Risk Subsidies and the Future of Nuclear Power in the U.S.

Should We Take Into Account Government Subsidies that Reduce the Risks Borne by the Nuclear Industry as We Consider Our Energy Future?

As I’ve written about before, U.S. law massively subsidizes the nuclear power industry.  In particular, a law called the Price-Anderson Nuclear Industries Indemnity Act dramatically skews the incentives to develop nuclear plants, and to site them in places where there is a lot of risk, because it requires the public to bear much of the financial risk associated with physical risk from nuclear facilities by capping operators’ liability. The Act also pools all operators’ risks under the cap, creating no incentive for any operator to consider the financial cost of safety risks and hazards unique to particular plants’ operations or location. On top of this, the U.S. also has committed to pay the costs of managing spent nuclear fuel, recently estimated at $21.4 billion through 2071 (though the GAO believes this estimate may be greatly understated). While there are many reasons that nuclear power development has mostly been moribund in the U.S. over the past several decades, it’s important to keep in mind that all the while, these subsidies have put a thumb on the scale in the opposite direction, by removing and shifting costs relating to nuclear power development and operation away from plant operators.

The Price-Anderson Act was designed to support the industry in the 1950s, when it was just getting off the ground and advocates believed that the liability for then-nascent risks was a huge barrier to innovation.  It was not designed to be a permanent subsidy.  But it has never been repealed.

One major impact of Price-Anderson is that it makes it unnecessary for plant developers and operators to procure adequate insurance, making it virtually unique among business ventures that pose risks of significant magnitude.  Insurers are quite rational in their approach to risk. Anyone who believes that nuclear power’s risks are low, or manageable, should be willing to let insurers price that risk.  Indeed, analysts with a pro-regulatory/anti-nuclear position and libertarians alike, as well as other academic commentators, have articulated these concerns.   The concerns, and the wide-ranging support for removing public subsidization of this risk, mirror similar concerns relating to other risks where government subsidies provide perverse incentives, such as flood and crop damage risks.  Aside from the incentives and benefits it provides to the industry, Price-Anderson means that every American has a stake in the safety of nuclear power everywhere in the country, in the sense that we collectively bear that risk, whether or not we get the benefits.

Earlier this week, Michael Shellenberger, an ardent pro-nuclear energy advocate (who recently commented here on a post about the Diablo Canyon plant by Steve Weissman), defended nuclear power in a debate here at UCLA that also featured NRDC’s Dale Bryk as well as Ken Caldeira and Mark Jacobson from Stanford (video is at the link; the program is part of the UCLA Institute of Environment and Sustainability’s Oppenheim Lecture Series).  I was struck – as I often am when I hear advocates of nuclear power speak – at the absence of any acknowledgment of the U.S.’s subsidization of nuclear risk.  Mr. Shellenberger has vehemently supported nuclear energy for the reason, among others, that it is cheaper than other zero-carbon energy sources in other countries, and has the potential to be cheap in the US.  But in all the writings by him and his colleagues on why nuclear energy in the US is more expensive than it should be, and how to make it cheaper, I haven’t found any evidence that he or others among current nuclear energy advocates have grappled with the implications of Price-Anderson, or other subsidies to the nuclear industry and costs attributable to that industry that our government has committed to pay.

I won’t rehash here all the detailed observations from my prior post, but I urge anyone who cares about this issue to learn about it, in order to better consider the costs and benefits of nuclear power in light of the nature and magnitude of liability that the U.S. government has removed from the nuclear power industry.  I am agnostic about nuclear power’s role in our energy future, but if we are looking at costs, benefits, and risks, we need to consider all the costs and risks that our laws allow the industry to externalize.  Now that Shellenberger and others are making a serious argument that we need nuclear energy in order to address climate change, it’s especially important.  Perhaps it may turn out to be good policy to subsidize some of those costs or risks, but I haven’t really heard anyone try to defend it.


[Update: I have edited the post to be more accurate, in response to Michael Wara’s comment below.  I have added the word “adequate” to this sentence: “One major impact of Price-Anderson is that it makes it unnecessary for plant developers and operators to procure adequate insurance…”]

, , , , , , , , ,

Reader Comments

7 Replies to “Risk Subsidies and the Future of Nuclear Power in the U.S.”

  1. Sean,

    I enjoyed your post (and the previous posts by you and by Steve). And I agree with your views that (1) we need to have a serious conversation about insurance in the same way we talk about the ITC/PTC and RPSs (and other renewable subsidies) if nuclear is going to have a place in the new energy future and (2) all costs/benefits should be weighed in trading off different energy mixes.

    I wanted to make a few comments about Price Anderson and risk in the nuclear industry in response to your post. First, as you correctly point out in your earlier post, the nuclear industry does self-insure for some portion of the risks of an accident via individual insurance (up to $375M cap per plant) and the pooling mechanism (up to $12B shared across the operating units). My understanding is that the nuclear industry also has to provide and maintain some financial assurances that in the event of a call on the extra money that plants have to contribute, they would be able to do so. As you also correctly point out in the earlier post, recent evidence from Fukushima indicates that were a serious accident to occur, the damages would easily blow through the second cap and leave the taxpayer on the hook.

    But in any case, it’s not quite right to say that “One major impact of Price-Anderson is that it makes it unnecessary for plant developers and operators to procure insurance.” I think the more accurate view is that P-A allows the industry to carry inadequate insurance. And that there is implicit guarantee on the part of the taxpayer for any coverage above the $12B cap.

    Second, you say (mostly in the earlier post) that without the requirement to carry (adequate) insurance, the industry lacks incentives to mitigate risks of accidents. I also view this argument as only partially correct. First off, there is the insurance up to the liability cap. Presumably, this, especially the initial $375M which is plant specific, has some impact on operational choices. Second, and perhaps much more important, there is the “hostages of each other” phenomenon. Nuclear operators know – or think they know – that were there ever to be a LOCA along the lines of Fukushima in the US, that would be the end of the industry. Full stop. Thus they have VERY strong incentives to avoid accidents and maximize operational safety. The best evidence for this is the existence and effectiveness of INPO. It’s not every industry that has a self-regulatory regime that requires CEOs to show up at an annual meeting to be publicly shamed by their peers for any safety lapses. But that’s how the nuclear industry rolls. And its how the industry has driven down accident rates while dramatically increasing capacity factors over the past 30 years.

    All this being said, I appreciate the discussion and agree completely with your main point – that nuclear’s costs and benefits need to be fully assessed right along side other technologies, and that the insurance subsidy is an important if hidden part of the cost side of the ledger for nuclear. All things being equal, I favor keeping the existing units open for the time being – if only to buy us time to further drive costs of renewables and storage down to ultimately replace them. But for new units, it would perhaps make much more sense to force companies into the private insurance market but also provide them with similar incentives to wind. It would be very interesting to see whether or not new SMR units would be financible without P-A but with a PTC.

    Lastly, an important point to make about SMRs is that its not clear to me how they will be compatible with P-A, which assumes that all units are of roughly the same size. A $375M insurance policy plus the cost of whatever credit line is required to meet the industry cap is probably a much bigger item on the budget for a 125MW reactor than for a 1GW reactor. But P-A would appear to treat them the same.

    1. Those are all fair points, Michael. Thanks for your thoughtful comments.

      I agree that the industry’s insurance arrangements are really just inadequate, rather than absent. I’ve edited my post to reflect that.

      And of course you are also correct that there are extremely strong societal reasons – as well as regulatory reasons – for the industry to mitigate its risks. My point really is that from an economic perspective, those incentives are small. And given that so much of the conversation about environmental regulation and future choices these days is based on economic incentives, as well as costs and benefits, I think it’s fair to point out that Price-Anderson gets it wrong. It’s remarkable how people who rely on economic analysis as the starting point for all decisions are willing to abandon it here. Moreover, while your point may be well-taken with operational controls, I’m not sure that operators and developers really do consider liability-related financial risk in their siting decisions – and certainly, they didn’t in older siting decisions. Does anyone really think that Indian Point (25 miles north of New York City), or San Onofre (45 miles north of San Diego), would have been sited in the same place if liability had been taken into account? That seems like a stretch to me.

      1. Hi Sean,

        Thanks for your kind words.

        My experience has been that there are two kinds of people still interested in building new nuclear. Those that like it from an engineering/innovation perspective – my colleague Burt Richtor comes to mind – and those who are protected from really engaging in the short-term economic forces – for example Georgia Power. If you hear anyone from Exelon talk, they will tell you about the risks their existing nuclear assets pose to the rest of the firm’s economics given current regulatory realities (ie no carbon price, PTC extension, RTO market dynamics).

        So the people who support nuclear tend to come from the cohort who favor engineered, planned decisionmaking in the power sector that takes a very long-term view, rather than the short-term, economic incentive focussed crowd.

        Regarding siting, I totally agree regarding the older plants. It would be an interesting question to ask of the Vogtle and SCANA reactors, currently under construction. It seems like the only place you can put a new nuclear unit these days is at the site of an existing nuclear unit. Otherwise it’s basically impossible to entitle the land use. And that says something – not sure whether its sensible or not – about public acceptance of these facilities.

        In any case, my guess is that the (essentially infinite) cost of dealing with siting in a greenfield location, even if that place would be less risky than alternative existing reactor sites, outweighs any possible savings that might accrue from reduced insurance premiums.

        1. That’s an interesting perspective, Michael. I am not nearly so engaged with either the engineering/innovation people or the power sector in general, so it’s useful to hear that. One other thing to add: it’s possible that with siting of new nuclear facilities near large population centers, the issue wouldn’t be the magnitude of the premiums, but the availability of insurance at any price (or alternatively, the uncertainty premium could be so unthinkably astronomical that it would effectively make the insurance unavailable). That is a challenge where insurance markets are faced with extraordinarily high-magnitude, low-probability risks.

  2. Not sure what you purpose is, Sean.

    This is a bit odd… “…creating no incentive for any operator to
    consider the financial cost of safety risks and hazards unique to
    particular plants’ operations or location.”

    If it were true, then we might expect things like hospitals to be safer
    for the public than nukes. But that’s observationally false, for

    And, a detailed analysis of “nuclear subsidies” is below*.

    I’m wondering if there’s an intellectual ‘game’ being played here,
    . For instance, you seem unhinged by Shellenberger’s statements
    and that some scientists are inconveniently advising increased nuclear
    rollouts. They’re right — late, but right.

    Ansel Adams supported nuclear power when thMy dear Sierra Club’s motto
    was wisely: “Atoms, not dams”.

    there are now >55,000 dams in the world and half the world’s rivers are
    in biological trouble because of damming. There are alos many millions
    of folks endangered by dams — about 600 in the Himalaya watershed, most
    poorly managed/constructed. Note the recent worries about Iraq’s Mosul
    Dam, coming apart at its base. Want to estimate the number of folks
    endangered by that single reality?

    So, when fossil-fuel-supported folks like Jacobson talk about using dams
    to make up for windmill or solar ‘farm’ unreliabilities, does not the
    endangerments from those dams get included in the WWS risk?

    We know very well that nuclear power has always been the safest form of
    generation, so I wonder how and why you mentally criticize the P-A Act,
    when it’s covering the safest form of power, including environmental,
    while other forms put groups at risk without financial protections.

    It must be an ‘inconvenient truth’ that so many climate scientists,
    other scientists and even the Dalai Lama advocate increased nuclear
    power, eh? See** — I personally agree with James Hansen’s advice to
    cut funds from anti-nuke orgs, just as I advise others to divest from fossil fuels.

    The happenings in cliamate & oceans will forever be an embarrassment to
    both the fossil-fuel and anti-nuke folks — if they can be embarrassed.


    * Subsidies…

    ** Lovelock 2013 — “We nuclear power soon”…
    http://energyforhumanity.org/climate-energy/nuclear-for-climate-declaration/ (50,000 scientists)
    http://tinyurl.com/kn22qcn (Hansen, Caldeira, Emanuel, Wigley)
    http://seekerblog.com/2015/03/10/james-hansen-calls-out-big-green-part-2/ (Hansen vs Big Green)
    “I also recommend that the public stop providing funds to anti nuke environmental groups. Send a letter saying why you are withdrawing your support. Their position is based partly on fear of losing support from anti-nuke donors, and they are not likely to listen to anything other than financial pressure. If they are allowed to continue to spread misinformation about nuclear power, it is unlikely that we can stop expanded hydro-fracking, continued destructive coal mining, and irreversible climate change.”


    Dalai Lama…

    Ben Heard (former Aussie anti-nuke)…

    Tom Blees…


    1. Nice post, thanks. Interesting that even here in the last comment, someone who is pro-nuclear basically changes the subject when confronted with the liability subsidy. Not many non-lawyers even know about the issue.

      What I’d be interested in but know nothing about is how other countries handle this liability issue. My assumption is that their governments say they will handle it, but do very little in terms of imposing a charge on nuclear power to help self-insure. But maybe that’s wrong, and if some country does account for the entire liability without finding it to be a major cost, that would be something that pro-nuclear activists could use.

      1. I don’t know much about what other countries do, but this is an interesting discussion of liability across various countries. It is strongly pro-nuclear in its slant (that is clear from its discussion and analysis of Price-Anderson), and it’s not footnoted so I can’t vouch for its accuracy in detail. But I have no specific reason to doubt the accuracy of the basic facts it contains about liability structure and conventions.
        And here is information specifically on liability relating to Japan’s accident: http://www.oecd-nea.org/news/press-kits/fukushima.html#4

Comments are closed.

About Sean

Sean Hecht

Sean B. Hecht is the Co-Executive Director of the Emmett Institute on Climate Change and the Environment, Evan Frankel Professor of Policy and Practice, and Co-Director o…

READ more