Jump to ContentJump to Main Navigation
What Will WorkFighting Climate Change with Renewable Energy, Not Nuclear

Kristin Shrader-Frechette

Print publication date: 2011

Print ISBN-13: 9780199794638

Published to Oxford Scholarship Online: January 2012

DOI: 10.1093/acprof:oso/9780199794638.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 23 October 2019

Trimming the Data on Nuclear Costs

Trimming the Data on Nuclear Costs

(p.69) Chapter 3 Trimming the Data on Nuclear Costs
What Will Work

Kristin Shrader-Frechette

Oxford University Press

Abstract and Keywords

Chapter 3 reveals that atomic energy is also very expensive. This chapter surveys all 30 recent nuclear-electricity cost analyses to show that all the industry-funded studies (but not university-funded research) tend to violate standard conflict-of-interest guidelines and to illegitimately trim relevant cost data by making counterfactual assumptions. Most such studies exclude costs of full-nuclear-liability insurance, underestimate interest rates and construction times by using “overnight” costs, and overestimate reactor load factors and lifetimes. Even if one ignores taxpayer subsidies, fission costs are at least triple those of wind, and they are much higher than those of solar PV. Nuclear energy likewise imposes many costs on the public, in the form of government-mandated liability limits that make nuclear-accident victims, not industry, responsible for more than 98 percent of worst-case-accident damages, even those caused by industry's intentional safety violations. Indeed, the chapter shows that market proponents agree: nuclear fission is uneconomical. Consequently, all credit-rating firms downgrade the credit of utilities with a nuclear plant. This costliness is one reason the nuclear industry cannot build any reactors without massive taxpayer subsidies, including covering half the costs of each $12 to $20 billion plant. Thus the chapter shows that atomic energy only “appears” economical—because most nuclear-cost studies (nearly all done by the reactor industry) ignore costs from the full, 14-stage nuclear-fuel cycle, and they erroneously include only costs from 1 stage (reactor construction and maintenance). When these flawed methodologies are corrected, market costs (excluding subsidies) of fission-generated electricity can be shown to be roughly 6 times higher than most nuclear-economics studies claim. After answering several objections, the chapter shows that nuclear fission is actually far more expensive than conservation, efficiency programs, and renewable-energy sources, like wind and solar photovoltaic (PV).

Keywords:   atomic energy, construction times, cost data, insurance, interest rates, reactor load factors, taxpayer subsidies

Merck Pharmaceuticals suppressed data on harmful effects of its drug Vioxx. As a result, many people died. Guidant Corporation suppressed data on electrical flaws in one of its heart-defibrillator models, and it too caused many patient deaths. Many similar examples reveal how financial conflicts of interest (COI) can skew biomedical research—especially when companies suppress data that could jeopardize their pharmaceutical profits. As a result, scientists and ethicists have long recognized the harm done by such COI. An Annals of Internal Medicine study recently showed that 98 percent of papers based on industry-sponsored studies reflected favorably on the industry's products; a Journal of the American Medical Association article likewise concluded that industry-funded studies were 8 times less likely to reach conclusions unfavorable to their drugs than were nonprofit-funded studies.1 Does something similar happen in energy studies done by electric utilities?

Both coal and nuclear utilities appear to massively underestimate the costs of their activities. For instance, an association of coal utilities and producers, the World Coal Institute, said in 2009 that coal is “cheaper per energy unit than other fuels,” but as chapter 1 documented, the US National Academy of Sciences says that annual costs of coal-generated electricity often exceed their benefits, especially for older, dirtier coal plants. The academy noted that annual US health-related and CC damages from US coal plants are more than $120 billion annually, including tens of thousands of coal-induced deaths per year. Even in market terms, coal benefits often do not pay for its costs. A recent report of the Mountain Association of Community Economic Development showed that, for a typical coal-producing state, like Kentucky, annual state revenues from coal are $528 million, but they cost $643 in state expenditures—and these figures do not include any health damages.2

Nuclear-industry cost estimates are just as misleading. Jonathan Porritt, chair of the UK Sustainable Development Commission and adviser to Gordon Brown, says, “Cost estimates from the [nuclear] industry have been subject to massive underestimates—inaccuracy of an astonishing kind consistently over a 40-, 50-year period.”3 A UK government commission agrees, claiming that these “massive underestimates” have arisen because virtually all nuclear-cost data can be “traced back to industry (p.70) sources”; the main US oversight agency, the Government Accountability Office, says something similar, repeatedly faulting the nuclear industry for greatly underestimating the full costs of its activities.4 Why do these flawed nuclear-fission cost estimates occur? The preceding chapter gave some of the historical reasons, namely the military legacy of secrecy and data falsification. It also noted that typically only the industry is privy to full nuclear-cost data. University of Greenwich business professor Stephen Thomas says the same thing: because the nuclear industry controls virtually all the economic data, it is difficult for others to check it or even obtain it; fission companies “are notoriously secretive about the costs they are incurring.”5 If these government and university charges are correct, they suggest the need to scrutinize nuclear-industry claims that, to address climate change (CC), fission is “the most cost-effective power source.”6 Is it?

Overview of the Chapter

This chapter argues that nearly all nuclear-fission-cost estimates (most of which are produced by the industry) are examples of grossly flawed science. The studies not only violate standard COI guidelines, widely accepted in scientific research, but also “trim the data” on nuclear costs. Just as chapter 2 showed that industry calculations of greenhouse-gas (GHG) emissions err in usually including only emissions from 1 (reactor operation) of the 14 stages of the nuclear-fuel cycle, this chapter likewise shows that industry calculations of nuclear costs err in usually including only costs from 1 (reactor construction and maintenance) of the 14 stages of the nuclear-fuel cycle.

Attempting to show that increased atomic energy can help address CC, many industry advocates claim reactors are an inexpensive way to generate low-carbon electricity. However, surveying all 30 nuclear-electricity analyses, post-2000, this chapter shows that all of the industry-funded (but not the university-funded) studies appear to fall into COI and to illegitimately trim relevant cost data in several vital ways. Most exclude costs of full-liability insurance, underestimate interest rates and construction times by using “overnight” costs, and overestimate reactor load factors and lifetimes. Yet if these studies correct only these 5 false cost assumptions, market costs (excluding subsidies) of fission-generated electricity can be shown to be roughly 6 times more expensive than most nuclear-economics studies claim. Although there are legitimate situations in which scientists ought to trim scientific data, the chapter shows that trimming nuclear-cost data fails to satisfy standard scientific guidelines. After answering several objections, the chapter shows that nuclear fission is actually far more expensive than conservation, efficiency programs, and renewable-energy sources, like wind and solar photovoltaic (PV).

(p.71) The Economics Argument of the Nuclear-Fission Industry

As chapter 1 revealed, scientists agree that there is no safe, non-zero dose of ionizing radiation, and as chapter 4 shows, there are higher cancer rates around nuclear plants, given their normal radioactive emissions. Safety concerns, however, are not what caused no new US nuclear reactors to be ordered since 1974. Instead, naming “financial hurdles” as the biggest contemporary obstacle to new atomic-energy plants, economists say the “rebirth of the US nuclear power market will be greatly influenced by its associated costs.”7 Hence it is not surprising that when nuclear proponents promote using fission to help address CC, they make the economics argument. That is, they say atomic energy is inexpensive, “the most economical way to generate baseload electricity,”8 “some of the cheapest power available,”9 “very high yield,”10 and “cost effective.”11 A top business-management-consulting firm, McKinsey, praises nuclear energy for “low operational and maintenance costs.”12

Current CC concerns are one reason many hope the economics argument is correct. However, past nuclear performance is one of many reasons to doubt it. Nuclear proponents, economic consensus, and credit-rating firms agree that the “excessive costs” of “uneconomic” nuclear plants are what caused the industry to cancel hundreds of reactors and to order no new US plants since 1974.13 Over 2 decades, the 2 top US reactor vendors, GE and Westinghouse, each lost money on every reactor that they delivered for a fixed price,14 yet those prices have risen substantially, and no reactor vendor still offers fixed-price contracts—a telling fact. Industry proponents say the latest Western nuclear plants each will cost at least $12 billion,15 while future reactors will cost even more because of new, untested designs. Nuclear advocates also admit that consumers are still paying for the reactor problems of decades ago. Many plants were canceled after billions of dollars had been spent on them, and some were completed but never opened because of safety problems. For instance, decades ago, the $5.5 billion Shoreham (New York) plant was closed the same year it opened, and the $9 billion Watts Bar (Tennessee) plant—the last constructed in the US, and completed in 1996—took 23 years to build. Because of numerous financial problems, nuclear proponents admit that US nuclear “ratepayers were left responsible” for “some of the highest electric rates in the country.” However, the same proponents say that things have changed, that nuclear power is now one of the cheapest electricity sources available.16 Are they right? This chapter shows that they are not.17

Nuclear-Cost Studies

To assess the economics argument for nuclear fission, consider all 30, post-2000 prominent, international, nuclear-cost studies that are original economic analyses. (p.72) None is merely a summary or derived from earlier reports.18 These 30 analyses include all original nuclear-cost studies that are publicly available in scientific journals, books, nongovernmental-organization (NGO) analyses, industry reports, and government documents since the year 2000. The list of 30 appears both balanced and comprehensive, as it includes all 7 studies that are reviewed by the global-nuclear-industry lobby group,19 the World Nuclear Association (WNA).20 The list also includes all 9 nuclear-cost studies reviewed21 in a prominent 2006 UK government report,22 and all 12 nuclear-cost studies reviewed23 in a 2007 Greenpeace International report.24

One interesting fact about these 30 prominent nuclear-economics studies is that a majority appear to trim nuclear-fission costs in at least 3 ways. Subsequent paragraphs argue that they ignore taxpayer subsidies, ignore long reactor-construction times and interest costs, and inflate reactor capacity (or load factor) and lifetime data. As a consequence, they grossly underestimate the real costs of fission-generated electricity.

Ignoring Taxpayer Subsidies for Nuclear Fission

Consider first the way that most nuclear-cost studies ignore taxpayer subsidies. The largest ignored subsidies are those for nuclear-liability insurance. The European Commission (consistent with WNA and Cato Institute figures) recently showed that, if commercial reactors had to purchase full-insurance-liability coverage on the market, this would triple fission-generated-electricity prices.25 As already mentioned in chapter 2, this means a majority of nuclear-economics studies underestimate costs by a factor of 3, merely because they exclude full-insurance costs, presumably because these insurance costs are not market-related, but mainly paid by governments and taxpayers. Yet without these subsidies (and resulting liability protection), utilities vehemently agree they would never use risky atomic energy.26 Why not? Fission-insurance rates, available on the market, accurately reflect high nuclear-accident risks. The US-government-calculated, lifetime-core-melt probability for all US commercial reactors is 1 in 5,27 and government says a worst-case accident could cause roughly $660 billion in damages, excluding medical costs.28 Given these high risks and costs, utilities, governments, and credit-rating firms universally confirm that no nuclear plants, anywhere, operate on the market, and none could exist without massive government subsidies, including an accident-liability limit.29 In effect, utilities have said, “Give us subsidies, or give us death.” As chapter 1 revealed, US government subsidies for nuclear fission—alone—cut nuclear electricity prices by 50–90 percent. Yet, almost no government, industry, or university analyses of nuclear-electricity costs take account of subsidies, and almost no studies include the costs of industry's transferring (through nuclear-liability limits) its serious atomic-energy risks to the people.

(p.73) The world's 443 commercial reactors fall into 3 camps regarding nuclear-accident-liability coverage. The vast majority of reactors are in the first camp (e.g., in China, India, Iran, and Pakistan), where operator nuclear liability is 0, and accident victims could bear 100 percent of nuclear-accident costs. One-third of reactors (many in Western Europe and the US) are in the second camp, where operator liability is minimal. US reactors are protected by the Price-Anderson Act, and they have the highest (minimal) liability, $10.8 billion—roughly 1.5 percent of government-calculated, worst-case-accident damages of $660 billion.30 In this second case, accident victims could bear roughly 98 percent of nuclear-accident costs. The third camp includes 13 percent of reactors (in Germany, Japan, and Switzerland), all having government-guaranteed, unlimited liability.31 There taxpayers and utilities, not only victims, would bear all nuclear-accident costs. The upshot? All global reactors have either direct (87 percent of global reactors, those in camps 1 and 2 above) or indirect (13 percent of global reactors, those in camp 3 above), government-mandated nuclear-accident-liability protection. In the direct-liability case, government transfers either 100 percent or 98 percent of nuclear-industry-accident risks, costs, and negligence mainly to potential accident victims. In the indirect-liability case, government transfers industry risks, costs, and negligence to taxpayers through guaranteed government subsidies.32

Although nuclear-liability limits mean that 87 percent of global nuclear ratepayers bear financial and health risks from nearby reactors, surprisingly none of the major government, industry, or university nuclear-cost studies include these liability-related costs. Not even the classic 2004 MIT and University of Chicago studies include them.33 Neither do assessments by the main international nuclear-lobby group, the WNA,34 or the UN's Nuclear Energy Agency.35 In fact, energy-cost studies done by Rice University,36 Lappeenranta University,37 the UK,38 the UK Royal Academy,39 the US Department of Energy (DOE)–funded Scully Capital,40 OXERA Consultants,41 and most nuclear economists trim nuclear-liability-related cost data.42 What is their rationale?

They have 3 main arguments for trimming nuclear-insurance costs. The first, their market argument, is that neoclassical economic analyses should include only market costs;43 because government allows industry to buy no full nuclear-accident insurance on the market, they claim the nuclear-accident-liability limit is an externality, not counted in market calculations. The second, their responsibility argument, is that industry bears no financial or legal responsibility for full nuclear-accident-liability coverage because government requires “no payment” for full coverage.44 The third, their subsidy argument, is that, because nuclear-liability limits require no full-nuclear-insurance coverage, no taxpayer costs are involved: the limits are a subsidy from potential radiological victims.45

How reasonable are these industry arguments for nuclear-liability cost-trimming? The market argument fails because, even if cost-trimming is consistent with neoclassical economic procedures, energy-policy decisions require total-cost (p.74) accounting, not merely market prices. For instance, the real cost of gasoline is not merely its market price, but its market price plus all taxpayer subsidies, which equals more than $15 per gallon; in the US, these subsidies include percentage-depletion allowances, fuel-production credits, enhanced oil-recovery credits, foreign-income deferrals, export-financing subsidies, and so on.46 If policymakers fail to consider the full costs of goods (like gasoline), externalities can cause market exchanges to generate false economic signals, cause misuse of resources, and jeopardize efficiency, societal welfare, sustainability, distributive equity, fairness, reasonableness, and other values.47 Because of nuclear-liability-limit subsidies or externalities, accident victims could face death, injury, financial harm, injustice, and other uncompensated costs. Besides, if atomic-energy investments were economically efficient, it would be unnecessary to subsidize and therefore socialize nuclear-industry risks and costs—transferring them to taxpayers—while allowing privatized nuclear-industry profits.48

As chapter 2 already suggested, the main problem with the responsibility argument (for trimming nuclear-insurance-subsidy costs) is that it begs the question against nuclear-industry financial responsibility for its accidents, contrary to virtually all liability law. This argument also assumes that legal actions (like government-allowed, nuclear-liability-cost limits) are always ethical. Yet obviously this assumption errs. Much ethics is not covered by law, and the law can be ethically wrong (as in government-mandated segregation or sexism). Consequently, people are ethically responsible for harms and costs they impose on others, regardless of what the law says. But if so, nuclear industries are ethically responsible for accident costs, and therefore these costs ought not be trimmed, even if the law limits liability. Because victims’ nuclear-accident costs are not limited to roughly 1.5 percent of worst-case-accident coverage, up to the $10.8 billion for which industry is liable,49 cost calculations ought not be limited to this amount.

The subsidy argument likewise errs, in begging the question that taxpayer subsidies should be excluded from nuclear-electricity-cost calculations simply because they are subsidies. For all the reasons already stated, nuclear-accident-insurance subsidies should be included in nuclear-economics calculations because they are real costs. In fact, if they were not real costs, the nuclear industry would not demand protection from 98.5 percent of worst-case-accident liability as a condition of operation, as already mentioned. Without protection from accident costs, industry and government admit, nuclear-electricity generation would stop.50 Yet if these costs are so substantial that industry requires protection from them, it is inconsistent to trim them from cost calculations merely because citizens, not industry, is forced to bear them. Because someone must bear nuclear-accident costs, because even pro-nuclear governments say nuclear-liability limits impose “potential” liability on taxpayers,51 and because (as mentioned earlier) both government-calculated, nuclear-accident probabilities and nuclear insurance are very high, full-liability costs arguably ought not be trimmed from nuclear-economics assessments. After all, prospective investors (p.75) would not want nuclear-liability-cost data trimmed, because trimming could obscure the difference between desirable and undesirable investment risks.

At least 4 other considerations likewise suggest that nuclear-liability-data trimming is not reasonable. I call these, respectively, the rationality, misleading-science, sensitivity, and economic-risk arguments. According to the rationality argument, rational energy choices presuppose full cost-benefit accounting because decision-makers should minimize economically inefficient choices, potential risk victims need full information to protect themselves, and investors need it to make economically efficient decisions. For instance, if people were offered job A or job B, but only job A provided health-insurance coverage, obviously the prospective employees would not want job offers to trim insurance information, because such information is needed for making reasonable job choices. The same is true for energy choices.

According to the misleading-science argument, full insurance-liability costs ought not be trimmed from nuclear-cost calculations, because this could mislead the public. The public probably is unaware of the nuclear-liability limit and how the US Price-Anderson Act sanctions a financial and medical risk-transfer from nuclear utilities to taxpayers. Marvin Fertel, senior vice president of the Nuclear Energy Institute, has admitted that because this law removes “the cost for insurance against the liability” from utilities, this cost is “passed on in part to consumers.”52 Yet, since at least 1987, the American Public Health Association,53 scientists, and bioethicists all have reaffirmed the public's right to know about science-related health risks imposed on it. This right requires that nuclear-liability costs not be trimmed from energy assessments.

According to the sensitivity argument, costs of nuclear-liability insurance arguably ought not be trimmed, both because they are substantial and because nuclear-cost calculations are extremely sensitive to them. On the substantial point, as already mentioned, the European Commission showed that requiring full-nuclear-liability coverage would triple fission-generated-electricity costs.54 The global nuclear-industry-lobby group, the WNA, admits these EU calculations are correct. WNA says average annual-liability-insurance premiums for typical Western commercial reactors are each about $400,000 per $300 million in coverage,55 and the US government says a worst-case nuclear accident could cost $660 billion. If additional nuclear-liability-insurance coverage were purchased at the same ($300 million) rate, full $660 billion coverage could cost about $880 million (400,000 × 2,200) per reactor per year. This figure is consistent with Canadian economists’ claims that full nuclear-liability insurance equals “half the capital costs of nuclear reactors,”56 consistent with inflation-adjusted Cato Institute estimates of $200 million per reactor, per year57 (but higher than the $33 million per reactor, per year, estimates of US economists Jeffrey Dubin and Geoffrey Rothwell).58 Regardless of the exact figure, because nuclear-insurance data-trimming is substantial, because law has transferred this industry risk to taxpayers, because insurance rates provide reasonable societal-risk estimates, and because nuclear-electricity prices are extraordinarily sensitive to (p.76) these high-liability-insurance costs, the public arguably ought to know them. Besides, US Securities and Exchange Commission rules require the absence of a nuclear-liability limit (and resulting financial costs) to be disclosed to nuclear investors.59 Potential victims deserve the same protection. Consequently, nuclear-liability costs arguably ought not be trimmed from fission-cost calculations.

According to the economic-risk argument, because all pro-nuclear cost assessments trim full nuclear-liability-insurance costs, they ignore various economic risks associated with serious accidents: planned reactors might be scrapped, others might be shut down, and people might be vulnerable to baseload-electricity shortages. A single event (an earthquake, terrorist attack, or accident) could financially derail the entire nuclear industry.60 Trimming nuclear-insurance-cost data thus errs because it encourages assessors to ignore accident-related economic risks.

In summary, because a majority of the 30 nuclear-cost studies mentioned above trim taxpayer-subsidized, nuclear-liability-insurance costs from their energy-cost calculations, they encourage flawed economic signals, inefficient markets, questionable research ethics, and unequal treatment. Moreover, it seems inconsistent and unethical for assessors to trim (and not disclose) full-nuclear-liability costs that increase taxpayer risks,61 while because of the associated financial risks, the US Securities and Exchange Commission requires disclosing lack of nuclear-liability limits to investors.62

Trimming Nuclear Costs by Ignoring Nuclear-Interest Rates

Another strategy of most nuclear-economics studies is to assume “overnight” plant-construction capital costs, currently at least $12 billion in the US.63 Overnight costs assume 0-percent interest rates and 0 construction times. For instance, although the WNA says “the case for nuclear energy is now solid on economics alone,” its economics calculations include only “overnight costs”—costs that exclude construction time, finance, and interest charges on construction capital—as if the reactor were built overnight, without any construction inflation or interest charges on capital. The WNA says overnight-cost estimates of at least “$2000 per kW of capacity” “have been produced by [nuclear-reactor] vendors and their partners.” Attempting to justify this cost-trimming (i.e., using only overnight costs), the WNA says “most studies of the competitiveness of nuclear power base their estimates of capital costs on … recent reactors in Asian countries [whose safety standards are weaker than in the West, as later paragraphs show] and use overnight costs.”64

Likewise, perhaps because official US national policy and relevant federal-agency policies are pro-nuclear, even US government agencies trim cost data on nuclear plants, as the Tennessee Valley Authority did recently. It used “overnight costs only” (p.77) to quote prices for its reactors.65 Following most nuclear-cost analysts, the authors of the 2009 MIT study also quote total nuclear-plant costs as “overnight costs”—then say “this total [nuclear-power-plant] cost, which is exclusive of financing cost, is $4,706/kW.”66 Noting that earlier MIT analyses employed overnight costs, “as described in the MIT (2003) Future of Nuclear Power study,” the 2009 MIT authors attempt to justify cost-trimming by saying that using overnight costs “represents the standard basis for quoting comparable costs across different plants.”67

This “standard” procedure of the nuclear industry, however, is deeply flawed economic science. There are problems with assuming both that nuclear-interest rates are 0 and that construction times are 0. First consider interest rates. As already noted, the latest Western nuclear plants each cost at least $12 billion.68 Because construction costs account for three-fourths of ratepayer prices for nuclear electricity,69 these prices are extremely sensitive to data-trimming assumptions about construction-interest costs. By how much do economics-argument proponents trim the cost of capital, the interest rates on nuclear-construction loans? To answer this question one needs to know (1) typical interest rates for risky projects such as nuclear plants; (2) interest rates that nuclear proponents presuppose in their calculations; and (3) full economic effects of ignoring differences between actual and assumed interest rates.

Regarding (1), typical risky-project interest rates, most private investors and banks are unwilling to invest in fission except at rates of at least 15 percent.70 Citing poor credit ratings, high construction costs, numerous plant cancellations, a competitive energy market, and a long history of cost overruns, delayed plants, and equipment malfunctions, the World Bank, European Bank for Reconstruction and Development 2006, Asian Development Bank 2000, African Development Bank, European Investment Bank, Inter-American Development Bank, and others say nuclear power is “uneconomic”; as a matter of policy, they refuse nuclear loans or investments.71 Citigroup, Credit Suisse, Goldman Sachs, Merrill-Lynch, Morgan Stanley, and virtually all other investors also refuse long-term credit to the financially risky nuclear industry.72 Consequently, if nuclear utilities can obtain private-market loans, they typically pay rates of 15 percent, a telling admission that nuclear economics is very fragile. Credit-rating companies, such as Moody's and Standard and Poor's, also downgrade credit ratings of utilities with reactors, claiming that even massive taxpayer subsidies often cannot make reactors economical.73 Moody's recently calculated that, even if one ignores costs associated with reactor decommissioning, reprocessing, and permanent waste storage, nuclear-generated electricity still costs 3 times more than that from new natural-gas plants and double that from scrubbed, coal-fired plants.74 Nuclear proponents admit that decommissioning one reactor costs $2–6 billion,75 that reprocessing raises nuclear costs 33–58 percent,76 and that permanent US radioactive-waste storage will cost $1 trillion—that is, $50 billion for each of 104 US nuclear plants.77 Although none of the preceding, additional, trimmed costs (mentioned in this paragraph) will be included in this chapter's analysis, nevertheless one can easily see that (p.78) including these additional costs would make atomic energy even more uneconomical than this chapter argues—and yet even without including them this chapter is able to show that nuclear costs are underestimated by 600 percent.

Regarding (2), interest rates that economics-argument proponents typically assume, they never assume 15 percent, the “going rate” for nuclear utilities.78 Even (normally more reliable) university nuclear-cost studies, like those from MIT79 and Chicago,80 assume nuclear-interest rates of 11.5–12.5 percent, and these too are underestimates. Instead, the main international-nuclear-lobby group admits “most studies” of nuclear costs assume 0 interest costs (0 interest rate)—and only the lower capital costs for cheaper, more dangerous, poorly designed Asian reactors, not Western ones.81 They trim nuclear costs by including only a cheaper plant's “overnight costs,” those occurring if the reactor were built overnight, without interest charges and construction-cost increases. Yet the latest US commercial reactors have taken 23 years to build, during which interest charges accrue.82 When nuclear proponent Steve Berry,83 a former US Argonne National Laboratory adviser, calculated fission costs in 2007, he presupposed year-2002 dollars and included only overnight costs. Nuclear proponents say studies by the World Nuclear Association,84 Rice University,85 US DOE–funded Scully Capital,86 OXERA Consultants,87 and the US Energy Information Agency routinely use only overnight costs.88 Yet because atomic power has the highest construction costs and longest construction times of any energy technology, using overnight costs substantially trims cost data. Even the pro-nuclear lobby admits that “overnight costs” underestimate real fission expenses, which “mainly depend” on interest costs and construction time—neither of which is included in overnight costs.89 Nevertheless, the WNA uses overnight costs.

Regarding (3), differences in nuclear economics, when one omits typical 15-percent nuclear-interest rates, standard amortization formulas show this data-trimming cuts nuclear-construction costs by 250 percent. If a utility had 15-percent interest rates, made quarterly payments, and had typical, 15-year loans, its construction costs (principal and interest) would be about $30 billion, not $12 billion in overnight costs. This conclusion is consistent with the WNA admission that even very low interest costs double nuclear-construction costs.90 If all other operating costs remain the same—and if capital costs (principal plus interest) are 75 percent of operating costs, as most nuclear advocates indicate—then moving from 0- to 15-percent interest rates would increase operating costs, thus consumers’ nuclear-electricity prices, by at least 188 percent.91 This conclusion is consistent with the pro-nuclear International Energy Agency's admission that each 5-percent interest increase raises generation costs 50 percent,92 as other pro-nuclear studies confirm.93

Interest costs are not the only data trimmed when industry uses “overnight costs” to assess nuclear-electricity economics. The main international-nuclear-industry lobby group admits the US had a 50-percent increase in construction material, equipment, and labor from 2004 to 2008; that it had a 100-percent increase from 2000 to 2008; and that nuclear-construction costs increase through time.94 (p.79) Proponents likewise say that using new reactor designs, with un-worked-out “bugs,” will drive up costs.95 World Energy Council studies also show trends of increased nuclear-construction time, cost overruns, and interest charges.96 Despite multi-billion-dollar government subsidies, every nuclear plant ever built thus has run over budget—often by 400 percent—and had longer-than-predicted-construction times.97 This is why no reactor vendors offer fixed-price, or “turnkey,” plants.98 Given all these factors, nuclear-plant costs are likely more than $30 billion for principal and interest. Yet the pro-nuclear US DOE–Scully Capital study showed that fission plants provided typical electric-utility rates of return on their investment99 only if nuclear-construction costs (principal and interest) did not exceed $1 billion.100 If the pro-nuclear DOE–Scully Capital study is correct, it explains nuclear-proponents’ incentives to include only overnight costs in fission assessments.

The reluctance of private banks and investors to lend to the nuclear industry, and the DOE–Scully Capital study's warning about low returns on nuclear investments, are both consistent with findings of the US government oversight office, the DOE Inspector General. It said taxpayer-financed nuclear-loan subsidies impose significant risks on American taxpayers.101 Nevertheless, government gave nuclear-loan subsidies of $18.5 billion in the 2005 US Energy Act,102 and President Obama promised another $54 billion in his 2011 budget, mainly because Wall Street would not do so. Similarly, the UK government repeatedly has bailed out nuclear utilities when they went bankrupt.103 Likewise, the French government, which holds majority interest in the French nuclear industry (e.g., Areva), has had to finance French nuclear-construction programs.104 As a result, Finland's Areva (generation-III+) nuclear plant has a 2.5-percent, French-taxpayer-subsidized interest rate on its construction loan,105 when the “going rate” is 15 percent. As already noted, no nuclear plant has ever operated without massive taxpayer subsidies. Thus the industry itself admits that high nuclear-capital costs (principal and interest) will rule nuclear fission “out of consideration” in the future; it admits that because of escalating costs, by 2030 the amount of electricity supplied by atomic energy will decrease from its current 16 percent to 9 percent globally,106 and from 25 to 20 percent in the UK.107 The preceding data mean nuclear proponents are trying to trim (or cover up) the very capital costs that are killing the industry, presumably to encourage nuclear investment.

Trimming Nuclear Costs by Ignoring Nuclear-Construction Time

As just noted, because nuclear-generated-electricity costs are so sensitive to interest rates, they also are sensitive to construction times. Yet long nuclear-plant-construction times are a global phenomenon. The most experienced nuclear operators, such as Florida Power and Light, say current US new-nuclear-plant construction time is (p.80) 12 years.108 A pro-nuclear US National Academy of Sciences report estimates at least 11 years.109 A pro-nuclear business-management-consulting firm, McKinsey, estimates 9–11 years.110 Nuclear advocates admit the last US nuclear plant, Watts Bar (Tennessee), took 23 years to construct and went billions of dollars over budget.111 Comanche Peak (Texas) went billions of dollars over budget and took 16 years; Nine Mile Point (New York) went billions of dollars over budget and took 14 years; Seabrook (New Hampshire) went $6 billion over budget and its delays caused utility bankruptcy;112 and so on.113

In the UK, average nuclear-plant-construction time is 11 years.114 The 4 French plants completed between 2000 and 2002 averaged more than 10 years to build and 14 years to produce commercial electricity.115 The 5 Japanese nuclear plants completed in the 1980s averaged 17 years to build, whereas the one built in the 1990s took 26 years.116 Eastern European nuclear plants, completed after 2002 and using US technology, have taken 15 years to build,117 as have Eastern European plants using Soviet technology.118 Moreover, several factors, already noted in discussing nuclear-interest rates, suggest new reactors will have longer construction times. Because no US plants have been ordered since 1974 and nuclear technology has been declining globally,119 new fission technologies are untested.120 Infrastructure, manufacturing facilities, and trained workforces are lacking. Costs for basic materials, like steel and concrete, also are spiraling upward.121

Despite the preceding data, industry lobbyists say most nuclear-economics studies assume overnight costs and therefore trim interest rates and construction times to 0,122 as already mentioned. For instance, Rice University,123 University of Lappeenranta,124 UK,125 and OXERA studies assume 0 nuclear-construction time.126 The pro-nuclear MIT authors assumed 5-year nuclear-construction times,127 as did Scully Capital,128 the Royal Academy,129 and the International Energy Agency.130 The pro-nuclear University of Chicago assumed 7 years,131 while other studies used 3 years.132

The economic effects of such nuclear-construction-time trimming are to erroneously lower calculated fission costs. Yet nuclear proponents admit that each 5-year nuclear-construction-time increase raises capital (principal plus interest) costs 100 percent.133 This suggests that assuming 10, not 0, years of nuclear-construction times increases capital costs 200 percent. But if capital costs are 75 percent of operating costs,134 then assuming 10-year reactor-construction times increases nuclear-electricity-operating costs 150 percent (0.75 times 200).

Given nuclear-cost sensitivity to construction times, it is disconcerting that official WNA nuclear-construction-time tabular data err. Instead of averaging construction times of all plants being built, WNA data do “not include plants on which construction has stalled.”135 Yet 22 plants are being built, and construction is “stalled” on 12; this suggests that construction has about a 50-percent probability of stoppage. Given how the WNA trims nuclear-construction-time data by including only 10 of 22 plants,136 its construction-time study is comparable to a (p.81) pharmaceutical-industry drug study that throws out all cases showing harmful drug effects, yet claims the drug is beneficial.

Accurate, untrimmed nuclear-construction-time data are crucial to energy choices because even proponents admit fission is an interim technology, useful only until efficiencies and renewable energy supply all needs.137 Thus, as chapter 6 shows, if reactors cannot be built quickly, they are of little use, especially because the classic Princeton study, published in Science, shows that only 6 of 9 renewable or efficiency technologies, “already deployed at an industrial scale,” could easily solve the climate problem by 2050; a 2006 study by the pro-nuclear US DOE claims that already-available renewable technology can provide 99 percent of US electricity generation even earlier, by the year 2020.138 As chapter 6 reveals, already by 2005 the annual global-growth rate of non-hydro renewable energy was 7 times greater than nuclear,139 partly because renewables like wind are inexpensive and their investments can be paid off in 10 years.140 In the last year, 60 percent of new added US electricity capacity was wind, as measured by peak summer demand,141 Trimming nuclear-construction-cost data thus could mislead energy decision-makers, especially given that nuclear market costs are 3–4 times higher than those for wind, as chapter 6 shows.

Trimming Nuclear Costs by Inflating Reactor-Load Factors

A fourth fission-cost-trimming strategy, used in a majority of the 30 studies, is overestimating reactor-load or capacity factors (plant-output percentages, compared to 100-percent output). For instance, WNA claims that “capacity factors of nuclear plants around the world have increased …. Levels of 90% and above have been achieved by many plants in Europe and Asia for many years.”142 Likewise, the 2009 MIT study presupposes a “capacity factor of 85%.”143 Yet what do actual empirical data say about nuclear-load factors?

The first atomic-energy plants began generating electricity in 1955. After 30 years of commercial experience, nuclear proponents say they ran about half the time and had average load factors of 50 percent.144 Low-load-factor reactors—like Fort St. Vrain (Colorado), at 14 percent—closed early because they were uneconomical.145 With more nuclear reactors (104) than any other nation, all US plants have an average-lifetime nuclear-load factor of 71 percent.146

Atomic-energy load factors are poor, proponents say, because even flawless performance and perfect components allow, at best, 90-percent load factors over a very short time, given needed debugging, refueling, and maintenance.147 Yet no nation's average-lifetime nuclear-load factor has ever been close to 90 percent because of leaks, equipment failures, and human error. In 2000, generator-tube ruptures at Indian Point reactor shut it down for 10 months, and replacement power averaged (p.82) $600,000 per day; boric-acid leaks and a football-sized corrosion hole in Ohio's Davis-Besse steel cap shut down the reactor for 2 years; replacement-power costs were $450,000 per day.148 Japanese, Canadian, UK, and other authorities likewise have ordered fleetwide reactor shutdowns in the face of systemic safety lapses.149 For instance, in 2005, when cracks in the graphite cores of AGR reactors caused safety compromises, 14 (of 23 total) UK reactors were shut down prematurely.150 Steam generators and reactor-pressure vessels, each costing $100 million, are just 2 of the “most-often-replaced parts” of a nuclear plant. Pro-nuclear, US DOE economics estimates are that 10 percent of all reactor components must be replaced every 10 years.151 On average, this means 1 percent of reactor components must be replaced each year, costing $120 million annually (assuming principal = $12 billion). Because the US GAO, the government oversight office, warns that nuclear-plant managers often defer such maintenance in order to keep reactors running and to save money, despite the resulting safety risks,152 only lifetime, national, fleet-average nuclear-load factors provide reliable data.

Will new, improved reactor designs increase nuclear-load factors and reduce shutdown times? Three facts suggest the opposite. First, standard, pro-nuclear IAEA databases reveal that new-reactor designs always have lower initial load factors (like 50-percent load factors for the first 30 years of US plants, already mentioned), although these often improve with reactor-operator experience and “debugging.”153 Second, because even vendors admit new-reactor designs typically have 50-percent load factors, at least for 4–5 years, they never guarantee specific load factors for clients.154 Third, industrial history reveals that after 50 years of commercial implementation, most technologies have achieved all likely design improvements, efficiencies, and cost savings; because commercial reactors have operated for more than 50 years, it is unlikely their load factors will improve significantly.155

Despite the preceding 3 facts, most fission-cost assessments assume much higher nuclear-load factors (thus lower costs) than the current 71-percent US average and 79-percent global average.156 Yet as already mentioned, the pro-nuclear IAEA excludes load-factor data for stalled or already-closed plants.157 It misleadingly calculates load-factor averages (US and global) by basing them only on the most economical nuclear facilities. Also, the 79-percent global nuclear-load-factor average is misleadingly high because of data from less-safe reactors (not allowed in the US), which are operated under lax regulations in the developing world.158 Consequently, US, UK, and French load-factor data are more accurate.

Past experience of Western nations also confirms that new-reactor designs (generation-III and -III+) do not have improved load factors. No generation-III+ plants in the West have been completed, as of 2010. The load factor for Finland's generation-III+ plant (being built by the French) will not be known until after many years of operation. Besides, previous generation-III nuclear plants (e.g., the US AP600) already have been abandoned as uneconomical159 or have suffered numerous breakdowns, (p.83) months-long technical shutdowns, and operation at 15 percent below their standard load factor (e.g., the Japanese ABWR-III's).160

Although load factors associated with some technological designs improve over time, the latest reactors, now fully implemented commercially (generation-II designs), have worse load factors than their predecessors. In France, which supplies more electricity (75 percent) from atomic energy than any other country,161 generation-II reactors have worse load factors than the French nuclear-fleet average. Given generation-II design and corrosion difficulties, most UK reactors (and all newer AGR-II reactors) operate at 71-percent load factors; half of UK reactors have load factors averaging 54 percent. Globally, three-fourths of the 414 operating reactors (having at least 1 year's service, and therefore higher load factors) have lifetime load factors below 80 percent. Only 7 of all 414 global reactors (1.7 percent)—mostly those with lax design, standards, and enforcement—and delayed maintenance in the Third World — have individual load factors of at least 90 percent.162

How close are pro-nuclear load-factor assumptions to the 71-percent US average or the 79-percent global average? As already noted, nuclear-load factors assumed in most of the 30 current nuclear-cost calculations, 90–95 percent, are physically impossible for any sustained period that includes maintenance. Yet 2 recent studies, from Finland's Lappeenranta University163 and OXERA,164 respectively, assume load factors of 91 and 95 percent. Five other recent, prominent, pro-nuclear studies—from the nuclear-lobby group WNA,165 US DOE–Scully Capital,166 the Royal Academy,167 and the Canadian Nuclear Association168—assume load factors of 90 percent—something never achieved by the average performance of any nation's reactors. Five other prominent studies also assume average load factors never achieved by any nation. The later WNA,169 UK,170 MIT,171 University of Chicago,172 International Energy Agency,173 and UK studies all assume average load factors between 80 and 85 percent.174

The most-commonly-assumed load factors, 90–95 percent, have been attained, only temporarily, by only 1.7 percent of all reactors, after deferred maintenance and many years of operation. They do not represent all-reactor, lifetime, national averages. Based on the current average load factor in the US (71 percent), these studies overestimate nuclear reliability by 19–25 percent. Even the 5 “more realistic” nuclear-load-factor assumptions (80–85 percent) overestimate fission reliability by 9–14 percent. Obviously, though, nuclear-economics studies should use average, lifetime, national load factors, not single-year load factors for the top 1.7 percent of non-US reactors. Using this fallacious 1.7-percent strategy, one could just as well trim all but the lowest load factors, then claim a 14-percent US nuclear-load factor, based on the closed Fort St. Vrain plant.175

What are the erroneous cost-savings of massively overestimating nuclear-load factors? The pro-nuclear MIT study calculated that, if nuclear-load factors decreased 10 percent, this alone would increase overall ratepayer costs 10–15 percent.176 MIT thus estimates roughly 1–1.5 percent ratepayer-price savings per 1-percent (p.84) load-factor improvement. All other things being equal, this suggests that using the most common load-factor assumption (90–95 percent), not the average US nuclear-load factor (71 percent), erroneously cuts nuclear-ratepayer prices 19–36 percent.

This load-factor data-trimming not only assumes that temporary performance of top-performing 1.7 percent of third-world reactors are typical of all reactors, but also obscures the fact that fission faces the same power-intermittencies as renewables like wind or solar. Correcting for load-factor data-trimming alone could raise nuclear-ratepayer prices 19–36 percent; yet the pro-nuclear UK government says wind-intermittency increases wind-generated-electricity costs only 6 percent, and wind is several times cheaper than nuclear energy.177 According to credit-rating firms, nuclear-energy prices are more than 15 cents per kWhr,178 while the pro-nuclear US DOE says actual US wind prices, on average over the last 7 years, are about 4.8 cents per kWhr.179 Because decision-makers need reliable cost data on energy-intermittencies, overestimating nuclear-load factors appears unjustified—and thus likely to lead to poor energy choices.

Trimming Nuclear Costs by Inflating Reactor Lifetimes

Estimated nuclear-fission costs also artificially drop when assessors overestimate reactor lifetimes. Given high-temperature, heavy-radioactive bombardment of reactor materials, current plants were designed to last 30 years. To save utilities money, some licenses have been extended longer,180 because it can be cheaper (assuming no major repairs) to spread nuclear-electricity costs over longer lifetimes.181 However, the global-average lifetime of all 119 already-closed reactors is 22 years.182 Safety and economic problems caused 19 US fission plants (20 percent) to retire early, and more than $20 billion was spent on 121 plants that were later canceled.183 Thus more US reactors (140) were closed prematurely or canceled (during construction) than are now operating (104).

Rather than actual, global-average reactor lifetimes of 22 years, however, virtually all nuclear-cost studies counterfactually assume much longer lifetimes. The WNA claims, for instance, that nuclear plants “can offer electricity at predictable low and stable costs for up to 60 years of operating life.”184 Such industry data-trimming obscures the 22-year global-average nuclear lifetime for at least 3 reasons. One reason is that published industry tables of reactor lifetimes omit canceled and early-retirement plants, then calculate average lifetimes only for remaining plants.185 Another reason is that all nuclear-industry and government cost assessments assume hypothetical, counterfactual longer lifetimes. The Lappeenranta University study assumed a 60-year lifetime;186 the WNA,187 US DOE–Scully Capital,188 Royal Academy,189 University of Chicago,190 International Energy Agency,191 OXERA,192 (p.85) and UK studies assumed 40-year lifetimes.193 MIT studies assumed 25–40 years.194 The vast majority of nuclear-cost studies assume reactor operating-lifetimes of 40 years or more. A third reason nuclear-industry data-trimming obscures accurate data is that, even if reactors lasted 30 to 50 years rather than 22, University of London economist David Fleming shows that, at best, reactors can “produce electricity at full power for no more than 24 years.” This 24-year limit occurs because of 4–8 years of start-up debugging and because corrosion and intense radioactivity soon make reactors “impossible to repair.” Consequently, Fleming says new-reactor designs will not have longer lifetimes.195

What are the economic effects of artificially overestimating reactor operating-lifetimes? The pro-nuclear MIT study calculated that increasing plant lifetimes from 25 to 40 years would reduce overall nuclear-electricity costs 5 percent.196 Erroneously trimming reactor lifetimes, however, not only underestimates costs but also ignores ratepayer economic burdens from early closures. New York's Shoreham plant, for instance, was shut down the year it opened, and taxpayers paid the bill. Washington Nuclear defaulted on $2.25 billion in bonds for a reactor,197 and ratepayers again paid. As already mentioned, typical nuclear-cost studies—which illegitimately trim expenses through counterfactual assumptions, such as longer plant lifetimes—are like pharmaceutical-company studies that claim drug safety, yet rely on trimmed drug-trial data that exclude subjects who were forced to quit the trials early because of adverse drug effects.

Effects of Trimming Nuclear-Cost Data with Five Counterfactual Assumptions

How does economic-data-trimming affect fission-electricity prices? The preceding data show that including full nuclear-liability-insurance expenses could—alone — increase atomic-energy costs 300 percent above most published nuclear-cost estimates, which were above $0.15/kWhr in 2008, according to credit-rating firms,198 and roughly $0.21/kWhr in late 2009.199 Including full 15-percent nuclear-interest charges rather than assuming 0-percent interest could alone raise costs 188 percent. Including 10-year rather than 0-year reactor-construction times could alone increase costs 150 percent. Using historical-average (71-percent) rather than hypothetical (90- to 95-percent) nuclear-load factors could raise costs 19–36 percent. Finally, using actual historical (22-year) rather than hypothetical (40-year) nuclear-plant lifetimes could increase costs 5 percent. Provided effects of these 5 types of data-trimming are independent, correcting them could increase atomic-energy costs nearly 700 percent—more precisely, 662–679 percent (300 + 188 + 150 + (19 – 36) + 5)—to about $1.47/kWhr (or 7 × $0.21/kWhr). Although this cost is far above all published estimates, it may be too low because it excludes expenses such as full nuclear-waste storage, reactor decommissioning, and the 15-percent annual increase in nuclear-construction costs caused by labor and materials increases.200

(p.86) Guidelines for Scientific Data-Trimming

Given these massive economic effects of atomic-energy data-trimming, is data-trimming ever defensible? What do scientists say? Methodological controversies over data-trimming have a long scientific history. Mathematician Robert Babbage long ago warned of scientists who “cook” and “trim” data, yet many researchers, even famous scientists, appear to have done so. Robert Millikan conducted experiments with oil drops to measure the smallest electrical charge on an electron. He dripped oil through electrically charged plates and measured the effect of plate charges on the oil drops. Later, in a published paper on the charge of an electron, he reported only 140 of the 189 oil-drop observations recorded in his lab notebooks. Consequently, some scientists say Millikan unethically trimmed his data. Others disagree. The historical record likewise suggests that Mendel's peas had some assistance in sorting themselves in ways that agreed with Mendel's theory. Today, in areas like neuroscience, researchers debate whether some data plots have been omitted, smoothed, or shaped.201

On one hand, data-trimming proponents say not deleting outliers could lead to misleading conclusions,202 or to indefensible variability in the data.203 They note many largely accepted techniques for case-specific, scientific-data trimming.204 In statistical studies, proponents say data-trimming often can give higher-power, more-precise results;205 reduce error in the mean; or make sample means less variable in heavy-tailed distributions.206 Besides, they say that “with a normal distribution … the power loss” is quite small, that data “trimming shows great promise and should have wide usefulness,” and that sometimes it “can make a silk purse from a sow's ear.”207

On the other hand, although most scientists say data-trimming to fit some predetermined theory is unacceptable, some say data-trimming is always wrong.208 They claim it makes many statistical inferences inoperable,209 often lowers statistical power, may delete important data, and is an “extreme” measure that ignores the theoretical importance of outliers. They say one should never trim data but should instead attempt to minimize outliers’ misleading effects.210

If one assumes there are at least some circumstances in which data-trimming might be legitimate, the key questions are whether, when, and how to trim data. The scientific literature provides 5 prominent guidelines.211 The first 2 guidelines suggest when and whether to trim, and the next 3 suggest how to do so.

  1. (1) Typically data should be trimmed only when they are rare, extreme data points; all other things being equal, “the more nonnormal, the greater should be the trimming,”212 because extreme data points may reflect measurement error.

  2. (2) Typically data should be trimmed only when doing so is necessary to prevent error: for example, when outliers overly influence the mean or come from a different population.213

  3. (3) Data-trimming typically should be accomplished by means of cuts that are symmetrical: for example, removing the upper and lower x percent of the data (p.87) set, then finding the arithmetic mean of what is left. One ought not “trim just the high values or just the low values,” because this would bias results.214

  4. (4) Data-trimming typically should be accomplished by means of techniques that allow documenting and evaluating associated error and uncertainty. Techniques include using standard formulae to calculate standard deviations or error bars for trimmed means, comparing confidence intervals for trimmed and untrimmed data, assessing different types of trimming by giving error bars for alternative degrees of trimming, and reporting “auxiliary information” on how well trimming worked.215

  5. (5) Data-trimming typically should be accomplished by means of techniques that include communicating trimming results to ensure they do not mislead: for example, noting the presence of outliers and the effects of trimming on results.216

Obviously there are many questions about how to use and interpret the preceding guidelines. How can one satisfy both guideline (3), regarding symmetrical cuts, and guideline (2), if measurement error apparently affects only upper-tail data? When should one employ guideline (1), if outlier data can reveal evidence of low-probability, high-consequence risks? While such questions deserve answers, they are beyond the scope of this chapter. The point here is to give a preliminary answer regarding whether nuclear-related data-trimming appears consistent with noncontroversial aspects of the guidelines.

Economics-Argument Data-Trimming and the Five Guidelines

Is nuclear-liability-related data-trimming defensible? Although guideline (1) suggests trimming only extreme data points, the earlier analysis shows that full nuclear-liability-insurance-cost data are not rare outliers, but consistent data possibly warning of needed liability protection. Thus, this trimming appears not to follow guideline (1). Nor does it follow guideline (2); eliminating liability-related costs is not necessary to avoid nuclear-cost errors. Rather, the trimmed costs are real, transferred from industry to the public. With the exception of parts of MIT analyses, most pro-atomic-energy studies likewise fail to follow guideline (3), regarding liability-insurance costs, because their trimming is not symmetrical: they trim only high-end nuclear-insurance costs. They also fail to follow guideline (4), which requires documentation of trimming-related error and uncertainty, because the majority of studies provides no documentation. Nor does this trimming follow guideline (5), which requires public communication of trimming effects so that taxpayers are not misled. The majority of studies includes discussion of neither nuclear-liability limits nor cost-trimming effects on citizens when nuclear-liability risks are transferred from industry to the public. One of the only pro-nuclear studies that (p.88) attempts to follow guidelines (4) and (5) is the 2003 (but not the 2009) MIT analysis,217 although it errs in other ways. The 2003 MIT authors justify trimming liability-related-cost data by erroneously claiming that (i) the nuclear-insurance subsidy “is very small” and (ii) US law does not require firms to carry full insurance.218 However, if (i) were correct, the nuclear industry would not require liability-insurance subsidies as a prerequisite for operation, yet obviously these subsidies are massive, half of reactor capital costs, as discussed earlier. Similarly, (ii) fails to justify insurance-cost data-trimming, because the law's not requiring something (full-liability coverage) does not mean analyses ought not include it. Legal and economic requirements are 2 different things; otherwise, legal and economic analyses would be identical, and they are not. Besides, as previous paragraphs showed, the US Securities and Exchange Commission follows guideline (5); it requires that investors be told whether there is a nuclear-liability limit (which reduces their investment risk). This requirement suggests the public deserves the same risk-information cost-trimming effects. Thus, almost all nuclear-liability data-trimming fails to follow guidelines (1)–(3), while most fails to follow guidelines (4)–(5).

Regarding the second and third types of nuclear-cost data-trimming—assuming interest-rate and construction-time data are 0—all pro-nuclear economic analyses likewise run afoul of data-trimming guidelines. They fail to follow guideline (1) because trimmed interest-rate and construction-time data are not rare outliers, but typical; all plants have non-0 values for both. Likewise, interest- and construction-time trimming fails to follow guideline (2) because it does not prevent, but actually causes, error—a 250-percent underestimate of nuclear costs when interest is trimmed to 0, as is typical. The trimming also fails to follow guideline (3) because these data cuts are not symmetrical. As previous paragraphs showed, the trimming is biased in cutting only lower-end interest-rate and construction-time data. Nearly all pro-nuclear studies likewise fail to follow guideline (4)—which requires the documentation of trimming-related error and uncertainty—and guideline (5)—which clarifies the effects of data-trimming. As already mentioned, however, 2 studies followed guidelines (4) and (5) and showed precise, erroneous cost decreases because of failure to use actual, 15-percent interest rates.219 Thus, virtually all interest- and construction-time data-trimming fails to follow guidelines (1)–(3), while most studies likewise fail to follow guidelines (4)–(5).

With respect to the fourth type of nuclear-cost data-trimming—assuming higher, counterfactual load factors—almost all atomic-energy-cost studies likewise fail to follow guideline (1) because they trim actual load-factor data, rather than rare outliers. They also fail to follow guideline (2) because using hypothetical load-factor data is not necessary to prevent errors, but likely generates them. Likewise, this trimming fails to follow guideline (3). The data cuts are not symmetrical: they trim only low-end load factors, biasing nuclear-cost calculations. Almost all nuclear-economics studies likewise fail to follow guideline (4), which requires the documentation of trimming-related error and uncertainty, and guideline (5), which warns against misleading the public about trimming effects. Ignoring guideline (5), (p.89) most of these studies fail to analyze the cost effects of using counterfactual nuclear-load factors. One exception is the MIT study, as already noted; it showed that each 1-percent load-factor decrease caused a 1.5-percent nuclear-ratepayer-cost increase.220 Thus virtually all pro-nuclear load-factor data-trimming fails to follow guidelines (1)–(3), while most fails to follow guidelines (4)–(5).

Finally, regarding the fifth type of nuclear-cost-data trimming, as earlier paragraphs revealed, almost all fission-economics studies trim costs by assuming longer, contrary-to-fact nuclear lifetimes of 40–60 years, not the 22 years shown by historical data. This trimming fails to follow guideline (1), because instead of deleting rare outliers, it omits factual, in favor of counterfactual, data on lifetimes. This data-trimming also violates guideline (2) because it is not necessary to prevent error, but rather generates it by using contrary-to-fact lifetime data. Lifetime-related data-trimming likewise fails to follow guideline (3) because instead of deleting symmetrically, assessors cut only low-end lifetimes. One exception is the UK (2002) Performance Unit study, which employs 15- to 30-year lifetimes. However, almost all nuclear-economics studies fail to follow guideline (4), which requires the documentation of trimming-related error and uncertainty, and guideline (5), which warns against misleading the public regarding trimming effects, because (with the exception of MIT scientists, as noted) nearly all failed to analyze load-factor-trimming effects on nuclear cost. MIT authors calculated that, if average-nuclear-operating lifetimes increased from 25 to 40 years, this would reduce overall-nuclear-electricity costs 5 percent,221 an admission that follows guidelines (4)–(5). Thus virtually all pro-nuclear, load-factor data-trimming fails to follow guidelines (1)–(2), while most fails to follow guidelines (3)–(5).

If preceding paragraphs are correct, the least-followed guidelines are (1)–(2), and assessors trim typical, not rare, data points. In trimming 4 of 5 types of nuclear-cost data (liability, interest-rate, load, and lifetime-related) nearly all these analyses also trim asymmetrically and thus fail to follow guideline (3). This suggests biased data-trimming, perhaps done to lower apparent costs. Although most pro-nuclear studies appear flawed in all 25 ways (failing to follow 5 guidelines regarding 5 types of data-trimming), the 2003 MIT study does follow 6 guidelines, namely guidelines (4)–(5) for 3 of the 5 types of data—liability, load-factor, and reactor-lifetime.222 Thus despite its serious shortcomings (failure to follow 19 apparent guidelines for scientific data-trimming), this MIT study is less flawed than the other 29 studies that trim nuclear-cost data in 25 questionable ways.

Studies That Do Not Trim Nuclear Costs

Because so many nuclear-economics analyses trim cost data, and do so in illegitimate ways, at least 2 questions arise: “Do any studies get nuclear costs right?” and “What can explain why so many nuclear-cost studies have erred so badly?”

Regarding the first question, many analyses do include more complete nuclear costs. For instance, although most studies ignore massive nuclear subsidies (e.g., for (p.90) accident-liability insurance), analyses done by university professors and by NGOs often include them.223 Authors from the University of New South Wales and the University of Melbourne, for example, claim that nuclear subsidies are “market distortions” and that taxpayer subsidies cover 60–90 percent of new nuclear-construction costs.224 In Britain, they say, nuclear subsidies are about $2.6 billion annually. As a result, these university authors note that “ignoring the huge subsidies from government to nuclear energy also makes the technology look less expensive,” although “the current economics of nuclear power make it an unattractive option for new generating capacity.”225

Similarly, many studies done by NGOs or university professors neither trim interest costs and construction times nor use overnight costs.226 Instead they note that “choosing an unrealistically low interest rate can make nuclear energy look much less expensive.”227 Thus, these authors use full market interest rates for nuclear construction, in part because many of them obtain data from banks and from Standard and Poor's, Moody's, and other credit-rating agencies. As one University of California, Berkeley, engineering PhD claims, nuclear plants “pay more on the margin for credit. Federal support of construction costs will do little to change that reality.”228

Likewise, many university and NGO studies use empirical, historical-average load factors and nuclear-plant lifetimes. For instance, one Technical University of Eindhoven scholar criticizes the assumption of a “very long operational lifetime [for a nuclear plant], as MIT proposes” in its nuclear-cost analyses. He says instead that today “only a few reactors in the world reached an operational lifetime of 24.6 FPY [full power years] today. Extensive refurbishments are required to reach even this lifetime … . The reliability of the reactor vessel determines the operational lifetime of a NPP. The quality of the vessel deteriorates over time by corrosion and neutron capture.” Thus “the lifetime of one reactor,” even with extensive refurbishment, is “30–40 years” at best; “the average operational lifetime of the reactors to be decommissioned on the list of the UK Nuclear Decommissioning Authority … is 18.7 FPY.”229

As the previous examples show, university-based nuclear-cost studies often provide more reliable cost figures, based on historical-average data, bank estimates, and credit-rating figures. Reliable nuclear-cost data are difficult to obtain, however, for reasons that become clear once we address the second question.

What May Explain Nuclear-Cost-Data Trimming

Regarding the second question, what can explain why so many studies err so badly in underestimating nuclear-electricity costs? At least 7 reasons come to mind. One reason, obvious from chapter 2, is that those who calculate nuclear costs ignore both full-fuel-cycle GHG emissions and costs. Examining only part of the fuel cycle, (p.91) they grossly underestimate atomic-energy costs. Yet, as chapter 2 showed, the massive energy requirements of all 14 nuclear-fuel-cycle stages mean reactors produce only 25 percent more energy than that needed throughout their fuel cycles. These large energy inputs cause energy bankruptcy, which in turn translates into financial bankruptcy. Thus, failure to calculate full-fuel-cycle energy debts is one reason that atomic-energy economic analyses err so badly.230

A second reason for nuclear-cost errors is that current turnkey or fixed-price data for nuclear-plant construction are not available because no utilities have been willing to build turnkey nuclear plants, as they did with early reactors. Nuclear proponents say this unwillingness has arisen because every utility has had cost- and construction-time overruns on all reactors; every utility has lost money on turnkey plants.231 Instead, “cost-plus” contracts have become the norm.232 Yet, given various types of fission-cost overruns, it can be difficult to account for all construction costs. Thus, the Congressional Budget Office recently revealed that average US nuclear-plant-cost overruns have been 207 percent and that reactors cost triple their original estimates.233 Although the Finnish reactor (now being built by French-government-owned Areva) is a turnkey plant, it is mostly taxpayer-subsidized; only 3.5 years after its construction began, it already was 50 percent over budget and 3 years late.234 In February 2010, project manager Jouni Silvennoinen said the reactor's estimated start date will be delayed to June 2012. By 2010, market construction costs were $7.2 billion and rising—already 200 percent over budget.235 Because this and another Areva plant are the only latest-design nuclear plants under construction in Western Europe or North America,236 good nuclear-cost data are difficult to find.

A third reason most nuclear-economics studies underestimate costs is that utilities are allowed to quote different nuclear prices to different groups, at different stages of the nuclear-fuel cycle, and they know that low-price quotes help sell reactors. These different price quotes cause confusion and cost misstatement. For instance, when a utility applies to build a nuclear plant and tries to show its cost-competitiveness, it quotes highly trimmed economic data that omit many charges—for example, interest and transmission-system upgrades—that are not paid to the reactor vendor. These trimmed data are “how the cost is typically quoted by the [nuclear- plant] vendor”; however, once the plant is built, the utility quotes much higher costs to the Public Service Commission, including inflation, interest, upgrades, cost overruns, and so on, because “the regulated utility will be allowed [by the commission] to recover this total cost through customer charges.” Yet as revealed earlier, interest charges alone more than double “the [nuclear-plant] vendor's EPC [engineering, procurement, and construction] overnight cost” that was quoted before the plant began. Moreover, the trimmed nuclear-cost figures, submitted to the commission to gain building approval, typically have “all detailed information about this cost figure” redacted from filings. Thus nuclear-price data often are underestimated because government allows them to be quoted in different ways, depending on their audience and use (either to convince utilities to purchase allegedly (p.92) low-cost reactors, or to convince public-service commissions to allow full nuclear-cost recovery from ratepayers). Given no universally accepted, canonical studies for what nuclear-cost assessments ought to include and how they ought to be represented, obviously it is in the reactor vendor's financial interest to underestimate nuclear costs in order to sell reactors. As already mentioned, the 2009 MIT authors say that using overnight costs “represents the standard basis for quoting comparable costs across different plants”; they are “how the cost is typically quoted by the [nuclear plant] vendor.”237 Given this norm and earlier comments on different nuclear audiences and data uses, there are no universally accepted, canonical studies for how and what nuclear-cost assessments ought to include.

Moreover, a fourth reason for underestimated nuclear costs is that many industry analyses do not clarify precisely what is included in their cost tabulations. Given this fact, and nuclear-industry control of most nuclear-cost data, misrepresentation seems likely. For instance, one prominent UK government study, done at the University of Sussex, noted that although industry-funded studies exclude financing and other costs,238 often “it is not known” (e.g., in the pro-nuclear Finnish analysis)239 what costs are included or excluded.240 Consequently, Sussex University researchers claimed that even government nuclear-cost reports, such as the UK Department of Trade and Industry's 2006 study,241 “leave no clear audit trail.”242 Scientists from the University of Melbourne and the University of New South Wales agreed. They said “published [nuclear] capital-cost estimates … derive from studies … from vendors of reactor systems …. The data are supplied … by the nuclear industry itself and are not open to objective verification.”243 More generally, the Sussex University authors (of a UK government report) criticize the “[cost] appraisal optimism” of most nuclear studies. Under the heading of “capital costs,” they note that “all of the [nuclear capital-cost] data is traced back to industry sources, usually reactor vendors, and the number of these sources is very few …. Reactor vendors inevitably and legitimately have an interest in presenting costs in a way that maximizes their chances of commercial success.”244

A fifth reason for underestimated nuclear costs is that nearly all studies are either performed or funded by the nuclear industry, which has financial incentives to minimize costs. Of the 30 post-2000 nuclear-cost studies analyzed here, at least 18—or 60 percent—come directly from nuclear interests.245 Virtually all have been either performed by,246 or at least partly funded by, the nuclear industry247 or pro-nuclear government agencies, like the US DOE.248 For instance, the 2009 MIT study notes it has been funded by the pro-nuclear or lobby groups EPRI and INEEL, but admits, “None of the figures reported [by the nuclear industry] for these [nuclear] plants represent actual costs.”249 Instead all of the cost figures are too low and have trimmed data.

A sixth reason for underestimated fission-electricity costs—another result of studies’ being funded or performed by the nuclear industry—is that industry typically claims fission-cost assumptions are either “privileged information” (and thus (p.93) kept secret by the industry) or purely counterfactual. Regarding secrecy, the 2009 MIT authors reported that, in Public Service Commission filings before nuclear-plant approval, nuclear-industry-cost figures (e.g., those provided by Georgia Power in 2009, when it considered building 2 nuclear plants) typically have “all detailed information about this cost figure” redacted,250 even though taxpayers subsidize nuclear-electricity costs and ratepayers cover the remainder. Likewise, a recent University of Greenwich nuclear-cost assessment noted that, for industry-performed or industry-funded studies,251 “many of the [cost] assumptions are not fully specified, being classified as commercially sensitive.”252 The same university researchers charge that one pro-nuclear UK government-commission study errs;253 it merely “reports the forecasts provided by [the 2 nuclear companies] British Energy and BNFL” and “uses BNFL's assumptions … [although] many of the assumptions, such as for construction cost, are categorized as commercially sensitive and not published …. On load-factor, the figures are also confidential, although the PIU states the assumed performance is significantly higher than 80 percent.”254

Yet, as already noted, actual nuclear-load factors are about 70 percent—not “significantly higher than 80 percent,” as cited by the preceding pro-nuclear government study. Another UK government report, done by University of Sussex researchers but not funded by the nuclear industry, says many industry-dominated studies—including the earlier, pro-nuclear UK commission report255—use nuclear-industry data but keep them confidential, to protect the industry.256 For instance, the UK Royal Academy of Engineering report,257 done by nuclear-industry contractor PB Power,258 states that “an allowance for [reactor] decommissioning cost is included in the capital cost, but it does not specify cost assumptions.”259 Both this study and the PB Power study are based on Finnish nuclear-industry data,260 but even MIT-based, nuclear-industry contractors admit “there is no detail on what is included in this [Finnish-cost] figure, and so it must be handled carefully.”261 Such figures could be used to underestimate nuclear costs, particularly because they “leave no audit trail.” As the same UK government report also emphasized, many studies “provide estimates of the overall generating cost of electricity using their own input assumptions,” but “because [many industry-funded] published studies do not show the precise method by which different input costs are translated into generating costs, and because the assumptions made will vary and be of differing methodological quality, it is not possible” to evaluate their “robustness.”262

Moreover, when secret nuclear-cost assumptions are revealed, typically they are counterfactual in a way that is favorable to the reactor industry. For instance, in the industry-funded Finnish nuclear-cost study,263 done at Lappeenranta University of Technology, the authors assume a counterfactual 5-percent (rather than the actual 15-percent) interest rate, as already mentioned. They likewise assume a counterfactual 91-percent (rather than historical-average 71-percent) load factor, and a counterfactual 60-year (rather than historical-average 22-year) lifetime, as already mentioned. Consequently, the study arrives at low nuclear-power costs. Nevertheless, (p.94) as already mentioned, the Royal Academy and PB Power studies accept these implausible Finnish assumptions and their confidential, contrary-to-fact data.264

What do the preceding 6 reasons for fission-cost underestimates suggest? Recall that these reasons include (1) the absence of fixed-price or turnkey reactor contracts, (2) the price discrepancy between vendor-quoted versus utility-quoted (to the Public Service Commission) nuclear costs, (3) the vendor and industry standard practice of quoting trimmed prices, (4) the lack of independent confirmation of most nuclear-cost data, mostly from industry, (5) the dearth of alternative analyses, because most fission-nuclear-cost studies are funded or performed by nuclear interests, and (6) the use of counterfactual assumptions and confidential assumptions by most studies. The preceding reasons, especially (4)–(6), suggest that nuclear-industry financial conflicts of interest (COI) also may help explain fission-cost underestimates.

Possible COI in Nuclear-Cost Studies

To investigate whether financial COI appear to be at least partly responsible for some fission-cost underestimates, one needs to understand what constitutes a COI. As defined in a classic 2009 US National Academy of Sciences (NAS) report, “conflicts of interest are defined as circumstances that create a risk that professional judgments or actions regarding a primary interest will be unduly influenced by a secondary interest. Primary interests include promoting and protecting the integrity of research,” the quality of scientific education, and the welfare of the public, whereas “secondary interests include not only financial interests … but also other interests, such as the pursuit of professional advancement.”265 What happens when one applies this COI definition to nuclear-cost studies that are performed or funded by fission interests? If the nuclear industry performs or funds economic studies whose results could affect its profits, this may “create a [COI, a] risk that professional judgments or actions regarding a primary interest,” scientific integrity, may be “unduly influenced by a secondary interest,” fission-industry profits.

To assess whether nuclear-related COIs may be involved in these 30 studies, consider 2 questions: (1) whether nuclear-cost studies funded by fission interests appear to trim fission-cost data, and (2) whether studies not funded by nuclear interests appear not to trim fission-cost data. If the answers to (1) and (2) are negative, then no COI may be involved. However, if the answers to (1) and (2) are positive, then COI may be occurring.

Although complete conclusions require a more extensive analysis than can be given here, at least this chapter can provide some preliminary answers regarding (1) and (2). That is, regarding question (1), this chapter showed earlier that, of the 30 recent nuclear-cost studies, at least 18 (60 percent) were performed or funded by pro-nuclear interests (also see later analyses),266 such as the US DOE. Moreover, (p.95) most of these studies trimmed the cost data in the 5 ways discussed in this chapter. Regarding question (2), this chapter also showed earlier that most nuclear-economics studies—that were neither funded nor performed by the nuclear industry—include most fission costs and do not trim them. Building on this already-presented information, one can divide the 30 post-2000 nuclear-economics studies into 4 groups, A–D below, based on who funds them and what cost data they include or exclude. (Some studies do not list their funders, and they are categorized as such.)

  • Group A (nuclear funders, pro-nuclear stance) consists of 18 analyses that (to varying degrees) are pro-nuclear and have been at least partly performed or funded by pro-nuclear interests (either industry or government).267

  • Group B (unknown funders, pro-nuclear stance) consists of 1 nuclear-cost analysis that is pro-nuclear and whose funders are unknown because the study does not mention them.268

  • Group C (nonprofit-NGO funders, anti-nuclear stance) consists of 4 nuclear-cost studies that are critical of high nuclear costs and whose funders are nonprofit NGOs.269

  • Group D (university funders, anti-nuclear stance) consists of 7 nuclear-cost studies that are critical of high nuclear costs and whose (at least partial) funders are universities,270 given that the lead authors of these studies are or were employed by universities.

To answer questions (1) and (2) above, consider characteristics of the nuclear-cost studies in each of these 4 groups.

Group A: 18 Studies with Nuclear Funders and a Pro-Nuclear Stance

The Group A studies—which are funded or performed by nuclear interests—represent most (18 of 30, or 60 percent) of the nuclear-economics studies.271 What is especially interesting, however, is that typical industry, government, and NGO groups—even anti-nuclear groups—appear to take these 18 pro-nuclear, largely industry-funded studies as the nuclear-economics paradigm. For instance, a classic UK government report lists only 9 fission-cost studies,272 all performed or funded by nuclear interests.273 The WNA, a global nuclear-industry-lobby group, lists only 7 fission-cost studies, all performed or funded by nuclear interests.274 Even environmental-group studies, like those of Greenpeace,275 list only 12 fission-cost studies—and all of them were performed or funded by nuclear interests.276 Thus, not only are most fission-cost studies done by pro-nuclear interests, but government, industry, and even environmentalists take these studies as dominant. Other interesting facts (p.96) about these 18 nuclear-funded studies are that none of them includes nuclear-cost data (a) from credit-rating agencies, (b) that cover taxpayer nuclear subsidies, or (c) that correct counterfactual, industry-supplied nuclear-economics data. To see how flaws (a)–(c) affect typical Group A conclusions, consider the 2009 MIT atomic-energy-cost analysis.277

On the positive side, as already mentioned, these MIT authors admit their work is funded by the nuclear industry;278 they note that, because of cost-data trimming, “none of the figures reported [by industry] for these [nuclear] plants represent actual costs.” The authors thus deserve credit for blowing the whistle on fission-industry failure to report actual costs, to reveal assumptions used to calculate costs, to explain counterfactual nuclear-cost calculations, and to explain the doubling of overnight nuclear-construction costs during 2003–2008.279

Despite these strengths, however, the industry-funded 2009 MIT authors ignore problem (a), high fission costs calculated by credit-rating companies. Instead they follow overly optimistic nuclear-industry assumptions that lead to erroneously low fission-energy-cost conclusions, as will be shown below. Regarding problem (b), the MIT authors fail to take account of massive taxpayer subsidies that artificially reduce nuclear costs. They say their analysis “does not include any of the benefits from the production-tax credits or loan guarantees … of 2005,” that is, a specific class of 2005 US taxpayer subsidies.280 Yet they inconsistently incorporate economic effects of other cost subsidies that artificially lower their calculated nuclear-electricity costs. If the late MIT physicist Henry Kendall is correct, pre-2004 US nuclear-power subsidies amount to about $20 billion annually—all of whose effects were ignored by the MIT authors.281 For instance, they ignore the billions of dollars in taxpayer subsidies needed annually for nuclear-waste storage, perhaps because these are not market costs. Instead the MIT authors assume that total costs of spent-fuel and waste disposal will be only “the statutory fee of 1 mil/kWhr currently charged” by government to the utility.282 Over the last 10 years, this amounts to only $5 billion total.283 Given the average 22-year lifetime of nuclear plants (see above), this statutory fee means the total collected from current US nuclear plants amounts to roughly $11 billion. Yet this amount, assumed by the 2009 MIT authors, is only a tiny portion of permanent waste-storage costs, most of which will be borne by taxpayers, per government agreement.284 In 1996, the US National Academy of Sciences studies said the total was not $11 billion, but at least $350 billion,285 and these costs now are $1 trillion.286 Thus the MIT authors include only between 1 percent (assuming $1 trillion is needed) and 3 percent (assuming $350 billion is needed) of the total monies needed for US nuclear-waste management—because they ignore taxpayer subsidies. More generally, the MIT authors assume fission electricity includes no taxpayer-subsidized costs,287 although “federal subsidies cover 60–90 percent of the generation cost for new nuclear plants288 and, as already documented, US federal nuclear subsidies have already amounted to about $150 billion. The MIT failure to take account of nuclear subsidies in nuclear costs is troublesome both because (p.97) utility executives say that “without [low-interest, taxpayer-subsidized] loan guarantees, we will not build nuclear plants,”289 and because the study failure suggests its industry funding may have influenced its fission-cost underestimates.

Regarding problem (c), because the 2009 MIT authors use mainly uncorrected nuclear-industry data, they make many counterfactual and inconsistent assumptions that lower nuclear-cost estimates. They assume, for instance, that the “total cost” of a nuclear plant includes neither financing nor interest charges. Yet they inconsistently admit these charges can double construction costs, and they counterfactually assume that nuclear-plant construction takes only 5 years, although earlier paragraphs showed that historical-average nuclear-plant-construction time is 10–23 years. Likewise, these MIT authors assume a nuclear-load or “capacity factor of 85%,” although earlier paragraphs showed that historical-average capacity factors are 71 percent. Likewise, the MIT authors assume annual inflation rates for future nuclear construction are 3 percent, although they admit that over the last 5 years, costs have increased 23 percent per year. They also assume that for fission, “the costs of capital [are] equal to those for coal.” Yet this assumption appears wholly unrealistic; market-interest rates for nuclear loans, as already mentioned, are 15 percent, whereas coal loans are only about 25 percent of that figure. Moreover, as already noted, interest can add 250 percent to overnight reactor costs, whereas the MIT authors admit that coal-plant interest charges add only 17–21 percent to overnight coal-plant costs.290

The MIT authors likewise claim to “update the cost of nuclear power,” although their nuclear costs are only half of those calculated by credit-rating firms like Standard and Poor's and Moody's.291 Moody's says that, since 2008, it has taken “a more negative view for those issuers seeking to build new nuclear power plants” because of “the substantial execution risks involved.”292 The discrepancy between MIT and credit-rating-company figures should have caused the 2009 MIT authors to question their industry-friendly assumptions, which led to artificially low nuclear-cost conclusions.

Perhaps because the 2003 MIT nuclear-cost analysis likewise was partly funded by the nuclear industry, it too fell into counterfactual and inconsistent assumptions about nuclear costs.293 It claimed to be funded by the “Alfred P. Sloan Foundation, … MIT's Office of the Provost, and [the MIT] Laboratory for Energy and the Environment.”294 However, “funding for this [laboratory that sponsored the] work comes from a variety of sources, including DOE, EPRI … [and] INEEL,”295 all pro-nuclear interests. Like the 2009 MIT studies, this 2003 research includes no nuclear-cost data (a) from credit-rating agencies, (b) that takes account of taxpayer-provided subsidies, or (c) that justifies using uncorrected, nuclear-industry-supplied cost data. Regarding (b), this 2003 MIT report criticizes nuclear subsidies, yet proposes additional “modest” taxpayer subsidies for nuclear power, but excludes these subsidies from its cost accounting of nuclear power. Likewise, regarding (c), the 2003 MIT analysis assumes that nuclear-plant construction takes only 5 years, although (p.98) earlier paragraphs showed historical-average nuclear-plant-construction time is 10–23 years. It also counterfactually assumes a nuclear-load factor of 85 percent, although earlier paragraphs showed that the historical-average load factors is 71 percent. Similarly, the 2003 MIT study assumes an 11.5-percent interest rate, although earlier paragraphs showed that 15 percent is the market rate. Finally, it assumes a 40-year lifetime for reactors, although earlier paragraphs revealed the historical-average lifetime is 22 years. Such implausible, inconsistent, and counterfactual nuclear-industry assumptions appear to have compromised the quality of MIT analyses.296

Group B: 1 Nuclear-Cost Study with Unknown Funders That Uses Uncorrected Industry Data

What about the other 12 nuclear-cost studies, those neither performed nor funded by nuclear interests? Group B studies consist of 1 nuclear-cost analysis,297 done by Oxera Consultants in the UK, whose funders are not revealed by the authors. Like the Group A analyses, those in Group B appear to include no nuclear-cost data (a) from credit-rating agencies, (b) that take account of taxpayer-nuclear subsidies, or (c) that correct inconsistent, counterfactual, fission-industry-supplied cost data.

Regarding (a), the Oxera study says nothing about how credit-rating-company claims contradict Oxera's conclusions about low nuclear costs. Regarding (b), Oxera's authors say only that “economic investment [in fission] is likely to require government support”; regarding (c), they admit their data and assumptions are “according to industry sources,” like Westinghouse, that supplied “related assumptions.”298 Unsurprisingly, the Oxera authors employ uncorrected nuclear-industry assumptions: for example, they use a nuclear-load factor of 95 percent,299 although the historical-average figure is 71 percent, as shown earlier. They also assume reactor-construction “cost inflation per year” is 2 percent,300 not 23 percent annually, documented for the last 5 years.301 Likewise, they assume reactor construction takes 4 years, not the historical-average 10–23 years, shown earlier. Finally, they assume nuclear-interest rates are 5 percent, not 15 percent, as shown earlier. Because their assumptions are inconsistent with historical data, are biased, and minimize costs, Oxera authors conclude that “the potential investment in nuclear new build is likely to bring positive returns.”302

Group C: 4 Studies with NGO Funders and Completely Corrected Industry-Cost Data

What about Group C fission-cost analyses? They are consistent with market data, like that from credit-rating firms, perhaps because their funders have no apparent (p.99) COI and are nonprofit NGOs.303 Group C includes a study from a nonprofit think tank, the Rocky Mountain Institute (RMI).304 RMI says, “Half our support comes from individual donors and foundation grants …. The other half comes from earned revenue—from consulting for corporations and governments.”305 A second study,306 from the nonprofit NGO Nuclear Information and Research Service (NIRS), says it relies on “contributions from citizens across the world to support our efforts.”307 A third study comes from another nonprofit think tank, the Institute for Energy and Environmental Research (IEER).308 The IEER website says its work is “supported by grants from foundations, concerned individuals and public-interest consulting contracts. Foundation funders include Colombe Foundation, Ford Foundation, Livingry Foundation, New-Land Foundation, Ploughshares Foundation, Stewart R. Mott Charitable Trust, Town Creek Foundation, and the Wallace Global Fund.”309 The fourth study in this group is funded by the Maryland Public Interest Research Group (PIRG).310 To see how nonprofit NGOs approach nuclear-power costs, note that all Group C studies (a) include nuclear-cost data from credit-rating agencies (b) take account of taxpayer-nuclear subsidies, and (c) correct nuclear-industry-supplied cost data. For instance, consider the 2008 study done by RMI.311

Regarding (a), credit-rating data, RMI acknowledged poor fission-credit ratings and argued that “the private capital market isn’t investing in new nuclear plants, and without financing, capitalist utilities aren’t buying.” Instead, poor nuclear-credit ratings mean “the few [reactor] purchases, nearly all in Asia, are all made by central planners with a draw on the public purse.” RMI also shows that, when one relies on “evidence-based studies,” like those done by Moody's and Standard and Poor's, the capital costs of fission-generated electricity are more than 3 times higher than the MIT-estimated nuclear costs, based on industry data.312

Regarding (b), taxpayer-subsidy data, RMI notes that even massive government subsidies have failed to make fission cost effective and that, once such subsidies are counted, authors “approach full costs.” RMI thus shows that the nuclear industry relies mainly on taxpayer subsidies, not private investors:

Taxpayers, who already bear most nuclear-accident risks[,] … for decades have subsidized existing nuclear plants by ~1–5¢/kWh. In 2005, desperate for orders, the politically-potent nuclear industry got those US subsidies raised to ~5–9¢/kWh for new plants, or ~60–90 percent of their entire projected power cost, including new taxpayer-funded insurance against legal or regulatory delays. Wall Street still demurred. In 2007, the industry won relaxed government rules that made its 100-percent-loan guarantees (for 80%-debtfinancing) even more valuable—worth, one utility's data revealed, about $13 billion for a single new plant, about equal to its entire capital cost. But rising costs had meanwhile made the $4 billion of new 2005 loan guarantees scarcely sufficient for a single reactor, so Congress raised taxpayers’ guarantees to $18.5 billion. Congress will soon be asked (p.100) for another $30+ billion in loan guarantees, or even for a blank check. Meanwhile, the nonpartisan Congressional Budget Office has concluded that defaults are likely.313

Regarding (c), correcting industry-based data and assumptions, RMI challenges nuclear-industry-funded conclusions—like those of the MIT and University of Chicago studies—that assume as much as 85–95 percent nuclear-load factors.314 The RMI authors say actual load factors are much lower because “even reliably operating nuclear plants must shut down” for roughly 8% of the time, “for refueling and maintenance, and unexpected failures” cause additional shutdowns for another “8% of the time.” Thus the most reliable single reactors have average-load factors of 84 percent, but not all reactors are reliable. Why not? Only “132 … 52 percent of the 253 [reactors] originally ordered” were completed. Not-completed reactors (numbering 121) are excluded by industry from alleged load-factor averages. Also excluded are another 28 US reactors (21 percent of those built), because they were “permanently and prematurely closed due to reliability or cost problems.” Industry likewise excludes yet another 27 percent of US reactors that “have completely failed for a year or more. Although surviving US nuclear plants,” 68 in number and one-fourth of total US reactor orders, have short-term load factors of about 90 percent, RMI says this 90-percent figure is misleadingly quoted in industry studies. Industry studies fail to reveal that the load-factor figure of 90 percent represents only a short-term load factor, not lifetime-average load factor, and only for 25 percent of US reactors, not all 253 US reactors that were ordered. As already explained, the lifetime-average US nuclear-load factor is 71 percent, not 90 percent. RMI thus explains that erroneously high load-factor figures arise from excluding low-load-factor reactors, which comprise 73 percent of US reactors.315

Likewise, RMI shows that when industry-funded, nuclear-cost studies exclude reactor-construction time and interest charges, this data-trimming illegitimately represents nuclear-capital costs as more than 50 percent less than they are. Yet in reality, as RMI notes, quoting the Economist, nuclear power “is now too costly to matter.”316

Group D: Seven 7 Studies That Are Partly University Funded and Use Fully Corrected Industry Data

Other studies likewise challenge the counterfactual assumptions in many industry nuclear-cost studies. Group D consists of 7 such analyses whose (at least partial) funders are universities, as lead authors are or were employed by universities.317 A chemist at University of Greenwich, in London, did one study.318 Another was done by “a Press Fellow at Wolfson College, Cambridge, during 2007/08.”319 MIT physics PhD Brice Smith, chair of the physics department at the State University of (p.101) New York (SUNY) at Cortland, did another analysis.320 Unlike industry studies, these 7 university-funded analyses include nuclear-cost data (a) from credit-rating agencies (b) that account for taxpayer-nuclear subsidies and (c) that use corrected, nuclear-industry-supplied data. Consider the study by physicist Smith from SUNY.

Regarding (a), Smith repeatedly cites credit-rating-firm data in his cost analyses. He quotes Moody's or Standard and Poor's to show that if utilities build nuclear plants, their credit is downgraded, which prevents further construction; that reactors always have cost overruns; that even subsidies will not make fission economical; and that credit-rating firms reject atomic energy.321 Given poor credit rating, Cambridge University professor Paul Brown likewise confirms that the high “cost of borrowing capital [for nuclear construction] in the open market” means that, “without government guarantees to hold down interest rates for new nuclear build,” no new reactors will be built.322

Regarding (b), uncounted nuclear subsidies, SUNY physicist Smith notes that, throughout its history, nuclear power has had to be “pushed along by large government subsidies.” Yet he warns: “Despite the magnitude of these [additional] proposed subsidies, they would still not be large enough to fully overcome the higher costs of nuclear power.”323 Brown concurs, noting that “without subsidy no new nuclear power station has ever been constructed”;324 he also warns that a “major public subsidy [of fission] is insurance against accident and the increasing bill for security.”325 In France, he says, “the public pays for the nuclear industry twice, through its electricity bills and again through its taxes. The true cost of nuclear energy in France is a state secret and has never been disclosed.”326 As already mentioned, taxpayer subsidies account for 60–90 percent of the cost of proposed new US reactors.327

Regarding (c), correcting nuclear-industry data and assumptions, SUNY physicist Smith criticizes the “highly optimistic assumptions” of industry. He criticizes the MIT 5-year reactor-construction assumption, because the US National Academy of Sciences says the figure is 12.2 years. Smith also shows that, given flawed interest and construction-time assumptions, industry has underestimated reactor-construction costs by 75 percent. In the overly optimistic MIT study (which assumed a nuclear-interest rate of 11.5 percent), Smith says MIT claimed interest charges represented only 20 percent of nuclear-capital costs; however, Smith argues that, if one raises this nuclear-interest rate to 12.5 percent (not even the “going rate” of 15 percent, as shown earlier), interest charges become 40 percent of nuclear-capital costs and make fission even less economical. After correcting many erroneous assumptions, Smith concludes: “It is unlikely that any significant improvements to the economics of nuclear power could be sustained.”328

What follows from the 4 classes of nuclear-cost studies (Groups A–D), given that analyses in each group share comparable funding and make similar assumptions? One conclusion is that of the 30 studies whose funders are known, a majority (18 in Group A) appear to be performed or funded by pro-nuclear interests. A (p.102) second conclusion is that, perhaps as a consequence, studies performed or funded by pro-nuclear interests (Group A) are most of those that trim fission-cost data. A third conclusion is that studies performed or funded by pro-nuclear interests are most of those that exclude evidence-based credit-rating data and taxpayer-subsidy data on high nuclear costs. A fourth conclusion is that most of the 11analyses (Groups C and D) that include credit-rating and taxpayer-subsidy data and use actual empirical data for cost assumptions are studies that are at least partly university-funded (the 7 in Group D). A fifth conclusion is that 4 of these 11 analyses—the Group C studies, which are funded by nonprofit NGOs—also include credit-rating and taxpayer-subsidy data and empirically based cost assumptions. Thus, a sixth conclusion is that most nuclear-cost studies (the 18 in Group A) appear to underestimate nuclear costs, perhaps because they are either performed or funded by nuclear interests. A seventh conclusion is that a minority of nuclear-cost studies (4 in Group C plus 7 in Group D) more accurately assess fission costs because they use empirical data, not industry assumptions, and because their authors have no obvious financial COI.

Moreover, although most studies that “get things right” are university-funded analyses (the 7 studies in Group D), not all university work is unbiased. Rather, 6 of the 18 nuclear-funded studies were done at universities (Chicago,329 Lappeenranta,330 MIT,331 Rice332), yet all exclude credit-rating and massive taxpayer-subsidy data; all use industry-projected and not historical-data-based assumptions; and all draw unrealistic conclusions about positive nuclear economics. Therefore, an eighth conclusion is that university-performed studies cannot protect against bias if industries with financial COI fund university research. However, a tenth conclusion is that, if funders of university-based studies have no obvious financial COI, these studies appear most reliable.

Why Nuclear-Cost Studies Fall into COI

Why do a majority of nuclear-cost studies apparently succumb to data-trimming and biased policy conclusions in favor of atomic energy? The answer may not be merely that they are performed or funded by those with apparent financial COI. Rather, part of the answer may be that COI are difficult to handle, so as to protect all legitimate interests.333 Another answer may be that although traditional professional codes of ethics typically require professionals to protect the public, these codes do not require full public disclosure of financial COI. For instance, the US General Services Administration, in its Federal Acquisition Regulations (FAR), does not require that consultants publicly disclose COI when they perform government-sponsored research. Because federal officials are not required to publicly disclose COIs, they are handled privately, within government confines. Provisions like FAR thus may help explain problematic assumptions in many nuclear-cost studies, For (p.103) instance, rather than requiring public disclosure of COI, instead FAR section 9.504(a)(2) says government officials should “avoid, neutralize, or mitigate significant potential conflicts [of interest]” or obtain waivers for COI when acquiring work from consultants or scientists. Rather than requiring full public disclosure of COI, FAR section 9.506(b) requires only that if a “government contracting officer” decides a particular action involves “a significant potential organizational conflict of interest, the contracting officer shall … submit for approval to the chief of the contracting office” a written analysis of the COI and how to mitigate or avoid it, so that the approving official can “approve, modify, or reject the recommendations in writing.”334

Likewise, US National Science Foundation (NSF) policies do not require public COI disclosure in government-sponsored research. Instead NSF requires only that an institution such as a university have a written COI policy, “manage” all COI prior to expending NSF funds, and ensure that grantees disclose to a “responsible representative of the institution [e.g., university] all significant financial interests of the investigator,” so the representative can certify to NSF that all COIs have been mitigated or avoided. Scientists need make no COI disclosure in their resulting research, to the public, or to NSF. NSF requires only that it be informed about any “unresolved conflict” of interest and that, otherwise, the institutional “representative” handle COIs.335

Similarly, the ethics code of the National Society of Professional Engineers (NSPE) does not require public disclosure of COIs. Instead section II.4 of the NSPE code requires merely that engineers “disclose all known or potential conflicts of interest to their employers or clients by promptly informing them of any business association, interest, or other circumstances which could influence or appear to influence their judgment or the quality of their services.”336 The Accreditation Board for Engineering and Technology (ABET) likewise requires (Guideline 1) no public disclosure of COI—only that analysts do work “consistent with … the safety, health, and welfare of the public and … disclose promptly factors that might endanger the public.” Yet this disclosure does not include the public. Instead ABET says (Procedures) only that “copies of the conflict of interest records will be provided” to officers selecting the analysts and their evaluators. ABET also requires (Guideline 4) that analysts “keep confidential all … evaluations unless by doing so they endanger the public.”337

How Energy-Cost Studies Should Be Done

However, if analysts knew that their possible COI could be disclosed publicly, they might avoid counterfactual or biased assumptions in their work. Given that scientists and engineers deserve privacy and yet that clients and affected parties (including the public) deserve protection from COI, how could the 30 nuclear-cost analyses (p.104) examined in this chapter have avoided both COI and their questionable assumptions? As previous paragraphs explained, there is no canonical study of nuclear costs. Nevertheless, one can specify at least 5 necessary conditions that might make energy-economics analyses more reliable. Although specifying necessary and sufficient conditions is impossible, it clearly is wrong to exclude relevant data, as many nuclear studies have done.

Following the earlier brief discussion of COI, one necessary condition for reliable nuclear-cost analyses is attempting to avoid financial COI. In sections 9.505-2(b)(1)-9.508(a)-(e), US FAR for instance, clearly say that analyses ought not be done by those who have personal or organizational COI. These regulations specify that “contracts for the evaluation of offers for products or services shall not be awarded to a contractor that will evaluate its own offers for products or services”; that the same scientists who prepare “a work statement” for some system “may not supply the system”; that scientists who might “provide systems engineering and technical direction” to a government agency for a reactor “should not be allowed to supply any power-plant components”; or that the same consultants who “prepare data-system specifications and … criteria … should be excluded from” supplying any “information technology” for that system.338 In other words, US government COI guidelines require avoiding situations in which those who assess some technology are the same scientists with financial interests in it.

Yet in the majority of analyses examined here (18 of the 30 nuclear-cost studies done since 2000)—including the majority of nuclear-cost studies cited by the WNA,339 by the UK government commission,340 and by the Greenpeace study—the same pro-nuclear groups who have financial interests in atomic energy assessed fission technology by radically trimming nuclear costs.341 Moreover, virtually all of these COI-tainted studies were used to help justify partial government “acquisition” or subsidy of nuclear power. Thus it is not surprising, as one UK government study warned, that fission-industry “cost estimates need to be treated with some caution as the vendors’ commercial incentive is clearly to estimate optimistically.”342

A second necessary condition, for cost studies that cannot avoid COI, is to mitigate their effects by publicizing them. As the earlier discussion suggested, this requires at least reporting, in the study itself, the full personal and institutional financial ties of those who perform or fund the analyses. Obviously, this disclosure is not sufficient to help protect against COI, but it is necessary.343 The public has a right to know about the quality of studies used to evaluate possible government subsidies, which are taxpayer dollars. Yet few studies funded by nuclear interests reveal full funding sources in their pages. For instance, in one MIT case,344 the study said funding partly came from an MIT lab, yet failed to disclose who funded the lab—the nuclear industry. If relevant financial ties (as with the lab) are reported by scientists and their institutions, they more likely will perform studies that withstand COI scrutiny. As a recent US National Academy of Sciences committee put it: “The disclosure of individual and institutional financial relationships is a critical but (p.105) limited first step in the process of identifying and responding to conflicts of interest.”345

A third necessary condition, especially for energy-cost studies, is employing lifecycle-cost analysis. With fission, such analyses should at least include costs associated with uranium mining, milling, conversion to uranium hexafluoride (UF6), enriching UF6, fuel fabrication, reactor construction, reactor operation, waste-fuel processing, fuel conditioning, interim waste storage, waste transport, permanent storage, reactor decommissioning, and uranium-mine reclamation. Although this chapter showed that typical economic analyses include only several stages of the nuclear-fuel cycle (e.g., reactor construction and operation),346 traditional codes of professional ethics, such as ABET Guidelines 5(c)-6(a), prohibit professionals from “omitting a material fact,” from “distorting or altering the facts,” and from “any conduct that deceives the public.”347 Yet this chapter shows that data-trimming nuclear costs omits material facts, distorts the facts, and may deceive the public. Lifecycle-cost analysis may help reduce these problems.

Doing lifecycle-cost analyses of energy systems also is important to ensure using limited resources efficiently, to avoid flawed and misleading accounting, to obtain the highest energy and environmental values for the lowest cost, to make rational energy-policy decisions, and to make energy-decision-making transparent and cost-effective.348 In the US, the federal government says lifecycle-cost analysis is both necessary and the best way to implement Executive Order 13101 (“Greening the Government”) and Executive Order 13423 (“Strengthening Federal Environmental, Energy, and Transportation Management).349 In part to satisfy executive orders, the US National Institute of Standards and Technology (NIST) has long had a software program, BEES (Building for Environmental and Economic Sustainability), that enables government purchasers and policymakers to use lifecycle-cost analyses to make economic, environmentally sustainable decisions about acquiring products and services—as mandated by the International Organization for Standardization (ISO) 14040 requirements for measuring environmental and economic performance.350 For all these reasons, the US National Academy of Sciences says “many federal acquisition policies … require life-cycle costing,” and it too has recommended life-cycle-costing in order to achieve “performance standards,” “preferable products,” and “sustainable development and value engineering.”351 Without lifecycle-cost analysis, public decisions about energy choices may ignore externalities that dominate costs. For instance, the International Energy Agency (IEA) points out that, because nuclear subsidies and negative externalities (e.g., taxpayer funding of most reactor-construction costs and permanent waste storage) are typically not included in nuclear-cost calculations, this causes nuclear-cost underestimates. Likewise, IEA says that because the positive externalities that lower costs (e.g., no fuel waste, no electricity-generation emissions, no catastrophic insurance needed) are typically not included in renewable-energy cost calculations, renewable-energy costs are overestimates. Yet, “unrewarded [beneficial] environmental characteristics” (p.106) of renewable-energy technologies, such as wind, are “the principal barrier to increasing the market share for renewable energy.”352

A fourth necessary condition for reliable energy-cost assessment is ensuring that energy-cost estimates are consistent with credit-rating and market data. When one uses nuclear-cost estimates based on credit-rating data, as many apparently reliable studies do,353 they are up to 350 percent higher than industry estimates (e.g., from the WNA),354 and nearly 200 percent higher than those from groups (like MIT), funded by the nuclear industry.355 Regarding the importance of credit-rating data, section 9.506(a) of US FAR note that if information is “necessary to identify and evaluate potential organizational conflicts of interest or to develop recommended actions [to mitigate COI],” government officials should seek information from both government groups, such as “audit activities and offices,” and from nongovernmental sources, such as “publications … credit-rating services, trade and financial journals.”356

A fifth necessary condition also comes from section 9.506(a) of the US FAR. It mandates that cost analyses should undergo reliable peer review by technical specialists and that federal officers who contract for studies, services, or products “should obtain the advice of counsel and the assistance of appropriate technical specialists in evaluating potential conflicts [of interest] and in developing any necessary … contract clauses.”357 Yet, there is no evidence that any nuclear-cost-study contracting groups, such as the US DOE—which funded the University of Chicago,358 Scully,359 and other nuclear-cost studies—had “technical specialists” evaluate them. Part of the reason for this failure may be that, at least since 1990, as chapter 2 showed, the US DOE has been repeatedly criticized by the US Congress, by government oversight agencies, and by the agency's own inspector general for its pro-nuclear biases and poor science; the COI criticisms of DOE have been so severe that the US Office of Technology Assessment and the Congress repeatedly have recommended either abolition of the US DOE or its regulation by an outside agency. Neither has occurred.360

The Climate Importance of Nuclear-Cost Assessments

Given CC problems, getting nuclear costs “right” is especially important because society needs economically efficient, low-carbon solutions, not inefficient guesses based on trimmed economic-cost data. However, if faulty fission analyses (like those just investigated) dominate CC discussions, they could jeopardize the groundswell of support for further increases in cheaper, cleaner technologies like wind.361 As a recent UK government commission notes, “Plans for new nuclear generation can therefore be expected to depress the market's appetite” for renewable-technology (p.107) “investment, relative to what it otherwise would be,” because there would be “a trade off between funding of nuclear power and funding of renewable.”362 Given this trade-off, it is especially important to get nuclear costs right.

Even without considering the data-trimming assessed in this chapter, nuclear costs roughly triple those of wind. As already noted, credit-rating firms say nuclear-fission electricity cost more than $0.15/kWhr in early 2008,363 and $0.21/kWhr in late 2009.364 Yet the US DOE says US full-fuel-cycle wind prices, on average over the last 7 years, are $0.048/kWh,365 and global wind potential is 35 times greater than current global electricity use.366 As chapter 6 points out in more detail, IRA says wind costs have been dropping 18 percent (and solar-photovoltaic costs, 35 percent) for every installed-capacity doubling, while atomic-energy costs have been increasing.367 By 2015 (a decade sooner than reactors, ordered today, could be operational), even the pro-nuclear US DOE says full-fuel-cycle centralized-solar-photovoltaic prices will be $0.05–0.10/kWh (depending on location), “competitive in markets nationwide,” and much cheaper than nuclear.368 Solar photovoltaics on only 7 percent of US land currently used for parking lots and buildings could provide all US electricity.369 Given such data, it is not surprising that atomic energy has been losing private markets. As already noted, by 2005, non-hydro renewable energy was annually growing 7 times faster than nuclear.370 In 2006, global added new wind capacity was 10 times greater than global added new nuclear capacity; in 2006, carbon-free renewable energy added 40 times more capacity globally than did nuclear; in 2007, carbon-free renewable energy gained more than $90 billion in global investment, and in 12 nations, renewable energy now provides between 17 and 50 percent of total electricity needs.371 As already mentioned, US government data, for the latest year available, show that wind has been responsible for 60 percent of annual new-electricity capacity, as measured by peak summer demand.372 Also as noted earlier, the classic 2004 Princeton University study shows that 6 of only 9 renewable technologies, already deployed at an industrial scale, could completely solve the climate problem by 2050; the US DOE is even more optimistic, arguing that available renewable technology can provide 99 percent of US electricity by 2020.373 Given these other, cheaper, lower-carbon (see chapter 2) energy options, nuclear fission is especially questionable.


Objecting to the preceding conclusions, one might ask whether nuclear analyses should use actual, historical-average data (as done here), or projected data (as industry does). However, as already noted, governments of many nations, including the UK and the US, have repeatedly said nuclear-industry cost estimates have consistently (since the beginning of the technology) been underestimates by several hundred percent. This suggests that, while it might be reasonable to use some (p.108) projected-cost data, the nuclear industry's own cost projections are not reliable. Besides, using projected cost data would require using hypothetical, untested data that have been provided by those who would profit from optimistic or low price estimates—a clear COI. For instance, the nuclear industry admits that reactors always have a minimum of 4–6 years of lower (50-percent) load factors, until “bugs” are removed.374 If so, more optimistic, projected nuclear-fission price estimates make little sense, at least for the first 5 years. Moreover, if industry's projected-cost numbers were correct, most banks would make nuclear loans, and reactor vendors would guarantee specific lifetimes and nuclear-load factors for clients. Yet neither will do so.375 Thus vendors, utilities, and the market itself all challenge projected counterfactual nuclear-fission-cost estimates. Also, as already mentioned, after 50-plus years of commercial implementation (as with fission), most technologies already have achieved all likely improvements.376 The nuclear industry itself, at least in part, also appears to admit that future fission costs will increase, not decrease, because, despite taxpayers’ massive atomic-energy subsidies,377 the nuclear-industry lobby says high costs will rule nuclear “out of consideration” in the future; by 2030, the WNA says fission will decrease from its current 16 percent of global electricity to 9 percent.378 For all these reasons, nuclear-industry proponents’ low-cost projections seem practically unrealistic, economically unsound, and unethical, given the COI associated with these projections.

Second, one might ask why so many countries use fission-generated electricity if reactors are uneconomical. Likewise, third, one might ask whether this analysis ignores the intermittency of renewable technologies, like wind and solar. Finally, fourth, one might ask whether it seems plausible for nuclear-cost research to have erred as badly as this chapter argues. These additional objections, and several others, will be discussed in detail in chapter 7.


What follows from the preceding analysis? It showed that, although many industry advocates claim nuclear reactors are an inexpensive way to generate low-carbon electricity and thus address CC, nearly all nuclear-fission-cost estimates (most of which are produced by the industry) are examples of grossly flawed science. Surveying all 30 recent nuclear-electricity analyses, this chapter shows that all industry-funded studies fall into conflicts of interest and illegitimately trim relevant cost data. Most studies exclude costs of full-liability insurance, underestimate reactor-construction-loan interest rates and construction times by using “overnight” costs, and overestimate reactor load factors and lifetimes. Yet if these 5 false assumptions are corrected, market costs (excluding subsidies, which are billions of dollars per year) of fission-generated electricity can be shown to be roughly 6 times more expensive than most studies claim—and far more expensive than available renewable (p.109) energy. If this chapter is correct, the proposed nuclear solution to CC is as troublesome as CC itself.

This chapter shows that nuclear power is a market failure; yet governments should be subsidizing technologies that are, or will be, market winners. As the chapter reveals, it also makes no sense to use atomic energy when its fuel cycle produces only 25 percent more energy than what must be input to it. Nuclear power thus requires subsidizing an old, expensive, dirty, nonsustainable technology of the past. Wind, solar PV, and other renewable-energy sources allow us to embrace newer, cheaper, cleaner, sustainable technologies of the future. The market and common sense are on the side of the future.


(1.) P. A. Rochon, J. H. Gurwitz, R. W. Simms, P. R. Fortin, D. T. Felson, K. L. Minaker, and T. C. Chalmers, “A Study of Manufacturer-Supported Trials of Nonsteroidal Anti-Inflammatory Drugs in the Treatment of Arthritis,” Archives of Internal Medicine 154, no. 2 (1994): 157–163; E. G. Campbell, K. S. Louis, and M. D. Blumenthal, “Looking a Gift Horse in the Mouth,” Journal of the American Medical Association 279, no. 13 (1998): 995–999; hereafter cited as Campbell et al., “Looking.”

(2.) For NAS report, see chapter 1. World Coal Institute, Coal (Richmond, UK: WCI, 2009), p. 2. Ben Evans, Stephanie Pistello, and Jeff Biggers, “Coal-Free Future Begins in Kentucky,” Louisville Courier-Journal (February 1, 2010), A7.

(p.267) (3.) J. Porritt, Chair of the UK Sustainable Development Commission, quoted in House of Commons Trade and Industry Committee, New Nuclear? Examining the Issues, vol. 1 (London: House of Commons, Fourth Report of Session 2005–2006).

(4.) UK Sustainable Development Commission (UK SDC), The Role of Nuclear Power in a Low-Carbon Economy (London: UK SDC, 2006), pp. 1, 4, 17; hereafter cited as UK SDC, The Role of Nuclear Power. US Government Accountability Office, Nuclear Material (Washington, DC: US Government Printing Office, 2008), p. 3; hereafter cited as US GAO, Nuclear Material.

(5.) S. Thomas, The Economics of Nuclear Power (Berlin: Heinrich Böll 2005), pp. 5–6, 8, 19, 20, 26, 30; hereafter cited as Thomas, Economics, 2005.

(6.) European Atomic Forum, Nuclear Energy (Brussels: Foratom, 2006); accessed August 10, 2009, at www.foratom.org/index.php?option=com_content&task=view&id=219&Itemid=938.

(7.) A. M. Herbst and G. W. Hopley, Nuclear Energy Now (Hoboken, NJ: John Wiley, 2007), pp. 12, 26; hereafter cited as Herbst and Hopley, Nuclear, 2007.

(8.) Gwyneth Cravens, Power to Save the World: The Truth about Nuclear Energy (New York: Knopf, 2008), p. 365; hereafter cited as Cravens, Power, 2008.

(9.) Herbst and Hopley, Nuclear, p. 12.

(10.) Stewart Brand, “Environmental Heresies,” Technology Review (May 2005); accessed February 18, 2010, at http://www.technologyreview.com/read_articles.aspx?id=14406&ch=biztech.

(11.) World Nuclear Association (WNA), The Economics of Nuclear Power (London: WNA, 2008); accessed September 9, 2009, at www.world-nuclear.org/info/inf02.html; hereafter cited as WNA, Economics.

(12.) McKinsey and Company, Reducing US Greenhouse Gas Emissions (New York: McKinsey and Company, 2007), p. 62; hereafter cited as McKinsey, Reducing.

(13.) F. L. Toth, “Prospects for Nuclear Power in the 21st Century,” International Journal of Global Energy Issues 30, nos. 1–4 (2008): 6; hereafter cited as Toth, “Prospects”; Herbst and Hopley, Nuclear, pp. 5, 124; Cravens, Power, pp. xiv–v.

(14.) “Nuclear Dawn,” Economist 384, no. 8545 (2007): 24–26; hereafter cited as “Nuclear Dawn”; Herbst and Hopley, Nuclear, p. 136.

(15.) Mike Morris, “The Next US Challenge,” Address by CEO, American Electric Power, to Detroit Economic Club, June 23, 2008. Vital Speeches of the Day 74, no. 9 (2008): 420–422; hereafter cited as Morris, “Next”; see J. Peter Scoblic, “Nuclear Spring,” New Republic 238, no. 7 (2008): 18; hereafter cited as Scoblic, “Nuclear.”

(16.) Herbst and Hopley, Nuclear, pp. 12, 4–7, 36, 43–44, 179, 12.

(17.) Kristin Shrader-Frechette, “Greenhouse-Gas Emissions and Nuclear Energy,” Modern Energy Reviews 1, no. 1 (2009): 58–62; Kristin Shrader-Frechette, “Data Trimming, Nuclear Emissions, and Climate Change,” Science and Engineering Ethics 15, no. 1 (2009): 19–23; hereafter cited as Shrader-Frechette, “Data Trimming”; Benjamin K. Sovacool, “Valuing the Greenhouse Gas Emissions from Nuclear Power: A Critical Survey,” Energy Policy 36, no. 2940 (2008): 53.

(18.) Thomas, Economics; WNA, Economics; World Nuclear Association (WNA), The New Economics of Nuclear Power (London: WNA, 2009); accessed October 12, 2009, at www.world-nuclear.org/reference/pdf/economics.pdf; hereafter cited as WNA, New Economics 2009; World Nuclear Association (WNA), The New Economics of Nuclear Power (London: WNA, 2005), p. 14; hereafter cited as WNA, New Economics 2005; Scully Capital Services Inc., The Business Case for New Nuclear Power Plants: A Report Prepared for the US DOE (Washington, DC: DOE, 2002); hereafter cited as Scully, Business Case; J. Du and J. E. Parsons, Update on the Cost of Nuclear Power (Cambridge: MIT Center for Energy and Environmental Policy Research, 2009); hereafter cited as Du and Parsons, Update; PB Power, Powering the Nation (Newcastle, UK: PB Power, 2006); hereafter cited as PB Power, Powering; Royal Academy of Engineering, The Costs of Generating Electricity (London: Royal Academy, 2004); hereafter cited as Royal Academy of Engineering, Costs; (p.268) UK Performance and Innovation Unit (PIU), The Economics of Nuclear Power (London: UK Cabinet Office, Performance and Innovation Unit, 2002); accessed January 19, 2009, at www.strategy.gov.uk/downloads/files/Pii.pdf; hereafter cited as UK PIU, Economics; Baker Institute for Public Policy, Japanese Energy Security and Changing Global Energy Markets (Houston: Rice University, 2000); hereafter cited as Baker Institute, Japanese Energy; D. Beutier, EPR [European Pressurized Water Reactor] Background and Its Role in Continental Europe (Paris: Areva, 2005); hereafter cited as Beutier, EPR; Canadian Energy Research Institute (CERI), Levelised Unit Electricity Cost Comparison of Alternative Technologies for Baseload Generation in Ontario, prepared for the Canadian Nuclear Association (Calgary: CERI, 2005); hereafter cited as CERI, Levelised; R. Tarjanne and K. Luostarinen, “Economics of Nuclear Power in Finland,” in American Nuclear Society (ed.), International Congress on Advanced Nuclear Power Plants (Hollywood, FL: American Nuclear Society [ANS], 2002); hereafter cited as Tarjanne and Luostarinen, “Economics”; University of Chicago, The Economic Future of Nuclear Power (Chicago: University of Chicago, 2004); hereafter cited as U Chicago, Economic Future 2004; International Energy Agency (IAE) / Nuclear Energy Agency (NEA), Projected Costs of Generating Electricity (Paris: IEA, 2005); hereafter cited as IAE/NEA, Projected Costs; Direction Generale de L’Energie et des Matieres Premieres (DGEMP), Reference Costs for Power Generation (Paris: Ministry of the Economy, Finance, and Industry, 2003); hereafter cited as DGEMP, Reference Costs; OXERA, Financing the Nuclear Option (Oxford, UK: OXERA, 2005); accessed January 16, 2009, at www.oxera.com/cmsDocuments/Agenda_June%2005/Financing%20the%20nuclear%20option.pdf; hereafter cited as OXERA, Financing; UK Department of Trade and Industry, Nuclear Power Generation Cost-Benefit Analysis (London: HM Government, 2006); hereafter cited as UK DTI, Nuclear Power Generation; S. D. Ansolabehere, J. Deutsch, M. Driscoll, P. E. Gray, J. P. Holdren, P. L. Joskow, R. K. Lester, E. J. Moniz, N. E. Todreas, and E. S. Beckjord, The Future of Nuclear Power (Cambridge: MIT Press, 2003); hereafter cited as Ansolabehere et al., Future; John Deutsch, Charles W. Forsberg, Andrew C. Kadak, Mujid S. Kazimi, Ernest J. Moniz, and John E. Parsons, Update of the MIT 2003 Future of Nuclear Power (Cambridge: MIT Energy Initiative, 2009), pp. 5–6; hereafter cited as Deutsch et al., Update; A. B. Lovins, I. Sheikh, and A. Markevich, Nuclear Power (Snowmass, CO: Rocky Mountain Institute, 2008); hereafter cited as Lovins, Sheikh, and Markevich, Nuclear Power; M. Mariotte, D. D’Arrigo, M. Olson, A. Binette, and D. Keesing, False Promises (Takoma Park, MD: Nuclear Information and Research Services, 2008); hereafter cited as Mariotte et al., False Promises; Arun Makhijani, Carbon-Free and Nuclear-Free (Takoma Park, MD: Institute for Energy and Environmental Research, 2007), pp. 144, 182–184, 188, 190, 192; hereafter cited as Makhijani, Carbon-Free; M. Diesendorf and P. Christoff, “Economics of Nuclear Power,” Energy Science (November 2008); accessed October 1, 2009, at http://www.energyscience.org.au; hereafter cited as Diesendorf and Christoff, “Economics”; Brice Smith, Insurmountable Risks (Takoma Park: IEER Press, 2006), pp. 44–51, 68, 70, 194, 204, 414; hereafter cited as Smith, Insurmountable; Stephen Thomas, Peter Bradford, Antony Frogatt, and David Milborrow, The Economics of Nuclear Power (Amsterdam: Greenpeace International, 2007), pp. 6, 31; hereafter cited as Thomas et al., Economics; J. W. Van Leeuwen, Nuclear Power (The Netherlands: Ceedata Consultancy, 2007); hereafter cited as Van Leeuwen, Nuclear; P. Brown, Voodoo Economics and the Doomed Nuclear Renaissance (Cambridge: Wolfson College, Cambridge University; London: Friends of the Earth, 2008); hereafter cited as Brown, Voodoo Economics; International Atomic Energy Agency (IAEA), Global Public Opinion on Nuclear Issues (Vienna: IAEA, 2005); hereafter cited as IAEA, Global Public Opinion; University of Sussex and NERA Economic Consulting, The Economics of Nuclear Power (London: UK Sustainable Development Commission, 2006); hereafter cited as Sussex-NERA, Economics; T. Madsen, J. Neumann, and E. Rusch, The High Cost of Nuclear Power (Baltimore: Maryland Public Interest Research Group, 2009); hereafter cited as Madsen et al., High Cost.

(p.269) (19.) WNA, New Economics 2005, p. 14.

(20.) Royal Academy of Engineering, Costs; CERI, Levelised; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; DGEMP, Reference Costs; Ansolabehere et al., Future.

(21.) Scully, Business Case; Royal Academy of Engineering, Costs; Beutier, EPR; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; OXERA, Financing; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future.

(22.) Sussex-NERA, Economics.

(23.) Scully, Business Case; PB Power, Powering; Royal Academy of Engineering, Costs; UK PIU, Economics; Baker Institute, Japanese Energy; CERI 2004; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; OXERA, Financing; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future.

(24.) Thomas et al., Economics, pp. 6, 31.

(25.) European Commission (EC), Solutions for Environment, Economy and Technology (Brussels: EC, January 2003), p. 132; hereafter cited as EC, Solutions; World Nuclear Association (WNA), Civil Liability for Nuclear Damages (London: WNA, 2008); accessed May 13, 2009, at www.world-nuclear.org/info/inf67.html; hereafter cited as WNA, Civil Liability; A. Heyes, “Determining the Price of Price-Anderson,” Regulation 25, nos. 4–8 (Winter 2002): 26–30; hereafter cited as Heyes, “Determining.”

(26.) E.g., Scully, Business Case; Heyes, “Determining”; D. Spurgeon, “Nuclear Energy: We Must Increase Its Role in Our Future,” address by US Assistant Secretary for Nuclear Energy at the Second Annual Global Nuclear-Renaissance Summit (Alexandria, Virginia), Vital Speeches of the Day 74, no. 9 (July 23, 2008): 422–425; hereafter cited as Spurgeon, “Nuclear Energy”; T. Slocum, “Nuclear's Power Play: Give Us Subsidies or Give Us Death,” Multinational Monitor 29, no. 2 (September–October 2008); accessed May 19, 2009, at www.multinationalmonitor.org/mm2008/092008/slocum.html; hereafter cited as Slocum, “Nuclear's Power Play”; American Nuclear Society, The Price-Anderson Act (La Grange Park, IL: ANS, November 2005); hereafter cited as ANS, Price-Anderson; G. S. Rothwell, Does the US Subsidize Nuclear Power Insurance? (Palo Alto: Stanford Institute for Economic Policy Research, January 2002); hereafter cited as Rothwell, Does the US Subsidize; Energy Information Administration (EIA), Federal Financial Interventions and Subsidies in Energy Markets (Washington, DC: Department of Energy, EIA, 1999); hereafter cited as EIA, Federal; B. P. Brownstein, The Price-Anderson Act, Cato Policy Analysis No. 36 (Washington, DC: Cato Institute, April 17, 1984), pp. 39–45; hereafter cited as Brownstein, Price-Anderson.

(27.) Makhijani, Carbon-Free; Smith, Insurmountable; Kristin Shrader-Frechette, Taking Action, Saving Lives (New York: Oxford University Press, 2007), pp. 42, 51, 95–97; hereafter cited as Shrader-Frechette, Taking Action.

(28.) Smith, Insurmountable, 194; Jim Riccio, Risky Business (Washington, DC: Greenpeace, 2001); accessed November 1, 2008, at www.greenpeace.org/raw/content/usa/press/reports/risky-business-the-probability.pdf.

(29.) E.g., Spurgeon, “Nuclear Energy,” p. 423; Slocum, “Nuclear's Power Play”; ANS, Price-Anderson; Heyes, “Determining,” p. 28; Scully, Business Case; Rothwell, Does the US Subsidize; EIA, Federal; Steve Cohn, Too Cheap to Meter (Albany: SUNY Press, 1997), p. 80, hereafter cited as Cohn, Too Cheap; Brownstein, Price-Anderson.

(30.) Smith, Insurmountable; Shrader-Frechette, Taking Action.

(31.) WNA, Civil Liability; J. Schwartz, “International Nuclear-Third-Party Liability Law,” in Organization for Economic Cooperation and Development (OECD), Nuclear Energy Agency (NEA), and International Atomic Energy Agency, International Nuclear Law in the Post-Chernobyl Period (Paris: NEA, 2006), pp. 37–72.

(32.) EIA, Federal.

(33.) Deutsch et al., Update ; Ansolabehere et al., Future.

(34.) WNA, Economics; WNA, New Economics 2005.

(p.270) (35.) International Energy Agency (IEA), World Energy Outlook 2005 (Paris: IEA, 2005); hereafter cited as IEA, World Energy Outlook 2005.

(36.) Baker Institute, Japanese Energy.

(37.) Tarjanne and Luostarinen, “Economics.”

(38.) UK PIU, Economics.

(39.) Royal Academy of Engineering, Costs; PB Power, Powering.

(40.) Scully, Business Case.

(41.) OXERA, Financing.

(42.) E.g., Cravens, Power; Herbst and Hopley, Nuclear.

(43.) E.g., Ronald H. Coase, “Adam Smith's View of Man,” Journal of Law and Economics 19, no. 529 (1976): 46; Viktor Vanberg, “Spontaneous Market Order and Social Rules,” Economics and Philosophy 2, no. 75 (1986): 100.

(44.) Rothwell, Does the US Subsidize.

(45.) ANS, Price-Anderson; American Enterprise Institute for Public Policy Research (AEI), Renewal of the Price-Anderson Act (Washington, DC: AEI, 1985), pp. 14–15.

(46.) Andrew Kimbrell,  Joseph Mendelson, and M. Briscoe, The Real Price of Gasoline (Washington, DC: International Center for Technology Assessment, 1998).

(47.) See David Pearce, “Environmentally Harmful Subsidies,” in Organization for Economic Cooperation and Development (OECD), Environmentally Harmful Subsidies (Paris: OECD, 2003), pp. 9–30.

(48.) See Brownstein, Price-Anderson.

(49.) US Code of Federal Regulations (CFR), Energy, title 10, part 140 (Washington, DC: US Government Printing Office, 2009).

(50.) E.g., Spurgeon, “Nuclear Energy,” p. 423; Slocum, “Nuclear's Power Play”; ANS, Price-Anderson; Heyes, “Determining,” p. 28; Rothwell, Does the US Subsidize; Scully, Business Case; EIA, Federal; Cohn, Too Cheap, p. 80; Brownstein, Price-Anderson. See Kristin Shrader-Frechette, Nuclear Power and Public Policy (Boston: Kluwer, 1983), pp. 10–11.

(51.) E.g., UK Department of Trade and Industry (UK DTI), Meeting the Energy Challenge (London: UK DTI, 2007), p. 191; accessed January 19, 2009, at commodities-now.com/content/research/includes/assets/UKWPenergy.pdf; hereafter cited as UK DTI, Meeting.

(52.) Quoted in Heyes, “Determining,” p. 28.

(53.) American Public Health Association (APHA): e.g., APHA, Reducing Occupational Exposure to Benzene in Workers and Their Offspring, Policy 2005–6 (Washington, DC: APHA, 2005); APHA, Preserving Right-to-Know Information and Encouraging Hazard Reduction to Reduce the Risk of Exposure to Toxic Substances, Policy 2002–5 (Washington, DC: APHA, 2002); APHA, Trade Agreements and Environmental and Occupational Health Policy, Policy 9404 (Washington, DC: APHA, 1994); APHA, Toxic Reduction as a Means of Pollution Prevention, Policy 9206 (Washington, DC: APHA, 1992); APHA, Strengthening Worker/Community Right-to-Know, Policy 8714 (Washington, DC: APHA, 1987).

(54.) EC, Solutions, p. 132.

(55.) WNA, Civil Liability.

(56.) Heyes, “Determining,” p. 29.

(57.) Brownstein, Price-Anderson.

(58.) Heyes, “Determining,” p. 29.

(59.) Brownstein, Price-Anderson.

(60.) Makhijani, Carbon-Free, p. 192; Smith, Insurmountable, p. 230; Matthew Wald, “Interest in Building Reactors, but Industry Is Still Cautious,” New York Times (April 2, 2005); Nature Editorial, “Nuclear Test: Japan's Response to an Earthquake Highlights Both the Promise and the Pitfalls of Nuclear Power at a Critical Time for Its Future,” Nature 448, no. 7152 (2007): 387.

(61.) Heyes, “Determining”; UK DTI, Meeting.

(62.) Brownstein, Price-Anderson.

(63.) Morris, “Next.”

(p.271) (64.) WNA, New Economics 2005, pp. 5, 20; 14, 17–18. Most nuclear economists’ failure to include the cost of capital (interest for the reactor-construction loan for a period of 8–30 years), and their instead using a 0 interest rate, or “overnight costs,” obviously has the effect of artificially lowering nuclear costs. The need for accurate nuclear-cost information, and for using actual interest rates, should not be confused with arguments about discounting far-distant costs, as related to climate change. In the case of climate-related discounting, the interest rates are largely arbitrary, not real costs, as is the case of those for nuclear construction.

(65.) Du and Parsons, Update, p. 11.

(66.) Ibid., pp. 5–6.

(67.) Ibid., esp. pp. 11, 5–6.

(68.) Morris, “Next”; see Scoblic, “Nuclear,” p. 18.

(69.) “Nuclear Dawn.”

(70.) Thomas, Economics, p. 19.

(71.) International Energy Agency (IEA), World Energy Outlook (Paris: IEA, 2006); Antony Froggatt, Financing Disaster—How the G8 Fund the Global Proliferation of Nuclear Technology (London: The Royal Institute of International Affairs, June 2001); accessed January 14, 2009, at www.eca-watch.org/problems/fora/documents/G8_eca-nuclear-2001.pdf.

(72.) Slocum, “Nuclear's Power Play.”

(73.) Makhijani, Carbon-Free, pp. 144, 190; see Smith, Insurmountable, pp. 51, 97; John Kennedy, Andreas Zsiga, Laurie Conheady, and Paul Lund, “Credit Aspects of North American and European Nuclear Power,” Standard & Poor's (January 9, 2006), hereafter cited as Kennedy et al., “Credit.”

(74.) Moody's Corporate Finance, New Nuclear Generating Capacity (New York: Moody's, 2008); hereafter cited as Moody's, New Nuclear.

(75.) Herbst and Hopley, Nuclear, p. 103.

(76.) Ansolabehere et al., Future.

(77.) Cravens, Power, p. 321.

(78.) Thomas, Economics, p. 19.

(79.) Deutsch et al., Update ; Ansolabehere et al., Future.

(80.) U Chicago, Economic Future.

(81.) WNA, New Economics 2005, p. 17.

(82.) Herbst and Hopley, Nuclear, pp. 4–7, 36, 179.

(83.) R. S. Berry, “Tomorrow's Nuclear Power Will Be Different Than Yesterday's Nuclear Power,” Bulletin of the Atomic Scientists, Roundtable on Nuclear Power and Climate Change (2007); accessed November 9, 2008 at www.thebulletin.org/roundtable/nuclear-power-climate-change/; hereafter cited as Berry, “Tomorrow's Nuclear Power.”

(84.) WNA, Economics.

(85.) Baker Institute, Japanese Energy.

(86.) Scully, Business Case.

(87.) OXERA, Financing.

(88.) Herbst and Hopley, Nuclear, p. 46.

(89.) WNA, New Economics 2005, p. 21.

(90.) WNA, Economics.

(91.) E.g., Herbst and Hopley, Nuclear, p. 171; UK SDC, The Role of Nuclear Power; Thomas, Economics, p. 16.

(92.) IEA, World Energy Outlook 2005; see Thomas et al., Economics, p. 35.

(93.) UK PIU, Economics.

(94.) WNA, Economics; Matthew L. Wald, “After 35-Year Lull, Nuclear Power May Be in Early Stages of a Revival,” New York Times (October 24, 2008), sec. B3.

(95.) William Sweet, Kicking the Carbon Habit (New York: Columbia University Press, 2006), p. 188, hereafter cited as Sweet, Kicking; Thomas, Economics.

(96.) Alexandro Clerici and the European Regional Study Group, The Future Role of Nuclear Energy in Europe (London: World Energy Council, 2006).

(p.272) (97.) Olaf Bayer and C. Grimshaw, Broken Promises (Oxford, UK: Corporate Watch, 2007), p. 5.

(98.) Thomas, Economics, p. 8.

(99.) Scully, Business Case.

(100.) Thomas, Economics, p. 26.

(101.) Slocum, “Nuclear's Power Play.”

(102.) See Smith, Insurmountable, p. 414.

(103.) Thomas, Economics.

(104.) Herbst and Hopley, Nuclear, p. 102.

(105.) Gordon MacKerron, “The Economics of Nuclear Power—Has Government Got It Right?,” Sussex Energy Group Policybriefing 1, nos. 1–4 (2007): 2; hereafter cited as MacKerron, “Economics”; WNA, New Economics 2005, p. 19.

(106.) WNA, New Economics 2005, p. 14.

(107.) Thomas, Economics, p. 30.

(108.) Herbst and Hopley, Nuclear, p. 183.

(109.) Smith, Insurmountable, p. 47.

(110.) McKinsey, Reducing, p. 27.

(111.) Peter Bunyard, “Ecologist: Taking the Wind out of Nuclear Power,” Pacific Ecologist 11, no. 51 (2006): 7; hereafter cited as Bunyard, “Ecologist.”

(112.) Peter Bradford, “Follow the Money,” Bulletin of the Atomic Scientists, Roundtable on Nuclear Power and Climate Change (2007); accessed November 9, 2009, at www.thebulletin.org/roundtable/nuclear-power-climate-change/.

(113.) Herbst and Hopley, Nuclear, pp. 43–44.

(114.) House of Commons Energy Select Committee, Fourth Report: The Costs of Nuclear Power (London: UK House of Commons, 1990).

(115.) International Atomic Energy Agency (IAEA), PRIS Database (Vienna: IAEA, 2007); accessed January 15, 2009, at www.iaea.org/programmes/a2/index.html; hereafter cited as IAEA, PRIS Database.

(116.) Peter Stoett, “Toward Renewed Legitimacy: Nuclear Power, Global Warming and Security,” Global Environmental Politics 3, no. 1 (2003): 108.

(117.) International Energy Agency (IEA), Nuclear Power in the OECD (Paris: IEA, Organization for Economic Cooperation and Development, 2001).

(118.) Bunyard, “Ecologist.”

(119.) Makhijani, Carbon-Free, p. 30; American Wind Energy Association (AWEA), Congress Extends Wind Energy Production Tax Credit for an Additional Year (Washington DC: AWEA, 2006); accessed March 5, 2010, at www.awea.org/newsroom/releases/congress_extends_PTC_121106.html.

(120.) Smith, Insurmountable, p. 50; Kennedy et al., “Credit.”

(121.) Mariotte et al., False Promises.

(122.) WNA, New Economics 2005, p. 17.

(123.) Baker Institute, Japanese Energy.

(124.) Tarjanne and Luostarinen, “Economics.”

(125.) UK PIU, Economics.

(126.) OXERA, Financing.

(127.) Deutsch et al., Update; Ansolabehere et al., Future; see Herbst and Hopley, Nuclear, p. 149; Brice Smith, “Insurmountable Risks,” Science for Democratic Action 36, nos. 1–2 (2006): 7.

(128.) Scully, Business Case.

(129.) Royal Academy of Engineering, Costs; PB Power, Powering.

(130.) IEA, World Energy Outlook 2005.

(131.) U Chicago, Economic Future.

(132.) Herbst and Hopley, Nuclear, p. 149.

(133.) Cravens, Power, p. xiv.

(134.) E.g., Herbst and Hopley, Nuclear, p. 171; UK SDC, The Role of Nuclear Power; Thomas, Economics, p. 16.

(p.273) (135.) Thomas, Economics, p. 5.

(136.) IAEA, PRIS Database; Thomas, Economics, p. 6.

(137.) Cravens, Power, p. 286.

(138.) S. Pacala and R. Socolow, “Stabilization Wedges: Solving the Climate Problem for the Next 50 Years with Current Technologies,” Science 305, no. 5686 (13 August 2004): 968–972; hereafter cited as Pacala and Socolow, “Stabilization Wedges.” The 9 renewable options (only 6 needed to solve climate problems) are efficient vehicles, reduced vehicle use, efficient buildings wind power for coal power, PV power for coal power, wind for gasoline in fuel-cell cars, biomass fuel for fossil fuel, reduced deforestation, and conservation tillage. National Renewable Energy Laboratory, Near Term Practical and Ultimate Technical Potential for Renewable Resources (Golden, CO: US DOE, 2006); hereafter cited as NREL, Near Term.

(139.) Thomas et al., Economics, p. 6.

(140.) Sweet, Kicking, p. 154.

(141.) Energy Information Administration, Electric Power Annual (Washington, DC: Department of Energy, 2009); accessed July 1, 2009, at www.eia.doe.gov/cneaf/electricity/epa/epa_sum.html; hereafter cited as EIA, Electric Power.

(142.) WNA, New Economics 2005, pp. 21, 10.

(143.) Du and Parsons, Update, p. 18.

(144.) Sweet, Kicking, p. 182.

(145.) Mariotte et al., False Promises.

(146.) Herbst and Hopley, Nuclear, p. 176.

(147.) Ibid., p. 174.

(148.) Christian Parenti, “What Nuclear Renaissance?,” Nation 286, no. 18 (May 12, 2008): 14; Smith, Insurmountable, p. 193; General Accounting Office (GAO), Nuclear Regulation: NRC Needs to More Aggressively and Comprehensively Resolve Issues Related to the Davis-Besse Nuclear Power Plant's Shutdown (Washington, DC: GAO, 2004), p. 20.

(149.) Sweet, Kicking, p. 186.

(150.) Bunyard, “Ecologist”; Herbst and Hopley, Nuclear, p. 172.

(151.) US GAO, Nuclear and Worker Safety (Washington, DC: US GPO, 2007), p. 4; hereafter cited as US GAO, NWS.

(152.) V. M. Fthenakis and H. C. Kim, “Greenhouse-Gas Emissions from Solar-Electric and Nuclear Power: A Life-Cycle Study,” Energy Policy 35, no. 4 (2007): 2549–2557.

(153.) IAEA, PRIS Database.

(154.) See Thomas et al., Economics, p. 31.

(155.) Thomas, Economics.

(156.) IAEA, PRIS Database; see Toth, “Prospects,” p. 6; Herbst and Hopley, Nuclear, p. 176.

(157.) IAEA, PRIS Database; see Thomas, Economics, pp. 5–6.

(158.) WNA, New Economics 2005, p. 17.

(159.) Westinghouse, Westinghouse AP1000 Advanced Passive Reactor (Pittsburgh: Westinghouse, 2003); accessed January 19, 2009, at www.nuclearinfo.net/twiki/pub/Nuclearpower/WebHomeCostOfNuclearPower/AP1000Reactor.pdf.

(160.) Nucleonics Week Editors, “Merrill Lynch Global Power and Gas Leaders Conference,” Nucleonics Week 47, no. 40 (2006): 4.

(161.) World Nuclear Association, Nuclear Power in France (London: WNA, 2008); accessed January 21, 2009, at www.world-nuclear.org/info/inf40.html; hereafter cited as WNA, Nuclear Power in France.

(162.) Thomas, Economics, pp. 8, 20.

(163.) Tarjanne and Luostarinen, “Economics.”

(164.) OXERA, Financing.

(165.) WNA, New Economics 2005, p. 10.

(166.) Scully, Business Case.

(167.) Royal Academy of Engineering, Costs; PB Power, Powering.

(168.) CERI 2004.

(p.274) (169.) WNA, Economics.

(170.) UK PIU, Economics.

(171.) Deutsch et al., Update; Ansolabehere et al., Future.

(172.) U Chicago, Economic Future.

(173.) IEA, World Energy Outlook 2005.

(174.) UK DTI, Nuclear Power Generation.

(175.) See Mariotte et al., False Promises.

(176.) Deutsch et al., Update; Ansolabehere et al., Future.

(177.) Smith, Insurmountable, p. 68.

(178.) Moody's, New Nuclear; see Arun Makhijani, “Nuclear Power Costs: High and Higher,” Science for Democratic Action 15, no. 2 (2008): 2–3; hereafter cited as Makhijani, “Nuclear Power Costs”; Keystone Center, Nuclear Power Joint Fact-Finding (Keystone, CO: Keystone Center, 2007); hereafter cited as Keystone, Nuclear Power.

(179.) Cravens, Power, p. 253; Smith, Insurmountable, p. 70; J. Aabakken, Power Technologies Energy Data Book (Golden, CO: US DOE, National Renewable Energies Lab, 2005), pp. 37–39; hereafter cited as Aabakken, Power Technologies.

(180.) Smith, Insurmountable.

(181.) WNA, New Economics 2005, p. 7; Herbst and Hopley, Nuclear, p. 169.

(182.) Mycle Schneider, “2008 World Nuclear Industry Status Report,” Bulletin of the Atomic Scientists (September 16, 2008); accessed May 20, 2009, at www.thebulletin.org/web-edition/reports/2008-world-nuclear-industry-status-report/.

(183.) Herbst and Hopley, Nuclear.

(184.) WNA, Economics, pp. 21, 10.

(185.) E.g., IAEA, PRIS Database.

(186.) Tarjanne and Luostarinen, “Economics.”

(187.) WNA, Economics.

(188.) Scully, Business Case.

(189.) Royal Academy of Engineering, Costs; PB Power, Powering.

(190.) U Chicago, Economic Future.

(191.) IEA, World Energy Outlook 2005.

(192.) OXERA, Financing.

(193.) UK DTI, Nuclear Power Generation.

(194.) Deutsch et al., Update; Ansolabehere et al., Future.

(195.) David Fleming, The Lean Guide to Nuclear Energy (London: University of London, 2007), p. 7; hereafter cited as Fleming, Lean Guide.

(196.) Deutsch et al., Update; Ansolabehere et al., Future.

(197.) Herbst and Hopley, Nuclear, pp. 43–44.

(198.) Moody's, New Nuclear.

(199.) Lovins, Sheikh, and Markevich, Nuclear Power.

(200.) Deutsch et al., Update.

(201.) Daryl Chubin and Edward Hackett, Peerless Science (Albany: SUNY Press, 1990), p. 132.

(202.) E.g., John C. Reinard, Communication Research Statistics (London: SAGE, 2006), p. 54; hereafter cited as Reinard, Communication.

(203.) David Sheskin, Handbook of Parametric and Nonparametric Statistical Procedures (Boca Raton, FL: CRC Press, 2004), p. 403; hereafter cited as Sheskin, Handbook.

(204.) E.g., M. Wu and Y. Zuo, “Trimmed and Winsorized Means Based on a Scaled Deviation,” Journal of Statistical Planning and Inference 139, no. 2 (2009): 350–365; M. Wu and Y. Zuo, “Trimmed and Winsorized Standard Deviations Based on a Scaled Deviation,” Journal of Nonparametric Statistics 20, no. 4 (2008): 319–335; J. A. Cuesta-Albertos, C. Matran, and A. Mayo-Iscar, “Trimming and Likelihood,” Annals of Statistics 36, no. 5 (2008): 2284–2318; L. A. Garcia-Escudero, A. Gordaliza, A. Mayo-Iscar, and C. Matran, “A General Trimming Approach to Robust Clustering,” Annals of Statistics 36, no. 3 (2008): 1324–1345; (p.275) J. Karvanen, “Estimation of Quantile Mixtures via L-moments and Trimmed L-moments,” Computational Statistics and Data Analysis 51, no. 2 (2006): 947–959; Z. Leonowicz, J. Karvanen, and S. L. Shiskin, “Trimmed Estimators for Robust Averaging,” Journal of Neuroscience Methods 142, no. 1 (2005): 17–26.

(205.) Norman H. Anderson, Empirical Direction in Design and Analysis (Mahwah, NJ: Erlbaum, 2001), p. 355; hereafter cited as Anderson, Empirical.

(206.) E.g., Sheskin, Handbook, p. 403.

(207.) Anderson, Empirical, pp. 354–356.

(208.) E.g., Gerald Miller and M. Whicker, Handbook of Research Methods in Public Administration (Boca Raton, FL: CRC Press, 1998).

(209.) E.g., Claes Fornell, Sunil Mithas, and F. Morgeson, “The Statistical Significance of Portfolio Returns,” International Journal of Research in Marketing 26, no. 2 (2009): 162–163.

(210.) Sheskin, Handbook, p. 403; see R. R. Wilcox, Applying Contemporary Statistical Techniques (San Diego: Academic Press, 2003); hereafter cited as Wilcox, Applying.

(211.) Reinard, Communication; Sheskin, Handbook; S. P. Kothari, Jowell Sabino, and Tzachi Zach, “Implications of Survival and Data-Trimming for Tests of Market Efficiency,” Journal of Accounting and Economics 39, no. 1 (2004): 129–161; Wilcox, Applying; Anderson, Empirical, pp. 352–356.

(212.) Anderson, Empirical, p. 356.

(213.) Ibid., p. 353; Reinard, Communication, p. 54.

(214.) Reinard, Communication, p. 54; Anderson, Empirical, pp. 353–356; Sheskin, Handbook, pp. 403–404.

(215.) Anderson, Empirical, pp. 353–356; Sheskin, Handbook, p. 403.

(216.) Sheskin, Handbook, p. 403.

(217.) Ansolabehere et al., Future; Deutsch et al., Update.

(218.) Ansolabehere et al., Future, p. 82.

(219.) IEA, World Energy Outlook 2005; UK PIU, Economics; see Thomas et al., Economics, p. 35.

(220.) Deutsch et al., Update; Ansolabehere et al., Future.

(221.) Deutsch et al., Update; Ansolabehere et al., Future.

(222.) Ansolabehere et al., Future.

(223.) E.g., Thomas, Economics; Lovins, Sheikh, and Markevich, Nuclear Power; Mariotte et al., False Promises; Makhijani, Carbon-Free; Diesendorf and Christoff, “Economics”; Thomas et al., Economics; Van Leeuwen, Nuclear; Sussex-NERA, Economics.

(224.) Diesendorf and Christoff, “Economics,” p. 2.

(226.) Thomas, Economics; Lovins, Sheikh, and Markevich, Nuclear Power; Mariotte et al., False Promises; Makhijani, Carbon-Free; Diesendorf and Christoff, “Economics”; Thomas et al., Economics; Van Leeuwen, Nuclear; Sussex-NERA, Economics.

(227.) Campbell et al., “Looking”; Diesendorf and Christoff, “Economics.”

(228.) Makhijani, Carbon-Free, p. 169. As already noted, most nuclear economists’ failure to include the cost of capital (interest for the reactor-construction loan for a period of 8–30 years), and their instead using a 0 interest rate, or “overnight costs,” obviously has the effect of artificially lowering nuclear costs. The need for accurate nuclear-cost information, and for using actual interest rates, should not be confused with arguments about discounting far-distant costs, as related to climate change. In the case of climate-related discounting, the interest rates are largely arbitrary, not real costs, as is the case of those for nuclear construction.

(229.) Van Leeuwen, Nuclear, pp. 32, 37, 32, 43, 54.

(230.) Fleming, Lean Guide, pp. 16–19.

(231.) Morris, “Next,” p. 136.

(232.) MacKerron, “Economics.”

(233.) US Congressional Budget Office (US CBO), Nuclear Power's Role in Generating Electricity (Washington, DC: US CBO, 2008), p. 17.

(p.276) (234.) Mariotte et al., False Promises; Lovins, Sheikh, and Markevich, Nuclear Power; Thomas et al., Economics; Nucleonics Week Editors, “Olkiluoto-3 Costs Weigh on Areva 2008 Profits,” Nucleonics Week (December 25, 2008): 9.

(235.) Julio Godoy, ENERGY: Nuclear Does Not Make Economic Sense Say Studies (Berlin: International Press Service, February 21, 2010); accessed March 18, 2010, at www.ipsnews.net/news.asp?idnews=50308&utm_source=twitterfeed&utm_medium=twitter).

(236.) Teresa Weinmeister, New Plants Update (Bethesda, MD: AREVA NP Inc., September 2009); accessed March 18, 2010, at www.areva-np.com/common/liblocal/docs/presentation_statutEPR/New_Plants_Update_2009.pdf).

(237.) Du and Parsons, Update, pp. 5–6, 14.

(238.) E.g., Scully, Business Case; U Chicago, Economic Future; IAE/NEA, Projected Costs.

(239.) Tarjanne and Luostarinen, “Economics.”

(240.) UK SDC, The Role of Nuclear Power, p. 19.

(241.) UK DTI, Nuclear Power Generation.

(242.) UK SDC, The Role of Nuclear Power, pp. 11–12.

(243.) Diesendorf and Christoff, “Economics,” pp. 2–3.

(244.) UK SDC, The Role of Nuclear Power, p. 4.

(245.) WNA, New Economics 2009; WNA, Civil Liability; WNA, New Economics 2005; Scully, Business Case; Du and Parsons, Update; PB Power, Powering; Royal Academy of Engineering, Costs; UK Department for Business Enterprise and Regulatory Reform Meeting the Energy Challenge: A White Paper on Nuclear Power (London: Her Majesty's Stationery Office, 2008; hereafter cited as UK 2008;{ Baker Institute, Japanese Energy; Beutier, EPR; CERI 2004; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; DGEMP, Reference Costs; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future; Deutsch et al., Update; IAEA, Global Public Opinion.

(246.) WNA, New Economics 2009; WNA, Civil Liability; WNA, New Economics 2005; PB Power, Powering; Royal Academy of Engineering, Costs; Beutier, EPR; CERI 2004; IAE/NEA, Projected Costs; IAEA, Global Public Opinion.

(247.) Du and Parsons, Update; Baker Institute, Japanese Energy; Tarjanne and Luostarinen, “Economics”; Ansolabehere et al., Future; Deutsch et al., Update.

(248.) Scully, Business Case; UK PIU, Economics; U Chicago, Economic Future; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future; Deutsch et al., Update.

(249.) Du and Parsons, Update, pp. 5, 10.

(250.) Ibid., p. 14.

(251.) E.g., Du and Parsons, Update; Tarjanne and Luostarinen, “Economics.”

(252.) Thomas, Economics, p. 25.

(253.) UK PIU, Economics.

(254.) Thomas, Economics, p. 26.

(255.) UK PIU, Economics.

(256.) UK SDC, The Role of Nuclear Power, p. 110.

(257.) Royal Academy of Engineering, Costs.

(258.) PB Power, Powering.

(259.) Thomas, Economics, p. 27.

(260.) Tarjanne and Luostarinen, “Economics.”

(261.) Du and Parsons, Update, p. 9.

(262.) UK SDC, The Role of Nuclear Power, pp. 11–12, 19.

(263.) Tarjanne and Luostarinen, “Economics.”

(264.) Royal Academy of Engineering, Costs; PB Power, Powering.

(265.) Bernard Lo, M. J. Field, and the Institute of Medicine, Conflict of Interest (Washington, DC: National Academy Press, 2009), p. 6; hereafter cited as Lo et al., Conflict of Interest.

(266.) WNA, New Economics 2009; WNA, Civil Liability; WNA, New Economics 2005; Scully, Business Case; Du and Parsons, Update; PB Power, Powering; Royal Academy of Engineering, Costs; UK 2008; Baker Institute, Japanese Energy; Beutier, EPR; CERI 2004; Tarjanne and (p.277) Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; DGEMP, Reference Costs; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future; Deutsch et al., Update; IAEA, Global Public Opinion.

(267.) WNA, New Economics 2009; WNA, Civil Liability; WNA, New Economics 2005; Scully, Business Case; Du and Parsons, Update; PB Power, Powering; Royal Academy of Engineering, Costs; UK 2008; Baker Institute, Japanese Energy; Beutier, EPR; CERI 2004; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; DGEMP, Reference Costs; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future; Deutsch et al., Update; IAEA, Global Public Opinion.

(268.) OXERA, Financing.

(269.) Lovins, Sheikh, and Markevich, Nuclear Power; Mariotte et al., False Promises; Makhijani, Carbon-Free; Madsen et al., High Cost.

(270.) Thomas, Economics; Diesendorf and Christoff, “Economics”; Smith, Insurmountable; Thomas et al., Economics; Van Leeuwen, Nuclear; Brown, Voodoo Economics; Sussex-NERA, Economics.

(271.) WNA, New Economics 2009; WNA, Civil Liability; WNA, New Economics 2005; Scully, Business Case; Du and Parsons, Update; PB Power, Powering; Royal Academy of Engineering, Costs; UK 2008; Baker Institute, Japanese Energy; Beutier, EPR; CERI 2004; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; DGEMP, Reference Costs; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future; Deutsch et al., Update; IAEA, Global Public Opinion.

(272.) Scully, Business Case; Royal Academy of Engineering, Costs; Beutier, EPR; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; OXERA, Financing; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future.

(273.) UK SDC, The Role of Nuclear Power.

(274.) Royal Academy of Engineering, Costs; CERI 2004; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; DGEMP, Reference Costs; Ansolabehere et al., Future.

(275.) Thomas et al., Economics.

(276.) Scully, Business Case; PB Power, Powering; Royal Academy of Engineering, Costs; UK PIU, Economics; Baker Institute, Japanese Energy; CERI 2004; Tarjanne and Luostarinen, “Economics”; U Chicago, Economic Future; IAE/NEA, Projected Costs; OXERA, Financing; UK DTI, Nuclear Power Generation; Ansolabehere et al., Future.

(277.) Du and Parsons, Update.

(278.) Ibid., p. v.

(279.) Ibid., pp. 11, 10, 14, 15, iii.

(280.) Ibid., p. 9.

(281.) Kristin Shrader-Frechette, Environmental Justice (New York: Oxford University Press, 2002), p. 131; hereafter cited as Shrader-Frechette, Environmental Justice.

(282.) Du and Parsons, Update, p. 21.

(283.) Shrader-Frechette, Environmental Justice.

(284.) Ibid.; US Congress, Worker Safety at DOE Nuclear Facilities, US House of Representatives (Washington, DC: US Government Printing Office, 1999); hereafter cited as US Congress, Worker Safety.

(285.) US National Research Council, Building an Effective Environmental Management Science Program (Washington, DC: National Academy Press, 1996).

(286.) Shrader-Frechette, Environmental Justice; US Congress, Worker Safety.

(287.) Du and Parsons, Update.

(288.) Lovins, Sheikh, and Markevich, Nuclear Power.

(289.) Madsen et al., High Cost, p. 17.

(290.) Du and Parsons, Update, pp. 4–6, 18, 16, 19, 22.

(291.) Ibid., p. iii; Mariotte et al., False Promises; Moody's, New Nuclear.

(p.278) (292.) Moody's Corporate Finance, New Nuclear Generation: Ratings Pressure Increasing, Report 117883 (New York: Moody's, June 2009).

(293.) Ansolabehere et al., Future.

(294.) Ibid., p. vii.

(295.) MIT Laboratory for Energy and the Environment (LEE), MIT Reports to the President 2001–2002 (Cambridge: MIT LEE, 2003); accessed October 12, 2009, at http://web.mit.edu/annualreports/pres02/03.03.html.

(296.) Ansolabehere et al., Future, pp. 43, 8, 82.

(297.) OXERA, Financing.

(298.) Ibid., pp. 4, 2, 5.

(299.) Ibid., p. 3.

(301.) Du and Parsons, Update.

(302.) OXERA, Financing, pp. 2–4.

(303.) Lovins, Sheikh, and Markevich, Nuclear Power; Mariotte et al., False Promises; Makhijani, Carbon-Free; Madsen et al., High Cost.

(304.) Lovins, Sheikh, and Markevich, Nuclear Power.

(305.) Rocky Mountain Institute (RMI), Helping Businesses/Organizations (Snowmass, CO: RMI, 2009); accessed October 12, 2009, at www.rmi.org.

(306.) Mariotte et al., False Promises.

(307.) Nuclear Information and Resource Service (NIRS), About NIRS (Takoma Park, MD: NIRS, 2009); accessed October 12, 2009, at www.nirs.orgn/about/nirs.htm.

(308.) Makhijani, Carbon-Free.

(309.) Institute for Energy and Environmental Research (IEER), Funders (Takoma Park, MD: IEER, 2009); accessed October 12, 2009, at www.ieer.org/ieerinfo.html ww.ieer.org/ieerinfo.html.

(310.) Madsen et al., High Cost.

(311.) Lovins, Sheikh, and Markevich, Nuclear Power.

(312.) Ibid., pp. 1–2.

(313.) Ibid., pp. 1, 11.

(314.) Du and Parsons, Update; Ansolabehere et al., Future, U Chicago, Economic Future.

(315.) Lovins, Sheikh, and Markevich, Nuclear Power, p. 10.

(316.) Ibid., pp. 1–2.

(317.) Thomas, Economics; Diesendorf and Christoff, “Economics”; Smith, Insurmountable; Thomas et al., Economics; Van Leeuwen, Nuclear; Brown, Voodoo Economics; Sussex-NERA, Economics.

(318.) Thomas, Economics.

(319.) Brown, Voodoo Economics, p. 2.

(320.) Smith, Insurmountable.

(321.) Ibid., pp. 44–45, 50–51, 97.

(322.) OXERA, Financing, p. 32.

(323.) Smith, Insurmountable, pp. 7, 49.

(324.) Brown, Voodoo Economics, p. 24.

(325.) Ibid., pp. 24, 31.

(326.) Ibid., p. 32.

(327.) Lovins, Sheikh, and Markevich, Nuclear Power.

(328.) Smith, Insurmountable, pp. 53, 40–41, 46–47.

(329.) U Chicago, Economic Future.

(330.) Tarjanne and Luostarinen, “Economics.”

(331.) Du and Parsons, Update; Ansolabehere et al., Future; Deutsch et al., Update.

(332.) Baker Institute, Japanese Energy.

(333.) S. J. Bird and R. E. Spier, “The Complexity of Competing and Conflicting Interests,” Science and Engineering Ethics 11, no. 4 (2005): 515–517; J. K. Roberts, F. W. Beard, R. J. Haefeli, (p.279) P. E. James, R. W. Jarvis, P. E. Polk, and P. E. Thompson, National Society of Professional Engineers’ Board of Ethical Review Cases (Denton: Murdough Center for Engineering Professionalism at Texas Tech University, 2001); accessed October 1, 2009, at 88/case85-86.htm; hereafter cited as Roberts et al., NSPE Board; S. J. Bird and R. E. Spier, “A Conflict of Interest Disclosure Policy,” Science and Engineering Ethics 14, no. 149 (2008): 152; hereafter cited as Bird and Spier, “Conflict of Interest.”

(334.) Federal Acquisition Institute (FAI), Federal Acquisition Regulations, Subpart 9.5—Organizational and Consultant Conflicts of Interest (Washington, DC: General Services Administration, 2005); hereafter cited as FAI, Federal Acquisition Regulations.

(335.) Lee Stokes, “Key Issues in Conflict of Interest,” Journal of Research Administration 33, nos. 2–3 (2002): 19–25.

(336.) Roberts et al., NSPE Board.

(337.) Accreditation Board for Engineering and Technology (ABET), Standards of Conduct (Baltimore: ABET, 2009); accessed October 1, 2009, at http://www/abet.org/code.shtml; hereafter cited as ABET, Standards of Conduct.

(338.) FAI, Federal Acquisition Regulations.

(339.) WNA, New Economics 2005.

(340.) Sussex-NERA, Economics.

(341.) Thomas et al., Economics.

(342.) UK SDC, The Role of Nuclear Power, p. 13.

(343.) Bird and Spier, “Conflict of Interest.”

(344.) Ansolabehere et al., Future.

(345.) Lo et al., Conflict of Interest, pp. 1–2.

(346.) Shrader-Frechette, “Data Trimming.”

(347.) ABET, Standards of Conduct.

(348.) National Research Council (NRC), Sustainable Federal Facilities: A Guide to Integrating Value Engineering, Life-Cycle Costing, and Sustainable Development (Washington, DC: National Academy Press, 2001), hereafter cited as NRC, Sustainable Federal Facilities.

(349.) Ibid. B. C. Lippiatt, BEES 4.0: Building for Environmental and Economic Sustainability, NSTIR 6916 (Washington, DC: National Institute of Standards and Technology (NIST), Office of Applied Economics, 2007); hereafter cited as Lippiatt, BEES 4.0.

(350.) Lippiatt, BEES 4.0.

(351.) NRC, Sustainable Federal Facilities, p. 4.

(352.) Mariotte et al., False Promises.

(353.) E.g., ibid.; Makhijani, Carbon-Free.

(354.) WNA, New Economics 2005.

(355.) Du and Parsons, Update.

(356.) FAI, Federal Acquisition Regulations.

(358.) U Chicago, Economic Future.

(359.) Scully, Business Case.

(360.) Shrader-Frechette, Environmental Justice; US Congress, Worker Safety.

(361.) Mariotte et al., False Promises; Makhijani, Carbon-Free; Diesendorf and Christoff, “Economics.”

(362.) UK SDC, The Role of Nuclear Power, p. 45.

(363.) Moody's, New Nuclear.

(364.) Lovins, Sheikh, and Markevich, Nuclear Power.

(365.) Smith, Insurmountable; Cravens, Power; Aabakken, Power Technologies.

(366.) Lovins, Sheikh, and Markevich, Nuclear Power.

(367.) J. Robert Oppenheimer, “International Control of Atomic Energy,” in Morton Grodzins and Eugene Rabinowitch (eds.), The Atomic Age (New York: Basic Books, 1963), p. 55.

(368.) US Department of Energy (DOE), DOE Selects 13 Solar Energy Projects (Washington, DC: US DOE, March 8, 2007); accessed June 9, 2009, at www.energy.gov.news.4855.htm.

(p.280) (369.) Madsen et al., High Cost.

(370.) Thomas et al., Economics.

(371.) Lovins, Sheikh, and Markevich, Nuclear Power.

(372.) EIA, Electric Power.

(373.) Pacala and Socolow, “Stabilization Wedges”; NREL, Near Term.

(374.) IAEA, PRIS Database.

(375.) Thomas et al., Economics.

(376.) Thomas, Economics.

(377.) Smith, Insurmountable; Slocum, “Nuclear's Power Play.”

(378.) WNA, New Economics 2005.