Skip to content

Nuclear Moral Hazard

Recent sales of U.S. nuclear plants raise questions about safety, liability, and economic incentives.

(Today’s post is co-authored with Catherine Hausman, an assistant professor at the University of Michigan.)


Last week, a company called Holtec International received federal approval to acquire New Jersey’s Oyster Creek nuclear power plant. Except Oyster Creek shut down last fall and will never produce another kilowatt-hour. Holtec is buying it to tear it down. It will be responsible for decommissioning the site, including managing spent fuel and other radioactive waste.

The Oyster Creek sale is one of several such recent transactions in which a U.S. nuclear plant is being sold by a large publicly-traded company to a smaller privately-owned company specializing in decommissioning. Though the other sales are pending approval by the Nuclear Regulatory Commission, it is not too soon to consider the potential implications for safety and the environment.

The U.S. nuclear industry has a strong record of safe operations. Historically, most owners of U.S. nuclear plants have been large companies, with significant “skin in the game” in terms of profitability and reputation if something were to go wrong. Do smaller companies have the same incentives?

The Indian Point Nuclear Power Plant in Buchanan, New York. Source: here.


Big Sellers and Small Buyers

One of the biggest sellers in this market is Entergy. Entergy is selling the Vermont Yankee plant (closed 2016), the Pilgrim nuclear power plant in Massachusetts (closing 2019), the Indian Point plant in New York (closing 2020-2021), and the Palisades nuclear power plant in Michigan (closing 2022). Both Entergy and Exelon (the company selling Oyster Creek) are large, publicly-traded companies with a total market capitalization over $19  and $48 billion, respectively.

The buyers, in contrast, are smaller, privately held companies. Holtec International specializes in decommissioning, spent fuel storage, and other aspects of the nuclear power generation. For Pilgrim, the buyer is NorthStar Group Services, another smaller, privately-owned company specializing in facility service contracting.

These buyers may be better suited to manage the decommissioning process. Such specialized companies could bring expertise focused on safe and thorough decommissioning, rather than managing a large portfolio of projects and sites operating at different points in the electricity supply chain.

A nuclear power plant half deconstructed during decommissioning. Source: Here.


Old Regime – Incentive for Safety

But, these nuclear asset transfers could have a big downside. In the past, safe operations meant that nuclear plants could make more money. We suspect that these economic incentives partly account for the good safety record of U.S. nuclear plants. Nuclear power plant owners worked hard to avoid problems because plant shutdowns are costly for plant owners.

Take Entergy, for example. At its peak, Entergy owned eight U.S. nuclear power plants, over 9,000 megawatts of nuclear capacity. With a large portfolio on the line, Entergy had an enormous incentive to make sure all its plants kept running without incident.

In short, under the old regime, it was profitable for nuclear operators to be extremely safe.


New Regime – Less to Lose

But that argument applied in an era when plants were actually generating electricity. Once plants close, this mechanism is no longer relevant – there are no operating profits on the line. Now the way to maximize profits is to minimize costs; so companies specializing in decommissioning will be working hard to figure out how to perform these functions as cheaply as possible.

And the reputation-based incentives also change. Before, companies worried that any problem at any plant would risk their reputation and thus their whole business – including other plants they owned and possibly including non-nuclear assets. But what about the new owners?  Smaller companies have less to lose.

Bankruptcy protection is also an issue. For a company like Entergy with a $19 billion dollar market capitalization, only a large incident would put it out of business. Not true for a smaller company. Economists have long argued that bankruptcy protection raises a moral hazard problem – with “judgment proof” companies having less incentive to act safely.


Following in the Footsteps of Oil and Gas?

A similar moral hazard problem arises with oil and gas wells. Our colleague Judd Boomhower has written about how small oil and gas producers face adverse incentives for safety and environmental risk. If the small oil and gas producer declares bankruptcy, it is not responsible for accident clean-up costs. Judd’s research shows that this can lead to less safe operating practices. Relatedly, the American West has thousands of “abandoned” wells that have not been properly remediated, many “owned” by companies that have gone bankrupt. Similar problems have shown up also with coal mines and offshore oil infrastructure. And the stakes for the nuclear sites are tremendous – across the country, we’re talking billions of dollars in anticipated clean-up costs.

An abandoned oil well in southern Texas. Source: Here.


Big Role for Regulation

When economic incentives alone do not ensure safe and thorough decommissioning, regulation should play a larger role. The new owners of these plants are inheriting substantial decommissioning funds, and the Nuclear Regulatory Commission has stated that it will be monitoring financial viability closely throughout the decommissioning process. But what happens if the new owners run out of money? Where will the necessary funds come from if the decommissioning funds prove insufficient to cover costs?

This is new territory for the Nuclear Regulatory Commission. As the U.S. nuclear generation fleet heads toward retirement, the NRC needs to pivot away from regulating construction and operation, and toward regulating decommissioning and fuel storage. Given the incentive issues these sales raise, it is critical that the NRC get up to speed quickly on the emerging risks.

Decommissionings at several recently-closed plants are aiming for accelerated timelines, which could be good or bad for safety. Between that and the specialized expertise that the new owners are bringing, the sales could turn out to be a win for the public. But the economic incentives for proper decommissioning are not very reassuring, and it’s not clear that regulations are ready to fill that gap.

Rancho Seco Nuclear Generating Station, a decommissioned nuclear power plant in Herald, California. Source: Here.


Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.

Suggested citation: Davis, Lucas and Catherine Hausman. “Nuclear Moral Hazard”, Energy Institute Blog, UC Berkeley, June 24, 2019,





Lucas Davis View All

Lucas Davis is the Jeffrey A. Jacobs Distinguished Professor in Business and Technology at the Haas School of Business at the University of California, Berkeley. He is a Faculty Affiliate at the Energy Institute at Haas, a coeditor at the American Economic Journal: Economic Policy, and a Research Associate at the National Bureau of Economic Research. He received a BA from Amherst College and a PhD in Economics from the University of Wisconsin. His research focuses on energy and environmental markets, and in particular, on electricity and natural gas regulation, pricing in competitive and non-competitive markets, and the economic and business impacts of environmental policy.

36 thoughts on “Nuclear Moral Hazard Leave a comment

  1. Carl Wurz and Geoffrey Rothwell have already made some excellent points about safety records, regulatory experience, and fact that decommissioning reactors isn’t a new business in the US.
    I’d like to challenge the implied presumption that the public is better protected when decommissioning is under the purview of large publicly traded corporations. There are numerous examples in the history of the electric power business where large, publicly traded corporations managed to escape liabilities by declaring bankruptcy.
    Given the current status of PG&E, the large, publicly traded corporation that presumably supplies power to the Berkeley-based author of this blog, I am somewhat bemused by the hypothesis that bigger is better.
    Holtec may not have much of a public reputation to protect, but it is a large, rapidly growing company with a $5 B order backlog and several major new projects under development. It has plenty of reasons to do careful, high quality work because its business success depends on its reputation within the industry and with the regulatory bodies that provide oversight.
    Poor performance is not economically beneficial because both regulators and customers (existing and future) have numerous tools at their disposal to penalize shoddy work.
    Rod Adams
    Publisher, Atomic Insights

  2. It does not take long to learn while skimming the mcubedecon website to learn they are not supportive of nuclear power. Furthermore, they tout the alleged environmental and ratepayer advantages of solar and wind power. mcubedecon raises points below that are not directly related to U.S. nuclear power production. Instead, the examples refer to the U.S. atomic weapons production plant a Hanford, Washington. Furthermore the U.S. does not reprocess its spent nuclear fuel.

    Rather than repeat the rebuttals, please visit websites such as Californians for Green Nuclear Power, Inc. (CGNP) at CGNP is an intervenor before the CPUC. We have also been supplying detailed information as to how solar and wind are *increasing* California power sector emissions, instead of decreasing them. The key concept is the natural-gas-fired generation that “firms” the huge amount of California solar and wind must be dispatched intermittently and inefficiently. IMO, the latter topic is long overdue for discussion at the Energy Institute Blog. CGNP would be pleased to adapt some of their recent filings before the CPUC in R.15-02-007 (The ongoing Integrated Resource Plan – IRP – Proceeding) to become blog postings for further review and commentary.

    • Let’s be clear. Carl Wurtz had made a broad claim that no harm had come from nuclear power fuel waste handling. I didn’t have to dig far to find a case where fuel handling for a 860 MW nuclear power plant has caused harm.

      That the CPUC discounted the evidence presented by CGNP in several cases speaks sufficiently to its credibility.

      • Please substantiate your claim with a reputable source where this happened and provide actual detail.

      • Sadly, as it it presently constituted, the main mission of the CPUC appears to be to maintain the fossil-fired status quo in California. The aggressive promotion of solar and wind as being environmentally beneficial by the IOUs, the California Energy Commission, and in media messaging is more of an advertisement than policies based on sound power system engineering and economic principles.
        For an introduction to the reality, first read the 2016 Washington Post article “Turns out Solar and Wind have a Secret Friend: natural gas” This article references the research of Elena Verdolini of the Euro-Mediterranean Center on Climate Change and the Fondazione Eni Enrico Mattei in Milan, Italy and David Popp of Syracuse University. Their 2016 working paper was updated by their Energy Policy Paper, “Bridging the gap: Do fast-reacting fossil technologies facilitate renewable energy diffusion?”
        California Fossil-fired “ramps” are becoming absurdly large as a consequence of the large amounts of intermittent solar and intermittent wind generation present on the California power grid. The California Independent System Operator (CAISO) showed a greater than 20,000 MW fossil-fired ramp (non-solar and non-wind generation) on June 11, 2019. To put that ramp in perspective, it is helpful to view it as “Hoover Dam Equivalents” equal to 2,078 MW. The ramp is about 10 Hoover Dams during a 10 hour ramp. That means add one Hoover Dam the first hour, two Hoover Dams the second hour, and so on until 10 Hoover Dams are running at the peak. Those fossil-fired generators are operated in an inefficient stop-and-go fashion, which yields more fossil fuel consumption – and increased emissions relative to the generation mix in California before large amounts of solar and wind were forced onto the grid. This occurs for the same physical reasons that your vehicle gets much worse mileage in city driving with heavy traffic than on a freeway with light traffic.
        An informative study regarding the maintenance of the fossil-fired status quo is found in “How Big Oil Continues to Oppose the Paris Agreement” March 22, 2019 InfluenceMap, London, UK 12 instances of “California.” 3 instances of “Western States Petroleum Association.”

    • WHAT? Emissions are increased by closing nuclear? But I have been repeatedly assured that is un-possible because we are in a super climate crises where we must do everything we can to reduce emissions.

      Jokes aside, you have it exactly right. Oh, and it’s not cheap either – solar and wind market penetration has been shown over and over to be positively correlated with rapidly increasing energy prices. Intermittency has a cost, and that cost is very high. Between 2009 and 2017, the price of solar panels per watt declined by 75 percent while the price of wind turbines per watt declined by 50 percent. And yet — during the same period — the price of electricity in places that deployed significant quantities of renewables increased dramatically. 24% in California, 50% in Germany.

      Click to access Hirth-2013-Market-Value-Renewables-Solar-Wind-Power-Variability-Price.pdf

      The fundamental explanation, published in 2013. Basically the economic value of wind and solar rapidly decreases with market penetration. There are lots of these studies, all showing the same thing. Hirth (2013) was just the first.

      Look at PGE in bankruptcy court – they have publically stated that shedding PPA agreements with renewable energy providers would save $2 billion + per year. That means they have been paying, with your money, $2 billion more than necessary for years. So a climate change driven wildfire will bankrupt PG&E, which will turn around and cut PPAs with renewable providers, bankrupting them as well. Whereas if PG&E has not been forced to buy over-priced renewables, they could have invested that money in making power systems safer, and potentially avoided the fire and bankruptcy altogether.

      Nice work guys. Really, just a stellar job.

  3. Again, you guys are totally clueless. Again, because the paper that reportedly showed that deregulation improved safety was a baseless paper (I’m sure you remember my discussion of it at the CPUC; if you don’t, I will post my written comments on LinkedIn). When are you going to learn something, anything about the nuclear power industry? I started working on decommissioning economics in 1989 with a Special Issue of The Energy Journal in 1991. I recently co-authored a report on decommissioning at the Nuclear Energy Agency of the OECD. (You can download it from the NEA.) Did you know that Areva/Framatom is an owner of the NorthStar Group? This is not some little under-capitalized corporation. This is a firm whose major shareholder in France; yes, the French State!

    The primary problem with making decommissioning profitable is the lack of waste management capacity. If you are creating decommissioning waste (see my recent co-authored report with the World Nuclear Association) and don’t have management capacity, the waste manager will extract your potential profits. Therefore, the most successful decommissioning firms also own either (1) waste management facilities, such as EnergySolutions that decommissioned Zion in Illinois with waste storage in Utah, or (2) they are big players in building Spent Nuclear Fuel storage facilities, such as Holtec, which is proposing a Consolidated SNF Storage Facility in Texas, as is a firm in association with Framatom.

    You guys have not done you research! There’s not a single credible reference to back up your hypothesis that “small” decommissioners don’t have the same incentives as “large” utilities. I advise you to take down this article because it is pure BS!

  4. The links provided are not relevant to the topic at hand. The reports mention incidents at fuel processing\fabrication facilities which are the front end of the fuel cycle. This is like saying that decommissioning at a coal plant will have issues because the coal mine/packing facility had incidents. Hanford was a site used for weapons production, which also makes it a false equivalency. These have nothing to do with decommissioning of power plants. Confounding different industries to make a point is either uninformed or misleading.

    Carl makes a strong point about worrying about something that has historically never manifested itself.

    • “This is like saying that decommissioning at a coal plant will have issues because the coal mine/packing facility had incidents” Exactly. The environmental and health damages from coal mining are directly linked to coal fired power. Coal is used for little else than electricity production in the U.S. now. So we need to link the entire chain of fuel production for nuclear as well.

      While weapons production was one use, the Hanford power plant also ran for decades, supplying the PNW grid, which is why I raised the issue.

      • This is an impressive act of mental acrobatics. You are linking nuclear weapons production facilities (Hanford, Savannah River) to commercial power generation. Decommissioning of defence facilities cannot be attributed to commercial power since they are two distinct industries. Further, those sites are not under the purview of the NRC and do not adhere to the same guidelines as do commercial nuclear power plants. I have yet to see an improperly decommissioned commercial nuclear power plant in the US.
        Uranium is used for defence, commercial, medical, radio-isotope production purposes, and in some instances is co-mined with other minerals; clearly assigning all of its use and subsequent effects to commercial nuclear power is not accurate. But by following the same logic, we’d have to compare the decommissioning record of other alternative sources, such as wind and solar, which use heavy and rare earth metals, with a poor track record of actual recycling or remediation; there is no need to describe the impacts of coal, oil, or gas as those are pretty evident.
        In sum, Dr. Rothwell and Carl Wurtz make the correct points that this blog worries about something that has never manifested itself, with mischaracterization of companies that are under extremely heavy regulation and scrutiny, with a large decommissioning fund that comes along with the sale of the deactivated plants.

      • If that’s why you raised the issue, you’re wrong again. Though Columbia (nuclear) Power Generating Facility is technically located on the Hanford site, its spent fuel has been stored responsibly in dry casks since the plant opened in 1984.
        To point to contamination from the 1950s, from an atomic weapons manufacturing facility, which never supplied any power to any grid is…pointless. Especially, when Columbia still generates 1.1 billion watts of carbon-free electricity to the Pacific Northwest grid, day and night – 220 times as much clean electricity as all solar in the State of Washington combined.

        • You’re showing your unfamiliarity with the Western power grid. The reason why the Columbia plant (WNP #2) was built at Hanford was because an 860 MW nuclear power plant (with a 400 MW rated capacity used for PNW planning purposes) was already operating at that site. The power output was sold by BPA. It was closed in 1996 and currently being decommissioned. I have been working on Western grid issues since the mid 1980s.

          • With you extensive experience working on Western grid issues, you must have missed the fact WNP #1 opened in 1975 – at least twenty years after contamination occurred at Hanford. They had nothing to do with nuclear power, but nice try.

  5. Lucas, with the passage of S-2313 in May 2018, New Jersey adopted a zero emission credit (ZEC) to reward the avoided-cost-of-carbon contribution of its two remaining nuclear plants, Salem and Hope Creek, for at least the next ten years. Unlike renewable energy certificates (RECs), which effectively double-count the clean energy contribution from renewables (once for the generation, again when its certificate is used to greenwash dirty generation), ZECs only charge ratepayers for what they deliver. Instead of a carbon tax, they reward clean energy with a “carbon refund.”

    “This is new territory for the Nuclear Regulatory Commission. As the U.S. nuclear generation fleet heads toward retirement…it is critical that the NRC get up to speed quickly on the emerging risks.”

    Decommissioning is not at all new territory for the NRC, and your concern about perceived “emerging risks” is typical of exaggerated risks associated with nuclear power in the public psyche. With no evidence mishandled power plant waste has ever been responsible for a single death or injury, there’s little to support fears NRC won’t continue to take responsibility for monitoring its safe storage – or even that, in the long term, U.S. nuclear generation is headed toward retirement.

    We might instead question the wisdom/morality of allowing Oyster Creek to retire prematurely, and saddle current ratepayers with $billions in decommissioning costs – but what’s done is done.

    • This blog points out that decommissioning by corporate entities with financial assets less than potential risk IS new ground for the NRC. And the issue isn’t about mishandling of the decommissioning process but rather about the potential abandonment when decommissioning costs unexpectedly balloon, as happens too often in the nuclear power industry.

      RECs do not double count “clean energy”–they are just transferable between energy sources. The solar plant that sells its RECs loses those credits. The only difference with the ZEC is that the latter isn’t transferable (for now).

      Here’s an IAEA report that documents 58 incidents at nuclear fuel processing plants:

      Here’s a discussion of the hazards and long lasting damage at the Hanford Nuclear Reservation. There was an injurty there in 2017.

      • Holtec, since 1986, has had thousands of interactions with the NRC, not only about decommissioning, but with innovative new designs for reactors being developed in conjunction with GE-Hitachi. Decommissioning is new ground for Holtec, or the NRC, only in the fervent imaginations of anti-nuclear activists.

        ZECs aren’t transferable for that very reason: they prevent dirty generation. RECs? Believers pretend CO2 already emitted can be somehow sucked out of the air by generating more clean energy, an idea so logically inept it’s unworthy of further comment.

        Re: your links: fuel processing around the world has nothing to do with decommissioning in the U.S. (or anywhere else), and the reactors at Hanford were not used to generate power. They were used to manufacture bomb-grade plutonium.

        • Again, you miss the point of the blog–this isn’t about the competency of Holtec. It’s about the financial resources and eventual solvency of Holtec and others compared to the original utility owners. There simply is no comparison.

          You made a claim about mishandled plant fuel waste that not limited to decommissioning (although the decommissioning at both Hanford and Savannah River have had severe environmental consequences.

          And as I pointed out you are mistaken in your characterization of RECs–it is not double counting. You misunderstand the use of RECs and ZECs for regulatory compliance purposes. Neither removes GHGs from the air. But both prevent GHG emissions in exactly the same manner by limiting overall system wide GHG emissions. The total system GHG emissions will be the same whether RECs or ZECs are used. RECs allow more flexibility about who can claim credit for those reductions.

          • Grasping at straws. Further hypothesizing about the “eventual solvency” of a reputable company with a thirty-three year history, and other anti-nuclear fearmongering, will be dutifully ignored.
            No, I didn’t make a claim about mishandled plant fuel waste. I made a claim about mishandled power plant fuel waste, which you re-wrote to cover your confusion of power plant fuel with highly-enriched bomb making material.
            A solar farm and a coal plant each generate 10 MWh of electricity. Do the math: 20 MWh of electricity was generated and 3.4 tonnes of CO2 was emitted. Does a fancy REC certificate make those emissions go away? Of course not, and any hooey about “limiting overall system wide GHG emissions” still gives renewables exactly twice as much credit as they deserve.

          • Hanford fuel is power plant waste. The technology used to produce that power is irrelevant. You made a false broad claim.
            Again, you don’t understand how RECs work. The solar and coal plants between the two of them produce 30 MWH of electricity and they have 10 MWH of RECs to share between them. For regulatory compliance purposes, the owner of the solar plant can sell those 10 MWH of RECs to the coal plant owner to allow the coal plant owner to comply with a regulatory standard, e.g., RPS. This is exactly the same as how GHG cap and trade programs work. There’s absolutely no reason that a ZEC program can’t be set up to operate this same way.

          • I understand exactly how RECs work: fossil fuel generation is allowed to qualify as renewable energy just because a solar farm has generated the same amount of electricity somewhere else. Its purpose is to comply with a regulation that has half the positive environmental impact its supporters claims it has.

            What use would nuclear plant owners have for the Monopoly money being traded by renewables/gas entrepreneurs? ZECs allow generating plants a legit monetary credit based on the social cost of carbon – commensurate payment for generating clean energy. One might think renewables promoters would jump at the chance to be part of such a program, but they’re not – though both solar and nuclear are included in a proposed ZEC in Ohio (HB 6), solar interests are fighting it tooth and nail, calling it a nuclear “bailout.” The reason? Because it doesn’t double-count their contribution, they’d have to compete on a level playing field with carbon-free, ’round-the-clock, dispatchable nuclear generation. And it’s obvious: there’s no competition.

          • No, you don’t understand RECs. The way they work is that the solar and gas plants switch places in the regulatory accounting, so that the renewable MWHs count does not change. And there is no reason why the ZEC programs can’t change in the same way. You are being naive if you don’t think that regulatory accounting can’t be changed. The reason that renewable developers don’t want nuclear counted is obvious from an economic standpoint–they don’t want the competition from another generating source. There’s nothing about a fiction of double counting.

          • Your bogus “regulatory accounting” has no relationship to environmental impact. So the “solar and gas plants switch places,” do they? Do we switch the carbon emissions from the gas plant to your solar plant too? Just the credits, no debits? That’s a laugh – try taking out a loan on those terms. See how far you get.
            “The reason that renewable developers don’t want nuclear counted is obvious from an economic standpoint–they don’t want the competition from another generating source.”
            Obviously. Why should ratepayers be forced to spend more to finance RECs when they provide half the benefit?

          • And ZECs will likely work so that the clean energy credits from the nuclear plants will be separated made transferable to coal plants to meet compliance requirements. There is no fundamental reason why ZECs must be tethered to the generator, just the same as RECs.

%d bloggers like this: