The cost of displacing fossil fuels: some evidence from Texas

by Peter Hartley

Although technological progress can alter the relative costs of different energy sources, depletion inevitably must raise the costs of fossil fuels leading to their displacement by alternative energy sources. What energy technology will likely be used when fossil fuels are no longer dominant? More importantly, what will be the cost of that alternative “backstop” energy supply?

The higher the cost, the longer will fossil fuels remain viable as an energy source. A higher cost of energy at the time of transition also implies that more total fossil fuels will have been used by then, regardless of the prior trajectory of fossil use. This has implications for the total stock of emissions released by fossil fuel combustion. Finally, a higher cost of the backstop technology will imply there is more of an “energy crisis” around the transition time as resources devoted to energy production and investments are taken from other productive activities.

Currently, wind or nuclear are the most feasible alternatives to fossil fuels for electricity generation. Bulk electricity storage will be required to enable wind to meet load, and it may also assist nuclear by allowing plants to operate at full capacity at all times. While many believe that batteries are the storage technology of the future, pumped hydroelectric storage represents more than 99% of current worldwide bulk storage capacity. It is the most mature technology, and is capable of storing large amounts of energy for substantial periods of time. Capacities of operating systems range up to more than 3GW. The plants have 70-85% cycle efficiency and a lifetime of more than 40 years. Batteries currently have limited capacity, limited lifespan, high self-discharge rate, and high maintenance costs. Data from the Energy Information Administration (EIA) implies that the levelized cost of the lowest cost battery system is currently at least 50% higher than the cost of pumped storage. We therefore used the cost of the latter as a reasonable lower bound for costs that could be attained by other bulk electricity storage technologies after further R&D.

The ERCOT ISO in Texas provides a realistic model for examining the costs of replacing fossil fuels by wind generation and storage, and for comparing wind power with generation based on nuclear and storage. ERCOT is relatively isolated from neighboring grids, and wind power was almost a quarter of its total generating capacity at the end of 2016. The paper contrasts the cost of meeting the ERCOT hourly loads for 2016 using wind plus pumped storage versus using nuclear plus pumped storage. Both systems also included natural gas open cycle turbine (GT) capacity equivalent to 10% of maximum hourly demand as reserve capacity and to provide ancillary services.

We used data on capital and operating costs of different types of generation from the EIA report, Capital Cost Estimates for Utility Scale Electricity Generating Plants. The cost estimates presented in the report use, where possible, data on actual or planned projects in the United States combined with generic assumptions for labor and materials costs. The cost estimates were developed using a common methodology across technologies. They represent the costs of a generic facility in a location that does not have unusual constraints or infrastructure requirements (including needed transmission upgrades). The EIA uses the estimates to develop energy projections and analyses, including forecasting retirements of old plants and the mix of generating capacity additions needed to serve future electricity demand.

We found that the wind plus storage system required almost double the storage of the nuclear plus storage system. The reason is that storage has to serve two functions in the wind system. As in the nuclear plus storage system, it has to cope with variations in demand. In the wind system, however, storage also is needed to offset the large daily and seasonal fluctuations in wind output. The high cost of storage then implies that the wind plus storage system is more costly even at unrealistically high real weighted average cost of capital (WACC) values of 10% per annum. Such high discount rates normally would be expected to greatly disadvantage capital-intensive nuclear generation.

This result has an important implication. It is often argued that storage would solve the problems with wind generation – its intermittency, non-dispatchability, and generally negative correlation with system load. Our result implies, however, that far from making highly variable and uncontrollable sources of generation more competitive, storage would in fact better advantage stable and controllable generation. With storage, such sources can be used to continuously and reliably supply the average load at low cost.

When we allow natural gas combined cycle (CC) and open cycle turbines (GT) to be included in the cost minimizing system, the natural gas plants can provide backup for the wind or nuclear plants at lower cost than pumped storage. Overall system costs are, of course, much lower, especially for recent natural gas prices in the United States, which are very low by historical standards. In fact, the cost of natural gas generation is currently so low that, at a realistic real WACC for electricity investments of 7.5%, natural gas prices would need to be more than triple the 2016 average value before wind or nuclear capacity would be included in the minimum cost system.

At a WACC of 7% or above, the cost minimizing solution over some range of natural gas prices involves wind and natural gas with no nuclear or storage. At still higher natural gas prices, nuclear is added. Eventually, as already stated, once natural gas prices are high enough to eliminate natural gas from the system (except for GT contributing emergency backup capacity), only nuclear and storage remain. The fact that the cost of a wind plus natural gas system is less than the cost of nuclear plus natural gas, but nuclear plus storage is less costly than wind plus storage might appear contradictory. The explanation, however, is that wind needs much more backup capacity than does nuclear. When that backup is expensive storage, the system with wind has higher cost, but when it is less costly natural gas plant, the combined system including wind can have lower cost.

The fact that a wind plus natural gas system includes more natural gas to backup the variable wind output also means that wind capacity is much less effective at reducing natural gas use, and thus emissions, than is nuclear generation. At a WACC below 7%, nuclear rather than wind displaces natural gas generation as natural gas prices rise. Where the transition is from natural gas to wind, the reduction in natural gas use is on the order of 30%, while when nuclear displaces natural gas the decline in fuel use is more like 50%.

A related point is that we find that increases in the cost of natural gas have very little effect on the total amount of fuel used until the price is high enough to trigger the entry of wind, or especially nuclear, generation into the system. A given percentage increase in the price of natural gas raises costs by around 40 times the percentage that it reduces natural gas use.

We also found that wind was included in the minimum cost system for a fairly narrow range of natural gas prices (about 3.15–3.7 times 2016 prices for a realistic real WACC of 7.5%). Furthermore, for these prices, constraining wind capacity to zero does not raise total costs by very much. Although it would slightly delay the exit of natural gas capacity from the system, it also advances the use of nuclear. As a result, the ultimate effect on natural gas use and CO2 emissions (especially cumulative emissions) would be trivial to non-existent.

Finally, greater uncertainty about the potential marginal climate impacts of CO2 emissions might also favor the use of more nuclear power in the short term by increasing its option value. In particular, if new information reveals a greater urgency to transition away from fossil fuels for environmental reasons, this will be much easier if there is more nuclear and less wind capacity in the system.

Link to the forthcoming paper (unfortunately behind paywall).  An earlier working paper version can be found [here].

Moderation note:  As with all guest posts, please keep your comments civil and relevant.


100 responses to “The cost of displacing fossil fuels: some evidence from Texas

  1. Standard Economics takes no cognizance of and is not in conformity with 2nd Law of Thermodynamics!!

    Dr.Charles A. S. Hall
    Economics Is Not A Science Because It Doesn’t Use the Scientific Method

    Charles A. S. Hall discusses the faults of the “Dismal Science”

    “Economics is not a science because it doesn’t use the scientific method”
    “ Don’t tell me dollars. Tell me energy. Because Dollars are only a lien on energy. That’s all they are”
    “Encourage us not to teach fairytales in economics classes. We teach a million young people fairytales in our Economics classes”
    “I had a wonderful talk at our biophysical economics meeting last week. And the speaker was an historian. He said the discovery of the 2nd law of thermodynamics absolutely transformed chemistry first, then physics, then all of the.. geology.. all of the sciences.. ecology.. Except one.. Economics. ”

    • David Springer

      The answer you’re looking for is decentralized harvest of solar power and storage thereof via hydrogen bonds in synthetic fuels which are direct replacements for those in use today.

      Leading candidates are genetically modified photosynthetic bacteria which excrete ethanol and fuel oils as a metabolic waste product and artificial leaves. I’m betting on the former being first to commercial success but there’s probably 20 years before a winner emerges.

      Thanks for asking!

  2. For a more detailed explanation of the factors Professor Hartley mentions in his opening paragraph, see this chapter from Sir Ronald Prain (1975’s classic text Copper the anatomy of an industry.

    Mineral demand, mineral resources, and technology are in a complex dance, with price the mediating factor. History — both the past decade (“peak oil”) and the past two centuries — shows us that accurate predictions are difficult — perhaps impossible.

    We see that in the widely cited RCP8.5 scenario — the worst-case scenario used in AR5 — which assumes that coal will be the fuel of the future in the late 21st century. My guess is that is not our future.

    “Currently, wind or nuclear are the most feasible alternatives to fossil fuels for electricity generation.”

    True, of course. But are they the most likely alternative over the time horizon in which “depletion inevitably must {substantially} raise the costs of fossil fuels”? Note the inserted word. That’s probably several decades.

    Much R&D is going into solar. Looking further out, fusion is attracting venture capital. These are smart people with relatively short time horizons. Tri Alpha Energy (now TAE Technologies) has reportedly raised over $150 million (see Wikipedia and their website). In July they passed a major milestone — with a long road ahead of them.

    Interesting times lie ahead of us. Expect the unexpected.

    • And see the current WUWT thread on breakthroughs in boron-hydrogen fusion power at:

      • The most promising experimental fusion device is FF-1 from LPPFusion in New Jersey. This reactor has already hit two of three of the Lawson Criteria to achieve fusion (temperature and sustained reaction time). The company is rapidly closing in on achieving the final criteria: plasma density. Currently #5 on the Fusion Leader Board, LPPFusion relies on Dene Plasma Focus theory, rather than tokomaks (magnetic confinement) or lasers (inertial confinement).

        Fusion Leader Board:

        How it works


    • Nuclear fuel is effectively unlimited.

      Uranium at extractable concentrations in the upper continental crust could supply all the energy for 10 billion people consuming energy at the current US per capita rate for 20,000 years. There’s 4 times more thorium than uranium. Then there fission.

    • Peter,

      “Nuclear fuel is effectively unlimited.”

      Yes, but uranium etc have the relationship that Pain describes: an inverse relationship between ore quality (broadly defined) and quantity. Lower quality ore (difficult to reach and-or difficult to refine) is more common than high quality ore. This pushes prices up over time.

      Improved tech (mining, refining, transportation, etc) pushes it down.

      Also — oil availability is a key factor in how much oil we burn. But U prices are a less important factor in the desirability of nuclear power. While safe in theory, the frequency of scary accidents makes many people worry about its utility in practice.

  3. After the Texas storms this year, should hurricane/wind/rain damage be considered for any wind (or solar) stability costs? Are the wind variability issues only one portion of the risk to supply costs?

    If a storm destroys 20% of the supply overnight, is there the backup and for how long?

    • The #1 reason for loss of electrical power are transmission line failures. Most are quickly repaired by the small army of line crews the utility companies keep on retainer. Utility line workers are one of the top 10 most dangerous professions in America.

      • Power went out a hour ago here in Ft. Worth TX (9:02 AM CST, 12/15/2017).
        UPS kicked in for my office equipment and satellite TV.
        Had to tether my internet off my cell phone to check ONCOR outage map which says power might be off till 2:00PM :(
        So I switch my AC main panel to my Chevy Volt’s 1500w AC inverter till power gets restored. Too bad my solar panels automatically disconnect the grid goes down when they loose the reference frequency/voltage.

      • This is interesting strategy. We just installed a natural gas back up generator, with a propane backup to that (the old fashioned way). Didn’t install solar, since it doesn’t help in power outage situation

      • My power is back on! Only offline for 2 hours so the ONCOR crews are pretty fast. I knew something wasn’t right since there have been several small power ‘blips’ since Monday and I guess it finally failed today. Nice thing about smart meters is the repair crews know exactly where to go when just a small local area goes out.
        I didn’t have the Volt when I installed the PV array or I would have wired it to get the reference freq/voltage from my pure sine wave inverter when I switch to backup power. Someday batteries will be cheap enough to install with auto switching and I’ll be ready for extended blackouts.
        Did you go with the Generac system? They have very good reviews but don’t forget to have it serviced and change the oil every 200 hours/2 years because of the exercise cycle (it runs 5 minutes weekly on some models).

      • Utility companies do not keep line crews on retainer. When storm recovery goes into effect, all utility crews are pulled off their projects and assigned to restoration. When the extent of the damage is large, utilities activate agreements with neighboring utilities for supplying outside crews. When an event impacts more than one regional utility, the utilities go further afield. We have sent crews to Conneticutt, Florida and Texas from the PNW. Utilities (or the service providers) may also try to hire crews from outside the region. Earlier this year I heard a neighboring utility lost several crews working on small cell installations to Texas. The bonus pay was worth them quitting steady work. Not that it was a big risk in the first place. Qualified linemen are in short supply.

    • USA grids are flimsy but can be beefed up. For example Japan has a much sturdier grid, because they have typhoons.

  4. Curious George

    Professor Hartley, could you please supply some data on the cost of displacing fossil fuels, like the impact on the cost of electricity or on the cost of gasoline or diesel or natural gas?

  5. Just remember that wind and solar don’t need water to produce power. All this worshiping over thermal power generation tends to overlook this key dependence on the availability of water. Water that must be must be clean and available in great quantities between specific temperature ranges. If the intake water is too hot or too cool the power plant will have to throttle back or shut down. The last time there was an actual blackout on the ONCOR grid was because of the water supply, not fuel.

    FYI: Texas is already in drought conditions and burn bans are in effect across much of the state.

    • Jack,

      (1) Texas!

      That’s true, and this trend deserves attention — with 45% “abnormally dry”, a condition that began in November. But only 26% of the State in “moderate” or worse drought.

      It would be tough on Texas to relapse back into drought after a round of Big Drought – Big Flood.

      (2) “Water that must be must be clean and available in great quantities between specific temperature ranges.”

      They can use closed circuit cooling systems instead. Higher installation costs, higher operating costs — using much less water, and less sensitive to temp of water input.

    • “Just remember that wind and solar don’t need water to produce power. ”
      Neither do thermal plants. They just are more efficient with water cooling instead of air cooling. (the water in the power loop is recirculated, not dumped.)

    • Can use air cooled condensers, which are becoming more common on combined cycle natural gas plants. Your theory does not “hold water”!

    • APS has three units west of Phoenix – i.e. in the middle of the desert. Cooling water is provided by the City of Phoenix waste water system. An on site water treatment facility gets the relatively pure water to the level required for reactor cooling systems.

  6. Pumped storage: with low wind availability the required pumping capacities are very high. Pumping must be fast for the short time as windpower is on; and then power can be slowly generated with the turbines for the long remaining time. Investments are disproportionate because of this asymmetry factor.
    Interseasonal storage: where to put the required water volume (at a useful altitude)? Is it possible to build dams and flood entire valleys for this energy losing proposition? (in Europe, permits would be impossible to get).

    It follows that to catch intermittency of windpower (or photovoltaic), standby generation with gas may well be the only manageable solution. But if you have to have the standby, why should you build wind turbine in the first place?

    • With the exception of the proposed Eagle Crest facility in Southern California (two abandoned open pit iron mines), there are almost no viable pumped storage sites left in the United States. Similar to Europe. So US load leveling is increasingly done with CCGT that can run on roughly equivalent thermal edficiency anywhere from 40% to 100% of rated output.

      • Battery storage (unlike pumped hydro) is inverter based and does not as readily supply the benefits of synchronous generation resources (Hydro, coal, gas, nuclear and biomass). When/where pumped hydro is off the table, gas generation gets a strong boost over storage as a “backup” for wind.

        Question to ponder – If you start with a wind system and see how much gas you need to supplement it, do you design a similar system as you would if you started with a gas system and added wind seeking to save fuel cost by displacing gas gen with incrementally “free” wind?

      • Curious George

        “is inverter based and does not as readily supply the benefits of synchronous generation resources” – is it just because the technology is not there yet, or it a limitation in principle?

      • See comment below about Duke Energy pumped storage facility.

      • Curious George – At increased costs inverter based generation can emulate synchronous generation. So may be the technology can get there, but the beneficial physics is integral to machinery rotating in synch with the grid. I’ve written on this a bit here over the years. Here’s one link.

  7. Read the draft working paper with great interest. Interestingly, Planning Engineer and I used the ERCOT grid to help estimate the true cost of wind (in the guest post here of that title some time ago) because the EIA LCOE was so obviously off. At current natural gas prices, CCGT is about $56/mwh and wind is actually about $146/mwh. A very bad economic deal.
    So low cost is neither wind nor nuclear, it is HELO coal or CCGT. And in both cases, it can be shown there is likely low to no offsetting ‘cost of carbon’. About 35% of all the CO2 increase since 1957 (Keeling curve) occurred this century. Yet except for the now cooled 2015-16 El Nino blip, tempurature has not increased except by Karlization. And the planet is greening, polar bears are thriving, and sea level rise has not accelerated.

  8. Speaking of energy storage I found this to offer a interesting twist on the idea of pumped hydro which is in essence a gravity based system.
    ARES energy storage technology has a system that is scalable in power and energy ranging from a small installation of 100MW with 200MWh of storage capacity up to large 2-3GW regional energy storage system with 16-24GWh energy storage capacity.
    They are currently building a medium size test system in Nevada.
    No emissions
    No fossil fuels used
    Water not required
    No hazardous waste created
    No harmful extraction of materials
    Clean decommissioning process with no lasting impacts to the natural environment.

  9. Reblogged this on Climate Collections.

  10. Peter Hartley, thank you for the essay.

  11. Peter Hartley,

    What energy technology will likely be used when fossil fuels are no longer dominant?

    There is only one that can replace fossil fuels. It’s nuclear power.

    Nuclear power:

    – can provide all the world’s energy demand, effectively indefinitely, because nuclear fuel is effectively unlimited (uranium, thorium; fission, fusion)

    – Can supply all the world’s transport fuels demand as well – petrol/gasolene, diesel, jet fuel and any other from sea water, and nuclear power generated electricity and hydrogen

    – Reliable, fully dispatchable power supply (yes, can follow demand load too)

    – The most environmentally benign way to generate power

  12. The take I get is that nuclear with storage is a lot better over the long run than wind with storage for a given area. I suspect the wind advocates will argue that wind will be more valuable over much larger areas if they can be inter connected.

  13. Coal has been priced out of the market by the endless war against coal waged by the EPA. A coal plant that was already built and running would beat the heck out of building windmills or solar or even new gas plants. More than necessary regulation and requirements to reduce emissions that were not causing any problem has unfairly penalized coal and put many people out of work for no good reason.

  14. Duke Energy has a pumped storage facility in the South Carolina mountains near where we used to vacation.

  15. In principle the inevitable 21st century energy transition will be cost driven and should occur in the absence of production – as opposed to first of a kind development incentives – subsidies. The transition to cheaper and abundant energy when it occurs will be rapid and far reaching as a result of the creative destruction of capitalism.

    There are cost competitive alternatives available now. Hydro, geothermal, landfill gas are niche generation that are cost competitive and dispatchable and should rationally be used where available. These may usefully balance wind and solar intermittency. And it is very unlikely that the exponential decrease in solar costs is exhausted yet – low cost solar can have a role in pumped storage, chilled water air conditioning or hydrogen production. The latter can be injected into natural gas lines to reduce demand and extend a very limited supply. The limit is human creativity.

    If we are talking emissions reduction – a step back to a multi-sector and multi-gas strategy is required to make much progress. The most significant of the other sectors is land use emissions. This may be reversed with practices that rebuild organic content of agricultural soils, with restoration of forest and rangeland and with the reclamation of deserts. With great benefits for – inter alia – food security, flood and drought resilience and biodiversity conservation. The land use sector in the US is a net negative emitter with great scope for further progress.

    Here are both sector and gas emissions as percentages of total global emissions. The need for a multi-gas and multi-sector strategy seems evident.

    Emissions of black carbon and sulfate must be included. Both have significant climate impacts. Sulfate may in fact warm the planet depending on the mixing ratio with black carbon of the source. “The best estimate of industrial-era climate forcing of black carbon through all forcing mechanisms, including clouds and cryosphere forcing, is +1.1 W/m 2 with 90% uncertainty bounds of +0.17 to +2.1 W/m 2. Thus, there is a very high probability that black carbon emissions, independent of co-emitted species, have a positive forcing and warm the climate. We estimate that black carbon, with a total climate forcing of +1.1 W/m 2, is the second most important human emission in terms of its climate forcing in the present-day atmosphere; only carbon dioxide is estimated to have a greater forcing…’ (Bond, T. C. et al, 2013) Black carbon and sulfates are a technology and economic development problem. Emissions of these substances can be virtually eliminated in the electricity sector with high efficiency low emissions coal technology – an economically attractive option in many settings as an interim energy source.


    There are low hanging fruit in emissions reductions that are happening in the real world – and multiple technological innovations possible. As suggested in the post – the crunch comes with rising fossil fuel costs and declining costs of alternatives.

    Advanced nuclear cost estimates are broader than suggested in the article – but promising.

    General Atomics says that it’s energy multiplier module will break even with gas at $6-7/MMBTU. This will happen within a decade as global energy demand grows and fracked gas field production falls off the precipice.

    It is not clear either why storage is required with nuclear – the necessity is for sufficient generating capacity to meet peak annual demand with some essential reserve capacity. The strategy then is to schedule the relatively minor down times outside of peak demand periods where possible. Excess supply capacity may technically be used for high temperature hydrogen production that can be catalyzed with CO2 to form liquid fuels.

    With small, modular reactors – typical of advanced designs – capital costs would seem to promise to be significantly less than light water reactors for obvious reasons. This is not new technology in large part – fast neutron plants have hundreds of years of operational experience and the race is on to bring 21st century designs to market.

    Advanced nuclear has other advantages over light water reactors. “To provide [electricity] in today’s world, an ‘advanced reactor’ must improve over existing reactors in the following 4-core objectives. It must produce significantly less costly, cost-competitive clean electricity, be safer, produce significantly less waste and reduce proliferation risk. It is not sufficient to excel at one without regard to the others.” Dr. Christina Back, Vice President, Nuclear Technologies and Materials for General Atomics, May 2016 testimony before the US Senate Energy and Natural Resources Committee hearing on the status of advanced nuclear technologies.

    Although a mixed bag of generating sources should not be ruled out – the backbone of any future system seems likely nuclear. Indeed as is the Obama energy transition
    strategy – that Trump should endorse. Small, modular nuclear as well have huge cost advantages in regions with an undeveloped transmission network.

  16. Somewhat related, US shale oil is keeping a lid on prices. ME and Russia cutting back has given it a great shot in the arm.

  17. What about the rest of the world? They understand the benefits of free enterprise even if they do not subscribe to the American ideal of the individual’s God-given right to life, liberty and the pursuit of happiness. They see things their way and their plan is to use more energy to create more not less economic wealth–e.g.,

    China may achieve its goal of reducing the carbon intensity of its economy (the ratio of carbon dioxide emissions per unit of output) by 45 percent below its 2005 level by 2020. But the country is growing so fast that its coal consumption and greenhouse gas emissions are expected to continue to increase. China is expected to add the equivalent of a new 500-megawatt coal-fired electric plant every 10 days for the next decade, according to projections by the United States government. ~Robert Stavins, ‘Climate Realities,’ NYT

    • The Paris reality is not 1.5 degrees C – just one absurd notion before breakfast – but a 3 Gt increase is CO2 emissions by 2030. Soem people haven’t got the IEA memo.

      This is of course denier logic down the rabbit hole. A comfortable place to be given the non-linearity of freakin’ everything.

      • Are we are almost at the point where Western academia believes humanity’s only chance is investing heavily in nonexistent technology?

  18. Doctor Hartley: it would be really interesting if you did a study for Jamaica. The island is mountainous, relatively poor, and is isolated by a very deep ocean trench, therefore a connection to a larger grid by cable is nearly impossible. I mention this because the Word Bank has committed not to loan funds to fossil fuel developments, and they seem to be ignoring that an island nation like Jamaica needs alternatives beyond solar and wind.

    • My thoughts are nuclear or coal are your best bet. Coal is cheaper infrastructure wise if you are purchasing it yourself. For nuclear the option might be for Chinese loans, or them making a company to perform the infrastructure requirements. Solar and wind with LNG backup is very expensive. Coal is compressed already, LNG is expensive because of compression energy loses even from cheap US gas.

  19. Here in the United States, the biggest obstacle by far to an expanded use of nuclear power isn’t the opposition of the anti-nuclear activists. It is the enormous upfront capital cost, and the extraordinary financial risk, of building the latest generation of nuclear plants.

    The V.C. Summer project was canceled after its cost more than doubled, going from an estimate of 12 billion dollars in 2011 to an estimate of roughly 25 billion dollars in 2017. The Vogtle project has seen a similar increase in its estimated cost and is now on a clear pathway towards cancellation.

    Well excuse me, but 25 billion dollars upfront capital cost is an absurd price to pay for 2,200 megawatts of new generation capacity that could be delivered from a new gas-fired plant constructed for one-quarter the capital cost, in half the time or less, and with much less technical and financial risk.

    I’ve been in nuclear construction and operations for thirty-five years and can say with some authority that America’s nuclear power industry is its own worst enemy. We have the technology and the technologists, but what we are sorely lacking is honest, competent project managers who can build a nuclear plant to the public’s high safety and quality assurance expectations while still keeping the project’s cost and schedule under tight control.

    For those of us who have been in nuclear construction and operations since the mid-1970’s, the period of 2012-2017 was a near exact repetition of what we saw in the period of 1979-1985. The VC Summer and Vogtle nuclear projects have been on a path to failure starting in 2012 at the beginning of their construction phase.

    It didn’t have to be that way.

    For those of you who think that over-regulation of the nuclear industry is the fundamental cause of these nuclear project failures, you must acknowledge that the power utilities who were funding these two projects claimed in 2012 that their planning estimates were sufficient to cover all of the cost and schedule burdens government regulation and oversight imposes on the nuclear construction industry.

    Why did these projects fail so miserably to stay on cost and schedule?

    Some ancient history is in order here. At the end of the 1980’s, the nuclear industry had learned a number of hard lessons as to how to keep their construction projects on track for successful completion on cost and on schedule. The basic reason the latest projects have failed is that all of the tough lessons that were learned from the previous round of nuclear project failures from the late 1970’s and early 1980’s were ignored:

    — Over Reliance on Contractor Expertise: The power utilities did not have their own in-house technical and project management expertise, the kind of expertise needed to make themselves knowledgeable customers for the nuclear construction services they were buying.

    — Incompetent Prime Contractor EPC’s: The prime contractors (EPC’s) chosen by the power utilities in 2012 to manage their nuclear construction jobs did not have substantial previous experience in constructing other kinds of large-scale nuclear projects of a similar magnitude and complexity, and did not themselves have the technical and the managerial expertise needed to tackle these kinds of projects.

    — Ineffective Contractor Oversight: The power utilities did not do an effective job of overseeing the work of the prime contractors and the subcontractors; and they did not act quickly to deal with the serious issues and problems that were emerging in 2012, 2013, and 2014 while there was still time to deal with those emerging problems and issues.

    — Ineffective Subcontractor Oversight: The prime contractors did not do an effective job of overseeing the work of the subcontractors. They did not act quickly to deal with emerging issues and problems in the subcontractor’s shops while there was still time to deal with those issues and problems.

    — Inadequate Project Control Systems: The prime contractors did not impose effective control systems for contractor and sub-contractor design interface control; for configuration control and management of design documentation and associated systems and components; and for proper and up-to-date maintenance of inter-contractor cost and schedule information.

    — Quality Assurance Deficiencies: The sub-contractors who performed the detailed construction and fabrication work did not do an effective job of in-house quality assurance, of in-house quality control, and of in-house configuration management & design control. They did not act quickly to deal with their own emerging issues and problems. Smaller problems were allowed fester until they evolved over time into bigger problems, ones that were much more difficult to solve.

    — Inaccurate and/or Dishonest Status Reports: The sub-contractors were not giving honest, accurate, and up-to-date information to the prime contractors concerning the true status of their work activities. The prime contractors ignored early and obvious signs that their sub-contractors were not performing to expectations; and the primes did not move swiftly to deal with subcontractor quality control issues and with subcontractor cost & schedule performance problems.

    — Cost & Schedule Control Deficiencies: The prime contractors did not maintain a proper cost & schedule control system for their overall project as a whole. Many activities listed on their project schedules were seriously mis-estimated for time, cost, scope, and complexity. Other project activities covering significant portions of the total work scope were missing altogether, making it impossible to accurately assess where the project’s cost and schedule performance currently stood, and where it was headed in the future.

    — Regulatory Oversight Deficiencies: In 2013 and 2014, federal and state regulatory agencies failed to act on early and substantial evidence of serious technical and management problems at the VC Summer and Vogtle projects. They were ineffective in pressing the power utilities and the prime contractors to deal with their project’s emerging issues and problems in a timely, proactive, and decisive way.

    There you have it. The power utilities who were buying these nuclear plants had the responsibility and the public trust to manage their nuclear projects using a highly disciplined, well-coordinated approach. The regulatory agencies who were monitoring these projects had a responsibility to act in the public interest. Back in 2014, those regulatory agencies had every justification for demanding better performance from the power utilities, or else, but the regulators refused to act when action was called for. Both the regulators and the regulated completely failed to deliver on their commitments.

    What about the small modular reactors and the molten salt reactors?

    Sure, the small modular reactors and the molten salt reactors have theoretical cost advantages over the AP1000’s and similar large-scale conventional reactors. But those prototype SMR and molten salt technologies will never see commercial operation unless we can build their follow-on commercial plants at a reasonable cost and deliver them on a reasonable schedule.

    When it comes to managing complicated industrial construction projects, nuclear is different. It will always be different, because it’s the nature of the beast. That said, we have all the knowledge, all the tools, and all the processes needed to build a nuclear power plant to an honestly-estimated cost and within a reasonably-estimated project schedule. The problem here is that we don’t actually use those tools, we just give them lip service.

    What we don’t have in the nuclear industry today are corporate managers who have the necessary combination of project management skill, professional dedication — and let’s say it out loud here, the necessary personal integrity — needed to meet their commitments to the public in the areas of nuclear project quality assurance, nuclear project cost control, and nuclear project delivery schedule.

    Until America’s nuclear construction industry gets its act together and starts hiring honest and professionally competent top-level and mid-level managers — people whose personal focus is on getting the work done on cost and on schedule, and to levels of safety and quality assurance that meet the public’s high expectations – until we do this, nuclear power has no future in this country.

    • Thanks for insider’s view of the industry Beta.
      I can’t help but notice how much this parallels our spending on defense except you need to multiply by 3X. I help build the first F-16s and the per copy cost was around 4-5 million each including support services. Pound for pound the F-16 is still the best fighter plane ever built. Fact is when your aircraft can maneuver so fast humans can’t safely operate the plane without blacking out it’s time to switch to drones anyway. When I look at the costs of a F-35 or the V-22 Osprey they ended up costing 3 or 4 times the original quoted price and I just can’t understand where all the money goes. I suspect this is why there hasn’t been an audit of the DoD since Dick Cheney was Sec. of Defence back in 1990.
      Perhaps we should outsource this to the Chinese or Russians. No I’m not joking. Nuclear has strong support with the republicans and I hear Trump is on very good terms with Putin. Get it done!

      • The F-35 is a multi-role aircraft that would shoot the F-16 out of the sky before it was even seen.

        “The F-35’s integrated sensors, information and weapons systems give pilots an advantage over potential threat front-line fighter aircraft. Compared to 5th Generation fighters like the F-35 and F-22, legacy aircraft have a larger radar cross-section (RCS), which means they can be more easily detected by enemy radar. In aerial combat, legacy aircraft have relatively equal opportunities to detect and engage one another, while a 5th Generation fighter pilot can see enemy aircraft first and take decisive, lethal action from a stand-off distance. The ability to see and not be seen is redefining previous generation air-to-air tactics.”

        The decisive change is the avionics – and advanced technologies – allowing integrated whole of battlefield tactical offensive and defensive abilities. This is a dog of war designed not for dogfights but to seamlessly combine multiple land, air and ocean based battlefield capabilities in a single and overwhelmingly powerful system.

      • The F-35 is not a pretty beast.

      • Robert,
        I’m glad you like the F-35.
        Try a internet search on “F-35 & Australia”.
        Some of the top search results include:
        Huge cost overruns…
        Sep 23, 2017 – Australia’s F-35A stealth fighters may cost millions to bring up to a fighting standard. … Australia’s two shiny new F-35 Strike Fighters may never go to war.
        Jun 23, 2017 – The shooting down of a Syrian fighter jet may have exposed a fatal flaw with Australia’s next generation aircraft… The F-35 can only hold 4 missiles internally. Once you strap on external ordnance it looses it’s stealth.
        So far the F-35 spends more time in the maintenance hanger than almost any fighter ever made. What the point of having a fleet of billion dollar fighter jets when they spend more time in the shop than they do on the flight line?

      • I generally consider your comments to be irrelevant at best and disturbingly crazy at worst. This veers in the latter direction.

        “Meanwhile, to date the RAAF’s AU-1 and -2 have flown almost 1,100 flight hours in 700 sorties, Over said.

        “They’re flying at about the utilisation rate that a mature weapon system like F-16s or F-18s are returning at. So the airplanes have really been performing remarkably well.”

        Four Australian pilots are now certified instructor pilots on the F-35 and two more are in training at Luke.

        Overall, 235 production F-35s have now been delivered, with aircraft operating from 12 bases.

        F-35 System Design and Development (SDD) is on track to be finalised by the end of the year, Over said, with a last three per cent of all SDD test points to be completed.

        “Right now the development program is rapidly winding to a close. We’re within three per cent of the testing to complete the development program and we’re in the final stages of tweaking the Block 3F software that will be pushed to the field later this year,” Over said.

        “The full functional capability that we’ve promised with Block 3F is actually flying in [flight test] jets right now and so it’s identifying the things that don’t work quite as the pilots would like for them to work and we’re tweaking through those last little details now.”

        We have two training F-35 based in the US at the moment I believe – and a fleet of 72 block ordered at a considerable discount. Try reading more than the headlines.


        As far as I can tell Australia is nothing more than our DoD’s poodle. You do as we say, you buy what we tell you to buy and you spill your blood (in Afghanistan and Iraq) when I can see no tangible benefit to your own security. Sorry I called you a sucker… fool is more precise.

      • You believe The Conversation? Says quite a lot about you – your characterization of an alliance that has endured since the Second World War a great deal more. None of it good. With people like this – is it any wonder the US has so few friends.

        We in fact have a quite robust regional defense network – of which the US is lucky to be part. As well as strong trading links with China. You are lucky we like you. Oh… wait…

      • I’m a big proponent for keeping the US military at the top for military technology and lethality. The F-35 has been a frustrating weapon system to deploy. I’m no expert on the matter, but maybe its being awarded a defense contract for development was premature in the early years of platform competition. I’m sure geopolitical defense needs weighed in early, perhaps politics won the day, but think about this; near $1.5 trillion was spent on developing the F-35 since its inception, this represents over a 5% contribution towards the total US debt of $20 trillion for a single weapons system. That’s not sustainable spending no matter how hawkish one is. Over 5% of the total $20 trillion US debt for one weapons system, as GDP has declined, think about that.

      • But can they do the “cobra”?

      • At about AU$80 million each – for 72 F-35? I think that’s some AU$17 billion all up – cool. These machines are amazing. These machines could map in real time for flooding, tsunami, earth quake, rioting, etc.

        I think we need some more cruisers and destroyers now – and nuclear submarines. Yayyyy…

    • I have worked for Bechtel on a $10 billion gas export hub – without any of these problems.

    • Would help if design was completed before construction started. However, the basic problem is too much stuff with too much excessive regulations. To be blunt, the Nuclear Regulatory Commission bureaucratic swamp needs to be drained.
      It is to late to save the existing nuclear reactors. However there is no logical reason to inflict the passively safe advanced reactors to the same level of stunning bureaucratic “red tape”

      • Advanced designs provide the potential for factory construction and quality control and generic approvals. As well as much reduced capital cost and increased efficiency with higher operating temps. I don’t mean to be a shill for advanced nuclear – but the ‘theoretical advantages’ are obvious.

        What is required is that governments reduce the risk to to builders and utilities of first of a kind installations.

      • The financial risks associated with excessive regulations currently vastly exceeds the ability to make a profit. Again, there is no rationale reason for the government to regulate all aspects of nuclear power. Rather the focus should be on those items that directly impact potential undue radiation exposure by the public. In the case of the passively safe advanced reactors, the extent of those items are actually rather small. That also means the costs are not that great.

        Ultimately, can advanced reactors be profitable? Maybe, but only if the current regulatory swamp is drained.

    • I agree with your points but the new aircraft impact rule did have a significant impact on these plants (10 CFR 50.150).

  20. Leftists, liberals and environmentalists believe and agree we must be very concerned about global warming and, we all should be fearful enough of the consequences of it to do whatever we can to stop it. As that mindset plays out in a Eurocommie state like California is a government that takes money from the productive to fund the storage of electricity instead of water.

  21. Beta Blocker,

    Here in the United States, the biggest obstacle by far to an expanded use of nuclear power isn’t the opposition of the anti-nuclear activists. It is the enormous upfront capital cost, and the extraordinary financial risk, of building the latest generation of nuclear plants.

    The “the enormous upfront capital cost, and the extraordinary financial risk, of building the latest generation of nuclear plants.” is a result of the anti-nuclear protest movement scaring the hell out of the population over the past 50 years. That is the root cause of the costs. If not for their activism, the costs could be 10% of what they are.

    Read this, the relevant the Notes and the references cited in them, including Daubert and Moran (1985), and Wyatt (1978),

    • Peter, as someone who has been down in the project trenches doing the nitty-gritty day-to-day work of constructing a nuclear facility, I have to be blunt. Your funky macroeconomic analysis of the 1970’s nuclear construction industry is about as raw an example of pencil whipping a conclusion as any such example can get. .

      Do you honestly think that if we could just eliminate the Nuclear Regulatory Commission and the heavy-duty oversight responsibilities it now carries, we could somehow reduce the cost of a 1300 Mw nuclear reactor vessel to 10% of what it is now?

      What about the cost of structural steel and concrete? The cost of skilled labor? Land? Connections to civil infrastructure? The onsite project mobile equipment? Purchase and delivery of a myriad of project systems and project bulk consumables to the construction site? Project administrative overhead? Installation and testing of the reactor systems, small and large? Initial and site support engineering design services? (Etc. etc. etc.)

      What system of theoretical pie-in-the-sky cost estimating technique are you relying upon in predicting the costs of building a real-world nuclear facility on a real-world nuclear construction site using real-world nuclear workers to build the plant?

      • I have to agree – and laugh. Let’s see where this goes.

      • The paper is based on actual data and projects what might have been absent the anti-nuclear mania of the 1970s and 1980s. In general economic prediction exists primarily to lend an air of legitimacy to astrology, but there were many nuclear plants built in this country at reasonable capital cost. Activists, by their own admission, sought to involve themselves in the regulatory process to increase costs. It is hardly controversial to suggest they succeeded. They are attempting to do the same thing to pipeline construction today.

      • Well, actually the costs can be significantly reduced, but not in the manner you might think. Solution lies with hybrids that marry the exceptionally powerful and low-cost natural gas (combined-cycle) plant and exceptionally efficient gas reactor. Basically, the reactor’s helium turbine drives the gas (combustion) turbine’s air compressor.
        The solution is classic economies-of-scale: make your product better than the competition – more powerful & more efficient.
        End up with a 1000 megawatt (electric) power plant on 20 acres of land. Uses +80% of energy in fossil fuel, produces 10% of nuclear waste of conventional reactor and easily competes with natural gas power plants. This patented technology is being quietly developed in Kansas.
        Will this technology ever be deployed? Not under the current regulatory environment. Excessive costs of obtaining a license from current regulatory swamp are much too high. More likely to emerge from China.

        The problem rather graphically demonstrates the problem Trump is attempting to solve. Massive over regulation choking off innovation.

      • China is too bureaucratically hidebound to be true innovators. The are building a fast neutron prototype – but it is pebble bed. Other designs are for factory produced units – with standardization facilitating generic approvals. Ship it by highway and plonk it in an underground bunker for 30 years.

      • They are building pebble bed reactors, but these are thermal reactors, not breeders. The fuel is embedded in graphite spheres and that slows down the neutrons to thermal levels. They are also working on molten salt fast reactors. Both designs do have various safety and operational issues. Not so sure how commercially viable these machines will ultimately prove to be.
        Conventional water reactors (big & small) probably make more sense than advanced reactors for those areas that (1) do not have access to low-cost fossil fuels and (2) are more reasonable in terms of regulations.

      • The fuel cycle commonly includes conversion of fertile to fissionable material – allowing use of nuclear waste or thorium as a fuel.

      • …. they are building a pebble bed prototype…

      • Do I think costs could be dramatically reduced if the NRC was actually reasonable? Yes, Maybe 50%, but that is not enough to be competitive with low-cost natural gas power plants.
        My assessment is based on about 50 years in the power industry, with about 1/2 that being in nuclear power.
        There is just no question that the industry is massively over regulated. I’ve seen it first hand on many occasions. The time and money wasted is counter productive and ends up detracting from safety. Fundamentally, the public is protected by the integrity of the folks running the stations, not bureaucrats in Washington.

    • BetaBlocker,

      You’ve used straw-man arguments. Intellectually dishonest – zero credibility. Your self-aggrandizement claims are irrelevant. Your comments about cost items is irrelevant and show you either didn’t read the link or didn’t take the time to understand it. If you find a genuine, significant error of fact in the analysis, please let me know. Not just silly insinuations and disingenuous statements.

    • Peter Lang: “Beta Blocker, You’ve used straw-man arguments. Intellectually dishonest – zero credibility. Your self-aggrandizement claims are irrelevant. Your comments about cost items is irrelevant and show you either didn’t read the link or didn’t take the time to understand it. If you find a genuine, significant error of fact in the analysis, please let me know. Not just silly insinuations and disingenuous statements.”

      I’ve read the full paper more than once. This isn’t the first time you’ve posted a link to it, and it isn’t the first time I’ve been highly critical of it here on Dr. Curry’s forum. I will not offer a detailed analysis of the paper for the simple reason that speaking as someone who has developed detailed cost and schedule estimates for a variety of nuclear facility construction projects, the paper doesn’t come anywhere close to passing the ho-ho test.

      Referring to my original commentary posted up above in an earlier thread, let’s take VC Summer and Vogtle 3 & 4 as modern examples of unbridled cost and schedule growth in nuclear construction projects.

      In 2006, the estimated cost to completion for adding two AP1000 reactors to the existing Vogtle plant site for the purpose of installing 2,200 megawatts of additional nameplate capacity was roughly 6 billion dollars. By 2009, the estimate had grown to 9 billion dollars. By 2012, when on-site construction began in earnest, the estimate was 12 billion dollars . By summer 2015, it was 16 billion dollars. The very latest estimate for Vogtle 3 & 4 is now 25 billion dollars.

      Extraordinary claims require extraordinary evidence. A claim that the cost our nuclear construction projects could be reduced to 10% of what it is today, if only the NRC’s tight regulation of the nuclear industry could be eliminated, is among the most extraordinary of claims.

      Peter, if you want to offer thoroughly solid evidence supporting your thesis that excessive government regulation is the fundamental driver for nuclear’s high capital costs, then let’s see a bottoms up feasibility estimate for constructing 2,200 megawatts of nuclear capacity in the US Southeast under the assumption that the regulatory requirements being applied are no more stringent than those now being applied to the construction of 2,200 megawatts of new gas-fired capacity.

      Ten percent of 25 billion is 2.5 billion, let’s say 3 billion dollars as a nice round target figure for your cost analysis. To be credible, your feasibility estimate must include at a minimum the following cost analysis elements:

      1: A list of the specific nuclear technologies to be used, including an analysis of their unique technical and cost control risks.

      2: A list of the specific locations where this additional 2,200 megawatts of nuclear capacity is to be installed, including an analysis of each plant site’s unique environmental, technical, and cost control risks if the nuclear capacity is to be spread across multiple locations.

      3: A list of the regulatory requirements, the industrial and manufacturing standard requirements, the component quality assurance requirements, the environmental compliance requirements, and the industrial safety and health requirements to be complied with, including an analysis of their potential impacts on cost and schedule if not properly applied or followed.

      4: An analysis of the state of the current nuclear industrial base in the United States, including an analysis of the potential impacts on cost and schedule if the industrial base isn’t currently strong enough to handle your specific project needs, or if the industrial base exists but can’t be properly utilized to its full potential.

      5: A description of the project organization and the project execution plan to be used, including an analysis of the potential impacts on cost and schedule if the project execution plan is not being followed and/or if the project’s cost and schedule risk analysis assessment and the associated risk management plan prove faulty in actual practice.

      6: A realistic project activity schedule with enough activity detail to allow for credible estimates of time dependent costs.

      7: An itemized bottoms up cost estimate tally sheet which includes these cost elements at a minimum:

      — Environmental analysis of the proposed locations and the state of current local conditions.
      — Regulatory permit analysis, permit preparation, and permit review and approval.
      — Emergency Management personnel and system costs for handling onsite and offsite emergencies specific to each proposed plant location and its surrounding region.
      — Land acquisition, site preparation, site security, ties to civil infrastructure, equipment mobilization and demobilization.
      — Acquisition costs for vendor-supplied major and minor reactor and plant support systems.
      — Construction material costs for concrete, structural steel, and for a variety of construction consumables.
      — Labor and associated ancillary costs for skilled craft workers, for project engineering services, for project administration, and for management overhead.
      — Energy and fuel costs for plant site construction operations, cost of onsite construction office space and shop facilities, transportation costs for large unitary systems and components.
      — System start-up and testing costs, including costs for redesign and rework of plant systems and components not performing to expectations.
      — Project closeout costs, including any cost penalties being assessed by the project owners or the project’s financial stakeholders.
      — Costs associated with project contingency allowances, time dependent cost escalation factors, and management reserve.
      — Last but not least, the cost of money as predicted to be extant during the term of the project.

      Peter, just to give you some leeway here, if you don’t think three billion dollars is quite enough money to deliver these 2,200 megawatts of nuclear generated capacity, then feel free to choose another figure you think is in better alignment with the basic tenants of your thesis.

      • <detailed analysis of the paper for the simple reason that speaking as someone who has developed detailed cost and schedule estimates

        Clearly you haven’t read the paper or haven’t understood it. Cost estimates of current plants have absolutely nothing to do with the analysis. As I said, you haven’t a clue!

      • Beta Blocker: “To be credible, your feasibility estimate must include at a minimum the following cost analysis elements ….”

        I know of an entirely credible and rigorous estimate that incorporated all of the cost analysis elements that you list. It doesn’t arrive at a 90% reduction of today’s cost, but does achieve a very respectable 81% reduction. You provide it yourself: $6 billion (2006) versus $25 billion (today) for Vogtle 3 & 4.

        I presume that the people who produced the estimate in 2006 knew their onions and went through a very thorough process to arrive at their $6 billion figure. It seems unlikely that they committed gross blunders in generating their estimate because although nuclear power technology has its complications it is not exactly bleeding edge stuff with unknown unknowns lurking in every corner. The blunders, therefore, must all have occurred in the execution. The question to be asked, then, is what reasonable assumptions got blown to smithereens out along the way? If the assumptions were reasonable, as I suspect, then the unreason must have lain in the blunders, which suggests they should be relatively easy to understand and fix (assuming their is a will to do so).

        Something occurred to me as an alternative solution to Vogtle 4 & 5, which although not as good in some respects, at least has greater certainty. I gather that a new Gerald E Ford type aircraft carrier costs around $9 billion to construct, which includes a 700MW nuclear power plant. The construction time is around 7 years. I presume the price and construction time could be significantly reduced if the ship was only fitted out minimally. So maybe just order a couple of those given that the people who’ve been tasked with building them have demonstrated that they are capable of doing it.

  22. Maybe I missed it, but I did not see where you broke out the extra generating capacity to operate the pumps to create the stored capacity. The efficiency of the system would also be of interest.

  23. Pingback: Weekly Climate and Energy News Roundup #296 | Watts Up With That?

  24. If I remember correctly, Finland had all the cost control problems mentioned here. Is there any place where the nuclear power plants have been built within reasonable overruns on the estimates? If CO2 was not considered a pollutant, then natural gas seems to be the way go in the near term. I’m sure Peter Lang can get some very good real estate deals in Fukushima. Also, why do no private insurance companies provide liability for nuclear plants?

  25. Hartley ==> Pumped hydro has always been my preferred tech for energy storage — it is so simple. We have pumped hydro storage in New York state — making a profit off of the difference between peak and non-peak electric prices.

    Any dam site and be turned into pumped hydro — we almost had a wonderful new lake and pumped hydro along the Hudson River at Storm King Mountain but the short-sighted enviro-nuts scotched it (and are so so proud of it…) — we could have taken advantage of wind and solar to feed power hungry NY City — but no, no changes to Nature allowed — though a beautiful new lake and recreation area would have been created by the project.

    Hydro really is the best and most “natural” way for produce electricity — the rain falls on the high places and flows happily down to the sea. Tapping that to turn generators is obvious — and man has been doing it for who-knows-how-long.

    Using intermittent power sources to pump water up into high lakes and then letting it run down producing a steady stream of dependable electricity is “a natural”.

  26. Tomorrow, Thursday December 21st, 2017, the Georgia Public Service Commission will announce its decision as to whether or not the Vogtle 3 & 4 AP1000 reactor construction project will be allowed to continue or else will be terminated. This decision, whatever it turns out to be, will be among the most consequential ever made by a regulatory agency in the long history of nuclear power here in the United States.

    Here in the US, including the options of nuclear, wind, solar, and hydro in the power generation mix is strictly a public policy decision. Left to its own devices, the power market in the US would swing decisively towards gas-fired generation given that among all the choices available for the next several decades, gas-fired generation has the least technical, environmental, and financial risks; and it also has the highest profit making potential for private investors.

    Beta Blocker: “In 2006, the estimated cost to completion for adding two AP1000 reactors to the existing Vogtle plant site for the purpose of installing 2,200 megawatts of additional nameplate capacity was roughly 6 billion dollars. By 2009, the estimate had grown to 9 billion dollars. By 2012, when on-site construction began in earnest, the estimate was 12 billion dollars . By summer 2015, it was 16 billion dollars. The very latest estimate for Vogtle 3 & 4 is now 25 billion dollars.”

    aporiac1960: “I presume that the people who produced the estimate in 2006 knew their onions and went through a very thorough process to arrive at their $6 billion figure. It seems unlikely that they committed gross blunders in generating their estimate because although nuclear power technology has its complications it is not exactly bleeding edge stuff with unknown unknowns lurking in every corner. The blunders, therefore, must all have occurred in the execution. The question to be asked, then, is what reasonable assumptions got blown to smithereens out along the way? If the assumptions were reasonable, as I suspect, then the unreason must have lain in the blunders, which suggests they should be relatively easy to understand and fix (assuming there is a will to do so).”

    A decade ago, in about 2006 when the initial cost estimates for pursuing a 21st century nuclear renaissance were being done, the 6 billion dollar estimate for a pair of new technology AP1000’s was thought by many to be too low. With twenty-five years passing without construction of a clean-sheet reactor design having been initiated, the US nuclear industrial base was in a deeply withered state. It was recognized that the steep learning curve for doing nuclear construction in the US had to be passed through for a second time, and that the cost estimates for initiating new projects had to include the costs of rebuilding the nuclear industrial base and of passing through the nuclear construction learning curve for yet another time.

    More realistic estimates for two AP1000’s were developed in 2009 and later in 2012 — 9 billion dollars and 12 billion dollars respectively. It cannot be emphasized enough here that the estimate of 12 billion dollars when onsite construction began in 2012 included the expected costs of full compliance with NRC regulations and of passing through the nuclear learning curve for a second time. These estimates also assumed that all the difficult lessons learned from the nuclear projects of the 1980’s would be diligently applied to the latest projects as they were being initiated and while they were in progress.

    How did 2012’s estimate of 12 billion dollars for two AP1000’s grow to 2017’s estimate of 25 billion dollars?

    The answer here is that all the lessons learned from the 1980’s were ignored. Thirty years ago, a raft of studies and reports were published which analyzed the cost growth problems and the severe quality assurance issues the nuclear construction industry was then experiencing, and made a series of recommendations as to how to solve these problems. Those studies had a number of common threads:

    Complex, First of a Kind Projects: Any large project that is complicated, involves new and/or high technology, has several phases, involves a diversity of technical specialties, involves a number of organizational interfaces, and has significant cost and schedule pressures—any project which has these characteristics is a prime candidate for experiencing significant quality assurance issues, cost control issues, and schedule growth problems.

    Strength of the Industrial Base: Nuclear power requires competent expertise in every facet of design, construction, testing, and operations. This kind of competent expertise existed in the early 1980’s but was not being effectively utilized in many of the power reactor construction projects, the ones that experienced the most serious cost and schedule growth issues.

    A Changing Technical Environment: The large reactor projects, the 1300 megawatt plants, were being built for the first time. They were being built without a prototype, and they were substantially different from previous designs. Those big plants had many new and significantly revised systems inside them, systems that had to be designed, constructed, tested, and subsequently operated.

    A Changing Regulatory Environment: In the late 1970’s and early 1980’s, there was a continual increase in the regulatory requirements being placed on power reactors. The Three Mile Island accident, the Brown’s Ferry fire, the Calvert Cliffs environmental decision, all of those events required the power utilities to change the way they were dealing with their projects in the middle of the game. Some power utilities were successful in making the necessary changes, others were not.

    Project Management Effectiveness: Those nuclear projects which had a strong management team and strong management control systems at all levels of the project organization generally succeeded in delivering their projects on cost and on schedule. Those that didn’t were generally incapable of dealing with the changing technical and regulatory environment and became paralyzed in the face of the many QA issues, work productivity issues, and cost control issues they were experiencing.

    Overconfidence Based on Past Project Success: Many of the power utilities which had a record of past success in building non-nuclear projects, and which were constructing nuclear plants for the first time, did not recognize that nuclear is different. Those utilities which did not take their regulatory commitments seriously and which did not do an adequate job of assessing whether or not the management systems and the project methods they had been using successfully for years were up to the task of managing a nuclear project.

    Reliance on Contractor Expertise: The projects which succeeded had substantial nuclear expertise inside the power utility’s own shop. Those utilities who were successful in building nuclear plants were knowledgeable customers for the nuclear construction services they were buying. They paid close and constant attention to the work that was being done on the construction site, in the subcontractor fabrication shops, and in the contractor’s technical support organization. Emerging issues and problems were quickly and proactively identified, and quick action was taken to resolve those problems.

    Management Control Systems: The nuclear projects which failed did not have effective management control systems for contractor and subcontractor design interface control; for configuration control and management of design documentation and associated systems and components; and for proper and up-to-date maintenance of contractor and inter-contractor cost and schedule progress information. Inadequate management control systems prevented an accurate assessment of where the project actually stood, and in many cases were themselves an important factor in producing substandard technical work.

    Cost & Schedule Control Systems: For those projects which lacked a properly robust cost & schedule control system, many activities listed on their project schedules were seriously mis-estimated for time, cost, scope, and complexity. Other project activities covering significant portions of the total work scope were missing altogether, making it impossible to accurately assess where the project’s cost and schedule performance currently stood, and where it was headed in the future.

    Quality Assurance: For those nuclear projects which lacked the necessary management commitment to meeting the NRC’s quality assurance expectations, the added cost of meeting new and existing regulatory requirements was multiplied several times over as QA deficiencies were discovered and as significant rework of safety-critical systems and components became necessary.

    Construction Productivity & Progress: For those nuclear projects which lacked a strong management team; and which lacked effective project control systems and a strong management commitment to a ‘do-it-right the first time’ QA philosophy, the combined impacts of these deficiencies had severe impacts on worker productivity at the plant site, on supplier quality and productivity at offsite vendor facilities, and on the overall forward progress of the entire project taken as a whole.

    Project Financing and Completion Schedule: As a result of these emerging QA and site productivity problems, many of the power utilities were forced to extend their construction schedules and to revise their cost estimates upward. Finding the additional money and the necessary project resources to complete these projects proved extremely difficult in the face of competition from other corporate spending priorities and from other revenue consuming activities.

    A Change in Strategy by the Anti-nuclear Activists: In the late 1970’s and early 1980’s, the anti-nuclear activists were focusing their arguments on basic issues of nuclear safety. They got nowhere with those arguments. Then they changed their strategic focus and began challenging the nuclear projects on the basis of quality assurance issues, i.e., that many nuclear construction projects were not living up to the quality assurance commitments they had made to the public in their NRC license applications.

    Regulatory Oversight Effectiveness: In the early 1980’s, the NRC was slow to react to emerging problems in the nuclear construction industry. In that period, the NRC was focusing its oversight efforts on the very last phases of the construction process when the plants were going for their operating licenses. Relatively little time and effort was being devoted to the earlier phases of these projects, when emerging QA problems and deficiencies were most easily identified and fixed. Quality assurance deficiencies that had been present for years were left unaddressed until the very last phases of the project, and so were much more difficult, time consuming, and expensive to resolve.

    Working Relationships with Regulators: The successful nuclear projects from the 1970’s and 1980’s, the ones that stayed on cost and on schedule, did not view the NRC as an adversary. The successful projects viewed the NRC as a partner and a technical resource in determining how best to keep their project on track in the face of an increasingly more complex and demanding project environment. On the other hand, for those projects which had significant deficiencies in their QA programs, for those that did not take their QA commitments seriously, the anti-nuclear activists introduced those deficiencies into the NRC licensing process and were often successful in delaying and sometimes even killing a poorly managed nuclear project.

    If it’s done with nuclear, it must be done with exceptional dedication to doing a professional job in all phases of project execution from beginning to end.

    Once again, it cannot be emphasized enough here that the estimate of 12 billion dollars for two AP1000’s when onsite construction at VC Summer and at Vogtle 3 & 4 began in 2012 included the expected costs of full compliance with NRC regulations and of passing through the nuclear learning curve for a second time. These estimates also assumed that all the difficult lessons learned from the nuclear projects of the 1980’s, as I’ve described them above, would be diligently applied to the latest projects as they were being initiated and while they were in progress.

    For those of us who went through the wrenching experiences of the 1980’s in learning how to do nuclear construction right the first time, what we’ve seen with VC Summer and Vogtle 3 & 4 has been deja vu all over again. The first indications of serious trouble came in 2011 when the power utilities chose contractor teams that did not have the depth of talent and experience needed to handle nuclear projects of this level of complexity and with this level of project risk. That the estimated cost eventually grew to 25 billion dollars in 2017 should be no surprise.

    The project owners and managers ignored the hard lessons of the 1980’s; they did not do a professional job in managing their nuclear projects; and they did not meet their commitments to the public as these commitments are outlined in their regulatory permit applications. Just as happened in the 1980’s, the anti-nuclear activists and the government regulatory agencies are now holding these owners and managers to account for failures that were completely avoidable if sound management practices had been followed.

    • Beta Blocker

      Thank you for your detailed and informative comment.

      I wonder if the single word that most completely encompasses the various issues you’ve highlighted is ‘culture’ (more specifically, the lack thereof). I mean culture not only in the sense of mastery of techniques, practices, processes and procedures, etc, but also the associated set of values that together orientate those within the culture towards a common objective. The thing that marks ‘culture’ as something beyond mere skill and rule following is the manner in which practices are internalised, reinforced and perpetuated as a system of norms by those participating in the culture. It is something like ethics, but ethics made practical and habitual across and between the range of activities that constitute a coherent industrial capability. I suspect that is one of the most difficult things to rebuild if that capacity has previously been allowed to wither and die.

      It is interesting that you emphasised ‘personal integrity’ in an early comment because it raises the question of how personal integrity arises in a given practical domain. Human systems of organisation that require personal integrity must define what it is, create wide agreement, and then construct the social mechanisms to inculcate and enforce the requisite norms in that specific setting. My notion of culture in this context does not depend on personal integrity, as such, because the integrity does not arise from the individual, which is an inherently unreliable entity, but from the culture which defines, demands and enforces integrity to the very point that the individuals are fashioned accordingly, or at least most are, and those that are not are either made to comply or expelled. And so the culture is preserved and perpetuated.

  27. Beta Blocker
    I hope Judith Curry has recommended you to advise Trump regarding his statements suggesting that his administration would consider nuclear power as a possible source of future energy. If you have not already, you should write a book on your experiences. I would buy it in a minute.

  28. Peter Hartley

    Thank you for an interesting working paper.

    Your Table 2 shows an LCOE for Wind as $0.069/kWh which could be much higher than what seems to be the case for actual bids for pure Wind power which are as low as $0.02/kWh. I follow both this website and the skepticalscience website to get a view from “both sides”. On the sKs website, I was referred to an article by Christoper Arcus dated January 11, 2018 (on the CleanTechnica website) which states that Xcel Energy took bids to replace 660 MW of coal fired power which they are shutting down. The quotes for pure wind totaling 17,380 MW averaged $0.018/kWh. Unless someone is into losing money, this anecdotal evidence would suggest that your estimate of $0.069/kWh is too high.

    Unless the quotes received by Xcel Energy are not indicative of most wind power projects, would this not impact your Table 4 so that some combined wind/natural gas combination would be justified? When I referenced your article on sKs, another comment was that you have not considered a combination of wind/solar with natural gas backup. I think the point being made was that using pumped storage as the backup was unduly increasing the overall cost compared to using natural gas as the backup.

    Look forward to your comments.

  29. Beta Blocker

    From the above comments, I get the sense that you do not think it is realistic to even consider even “advanced nuclear” when it comes to the future power needs of the US. Your reference to new gas fired electrical generation at ¼ the cost probably answers that.

    But on a separate question relating to the viability of nuclear power to supplant fossil fuels, do you have any comments on two papers by Derrick Abbot, a professor in the Engineering Department of the University of Adelaide in 2010 and 2012 on the impracticality of using nuclear power given the limited availability of extractable uranium?

    Abbott quotes the World Nuclear Association (2011) as projecting 80 years of economically extractable uranium at the current rate of consumption using conventional reactors in place at that time. I know there are other suggested methods of more efficiently using the existing uranium but I understand that these have not been proven commercially.

    As well, Abbott suggests that there is a problem with the availability of various rare minerals used in the construction of nuclear plants if the world were to “go nuclear”.

    Given the political atmosphere today and the existing costs, it may be that this is all academic.