The Nuclear Mirage

The Nuclear Mirage

In 1979, I stood inside the containment building of a nuclear power plant. The shiny steel reactor vessel at the Diablo Canyon Power Plant outside San Luis Obispo had yet to be loaded with nuclear fuel. I was touring the facility as a staff member of the California Assembly Office of Research, so I could educate Assembly members about nuclear power. It was a hot topic. The worst commercial nuclear accident in U.S. history had happened earlier that year at Three Mile Island. I could see all around me the complexity and risk involved with generating electricity from nuclear fission.

Now, when I speak publicly about climate change, I am often asked if nuclear power is a solution to the crisis. Nuclear power plants currently provide about half of our carbon-free electricity. Since an essential goal for addressing climate change is to generate all of our electricity without emitting carbon, many—including some prominent climate change activists—argue for expanding nuclear power (e.g., see this video from the Years Project).

But even after 40 years of additional research, development and operating experience, nuclear power still has major problems related to cost, safety, weak regulation and inadequate oversight. These problems, routinely glossed over by advocates for the expansion of nuclear power, will continue to limit its role. Indeed, despite our growing need for carbon-free electricity, the contribution of nuclear power to our electricity supply continues to shrink.i Its resurgence will never be more than a tantalizing mirage.

Nuclear Sticker Shock. The technical complexity involved in protecting against the enormous risk inherent in nuclear power makes it very expensive, and the industry presently depends on privatizing profits while socializing the risks. The availability of private capital for nuclear power plant construction is nonexistent without either loan guarantees from the government or assumption of all construction risks by customers.ii A recent MIT study determined that “repeated failures of construction management practices” is the primary reason for the extraordinary cost overruns common in nuclear power plant construction. Such overruns led to the cancellation of the V.C. Summer nuclear project in South Carolina after $9 billion had been spent by ratepayers, and to the bankruptcy of Westinghouse Electric Company in 2017.

In addition to socializing the risk of capital investments (by spreading risks to taxpayers and ratepayers), nuclear power plants only operate because the liability of individual plant operators for major accidents is currently limited to about $450 million through the Price-Anderson Act (with the total liability of the nuclear industry for an accident limited to about $13 billion). Subsidies and government expenditures also reduce the cost of uranium mining, fuel enrichment and cooling water for reactor operations.

Moreover, it is the federal government that is responsible for the disposal of the highly radioactive spent fuel generated by nuclear power plants, some of which remains toxic for hundreds of thousands of years. Despite trying for decades, we have been unable to site and construct a repository for this waste, and so most spent fuel sits in pools of water at the plants where it was generated.iii Because the government failed to meet a statutory deadline in 1998 for taking the waste off the hands of nuclear power plant operators, the government is now paying nuclear licensees for storage. By 2020, these payments will have reached $20 billion, and will continue indefinitely unless a waste storage facility is constructed. While the cost of the facility is theoretically to be covered by ratepayers through the Nuclear Waste Fund, no one knows what the ultimate cost of waste storage will be. No other industry requires such a government commitment to clean up the waste from its operations.

For these and other reasons, electricity generated by nuclear power plants is relatively expensive. A recent estimate of the cost of various sources of electricity, calculated annually by the financial firm Lazard, shows that for utility-scale electricity generation new nuclear plants are becoming less economically attractive as the cost of other technologies (especially wind and solar) continues to fall. Indeed, power from new reactors will cost between two and three times as much as meeting the same needs through renewables, increased storage and grid enhancements (energy efficiency continues to be the cheapest “source” of all). A careful analysis of nuclear subsidies concluded in 2011 that the cost of nuclear power plants (when all subsidies are included) is actually larger than the value of the electricity they have produced.

An Ever-Present Danger. My professional career started when the Three Mile Island Nuclear Plant melted down in March of 1979. A plant of similar design was operating outside of Sacramento, where I was an intern with a California State Assemblyman. People were suddenly very worried about nuclear safety. The theatrical release 12 days before the accident of The China Syndrome, which centers on a nuclear power plant accident, had only heightened public concern.

These worries were not misplaced. The threat to public safety inherent in the use of nuclear power is significant and often addressed only superficially by nuclear power advocates. Nuclear reactors are exceedingly complex systems that have the potential to release lethal radiation to the local environment. The accident at the Chernobyl Nuclear Power station in the Ukraine resulted in a 1,000 square mile exclusion zone where human access is restricted for safety.iv

Reactors have been engineered to make the chances of accidents very small, and nuclear power advocates are quick to point out that a major radiation release has not occurred in the U.S. (although most of these advocates are unaware of the accident at the Santa Susana Field Laboratory, which released hundreds of times the radiation of Three Mile Island in July 1959 and caused an estimated 300-1,800 cancer deaths). While it is certainly good news that currently operating facilities in the U.S. have not accidentally released a large amount of radiation, there have been close calls. For example, in 2002, a severe incident occurred at the Davis-Besse nuclear power plant in Ohio, when a football-sized hole in the top of the reactor vessel was discovered by accident during routine maintenance. The reactor had to be shut down for two years for repairs.

The fact that a major accident could  happen is why taxpayers have had to insure the industry’s liability through the Price-Anderson Act (the accident at Three Mile Island turned a billion-dollar asset into a two-billion dollar clean-up job in about two hours). The final cost of the Fukushima-Daiichi accident, which was not as severe as it would have been had the wind not blown out to sea for most of the days following the accident, could be more than $500 billion. It is unknown who would pay such costs in the U.S., although with the legal cap on the industry’s liability the economic pain would likely be spread among ratepayers, property owners and taxpayers.

Even if engineered safety systems work as designed, a major accident could also be caused by a terrorist attack. People tend to think that nuclear power plants are “hard targets” because of the renowned strength of the containment buildings that house the reactors, a strength that was palpable when I stood inside the containment building at Diablo Canyon. But terrorists would not attack these reinforced structures, instead focusing on the control buildings or the waste storage pools outside the containment structure.v While the Nuclear Regulatory Commission (NRC) requires nuclear plant operators to prove that they can defend against terrorists through planning and “force on force” test attacks, operator performance is still And cyberattack now represents a new terror threat to nuclear power plants. The Wolf Creek Nuclear Operating Corporation in Kansas was attacked in 2017.

Although nuclear power plants will not explode like nuclear bombs, the nuclear fuel cycle and nuclear weapons are linked, and so expanding commercial nuclear power enhances the opportunities for the proliferation of nuclear weapons. We can see this safety problem unfolding as Iran begins enriching its fuel-grade uranium to bomb-grade uranium (as did Pakistan in the 1970s). The spent fuel from uranium-based commercial reactors contain plutonium, which is generated during operations. Extracting plutonium from spent reactor fuel is how India, Israel and North Korea became nuclear powers.

Underwhelming Oversight. Climate change impacts like more floods, intense storms, heat and sea level rise are only making matters worse for nuclear power plant safety.vii One would think that power plant operators would be required to address these new threats, but the NRC ruled in January that nuclear plants wouldn’t have to update facilities to deal with new, higher levels of expected flooding. Former NRC Chairman Gregory Jaczko notes this is akin to documenting “…the safety of seat belts and then not making automakers put them in a car.”

Such deficiencies in regulating the nuclear power industry are not new. In April 1979, I attended a hearing of the California Assembly Subcommittee on Energy that was focused on nuclear safety. In the wake of the Three Mile Island accident, people were seeking assurance regarding the safety of these enormously powerful facilities. A key witness was Dr. Robert Pollard, a Navy-trained nuclear engineer and legendary early critic of the nuclear power industry. Two months before the meltdown, Pollard and his colleagues at the Union of Concerned Scientists had demanded that the reactor at Three Mile Island and 15 others be shut down due to concerns about safety.

At the hearing, Pollard described in blistering detail how the nuclear industry was cutting corners on safety and pressuring the NRC to weaken safety and security standards. The Union of Concerned Scientists documents that this practice continues today.viii In response to a dangerous fire in 1975 at the Browns Ferry plant in Alabama, the NRC issued regulations to reduce fire hazards, but as of 2011 more than three dozen plants still had not complied with these regulations. This is one of many such examples of weak oversight.ix

Bill Gates to the Rescue? One hope for addressing the serious problems plaguing nuclear power has been to develop a “new generation” of nuclear power plants. This hope was buoyed by Bill Gates’ founding of TerraPower, when he said “the problems with today’s reactors, such as the risk of accidents, can be solved through innovation.” Despite this burst of confidence, Terrapower’s “innovative” traveling wave reactor continues to have problems. These problems have prompted Gates to question if nuclear power can ever be a climate solution.

There are also many promoters of “new reactors” based on thorium fuel, although thorium-based reactors have been studied for decades.x These reactors continue to have problems that have frustrated efforts to commercialize this technology. Another new approach to commercial nuclear electricity, small modular reactors, is also facing many development challenges despite extensive support from governments around the world.

Even with these problems, a strong climate-based argument can be made for extending the operating licenses of existing reactors that have demonstrated safe operation, competent management and operating costs that are actually cheaper than replacing them with other low-carbon sources.xi A recent analysis from the Energy Information Administration suggests even a modest carbon price of $15/ton limits nuclear plant retirement. However, nuclear plants that want to participate in such programs must be willing to open their books to document their economic performance.

While carbon pricing can help maintain nuclear power’s contribution to our carbon-free electricity supply in the near-term, nuclear power has not been able to overcome the safety and cost issues that have prevented its expansion despite decades of effort. The promise of nuclear power remains nothing more than a mirage, and my doubts are echoed by others (including the past Chairman of the NRC). Grave threats to public health and fundamental economic problems mean that nuclear power is unlikely to be a major component of our future carbon-free electricity grid. We are much better served investing in renewable electricity sources to promote safe and cost-effective decarbonization.

I wish to acknowledge the valuable review of a draft of this article by Peter Bradford, former member of the Nuclear Regulatory Commission. All statements and conclusions in the article are my own, not those of Mr. Bradford.



i More than one-third of our approximately 100 operating nuclear power units have closed since 2012 or are projected to close in the coming years, and there is only one two-unit facility currently under construction. If the electricity produced by these reactors is replaced by burning natural gas, which is likely in the short term with unplanned closings in today’s power markets, then our emission reduction goals will be even harder to reach.

ii In March of this year, the federal government authorized an additional $3.7 billion loan guarantee to the Vogtle nuclear power plant in Georgia, the only plant currently under construction in the U.S., bringing the total to $12 billion.

iii According to the General Accounting Office, 80,000 metric tons of spent fuel from commercial reactors is stockpiled at 80 interim storage sites in 35 states. The amount of this waste is expected to increase to about 140,000 metric tons over the next several decades.

iv The accident initially contaminated an area of about 60,000 square miles in Belarus, Russia and Ukraine.

v Most waste storage pools contain five times as much radioactive fuel as the reactors themselves.

vi For example, despite being located near water, plants do not have to consider attacks from the water as a design threat, and over 5% of “force on force” tests are failed (the NRC has resisted a recommendation to change the evaluation system from pass/fail, and so it cannot distinguish barely adequate performance from outstanding performance).

vii According to Bloomberg “…54 of the nuclear plants operating in the U.S. weren’t designed to handle the flood risk they face. 53 weren’t built to withstand their current risk from intense precipitation; 25 didn’t account for current flood projections from streams and rivers; 19 weren’t designed for their expected maximum storm surge. Nineteen face three or more threats that they weren’t designed to handle.” As surface water (either freshwater or ocean water) used for cooling gets warmer, nuclear plants have had to reduce their power output.

viii In this manner, the nuclear industry operates just as many other U.S. business alliances, but the stakes of weak protections are extraordinary because of the inherent danger of nuclear technology. Unfortunately, the NRC frequently operates like a captive bureaucracy, too often promoting the interests of the industry over public safety. At the time of Pollard’s testimony, nobody knew what would happen at the reactor near Philadelphia, except that some radiation had been discharged to the atmosphere. In a dramatic moment, the Navy-trained Pollard responded to a question by saying “Sir, my activities on nuclear submarines are classified information that I cannot discuss. But Sir, don’t think this kind of thing doesn’t happen on nuclear submarines.”

ix The NRC has also reduced mock terrorist attacks from three scenarios to one, required less rigorous safety upgrades than its own task force recommended in response to the Fukushima-Daiichi accident and has refused to require the transfer of spent nuclear fuel from overcrowded pools to safer dry storage casks. The NRC has just proposed to reduce the frequency of safety inspections at nuclear power plants.

x Thorium itself cannot sustain a nuclear chain reaction but must be combined with uranium or plutonium. While some claim thorium-based reactors are a panacea for the industry, the Department of Energy has concluded that the choice of using thorium- or uranium-based fuel is just a preference (the Oak Ridge National Laboratory ran a test reactor using thorium fuel in the 1960s).


xi Economic operation of nuclear plants will be enhanced by structuring electricity markets so that these operators get paid for the public benefit of producing carbon-free power (e.g., pricing carbon emissions or providing zero emission credits as done by the State of New York). An MIT study concludes it is possible to operate nuclear reactors in a more flexible way that can follow daily electrical load, which would lower power operating costs and increase revenues (such operating protocols are more common in Europe). However, the size of nuclear power plants and their relative inflexibility means that in any given area nuclear power can crowd out smaller and more flexible sources of electricity (that can also be cheaper). This is a key reason supporting Pacific Gas & Electric’s decision to close the Diablo Canyon nuclear power plant in 2025.