Britain's First Nuclear Power Plant In A Generation Isn't A Great Deal, But Neither Are The Alternatives

“…the strike price … this power plant is no different to that agreed for onshore wind farms, less than it is for solar power and far less than for offshore wind.”

There will be no guaranteed strike prices for solar and wind in UK. Solar and wind will be auctioned, so the lowest bidders will get strike prices. Those strike prices will be for only 15/20years and not inflation corrected. Taking into account that the circumstances for solar and wind in UK are about the same as in Germany, this will lead to strike prices that are 3 to 4 times lower than Hinkley C will get.*)

Hinkley gets £92.50/MWh at the 2012 price level, inflation corrected, during 35years. With 2%/year inflation that implies; €138/MWh in 2023 (at its start), and €193/MWh in 2040 (halfway the guarantee period).*)
In addition Hinkley gets loan guarantees worth ~€1billion/year = €37/MWh, accident liabilitiy limitation =€6/MWh, waste storage costs limitation, decommission costs limitation. Together a subsidy value of ~€60/MWh.

In Germany prices are now ~€90/MWh for solar during 20 years (going down with 1%/month); ~€80/MWh for onshore wind; ~€150/MWh for offshore wind (wind during 15years, price going down ~3%/year). The long time (~40years) price decrease for solar: 8%/year and wind 3%year, implies for the guaranteed prices in 2023: €43/MWh for solar, €67/MWh for wind and €114 for offshore wind.
For 2040 those prices will be €20/MWh for solar, €40/MWh for wind and €70/MWh for offshore wind.

So UK new nuclear costs ~4 times more than German new renewable…

Looking another way: Futures show that the German whole sale prices will be ~€30/MWh in 2023. That is >4times cheaper than UK will pay for Hinkley. **)

____
*) Exchange rate: £=>€ 1.20
**) UK whole sale prices (~€60/MWh) are already substantially higher than those in Germany (~€40/MWh), so NL exports German electricity to UK (France too).

Bas, how are you valuing Hinkley’s loan guarantee at “~€1billion/year”?

You realize that a loan guarantee is not a loan, and not a penny changes hands…don’t you?

Robert, this is excellent work.

Seldom do wind advocates address the footprint issue (which directly impacts cost, if wind must be built offshore), and you’ve laid it out in rather stark terms.

As Fukushima recedes in memory, I suspect antinuclear activists will employ more and more nonsensical reasoning to mask their phobia, and the argument for renewables will become proportionally less attractive by comparison.

The garantee concerns loans for at least £10billion = €12billion.

The normal interest rate for such investment loans is raised as the bank calculates a ‘risk’ premium, which for this type of investment is ~8%(=~€1billion/year). So while government now gets loans for 2%/year, Areva would have to pay >10%/year for this investment.

This 8% risk premium is now ‘paid’ by UK government (tax-payer); ‘invisible’ until the project goes wrong. Then the tax-payer may have to pay up to the full €12billion (worst case).

You may judge that the 8% risk premium is too much. Experience has led to the general practice that banks look into similar projects, in order to judge the risk.

So:
– check how many NPP’s were started to build in the past, how many were delivered roughly on time (NPP’s that took 25years to build, such as Watts Bar 2 in USA, probably did not pay the full investment interest to the banks neither payed the loans back);
– how many did run long enough with good P&L figures, so they did/could pay back the loans.

Substantial percentage of the started nuclear projects never could pay back their loans. Taking into account the crumbllng status of nuclear energy, the banks may estimate that prospects are even more grim. Consider also that general estimates are that whole sale prices will decrease further, roughly in line with falling cost prices of solar, batteries (so the Germans stopped building pumped storage), power-to-gas and reverse, and wind.

Check e.g. the world nuclear industry status report 2014 for important figures.
Nowadays no new NPP can be built without such government guarantees, as the banks judge it to be risky projects. So the risks are transferred to the tax-payers = subsidy by the tax-payers.

“So offshore wind subs are twice this proposed build”

You have to look into the situation in the 35 year period 2023 – 2058.

Offshore wind cost prices go down (check e.g. German FiT’s) and are now already 20% lower.
UK will auction offshore wind projects, so you may assume that those prices will reach German levels.
The last year stated wind strike prices are max. prices that will never be reached.
Furthermore those prices are not inflation corrected and last 15year only.

If you do correct you will find that even expensive offshore wind will be at least a factor two cheaper!
(check my previous post in thisthread).

“Those strike prices will be for only 15/20years and not inflation corrected.”

As usual you are wrong. Quoting from the CFD contract

“In summary, the approach to the standard CfD contract terms is as set out in August, with a 15 year contract for renewable technologies, with payments indexed to inflation (CPI), and obligations to deliver the contracted capacity in a timely manner.”

Pointless reading the rest.

Bas, I’m not going to waste any more time on this nonsense other than to point out your reference to a half-century-old nuclear plant, and your wild and unsupportable hypotheticals, only reflect on the desperation of the antinuclear movement.

In environmental terms, that’s an encouraging sign.

“Paying a French state owned company … relying on a state owned Chinese firm…A more rational policy would be for the UK government to do the investing.”

Robert do you have any idea why the UK has chosen to outsource its nuclear industry?  Don’t you have regulated public utilities (in the US, these are investor owned companies who own power plants and sell power to the public for rates which are set by government regulators, to provide a controlled profit; we also have merchant power producers who sell power on the free market, but the regulated companies build most of our nuclear plants) or did Thatcher get rid of them all?

Assume your statement comes from a 2013 governement document. As you can see here (table 1, page 7) the strike price will go down for wind and solar. But it is indexed for new projects. 
Once a project runs, no longer inflation correction. That is not strange as e.g. a PV solar provider has hardly any cost after the initial investment (so with high inflation the solar owner would otherwise get windfall profits).

But Hinkley gets inflation correction while it runs during 35 years in addition to the inflation correction during the building period from 2012 onwards.

Agree that UK government meshed things a litte
The CfD’s for renewable have changed several times in past year and will change anyway in 2015/’18 under influence of the EU (in Germany too). Then market bidding; only ‘cheapest’ renewable suppliers get a CfD.

That implies that market auctioning for renewable will be active when Hinkley will come on-line in 2023.
So to compare, you have to account for that market situation. As the costs factors are near the same in Germany and UK (~similar salaries, same PV-panels, same wind turbines, etc), it is reasonable to assume that wind & solar project bidders will then offer about the same low prices as in Germany.

Btw.
UK’s CfD’s for renewable are somewhat complicated and open for manipulation (under e.g. budget restrictions, the reference price may become unrealistic high, so government pay-out can be minimized, etc). This lawyers page explains some of it.

Right, the important thing is not the initial cost of power from a given plant, but rather the fleet average cost from all power plants.  Nuclear looks great by this metric, as long each generation of people builds their fare share of these very long lived pieces of infrastructure.  Of course the fleet average cost is better when new-plant incentives are shorter lived; maybe the next UK nuclear plant will have a 30 or 20 year price guarantee instead of 35 years (it will make a huge difference for our children, but very little difference for today’s prospective plant investors).

On-shore wind also looks good for average cost (at least in the windy central US plains), as long as the grid is fossil fuel dominated.  If energy storage is added to the windpower, the cost becomes unaffordable. 

It is simply incorrect to imply that inflation adjustment makes nuclear power less affordable in the future.  Hinkley’s power cost 30 years from now will be exactly equivalent to  £92.50/MWh in today’s currency.

It is absurd to compare nuclear cost to solar in the UK.  Nuclear provides power year around (with decreased average output during seasons of low demand due to refueling outages).  Solar provides the strongest output during the summer when UK electrical demand is lowest, and provides zero output in winter evenings when demand reaches the annual maximum (see this report which Robert linked in the article). Solar in northern climates only makes sense if the grid is dominated by expensive fossil fuel.

I wrote a DCF model for it. In my most pessimistic model, the loan and interest had been repaid after 18 years of plant operation (12 years for the optimistic model). Only two people I’ve talked to have, so far, attempted to justify the 35 year strike price with:

• (1) “they need high profits to repay the loan because the cost of borrowing will be high”.
• (2) “in future the market rate (~ current wholesale price) may increase 2.5 times above what it now is, so the strike-price protects the consumer against high future higher prices (such as £110/MWh wholesale prices)

Both justifications being wrong. Case (1) is rebuked because EDF believe they’ll pay just 3.5% interest on their loans. Case (2) is ultra-unlikely. Recent forecasts don’t foresee high price rises for gas generated electricity.

Here’s my findings, followed by my conclusion:

• After the owners have repaid the loan, the plant is making super-profits.
• Super-profits are returned even without a strike price.
• With a strike price (£89.5 / MWh), super-profits will be 3 times the normal super-profit.
• After EU’s intervention, our government will get a nice share of those super-profits. This is referred to in newspapers as “the consumer benefiting too”
• The long, 35 year, term is just a stealth tax.

@Robert,

You claim greens oppose this on the basis of fear of radiation. I’m not sure. Greens are politically fragmented into at least 3 main types:

• traditionalists – who might well oppose it because they’re afraid of radiation
• anti-capitalists – who use environmentalism as a stick to beat capitalism with. Their main anti-nuclear argument seems to be that ‘we can’t trust capitalists to run nuclear power safely’
• modernists – who favour it

The formal argument coming from green groups is that nuclear power is uneconomic, it will:

• cost 10 times more than ‘they’ claim,
• never make a profit.

This last argument actually helped EDF obtain such as ripoff deal!  If nukes will ‘never make a profit‘, a 35 year, or longer, strike price will be essential!

@Nathan

UK doesn’t have an industry able to design and build nuclear power reactors without a hitch. No reactor’s been built here for 20 years. We have a tiny university fission research sector, because we’ve spent £0 on nuclear fission research over the last 20 years [So said a recent Parliamentary report] The French have lots of operating reactors and large budgets for both private and state financed R&D.

the plant cost has been paid after 12 to 18 years. The remaining 17 to 23 years is the pig feeding in the trough.

“… I think the uk should wait 5-10 years before a nuclear build out. Let the Chinese bring back the manufacturing base first“

The way to get factories built is to place orders, and include local sourcing of components in the contract.  Remember that these nuclear builds take several years from the time the paperwork starts, so there is plenty of time work supply chain issues.

I think the UK should partner with one of the SMR companies and build 4-8 GWatts of SMRs.  As an early adopter, they could become a global supplier of parts.  Also, the smaller size of each project phase (around 500 MW) and more numerous phases would make each phase more managable and give the contractors more incentive to perform.

Bas bomb 

What rubbish.Please reread my statement and the quote I used. The price is inflation corrected over the life as the contract clearly sets out. Further I did not find the CFD terrible difficult to understand.

“Hinkley’s power cost 30 years from now will be exactly equivalent to  £92.50/MWh in today’s currency.”

No. It will be exactly equivalent in 2012 currency. Read the contract.

“..the reactor is built for a 60 year life and imo will get two 20 year extensions to give it a 100 year life…”

Check PV price developments (~8%/a price decrease during the last 40years), and the expert expectations for the future:
– 40% yield while present PV-panels have ~20% yield. So installations of similar size will produce twice the electricity.
– Continued improvements in costs and quality just as with chips; thinner layers, longer guarantees, convertors more integrated, etc.

Then you understand why experts are positive regarding similar (8%/a) continued long term price decrease during the next decades. That implies that the cost price for solar will be ~€25/MWh in 2030-2040.
Experience in USA (Vermont, etc) and Germany (GrafenRheinfeld) shows that even depreciated NPP’s cannot compete at that price level. Not even at a €30/MWh level.

So it is highly unlikely that Hinkley C will continue to produce after 2058 when the guarantee period of 35years ends (if the plant is operational in 2023)!

Note that:
Wind cost prices decrease with bigger and higher wind turbines (Denmark will install 8MW for its new wind parc, 20MW expected) as well as more massproduction.
The price decrease goes less fast (~3%/a). Still we see in USA already levels <$40/MWh. So in 2050 the price level of wind will also be <€20/MWh.

Battery prices are going down also substantial, as well as power-to-gas/fuel.

A few little drops of rain to spoil your parade here. First, the Hinkley contract starts at 92.40 pounds but escalates for the life of the contract, whereas the proposed contracts for difference for offshore wind will be flat. You conveniently neglected to mention that. Also, history tells us that the next UK nuke contract, and the one after that, and the one after that, etc. etc., will be no less than, and possibly more than, 92.40 per MWh. You might claim that will not be the case, and no one could prove you wrong, but you’d have to explain away over 50 years of commercial nuclear experience. In the meantime, offshore wind is a technology still in it’s early commercialization phase, and experience suggests the cost will decline steeply with increased commercial deployment. You say there’s no evidence that that will happen anytime soon, but they there’s no evidence it won’t, either, and to believe that it won’t you would have to explain away a couple of decades of experience with onshore wind and solar PV. So while there is a perfectly rational basis for deployment support for offshore wind, even if there’s a chance the bet won’t pay off, there is no particular justification for paying significantly over the odds for new nuclear plants. That’s not to say it doesn’t make sense to perhaps have one new nuclear plant in the 2030 mix alongside other technologies that are likely to be far less costly, but that’s a different argument than the one you’re making. Hinkley is just a bloody expensive undertaking, made even moreso by comically incompetent government policy (and isn’t that the sad tale of the “commercial” nuclear industry for the past 50 years?) even when compared to offshore wind.

Certainly any future nuclear plants should be evalutated against the cost of alternatives at that time.  But the Hinkley plant is basically guaranteed to provide cheap power after its 35 year contract expires, and any wind turbines built today are guaranteed to be decomissioned within 30 years.  

We benefit today from cheap power from long lived nuclear plants built and paid-for by our parent’s generation.  We should not complain about building expensive plants that will mostly benefit our children.

Also, note that the Hinkely power price is already guaranteed to fall to £89.50/MWh, if the sister EPRs are installed as planned at Sizewell (see here).  But it may be that other reactor designs are cheaper.  The Chinese have built both the AP1000 and the EPR, and seem to invest more in the AP1000 (which is being considered for Mooreside in Cumbri UK).  The Hitachi-GE ABWR is also being planned for Wylfa Newydd in Wales.

I didn’t say we shouldn’t build expensive plants that will benefit our children. That’s why I support deployment of offshore wind today, even at high cost, since our children and grandchildren will benefit from the improvements in cost and performance likely to follow as a result (I could even support building expensive new nuclear plant, though history tells us not to expect any significant improvement in cost and performance as a result of doing so). I said the claim made that Hinkley is actually cheaper than offshore wind as a deployment option is not supported by the facts. And you don’t know that offshore wind turbines will be decommissioned within 30 years. The history of wind power so far is that aging machines are retrofitted with upgraded equipment and power controls and continue life as refurbished and more powerful plants, in the same way that we’ve done with all other kinds of power plants throughout the 35 years I’ve been in this business. You have no basis for asserting it will be any different with offshore wind plants.

Michael, from personal experience – here is what happens to aging wind machines:

They break. They don’t get fixed. They rust, and remain indefinitely.

If the wind industry was half as environmentally responsible as the nuclear industry, a “decommissioning” fund would be set up for every new park, so when it goes under at least the hauling away of rusting junk would be taken care of.

You write:

I could even support building expensive new nuclear plant, though history tells us not to expect any significant improvement in cost and performance as a result of doing so…

as if we can use nuclear plants built 50 years ago as a basis for the performance of future plants. Can we do the same with 50-year-old wind turbines?

Pictures are nice, data are preferable. The wind turbine fleet in Europe is well maintained and constantly being retrofitted. I could show you some pretty gnarly photos of nuclear plant problems as well – not terribly informative but entertaining. As for the experience with cost of nuclear plants, the track record since the first one went online in Shippingport, PA is extremely clear and unambiguous –  the trend in real costs (that is, adjusted for inflation – you don’t need a picture of that do you?) is inexorably upward. The data are available from a number of reputable sources if you don’t believe me. Nathan mentions the Chinese plants. I’ve built power plants in China. Nothing you read about the cost of Chinese nuclear plants should be taken at face value – no one will ever know the true full cost of those plants. And would you want the Chinese to build a nuclear plant near you? I know what my answer would be. I’ll take skilled workforces and well inspected material anytime.
 

Michael, dredging up a 60-year-old U.S. commercial nuclear reactor (the first one, in fact) as an example for the future is ridiculous. And the reliability of nuclear plants, their regulation, misguided activist intervention, their generation, and yes, their price – all are headed “inexorably upward”. So what? And when did I become an advocate for Chinese labor?

Nonsensical grasping at straws – the last refuge of a dysfunctional, rusting argument.

Bob, you seem to be missing the point here. I simply mentioned Shippingport to illustrate that the US commercial nuclear industry has a long history. The data to which I referred (as opposed to lurid photographs) is the cost data for what has been built since. The “overnight” cost for the 83 plants built in the US between 1971 and 1993 rose steadily, in constant 2009 dollars, from a little less than $1000/kW to over $5000/kW. That’s a 500% increase in cost. Prior to TMI the cost had merely steadily doubled, so while the increases in cost accelerated somewhat in the post-TMI increase in regulatory oversight it wasn’t fundamentally different to what was going on “in the good old days.” Onshore wind between 1980 and 2009 went from about $250/kW to about $50/kW as a result of increased deployment. (The cost of PV in constant dollars between 1975 and 2010 declined by a factor of about 60 to 1.) Robert Wilson kicked off a debate about relative costs and the rationale for “expensive” government-backed long-term commitments. I was simply pointing out that the rationale for offshore wind is well supported by the history of policy-driven deployment of newly commercializing technologies, while the rationale for Hinkley is less clear since, if one looks at the track record for nuclear, it is not only a mature technology but one that has never exhibited any positive learning curve throughout its history. That might change, but it’s a sure sign of “misguided activism” (to use your phrase) that one would ignore one clear historical trend to support one’s pet technology and ignore another clear historical trend to trash competing technologies. The only “grasping” going on here is by you, Bob.

When you’re right you’re right. There is indeed a CPI escalation clause in the CfD for offshore wind – yet another example of incompetent government policy. Apologies. That does not, however, negate the point about expected learning effects.

“The history of wind power so far is that aging machines are retrofitted with upgraded equipment…”

Please provide evidence for this.  My understanding is that when a wind farm is upgraded, the old machines are not upgraded the way old nuclear reactors have been “uprated” (wherein the expensive reactor and buildings are retained but a new larger steam generator, turbine, and/or electric generator are added), but rather the old wind turbine is demolished, and replaced with a new one, including the foundation, and likely electrical transmission.  Only the land is reused.

“the claim made that Hinkley is actually cheaper than offshore wind as a deployment option is not supported by the facts.”

This data from the US government’s EIA shows states that off-shore wind is double the cost of new nuclear power.  

This report from the UK DECC also shows nuclear to be cheaper that off-shore wind (but by a smaller margin), but DECC uses a discount rate of 10% (which is very high, and indicates that future benefits are not valued very highly, which tends to disadvantage long-lived nuclear versus short lived wind farms).  The DECC report also predicts a substantial learning curve for nuclear, with costs dropping 12% between 2013 and 2019 (going from FOAK to NOAK – nth of a kind).

Additionally, when external grid costs are included, nuclear’s grid costs are nearly zero as penetration goes from 0 to 50%, and rises very slowly (due to curtailment), until penetration reaches around 80%.  Wind incurs external cost even at zero penetration for 100% fossil fuel backup capacity, and the cost rises quickly after around 30% penetration when the addition of storage starts becoming attractive.

I think if you look back at the source of your claim that nuclear costs more than off-shore wind, you’ll find that it comes not from government or energy industry analysists, but rather from committed anti-nuclear activists who won’t be swayed by any data that conflicts with their pre-conceptions.

A good report on wind repowering activity in Europe can be found at http://www.diva-portal.org/smash/get/diva2:644677/FULLTEXT01.pdf. Repowering of wind sites is still a relatively new business becase the vast majority of wind capacity is still within its warranty period, but all of the big players, including GE Wind, have identified it as a significant growth opportunity in onshore wind going forward. And what’s involved in repowering is, of course, particular to the industry – you retain what’s useful and valuable and decommission what’s not. In wind what remains valueable is the developed site, the wind resource, the infrastructure and the license to operate. What’s no longer useful, contrary to what Bob Meinetz’s photo purports to show, is decommissioned and removed from site, just as it is with any power plant repowering.

The EIA and DECC reports are forecasts, and they must therefore be taken with a grain of salt. In both cases they are from agencies that have historically been overly optimistic about the future costs of nuclear. EIA said a few years back they expected new nuclear to  to come in at a LCOE of $80/MWh, and a few years before that they were predicting it to be as low as $60/MWh. Olkiluoto 3 is coming in at about 15 cents/kWh based on current completion forecasts, and Flamanville is likely to be north of 10 cents, noteworthy given the policy of the French government to socialize many of the costs of EdF’s nuclear plant development program (including cost overruns). DECC’s forecast simply must be seen in context, given that they are the government arm tasked with carrying out (read: “sell”) the government’s commitment to build new nuclear in the UK whatever it takes. As I’ve already said, it’s popular for those who find themselves in DECC’s unenviable position to project healthy learning curves for nuclear, but history has been decidedly unkind to them. As for offshore wind, you’ll see that DECC put the cost well below what EIA does, which is consistent with EIA’s historical tendency to lag badly recent developments in the renewables industry. Their projection for PV, for instance, is higher than what PV is being installed for today in Germany. So I’m afraid I don’t put much store in EIA and DECC forecasts.

On “external grid costs” you’ve repeated the same old mistake all nuclear enthusiasts do. The idea that wind needs “100% fossil back-up” has been thoroughly discredited, and not just by your green bogey-men but by a number of credible independent analysts, most recently by the IEA in their 2014 report The Power of Transformation. And it is equally untenable to claim that nuclear incurs no external grid costs at high levels of grid penetration. In fact, as France’s experience has proven, high levels of nuclear market penetration are accompanied necessarily by high required investment in storage to deal with the problem of overgeneration and grid investments to deal with the large unit sizes and N-1 related issues that come with nuclear. IEA data show, for instance, that between 1974 and 2010 every 3MW of new nuclear investment was accompanied by 1MW of pumped hydro storage investment, and that doesn’t include, for instance, the French program to expand the country’s system of thermal storage facilities during the same period. So I’m afraid you share the same disease you attribute to little Green goblins – a tendency to see the problems you want to see with renewables and to overlook what you’d rather not see when it comes to nuclear. I am not anti-nuclear – indeed, I oppose the early retirement of the nuclear fleet in Germany – I’m anti-delusion. I apply the same standard to renewables – I insist on facing up to the challenges warts and all.

(1) In terms of available reactors the AREVA EPR is the only Gen III or later design currently approved by UK Office for Nuclear Regulation (ONR).  The other potential designs are GE-Hitachi ABWR and Toshiba AP1000, which are still undergoing regulatory approval. The 2nd hitch is that EDF own the site. They have their own technology roadmap, which, presumably doesn’t involve  their staff having to learn about too many diverse reactor designs. I don’t disagree with Robert’s suggestion, but it would require quite a bit of state involvement and steering. UK government of all complexions seem keen to avoid such involvement. Government affect a free market mindset. We don’t have a free market for energy supply, only EDF can commission it, only 1 design is currently approved.

(2)

== “Cost, however, is not what opposition to this power plant is really about” ==

True of greens but not all opposition is green. If it were half the price, greens would still oppose it, but other opponents would be all for it.

Michael, 

Building on the work of people like MIT economist Joskow, a recent series of peer reviewed papers by Hirth analyze in great detail the costs of integrating variable renewables. The papers seem (to me) to be extremely solid both on the empirical and modeling side. Two of the papers are here:

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2200572

http://www.feem.it/userfiles/attach/20131028948484NDL2013-090.pdf

The first paper shows that at around 20% market penetration, marginal wind costs are around twice the naive generating cost. This is mostly because of what Hirth et al call ‘profile’ effects – lots of thermal generation needs to be retained for when the wind isn’t blowing. Wind lowers residual (non-wind) peak load requirements only slightly. See the plot on page 14 for Germany. This is consistent with what Nathan claimed above. 

The second paper is even more interesting in some ways. It tries to calculate the optimal share of variable renewables in the generation mix. One of the most interesting results is that for moderate cost of carbon, optimal wind generation share increases. But past a carbon value of around 40 euros per ton, the optimal share of wind decreases, replaced by nuclear and CCS, the latter two technologies being far more effective at reducing carbon emissions.

It seems from your comments above that you disagree with the results of Hirth and the earlier papers by Joskow. If so, I am genuinely curious to know specifically where you think these papers go wrong.  Cost of integration or the cost of a whole energy system or whatever you want to call is an extremely important (and rather difficult) subject.  Any light you might be able to shed on why Joskow and Hirth and many others who have studied this subject are wrong when they argue that the cost of integrating variable renewables is quite high and monotonically increasing with penetration level would be genuinely interesting. 

UK Government wants to auction new renewable projects*). So the fancy prices of UK government are max prices that will never become reality (would generate windfall profits for the owners).

As it is a new EU rule, Germany reluctantly will follow restricting itt more than UK for (big projects only, etc).

My mistake, I should have said, “wind requires near 100% fossil fuel backup capacity”.  No amount of hand waving can change the fact that sometimes there is zero wind across very large areas.

Deployment of energy storage with nuclear is an optional cost optimization to control curtailment.  It is clearly the case that for any level of (fossil fuel + hydro) penetration below 50%, much less storage is required in a nuclear-rich grid than in a wind dominated grid.  Furthermore, in grid with a summer demand peak, nuclear and solar together will require less storage than wind+solar.

Hirth is a Vattenfall employee whos study results support the portfolio of Vattenfall (nuclear, FF) more than any other…

To both Nathan and Jeffrey, I am not only well aware of Hirth’s papers, I know Lion quite well. He builds on work from Ryan Wiser at LBNL, whom I also have known well for many years. Both have done very good work in this area, and both would tell you the same thing – which is that they’ve described the challenge posed by intermittent resources, not the necessary end state. The IEA’s work, which builds on their work as do other similar analyses and on which Lion was an advisor (as was I), describes all of the solutions one can implement using existing technologies, concluding that the amount of “back-up” as you call it (not a very apt term but I won’t quibble – what we’re talking about are the same suite of frequency, regulation, operating and contingency reserves all power systems require to some extent – an especially useful example in this case being the fossil plants that are kept in reserve for hydro systems in the Pacific Northwest) can be reduced by a factor of nearly 10. In other words, in a system that is transformed in sensible ways to deal with the growth of intermittent resources the need for reserves is slightly larger but not dramatically so than one that continues with “more of the same”, just as a system that is transformed in sensible ways to deal with the growth of inflexible nuclear requires far less in additional reserves than one that is not.

Further to my post below, if you want to read the analysis from the IEA of what one can sensibly do to address the integration issues described so well by Hirth and Wiser, rather than just sit there and act as if nothing’s changing, the report is called The Power of Transformation, published early this year. You’ll see that Lion is listed as one of the advisors to the report (as am I). The report also alludes to, though it’s not the focus, the integration issues posed by large shares of inflexible nuclear plant. They are solvable as well. No point in describing a problem unless one is going to ask the next question, which is whether or not there are sensible, readily available solutions. That’s what an engineer does. (Joskow, by the way, whom I also know, is a very good economist, but he’d be the first to tell you he’s not an engineer.)

And one more thing to you Nathan. Deployment of energy storage with nuclear is “optional” only in the same sense as the measures in the IEA study for renewables are optional – you can certainly have a reliable system without them, but it would be much more expensive. If the nuclear-intensive French system wasn’t a heck of a lot less expensive to build and operate with the extensive system of pumped storage hydro and thermal energy storage facilities they’ve built, I really doubt they would have spent the money. They certainly didn’t make those huge investments just for the fun of it. Nathan, you need to start facing up to the fact that you exhibit a positive bias when it comes to nuclear – things that for renewables are insurmountable obstacles are, for nuclear, just features of a system with a lot of nuclear that represent optional solutions. Nothing to be embarrassed about – renewables advocates suffer from the same positive bias – but it is what it is.

== in a system that is transformed in sensible ways to deal with the growth of intermittent resources the need for reserves is slightly larger but not dramatically so ==

The only renewables path sensibly proposed for the UK is wind and more wind. Other technologies are either unavailable or way too expensive. Wind requires 98% backup for it’s nameplate capacity. For example: 100 GWe of installed wind generates between 2GWe and 54GWe here, averaging 26GWe. UK demand varies between 28GWe and 59GWe. UK is subject to periods of calm. A cold, calm winter evening may see only 2GWe from 100GWe of wind, leaving a deficit of up to 57GWe from what?  If you want to contradict me, please post a worked out UK renewables electricity plan. UK capacity right now is ~ 78GWe (including 10.3GWe wind, with several plants offline being serviced).

PS: I’m not interested in the Pacific Northwest. I don’t live there.

Michael,

Thanks for your detailed response. I’m putting this response at the top of the thread because the text window is getting narrow below. I would very much like to read the ‘The Power of Transformation’. Is there a free version somewhere on the internet you could point me to? (I’m not sure I want to pay 80 euros for the report).

This whole wider discussion is about finding the optimal path toward a low carbon future where optimal means rapidly scalable (the climate change imperative) and as low a cost as possible (economic resources are scarce and shouldn’t be wasted). Finding this path requires accurate estimates of total system costs, as I’m sure you agree.

The best kind of data for this, when it is available, is real world data – what actually happens when you try something. In that context, we have clear empirical data on both the nuclear side and renewable side. France went 75% to 80% nuclear in a couple of decades, its electricity costs are average for Europe, and its and carbon emissions are very low.  On the renewable side there is Germany which despite spending huge amounts of money on renewables is still mostly powered by fossil fuels, has high electricity costs, has high carbon emissions, and a very large share of their renewables are not in the variable renewable category (biomass and hydro). Knowing just these two wildly divergent real world experiences, I know where I would put my bets for the most effective practical path toward rapid decarbonization.

Coming back to your specific comments, Hirth and Joskow make some definite claims in their respective papers. Just to be clear, which, if any, of the following statements from Hirith’s paper on ‘System LCOE:What are the costs of variable renewables’? do you disagree with?

Page 1 (Highlights):

• Integration costs of wind power can be in the same range as generation costs at moderate shares (~20%).
• Integration costs can become an economic barrier to deploying VRE at high shares.
• A significant driver of integration costs is the reduced utilization of capital-intensive conventional plants.
• An economic evaluation of wind and solar power must not neglect integration costs.

Page 11: “VRE contribute energy while hardly reducing the need for total generation capacity in the power system (reflected in the low capacity credit of VRE). Thus the average utilization of dispatchable power plants is reduced (reflected in decreasing full-load hours). This utilization effect (Nicolosi 2012) leads to inefficient redundancy in the system and higher specific costs compared to the hypothetic situation if wind and solar would not be variable.”

Page 12:“Second, VRE hardly reduce the need for reserve capacity especially during peak load times due to their low capacity credit.”

Regarding the cost of nuclear integration which you comment on, are you suggesting that the cost and difficulty of integrating nuclear at say 75% penetration is comparable to the cost and difficulty of integrating variable renewables at a similar penetration? I thought the general consensus was that the latter problem far, far more difficult and expensive (for example, unlike nuclear, it has never been done and does not seem to be easy – see Germany again), but perhaps I am misinformed.

Jeffrey, read the IEA report, which (again) Lion advised on and with which he largely agrees. If you don’t want to pay for it, try the various reports from the Western Wind and Solar Integration study by NREL, GE  Energy and others, which is also a very robust analysis by technically sophisticated organizations who have thought long and hard about these issues. The two key points about the report you quoted are (1) he says the integration costs “can” be very high, but he also agrees with the conclusions of the IEA report that the integration costs can be quite low, it all depends on how you adapt the system over time (or not) to account for the growth of variable resources; and (2) if I’m not mistaken those are fairly dated quotes at this point, made well before the various studies demonstrating that these issues are mitigable to quite a degree using various available strategies.

As for “real world experience” I have a few points in response. First, where France is today is the result of a massive state-funded investment program carried out beginning in the late 1960s that included not only the nuclear plants themselves but huge investments in the network and in various forms of energy storage. No one will ever know just how much it all cost because so much of it was socialized across the various sectors of the French economy. Where Germany is today is that they are paying – and yes, they are paying quite a bit – for a government decision early on to commercialize technologies they will need to achieve their climate and energy objectives. Many of the current PV facilities, for instance, were built at a time when feed-in tariffs were 45 euro-cents a kWh. The feed-in tariff for rooftop PV today is 13 euro-cents and continues downward. One big difference between where France was decades ago and where Germany is today is that in Germany we know exactly what it’s costing, whereas no one knew then or knows today exactly what France’s nuclear transition cost. For anyone else starting from where Germany has brought us it will cost far less simply because of the great gift the Germans have given the rest of us – cheap, commercial solar that continues to get cheaper.

As for placing your bets, if youi really believe the nuclear industry value chain can deliver the construction of 700 GW of new nuclear plants in the next 20 years I’ve got some land in Florida to sell you. No single technology – be it nuclear, solar, wind or any other of the low-carbon alternatives – will be able to do this alone. Nuclear may well be a part of the mix, but so will wind and so will solar. All of the reputable studies that have been done on the transition agree on some version of that conclusion.

As for comparing the cost and complexity of integration high shares of nuclear vs the cost and complexity of integrating high shares of renewables, I think it’s a good start that you’re asking the question. At least we finally have some acknowledgement that integrating large quantities of capital intensive, very low production cost, inflexible/intermittent resources into the system is a challenge, regardless of what they are. Speculating on comparing the two would be entertaining but I would simply encourage you to read the studies, as well as others that are out there.

Michael,

Thanks. “try the various reports from the Western Wind and Solar Integration study by NREL, GE  Energy and others”. Could you provide specific references or links? I really would like to read them.

“(2) if I’m not mistaken those are fairly dated quotes”. Paper published Dec, 2013, less than a year ago.

“As for placing your bets, if youi really believe the nuclear industry value chain can deliver the construction of 700 GW of new nuclear plants in the next 20 years I’ve got some land in Florida to sell you” 

This depends very much on how serious we are about climate change. If we were serious instead of just pretending to be serious – hosting endless summits and pointless international conferences – the world could build 700 1GW nukes in twenty years. A good start would be a significant, rapidly growing, revenue neutral  carbon tax. That would change the economics overnight. But beyond that, we would need to mobilize large state resources (e.g. loan guarantees, expedited regulatory processes, etc) to get it done in twenty years. But it’s clearly possible. France in the 1960s was far less wealthy and far less technologically sophisticated than we are in 2014; if they could do it fourty years ago, we clearly could do it now – if we wanted to. There would be large economies of scale. It’s very easy to have cost overuns if you build a handful of plants per decade. But if any government were to embark on an ambitious nuclear project like France’s, they would have plenty of competition between firms and lots of leverage in keeping costs down. (Parenthetically, I’m reading a book on history of the integral fast reactor. It’s amazing what a difference there is between the 50 and 60s and today in terms of nuclear progress. The authors talk about how a handful of twenty and thirty year olds built the first breeder reactor EBR II- also the first nuclear power generator – in a couple of years and then watched the thing run perfectly for the next 30 years, passive safety systems and all. Nowadays, doing something like this would be impossible. Why is that? Did we forget how to do it? No bright thirty year olds out there? Or did we throw up so many hurdles to progress based on irrational fears of radiation, that we’ve tied our own hands in dealing effectively with the much more serious risks of climate change?)

My real bet – which I hope is wrong –  is that we continue to do very little and that we will have an energy system that in 2050 which is still dominated by fossil fuels.

“Nuclear may well be a part of the mix, but so will wind and so will solar.” I don’t disgaree with that.  Wind and solar should be part of the mix. But I am skeptical – based on Mackay’s basic physics analysis – whether wind can ever be a dominant part of the mix. Solar has much more potential than wind, but suffers even more severely from integration problems.

“I think it’s a good start that you’re asking the question.” Well, it’s not really my question. You raised this point – that integrating nuclear is costly – and I asked you to quantify it, which your answer doesn’t do. Given the experience of France successfully integrating nuclear at 75%, I am skeptical that integrating nuclear (which can be dispatchable) is a huge hurdle even at very high penetration rates, unlike variable renewables where it clearly is a big deal even at modest penetration rates, but I am happy to be shown wrong. For a given level of penetration, how does the the integration cost of nuclear compare with that of variable renewables?

Jeffrey, I’m pasting part of an online debate which originally appeared on PhysicsWorld.com between physicists Peter Hodgson and Dennis Anderson. It was titled “Do We Need Nuclear Power?” and it made similar points over ten years ago.

Meeting the world’s energy needs is an urgent problem – and all practicable energy sources must be used to solve it. The exact mix in different regions will depend on many factors, particularly the indigenous fuels as well as local geography and economics. Developed countries must help developing nations to increase their energy supplies and curb existing wasteful habits. Continuing efforts must be made to reduce pollution and carbon-dioxide emissions. To make progress in discussions about energy production and the effects on the environment, it is essential to have numerical data. Without such information, it is impossible to know whether a proposed source or effect is important or negligible.

If we are to stabilize the emission of carbon dioxide by the middle of the 21st century, we need to replace 2000 fossil-fuel power stations in the next 40 years, equivalent to a rate of one per week. Can we find 500 km2 each week to install 4000 windmills? Or perhaps we could cover 10 km2 of desert each week with solar panels and keep them clean? Tidal power can produce large amounts of energy, but can we find a new Severn estuary and build a barrage costing £9bn every five weeks?

Nuclear power, however, is a well tried and reliable source, whereas the alternatives listed by Anderson are mainly hope for the future and have yet to prove themselves. At the height of new nuclear construction in the 1980s, an average of 23 new nuclear reactors were being built each year, with a peak of 43 in 1983. A construction rate of one per week is therefore practicable.

Bob, Thanks – I agree with their conclusions. I also agree that the huge footprint of wind and solar is often overlooked.

We should also reference Schalk’s excellent review of Hirth’s 2015 paper:

http://theenergycollective.com/schalk-cloete/333521/optimal-share-interm...

So now France targets further reduction of the share of nuclear produced electricity towards 50% in 2025.

The twin reactor Fessenheim power plant to be closed next year,  etc.

Fessenheim being the first, partly because it is unsafe as its basement is too weak to withstand an earth quake, while it is built on a fault in the earth crest. So an earth quake prone area.

Target of France is not only to close the most unsafe reactors they built in the great nuclear expasion years.

Their primary target is to reduce the volume of nuclear generated electricitiy to 50% max.
A rational reason for that can be seen when one looks to Belgium.

This summer Begium had to close half of their nuclear fleet due to technical problems. So they now have a problem to ensure their electricity supply this winter.
Especially since their interconnection capacity with Germany is not much.