall meet.

Wow - that wikipedia links lists in the neighborhood of 60 pumped storage facilities, each as large as a normal coal plant/nuclear reactor, and you talk about how tiny pumped storage is in terms of load balancing?

And you do know why pumped storage exists in Germany and France, right? Mainly for load balancing nuclear power plants, which are pretty much the definition of baseload supply.

Then, of course, there is - 'The Bath County Pumped Storage Station is a pumped storage hydroelectric power plant, which is described as the “largest battery in the world”,[2] with a generation capacity of 3,003 MW[3] The station is located in the northern corner of Bath County, Virginia, on the southeast side of the Eastern Continental Divide, which forms this section of the border between Virginia and West Virginia. The station consists of two reservoirs separated by about 1,260 feet (380 m) in elevation. It is the largest pumped-storage power station in the world.[3]

Construction on the power station, with an original capacity of 2,100 MW, began in March 1977 and was completed in December 1985 at a cost of $1.6 billion,[4] [5] Voith-Siemens upgraded the six turbines between 2004 and 2009, increasing power generation to 500.5 MW and pumping power to 480MW for each turbine.[1][6] Bath County Station is jointly owned by Dominion Generation (60%) and the Allegheny Power System (40%), and managed by Dominion.[3] It stores energy for PJM Interconnection, a regional transmission organization that serves 60 million people in 13 states and the District of Columbia.' http://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Station

But the interesting thing about pumped storage is that it is completely neutral when it comes to the electricity that power the pumps which lift the water - so all of the pumped storage originally built in Germany to increase the cost efficiency of mainly nuclear baseload power generation can now be used to store power generated through such things as wind parks. Cost for the utility? Pretty much zero. Problems in using pumped storage facilities as designed? Pretty much zero.

No bureaucrats needed - power companies can generally be trusted to use existing assets in a way that benefits them, after all.

>lists in the neighborhood of 60 pumped storage facilities, each as large as a normal coal plant/nuclear reactor, and you talk about how tiny pumped storage is in terms of load balancing?

Those other links show worldwide capacities of wind and solar to get some comparison of scales in question, not my fault you're innumerate.

>And you do know why pumped storage exists in Germany and France, right? Mainly for load balancing nuclear power plants, which are pretty much the definition of baseload supply.

They exist because it's cheaper to sell them excess electricity for nothing* than switch baseload plant to load follow. During peaks, baseload plants keep producing and there are other load-following plants, not just pumped-storage. On the contrary, significant share of "unreliables" (let's call them as they deserve) would mean that when their output drops, there's a gap of gigawatts or tens of gigawatts (depending on the size of affected region) of missing capacity, so no imaginary smoothing out the kinks but taking over large part of the load for some unspecified time (pumped-storage plants tend to have only few hours of capacity). Not my fault you don't know or understand basic facts of electricity production and transmission.

*Contrary to the Green delusions, grid can't sustain production exceeding or falling short of demand, excess electricity is dumped at low price, for free or for fee to "get you rid of it." That is the normal mode of operation and every operator had to carry the associated burden. For unreliables, it is now however mandated, that they are exempt of this rule - their output has to be bought as a first source, for mandated price (above market price, indeed), they can't be switched off when necessary and don't have to pay cost of overproduction during their peaks. This is usually not stressed enough and lack of understanding often confounds uninformed people who, with important pieces of puzzle missing, can't understand "why we don't do more."

'Those other links show worldwide capacities of wind and solar to get some comparison of scales in question, not my fault you’re innumerate.'

Well, it was this comment - ' but when your little bits are of values close to Planck’s constant and the way necessary to undergo to keep grid together is in meter' that seemed to display such a massive amount of ignorance why pumped storage exists (hint - because it is more efficient for utilities that operate coal and nuclear baseload generators to use pumped storage to even out generating capacity to handle normal peak loads).

'On the contrary, significant share of “unreliables” (let’s call them as they deserve) would mean that when their output drops, there’s a gap of gigawatts or tens of gigawatts (depending on the size of affected region) of missing capacity, so no imaginary smoothing out the kinks but taking over large part of the load for some unspecified time (pumped-storage plants tend to have only few hours of capacity). Not my fault you don’t know or understand basic facts of electricity production and transmission.'

'Gigawatts or tens of gigawatts'? - You really should read the link, tzo at least get an idea of the numbers you are so casually throwing around - http://reneweconomy.com.au/2014/opposite-energy-policies-turned-fukushima-disaster-loss-japan-win-germany-97060 Including this part - 'Many claim renewables could harm grid stability. So why do Germany, with 25% renewable electricity in 2013, and Denmark, with at least 47%, have Europe’s most reliable electricity, about ten times more reliable than America’s? '

'This is usually not stressed enough and lack of understanding often confounds uninformed people who, with important pieces of puzzle missing, can’t understand “why we don’t do more.”'

You really should become more familiar with how things work in Germany. Again, from the article - 'Japan’s economy wilted while Germany’s throve, adding several hundred thousand clean-energy jobs—part of the energy transition’s net macroeconomic benefit. Japan’s electricity prices soared while Germany’s whole­­sale electricity prices fell more than 60%—including 13% in 2013 alone, when year-ahead prices hit eight-year lows. That’s why French energy-intensive industries complain that they can’t beat their German competitors’ one-fourth-lower power prices. The latest manufactured myth of German “deindustrialization” is ironic because big German industries pay approximately those low and falling wholesale prices and are exempted from paying for the renewables that cause them, as well as from grid charges. Those burdens were instead heaped on households (whose bills are half taxes), though household tariffs have now stabilized as providers’ old contracts roll over.'

German industry loves 'unreliables' (Germany remains a reliable net exporter of electricity, by the way, even as it shuts down nuclear plants) - and the French are unhappy that their nuclear plants just can't compete.

Well, I've tried. Enjoy your delusion and have a nice day.

I live in Germany, and the delusion of the German electrical grid, its prices, its stability, and the mix of sources supplying the grid are all matters of public record - Bundesnetzagentur - is one shared by more than 80 million people.

Enjoy explaining to them how deluded they are in the future. I mean, as it is right now, Germany is already a socialist hellhole, right?


You seem to know what you are talking about so here's a question: Say there was no subsidy & you were a power generation company would you add any wind / solar capacity at all to your generation profile?

Just curious what the baseline economic scenario is.


Pumped storage helps, but it shouldn't be overstated. California has 2.943 GW of pumped storage capacity (three projects). Total peak summer demand is around 60 GW.

What is total summer average demand?

Thats sounds about right. UK is a bit smaller by %. It's nice to have but nowhere near enough for serious load balancing, at least once internmittent renewables are several times the % of hydro.

This is an interesting paper, but has a number of issues.

1) The deployed cost per MW of solar is already around 1/3 lower than in this paper and continuing to drop.

2) Tol (2011) finds the mean estimate in the literature of the social cost of carbon to be $177/ton, so the $185 estimated there is not far off. Factor in the already lower cost of solar than this paper uses, and it is already below the price necessary to be the cheapest carbon abatement method overall.

3) The impact of natural gas leakage on full-lifecycle natural gas emissions is not accounted for. Assuming only somewhat higher than the EPA estimates and a 50 year window, NG leakage eliminates a significant chunk of the NG carbon emissions savings.

4) Differential demand for electricity at different times of the day (as reflected by differential spot prices) isn't reflected in this model. Electricity demand peaks in late afternoon to early evening. At those times, spot electricity prices can as much as triple. This coincides with the peak of solar availability - something utilities are very aware of, and which makes solar more attractive than a simple levelized cost reflects. This isn't captured in this model at all. (That higher electricity price during peak demand also reflects a higher true carbon cost of the electricity being generated during the peak than naive math would have one believe, due to the low utilization of the assets creating that electricity.)

5) Finally, saying 'solar' and 'wind' overall in a blanket way is bound to be deceiving. Solar in Alaska is very different than solar in New Mexixo. Average capacity factors across the nation are less meaningful than the capacity factors of regions where projects are under consideration.

Richard Tol is an interesting source to cite.


Oodles of data errors & bad statistics. Sounds like a nightmare. I wouldn't trust any of his work.

1) Difficult to judge; the solar market is glutted such that wholesale price may be below manufacture costs. Installation costs are now siginificant and should ultimately put a floor under the price. Remember; price per MW solar depends on where on the planet you have placed your solar farm - please include interconnection costs to get the power somewhere useful! Also - does not include intermittency costs . These are very high, which is rather the point of the paper.

2) Social cost of carbon at $180/ton seems high....quick survey of market values and government pricing suggests values in the $20 - 60 range. I'd like this double-checked.

3) True.

4) Peak solar is midday, not early evening. Highest irradience is not the hottest part of the day.

5) True.

Reprice your solar to mwh demand satisfied. Otherwise it is snake oil sales.

Here is some information concerning two economies -

'Japan thinks of itself as famously poor in energy, but this national identity rests on a semantic confusion. Japan is indeed poor in fossil fuels—but among all major industrial countries, it’s the richest in renewable energy like sun, wind, and geothermal. For example, Japan has nine times Germany’s renewable energy resources. Yet Japan makes about nine times less of its electricity from renewables (excluding hydropower) than Germany does.

That’s not because Japan has inferior engineers or weaker industries, but only because Japan’s government allows its powerful allies—regional utility monopolies—to protect their profits by blocking competitors. Since there’s no mandatory wholesale power market, only about 1% of power is traded, and utilities own almost all the wires and power plants and hence can decide whom they will allow to compete against their own assets, the vibrant independent power sector has only a 2.3% market share; under real competition it would take most of the rest. These conditions have caused an extraordinary divergence between Japan’s and Germany’s electricity outcomes.

Before the March 2011 Fukushima disaster, both Germany and Japan were nearly 30% nuclear-pow­ered. In the next four months, Germany restored, and sped up by a year, the nuclear phaseout schedule originally agreed with industry in 2001–02. With the concurrence of all political parties, 41% of Germany’s nuclear power capacity—eight units of 17, including five similar to those at Fukushima and seven from the 1970s—got promptly shut down, with the rest to follow during 2015–22.' http://reneweconomy.com.au/2014/opposite-energy-policies-turned-fukushima-disaster-loss-japan-win-germany-97060

It's hard to believe that the real Petr Beckmann would have written "1MW per year" unless he was writing specifically for the great unwashed. Or economists.

I don't understand why the Republican party decided to cozy up to the Coal sector instead of the nuclear sector which would be the main beneficiary of a carbon tax.

Because coal is already rich, who do you think keeps the lights on turned out at Tyler's trollporium? Nuclear power was expensive, shitty, and scary in the 70s and so there are no Koch brothers of nuclear power.

The idea that human-produced CO2 will produce catastrophic global warming has now been dismissed by observational data.

Nevertheless, advances in solar tech are picking up speed and it's clear there is a point in the near future when it compares favorably with fossil fuels.

Wind is just a colossal waste of money. Biofuel is simply destructive, in terms of both raising food prices and destroying tropical forests in Indonesia and elsewhere.

Wind is already working pretty well in Colorado. They are solving the intermittency issue.



This should be highlighted, as another power company discovers that not paying for fuel is a great way to save money (what, you think coal/natural gas/nuclear plants don't have capital costs regarding construction and maintenance?) -

'Before the forecasts were developed, Xcel Energy, which supplies much of Colorado’s power, ran ads opposing a proposal that it use renewable sources for a modest 10 percent of its power. It mailed flyers to its customers claiming that such a mandate would increase electricity costs by as much as $1.5 billion over 20 years.

But thanks in large part to the improved forecasts, Xcel, one of the country’s largest utilities, has made an about-face.

It has installed more wind power than any other U.S. utility and supports a mandate for utilities to get 30 percent of their energy from renewable sources, saying it can easily handle much more than that.'

But that is just the free market talking - I'm sure some of the more astute anti-regulation commenters here are more than capable of explaining why a power company doesn't know its own business.

"supports a mandate" and "free market talking"

As a fan of absurd humor, I like this. As a critical thinker, I cringe.

In case you're not trolling, let me enlighten you - once this is mandated, they will produce as much as they can, balanced by surely plentiful hydro (it's mostly Rocky Mountains states, right?), possibly dump the excess on neighbors mandated to take it or dump it on neighbors and sell certificates of "green electricity" to others, who are either mandated or pushed to have some "green percentage." Nothing of this is some hidden wisdom, that's just following European development. That is, unless they'll just drown wind in their overall output:

There are 11 states that derive more than 7% of their energy from wind, Colorado, Idaho, Iowa, Kansas, Minnesota, North Dakota, Oklahoma, Oregon, South Dakota, Texas, and Wyoming. Two of those states have deregulated, Texas and Oklahoma, and electricity prices have been dropping. So, while your reasoning on the mandate is mostly correct, it is also clear that xcel has learned to make money with the technology more efficiently than other providers in the state. The improvements in the technology suggest it can compete favorably with carbon fired plants.

The initial mandate provided the incentive to learn how to be more efficient with wind power. I would agree with you that there is no need to increase it to 30%. It looks like it will be a viable tech in areas with lots of wind now that the intermittency problem is mostly resolved. There will still be a need for carbon/nuke/hydro back up, so it can't fill all the needs.


'Nothing of this is some hidden wisdom, that’s just following European development.'

Well, not German, because in Germany, the charm is the feed-tariif, meaning that about 50% of renewable energy sources are in essentially private or community hands - that is, those individuals and communities are profiting from selling power, and expect to do so long into the future. Which is a different framework from that proposed by xcel, which would retain its essentially monopolistic market position, and not be forced to buy electricity from independent electricity providers. In other words, xcel wants to redistribute profits among a small club of large utilities to its own benefit.

A major part why the Energiewende seems to attract so much attention is that several essentialy regional market monopolies are being crushed because they invested in the wrong technology - and with wind currently cheaper to build than a coal or gas plant, the power companies lost out on a chance to ensure the sort of profits that the financers of windparks did. Much like their opposition to the democratic decision to phase out nuclear power in Germany (a majority of Germans have been opposed to nuclear power for more than 2 decades at this point), those four major utilities have managed to lose billions of euros by betting that they would be able to change the EEG significantly.

"Catastrophic" is a bit of a weasel word, because it gives you the flexibility to imply that anthropogenic climate change is not happening (which is, as you know, inaccurate) but forces anyone who calls you on it to commit to some idea of catastrophic that clearly isn't happen, because whatever catastrophe they have in their head, you can trump with "that's not what I meant by catastrophe."

So I'll just say, if you consider the loss of trillions of dollars and millions of lives to be a catastrophe, the idea that anthropogenic CO2 has produced global warming and most probably will cause further, catastrophic global warming is incredibly well-attested to by observational data.

Your first paragraph was so good but the second made very little sense. While I agree that the data linking global temperature rise to anthropogenic emissions is quite strong, I can't say the same about hypothesized links between future warming and "catastrophe".


50m sea level rise? That's not catastrophic?

Read Mark Lynas '6 degrees' (Centigrade, note). The scientific papers he cites shows what we start losing at each level. Stern Review did the same.


The bottom line is we have had 0.8 degrees C so far, a similar amount is 'locked in' given current radiative forcings. That's our best current estimates (with wide error bars which could mean things are *worse*).

And when we get above 2 degrees C, the losses start to mount geometrically. Mass extinctions and mass migrations of human beings. Above 4 degrees C it starts to become difficult to imagine continuation of western civilization. Mass migrations finished off the Roman Empire-- the migrations (for reasons imperfectly understood) of various barbarian tribes (especially the Gothic tribes) across the frontier and into France, Spain, lower Danube, Tunisia etc.

Somewhere out there there are unstoppable positive feedback loops: melting permafrost leading to mass methane release (we'll leave aside undersea methane calthrates) leading to explosive global warming and possibly our extinction.

Remember there are c. 20 nations on this planet which either have nuclear weapons, or are nuclear weapons capable. If there are significant geopolitical disruptions arising from these environmental changes (Israel - Arabs, Pakistan - India) then the risks of a nuclear war increase.

Using the IPCC's models, what happens in the 98th percentile of bad things?

Damage which is still less than a world war.

Depends on where you think we stop. At RCP 8.5 (Representative Concentration Pathway -- 8.5 is the radiative forcing) then you are talking much worse than a World War.

Note so far we are tracking RCP 8.5 quite nicely, not the 'mid' cases of RCP 4.5 and 6.5 (RCP 2.5, the 'base' case is in fact probably no longer attainable-- it assumes improbable reductions in emissions by mid century).

And are we talking a relatively 'mild' World War like number 1 (and 2) or Number 3, which could be an order of magnitude more destructive (indicative atmospheric modelling of an India-Pakistan exchange of c. 200 weapons suggested ruined harvests globally for several years, on a planet with 7 billion people and stored food of c. 120 days; even slaughtering our food animals and (unlikely) global resort to starvation rations we are talking billions starve to death or die of hunger related diseases**).

Say 100m dead in WW2 (28m in USSR, c. 15m in China, + the rest). We are talking at least 1 bn dead if we refight it *and* we don't postulate a major nuclear exchange between (2 of 3) Russia, USA, China.

Remember the masterful novel 'A Canticle for Leibowitz' ends with the start of World War Four.

** Helen Dunsmuir's 'The Siege' is based on historical sources of the siege of Leningrad. The scene where the Soviet bureaucrat works out that if he increases the civilian ration by 150 calories per day he increases their life span in days in terms of war production by enough to justify the increase is unforgettable. And Leningrad had the advantage of a brutal police state which could enforce food rationing, shoot cannibals etc.

"The idea that human-produced CO2 will produce catastrophic global warming has now been dismissed by observational data."

No. The opposite happens to be true.

Plus of course basic physics.

Increase CO2 in a mass of air and you get about half the warming effect expected by our current models. Other GHGs add say another 10% (depending on when: ST methane has a much greater effect).

Then you add in water vapour.

It's not called the 'Greenhouse effect' for nothing. And Venus (and Mars) provide perfect demonstration experiments.

The picture for solar and wind is even bleaker if the goal is to use them to replace most of our fossil fuel use. On page 9, the author notes:

"In other words, it takes 4.28 MW of wind capacity to produce the same amount of electricity with the same degree of reliability as 1 MW of off-peak coal plant capacity.5 Similarly, it takes 7.30 MW of solar capacity to produce the same amount of electricity with the same degree of reliability as 1 MW of off-peak coal plant capacity."

What he is doing is looking at annual energy production. He says we need 7.30 MW of solar capacity to replace 1 MW of coal capacity so that with very high probability the solar will produce as much energy as the coal during the course of an entire year. That is a OK calculation if solar and wind penetration are small and fossil fuels are waiting in the background to cover the daily and shorter term fluctuations in demand that we care about. But if we want the grid to be mostly solar and wind (as some environmentalists and the entire country of Germany believe, mistakenly in my opinion, to be possible), then it is far from sufficient to just look at annual numbers. We need a continuous supply of energy every day (and at even shorter timescales); yearly averages don't cut it. This means that if we want to go all solar + wind, we a) need even more capacity (to cover the shorter term fluctuations) and b) we need storage - lots of storage. As people have noted above, only pumped hydro is anywhere near to being economical, and the availability of good hydro locations for this is quite limited. Battery technology is several order of magnitude too expensive for large scale storage and is unlikely to come down fast.

Nuclear can on the other hand does not require storage. (And it can also load follow - they do it in France). It is also a sufficient solution to getting us off carbon, at least in the developed world. It would cost around 2 trillion upfront to replace all the fossil generating capacity in the US with nuclear. A large number, but manageable spread over several decades.

What we could end up with - what even Germany may end up with - is spending a lot on solar, then realizing that it is only a very partial solution, that storage is not economical - and then having to go back to building nukes (or just keeping coal and gas). But if we have the nukes, as we should, we don't need solar - it's just a redundant expense (fuel costs for nuclear are very low).

1. The real benefit of solar is that it's already distributed. It makes the most sense in distributed applications where it reduces the burden on the grid.

2. Solar costs are now about 5c/kwh in New Mexico which is competitive with many forms of electricity generation. That's without tax breaks.

3. Solar clearly makes more sense where it's sunny. The cost of generation scales inversely with the solar resource. That's almost 2:1 across the US.

4. It's foolish to look at any energy technology as an all or nothing solution. The energy generation pie will always have a portfolio optimization, not corner solutions. Best technologies will be location and application specific.

5. To really scale solar above a few pct in appropriate regions some short term storage is necessary to be able to reduce base generation. That could be 24hr type storage for residential solar. IE the stuff Tesla is working on. Then the grid is the backup.

6. Please read point #4 again.

7. Please read point #4 again.

I definitely agree with your #4. Solar power is all about location, location, location (and sometimes month of the year), yet this article makes the broad statement "Solar farms run at only about 15% of capacity". The Brookings Institution folks had to make a lot of assumptions about capacity utilization to crunch their numbers, but those assumptions are so general that the resulting conclusions are meaningless for anybody managing a utility or directing public policy.

On your #4, add "regulatory regime specific" as well. Xcel Energy recently told the Colorado PUC that it can purchase wind power for less than the cost of new natural-gas fired generation. This is in substantial part because the Colorado Front Range is sufficiently isolated from other demand centers that Xcel is allowed to operate under somewhat different rules. In the case Xcel was describing, they could sign a 20-year contract under which they take every MWh the wind farm can generate, spinning down their own fossil-fuel generation as necessary to make that possible.

Such an arrangement is not possible in most (by population, if not area) of the US. The market model used there handicaps wind/solar substantially; the model assumes dispatchable sources, which wind and solar are not.

Environmentalism is an anti-human death cult.
The eco-fascists have no place in a modern, decent society.

Ah, I see: the contrarians are going to make nuclear power the darling technology.

My questions (which has nothing to do with any assumptions and calculations):
1. Are the Germans wrong, unwise etc to aim for something like 50% solar-generated electric power by 2030?
2. If they can do it and are doing it -- with far less solar insolation than we 2. do -- why shouldn't we?

Aim, very much like talk, is cheap. Some might say free and without value at all.

Check out what Germany is already doing before making such a statement as your.

As I recall, Germany is leaning heavily on neighbors France and the Czech Republic, and on coal for the reserve power to supplement their intermittent renewables. See this, this, this, and this for examples. It is not clear to me that they are succeeding in a sustainable way, and the article linked in the OP points to one area of concern.

A moderate amount of false alternatives is going on here. (Probably because this is being discussed by advocates rather than engineers.) While the underlying data supports the Economist article, all of this discussion ignores the most attractive alternative, which is efficiency improvement. The interesting metric there is dollars per ton of CO2 not emitted.

The ideologues ignore this metric, with good reason. The German subsidy of 0.54 Euro per KWh converts into $1350 / ton of CO2, with the current mix of generating methods in Germany. The Massachusetts subsidy of 0.50 dollars per KWh converts into $800/ton with New England power pool mix. Those subsidies are easy to convert subsidy into effectiveness. Other subsidies can be much more complex.

The pramatists include this metric in evaluating choices. The Dutch refuse to relax regulations or subsidize small wind projects because field tests of a dozen different small wind systems revealled that the real costs ranged from $1500/tom to $2500/ton. The Dutch won't even support relaxing zoning laws for small wind because they are viewed as purely symbolic emotional investments with no environmental value. Private corporations are making thousands of individual decisions using this metric. One such company is Maersk shipping, which reduced their total corporate fuel bills by 30% in the past five years. One of many fuel savings was a propellor retrofit for a few hundred ships (only some ship types are suitable) that cost about $20/ton. At $20/ton there is a financial benefit, since bunker fuel costs about $40/ton of CO2 generated. The TVA recently announced an amazingly low cost of $0.70/ton for re-design and retrofit of air pollution controls at a steel mill. That is unusual.

Estimates for the aggregate CO2 available with a net fuel cost savings vary widely. It ranges from 20% to 60% of the target CO2 reductions desired to reduce climate change risks. It's the ideologues on both sides who don't seem willing to accept that a very significant change can be made with profitable investments. Fortunately, the corporations tht pay those fuel bills are realizing how much money can be saved, and they are increasingly ignoring the ideologues.

'all of this discussion ignores the most attractive alternative, which is efficiency improvement'

Except that is explicitly not true for Germany (again, from http://reneweconomy.com.au/2014/opposite-energy-policies-turned-fukushima-disaster-loss-japan-win-germany-97060) -

'In 2010, those eight units produced 22.8% of Germany’s electricity. Yet a comprehensive package of seven other laws passed at the same time coordinated efficiency, renewable, and other initiatives to ensure reliable and low-carbon energy supplies throughout and long after the phaseout. The German nuclear shutdown, though executed decisively, built on a longstanding deliberative policy evolution consistent with the nuclear construction halts or operating phaseouts adopted in seven other nearby countries both before and after Fukushima.'


'Germany also uses energy more efficiently. In each of the past three years, German electricity consumption fell while GDP grew. During 1991–2013, i.e. since reunification, German real GDP grew 33% using 4% less primary energy and 2% less electricity, and emitting 21% less carbon. Even more ambitious savings are available and planned.'

It isn't either/or, after all. Not that would be considered too surprising in a country which tends to export engineers, not import them.

But seriously, did you just compare bunker oil fueled ship efficiency with electrical generation? It really, really isn't either/or.

I really did compare tons CO2 emitted. There are many different goals. Air pollution, geo-political vulnerability, etc. are all factors. Most of these discussions are driven by CO2 emissions. If you wish to consider only electrical power generation OK, but fungibility matters when you consider reducing CO2 emissions. Climate effects do not care whether the CO2 emission was motivated by transportation, process, or building uses nor by whether the fuel was burned to generate electricity or for some other reason.. Geopolitical vulnerability from fuel sources is considerably different, but doesn't seem to be a primary motive in these analyses. Fuel use fungibility matters in other ways too. Those ship efficiency improvements reduced bunker fuel use in the open ocean and diesel use when near Europe or North America. Reducing diesel use does matter, indirectly to electricity users and more directly to transportation users.

There are still huge savings from efficiency as your own figures show. Leaving efficiency out of the alternatives discussed is what I complain about. It's wrong to list the various forms of generation without including efficiency improvements as part of the mix. It's still easy to find electrical improvements for under $50/ton CO2, although the fungibility of power sources (including electricity) for building and industrial use does introduce some ambiguity. For instance, the steel mill that found $0.70/ton improvements was reducing electricity usage in their air pollution control systems.

Germany's improvements had two stages. They took emissions from 1,020 Megatons/yr in 1990 to 920 Megatons in 1995 and 870 Megatons in 2000. In the next 12 years it dropped to 807 megatons. The initial huge improvements were due to the replacement of grossly inefficient Soviet systems. The USA grasped improvement as a strategy a decade later. CO2 emissions were 5,870 megatons (2000), 5,940 megatons (2005), 5,500 megatons (2010), 5,200 megatons (2012), 5,000(est) megatons (2013). These improvements are primarily from efficiency improvements and primarily driven by corporate cost saving efforts.

This. I work in transportation- we don't include CO2 estimates (or any other emissions) in the cases for technology to reduce fuel spend. It's a nice bonus that goes in the sustainability report, but the dollars saved by not buying gallons of diesel justify the expense.

An average semi on the road today burns about $0.55 cents per mile in diesel, with an mpg around 6.1mpg. A new truck with a 2014 engine and a good driver can get 9 mpg, and a cng 11.9 liter engine can get about 5.9 miles per diesel gallon equivalent. Every mile per gallon improvement over average saves about ten cents per mile.

I just has solar installed on my parents' house here in Australia. It cost about $2 US a watt before any subsidy. It was a 2.5 kilowatt system and took two people about two hours to install. Using a 5% discount rate it produces electricity for under 10 cents a kilowatt-hour. That's less than a third of what my parents pay for grid electricity. As this example clearly demonstrates, Improved technology has quite simply resulted in substantial and effective competition for incumbant fossil fuel generators. This has upset a number of people, but so far they haven't been able to convince Parliment that putting solar panels on one's roof is a crime. However, I'm sure they're working on it.

So you guys pay 30 cents a kWhr for grid power? That sounds quite expensive to me.

Yes. Or no. All up I paid 44.5 US cents a kilowatt-hour on my last electricity bill. And don't worry, that sounds expensive to me too.

Australia has low population densities, making fossil fuel grid relatively expensive there, and in New Zealand. Whereas solar has really good insolation values, in most of the country. Of course, solar users are partially subisidised by having the grid there for intermittent backup, without paying the real costs of that back up.

Australian electricty was below average cost for developed countries in the past when the population density was lower. Mind you, this was before successive waves of privatisation.

Interesting. I've looked into getting solar installed recently, and prices in Austin are still about $4/watt before subsidy.

Having to pay $4 a watt is unfortunate, Komori. The good news is there is no reason why installation costs in the US can't get as low as they are here in Australia or in Germany. After all, installers aren't doing anything magic Down Under. You may not be able to get a price as low as in Australia since our installers have had plenty of practice to hone their skills and the paperwork here is extremely simple, but if you keep looking I expect you'll soon find something a lot better than $4 a watt.

It is easy to cook these books. A solar plant has high upfront cost, but never needs coal. A coal plant needs fuel for 30 years.

I better thumbnail would be amortized costs over full plant life.

When energy economists compare different energy production systems at a high level, they generally things out into three components: upfront capital costs, ongoing maintenance costs (just to produce the first Kwh), and marginal energy costs (once you make the first Kwh, how much for an additional?).

People may be interested to know that the cost of intergrating wind power in Texas is 17 times less than the cost of integrating conventional power plants:


Now I'm not particularly familiar with Texas, but I do know that here in South Australia as we increased the amount of electricity generated by wind from zero to one third we added zero additional ancillarly services (spinning reserve) to cope with it.

And I'll mention that South Australia gets almost 40% of its electricity from wind and solar. About a third from wind and over 5% from solar. And while South Australia is connected to the neighboring state of Victoria's grid we have the capacity to take an axe to the electricity interconnectors and run the state grid independently with no disruption in supply.

'we added zero additional ancillarly services (spinning reserve) to cope with it'

That spinning reserve is represented in the third of capacity that was idled - a certain amount was idled less, so to speak, than what was likely mothballed (if not actually sold).

The same applies to pumped storage - after all, the pumps don't care where the electricity comes from.

I am pretty sure that the turbines in idled or mothballed power plants don't spin.

Ah - I meant that basically, a natural gas powered plant that had been taken offline could be still maintained, exactly as before, to supply power quickly when necessary. Previously, that plant had been part of the normal system, though natural gas plants tend to placed more in the peak supplier/reserve category. That now offline plant remains part of the normal reserve, but is no longer part of normal generating capacity. In other words, nothing has actually changed.

What has changed is that one coal plant is now in mothballs, wholesale electricity prices are lower, and carbon dioxide emissions have been reduced.

You have to pay the construction and maintenance costs of your back-up CCGC! On a 1-to-1 basis! Plus you have spent to build and maintain the renewable!

Even if you have something like 100% renewables capacity, of whatever mix of wind and solar is most cost-effective, you will still be burning gas for 30% of your power on days the sun don't shine or the wind don't blow. That burn will not be at peak efficiency - overall you'll reduce your C)2 by maybe 50% and raise your total cost by maybe 100% relative to a fossil baseline.

Alistair, I spend to build and maintain the renewable because it is cheaper than relying solely on fossil fuel electricity. Those fossil fuel generators who can't make a profit under the new conditions will close down their operations. Those who can make money stick around. It's a process that occurs throughout our society and as a result nowadays it's very hard to find a slide rule or a cathode ray TV.

It would help if the article got the units of measure right! What does "1 MW per year" even mean? If you mean the average power produced over the year, the correct unit is simply MW. Power is unit energy per unit time, not unit energy per unit time per unit time. Is the article suggesting the cost to add solar capacity at the rate of 1 MW installed capacity each year..

My home's PV array produces over 8MW-Hrs in a year. Average power produced from the 6 KW rated system is 0.95 KW.

I am not sure if the error is from Tyler Cowen or the author of the paper, but this gets a failing grade.

Tyler is not an engineer, but can spot numbers favorable for mood affiliation.

Speaking of mood affiliation, it's interesting that the idea for generating your own power doesn't seem to get much support from some of the libertarian crowd. If there is anything that smacks of big government it's a public utility or big nukes. Aren't thee the same people who wanted to privatize roads? Odd.

I think most of the libertarian crowd would support a model where the market moves to the cheapest form of energy production...without government/regulatory/tax subsidies and/or chrony capitalism

Coal power plants would never stay in business in a libertarian society due to the legal costs related to the effects of pollution on peoples land and health.

The data I've seen suggests the 250PPM externality is sizeable, but not big enough to overcome their cost advantage.

The article perpetuates the myth that German investments dropped the price. In reality, the price had already been falling and then took a two year swing upwards as a result of the increased demand. It finally started coming back down once the Chinese volume came online.

@Mesa - you seem to be laboring under the illusion that your idea of libertarians encompasses all libertarians.

"....some of the libertarian crowd". Direct quote.

A friend of mine told me that Germany gets 50% of its energy from solar power. That seemed very high. So I checked wikipedia.

Germany actually only gets 4% of its electricity from solar power.

However, on one day in 2012 they managed to get 42% of their power from solar power. Must have been a rare day where all of Germany was sunny.

That should tell you something about how much capacity they have installed. Its a pity they didn't install it somewhere sunny. They could have probably powered countries in hot, sunny climes.

'However, on one day in 2012 they managed to get 42% of their power from solar power.'

That number is also absurd, by the way.

Possibly there is some confusion between wind and solar? Because several other commenters seems to be unaware of that distinction.

It's not unusual for South Australia to meet over a third of total demand at around noon with electricity generated from rooftop solar. That's kilowatts, not kilowatt-hours. Kilowatt-hour wise we get about 5+% of our electricity from rooftop solar.

I'm sure you can get good figures for wind in the middle of the night too. Max transient % of demand is not a useful statistic.

It is something that is useful to know if one is interested in changes that are occurring to the grid.

A little bit late, but Mr. Frank forgot to account for 3 unknown costs of nuclear energy. A) There is no site for final high radiation waste underground storage. Some government (taxes) has to built it first. So far, all the implemented solutions are temporal, spent fuel is stored in a water tank until it cools in a few years, later stored in a dry case. But neither the US, Germany, France or Switzerland have found and developed a final underground storage site for the spent nuclear fuel in dry cases. Since the solution has not been found and implemented, how can you estimate the COST to make a reliable assessment? B) Nuclear plant decommissioning: nuclear plants are not only expensive to built but also to demolish and manage the radioactive material. It seems the US developed a financial tool to save money for plant decommissioning (http://alturl.com/gax3r) but all the materials are stored in the old plant land waiting for the final underground repository. In Germany, they had a savings tools similar to US, but the decommissioning of nuclear plants before end of life cycle will mean a big bill directly to taxpayers because they want the plants gone and the sites clean in a few years. In France, since there's no final storage, the decommissioning of old reactors is slow and will take like 30 years, that way yearly expenses are kept low. They hope they'll have a underground storage by 2040 C) Accident insurance was not accounted and Mr. Frank is outright lying because there IS insurance for nuclear accidents. Relatively small radiation leaks like Fukushima can produce huge compensation bills. In the case of Chernobyl, well....it was Russia and compensation was 0. In the case of Fukushima TEPCO is partially insured but the government (taxes) are absorbing most (70%) of the evacuated population compensation bill. In the case of Germany, the 4 operators of nuclear reactors have created an insurance found of 2.5 billion Euros but they acknowledge is not enough and in the case of an accident the government would have to absorb the bill. ( http://alturl.com/4jchc). In the US, all the reactors owners have created a 12.6 billion fund, if the compensation for damages is greater the government pays. The price to setup of a 12.6 billion fund is not negligible as Mr. Frank stated. If people is really disgusted by subsides to renewable energy, which other industry apart from nuclear is helped this way to make insurance cheaper so private owned reactors are financially viable businesses?

In conclusion, Mr. Frank did not had and can't have the right numbers to make the cost comparison among energy sources.

"If all the costs and benefits are totted up using Mr Frank’s calculation..."

I did a quick back-of-the-envelope calculation, figuring one dollar per barrel of nuclear waste (1 barrel = approx. 200 lbs., which is admittedly a total guess on the actual weight of nuclear waste) / year x 10,000 years (since that is supposed to be the designed length of the Yucca Mountain project). That does not take into account transfer or security costs, of course.

My equation produces results that seem . . . different . . . than Frank's.


(1) Spent nuclear fuel does not come in barrels. It is solid.

(2) Almost all of the spent fuel is fertile U238, that can be turned into fissile Pu239 or simply returned underground as it is the same material that came out of the ground. Much of the remaining fraction is fissile material that can be burnt in an existing reactor. Much of the remainder of that is stable within a few decades. The remainder thereof produces much less radiation than what sits in spent fuel pools - remember, the longer lived the isotope, the less radiation it produces per unit time. It would also be less than 1% of the remainder. It could be stored in a facility that took up a very small area and pose no threat to anyone. Alternatively, it could simply be vitrified, rendering it immobile and buried miles deep, well under usable water tables. It would be separated from those water tables by thousands of feet of rock, and even if it came into contact with the shallow water table, it would be in glass (immobile). Earthquakes that broke that glass would just grind it into sand, so it would still be immobile.

(3) The federal government currently charges $0.001/kW-hr for the accumulation of waste, and this fund has almost $30 billion, but utilities are still required to store their waste at their own cost.

(4) Production of solar panels produces chemical wastes, which remain indefinitely unless treated. These can of course be recycled, but like most chemical processes it's energy intensive. Really any country that has a mix of renewable and nuclear energy sources can decarbonize while retaining it's ability to produce. Having abundant energy gives humanity the ability to recycle, desalinate water, produce steel, have transportation, and explore space. My personal preference would be to see the US able to do these things while stabilizing the environment.

Really any country that has a mix of renewable and nuclear energy sources can decarbonize while retaining it’s ability to produce

I think this sentence points to an error in the original post. If a single energy source is 'cheapest' then why do we have multiple energy sources to begin with?

Well why do we have bottled water when there's tap water just about everywhere? (And if you're really against drinking tap water you can buy a filter that will make your tap water as pure as what you get from plastic bottles). Clearly there's multiple criteria for evaluating an energy source leading us to have a mix of different types of energy.

This makes these types of posts a bit of a red herring. No one has proposed trying to replace all of our electric generation with solar panels or windmills anymore than people advocate using bottled water for washing our clothes or cars.

Comments for this post are closed