Hope for Nuclear Regulatory Reform?

April 22, 2015, 2:43PMANS Nuclear CafeJim Hopf

As I've discussed in many previous ANS Nuclear Cafe posts (see: How Can Nuclear Construction Costs be Reduced & Cost/Benefit Analyses of Nuclear Requirements), my belief is that the primary reasons for the lack of nuclear power's growth/success are the extremely burdensome requirements and regulations. They are vastly more strict than those applied to competing sources, in terms of dollars spent per life saved, or amount of environmental impact avoided. For various (non-scientific) reasons, nuclear is held to standards that are orders of magnitude higher than those applied to competing sources.

Many in the nuclear field believe that nuclear's future success can be attained through new, superior reactor designs and fuel cycle technologies that will be successful, and economic, even under this tremendously uneven regulatory playing field. I have always been less sanguine, and believe that without changes in the regulatory playing field, there will be little new nuclear, at least in the developed world. However, given the politics and public attitudes, I haven't been too hopeful that significant regulatory reform was possible.

Senate NRC hearing

Recent events, however, suggest a ray of hope, i.e., that some amount of regulatory reform may actually be possible. After the Republican takeover of the senate, Lamar Alexander (R., Tenn.) has been elevated to chairman of the Senate Appropriations Committee on Energy and Water Development. During a March 4 hearing on the the Nuclear Regulatory Commission's Fiscal Year 2016 budget, Alexander made reference to "excessive and unnecessary" regulations, and his desire to work with the NRC to address the issue. He also stated that federal law "doesn't say that it should be so hard and expensive to build and operate reactors that you can't do it," suggesting that NRC regulations and requirements may have reached that point.

In response to Alexander's remarks, NRC commissioner Kristine Svinicki admitted that recent cost/benefit analyses by the NRC, which had been used to justify recent additional (post-Fukushima) regulations, underestimated the costs of compliance by several orders of magnitude.

A video of the hearing and transcripts of prepared remarks can be found here. Of particular interest are Alexander's prepared remarks. This senate hearing is also discussed in this Atomic Insights blog post.

NRC cost/benefit analyses?

It should be noted that (by my understanding) the NRC cost/benefit analyses only apply to new regulations and/or required backfits. They do not apply for older/existing regulations, or compliance with requirements for new reactors under construction. It seems clear to me that cost/benefit analyses have not been applied to nuclear regulations overall. In other words, cost vs. benefit analysis appears to be something the NRC has either only recently started doing, or is something that it only does selectively. My own evaluation suggests that the overall body of nuclear regulations (and NQA-1 Quality Assurance requirements) has an overall cost that runs well into the billions of dollars per life saved, which is orders of magnitude higher than the cost of regulations applied to other energy sources, and orders of magnitude higher than the general government safety criterion of about $10 million dollars per life saved.

If cost/benefit analyses were done at all in past decades, it is likely that extremely unrealistic or inaccurate estimates of meltdown consequences were used as the basis. For decades, we have been told of official accident consequence estimates that involved thousands of short term deaths (from acute exposure) as well as tens of thousands of latent cancer fatalities in subsequent decades. Suffice it to say that Fukushima, and even Chernobyl, have shown those estimates and analyses to be wildly off the mark, by orders of magnitude. Fukushima showed that even a worst-case event, with the full meltdown of three large reactors along with a significant failure of containment, caused no deaths and is projected to have no measurable public health impact. It should also be noted that the maximum potential release from a small modular reactor would be more than an order of magnitude smaller still.

Given these facts, one has to wonder how one could come up with any finite cost-per-life saved for any nuclear safety measures (that would reduce the frequency or magnitude of reactor accident releases). It looks more like this is a purely financial risk on the part of the utility, which would lend itself to the utility (and their insurers) making a purely financial decision (that weighs cost vs. reduction in financial liability).

Needed changes

Not only does the NRC need to perform more realistic and accurate cost/benefit analyses for new and/or backfit-related regulations, but it needs to do a bottoms-up review of its entire body of existing regulations, requirements, and procedures. That review needs to be based on realistic accident consequences, based on what we now know from Fukushima, etc. I've discussed this in more detail in previous posts.

With respect to fabrication QA (e.g.. NQA-1), cost/benefit analyses need to be performed that are based upon realistic estimates of component failure frequencies (for various "Q" levels), realistic types and severities of component failure (no assuming that a failed component simply disappears), and realistic estimates of component failure consequences. With respect to component or design deviations, cost/benefit analysis should be considered instead of just requiring verbatim compliance. Can anyone argue that the rebar issue with the Vogle basemat really involved potential public health risks that were worth the huge cost of the long process that was required to "address" the issue? Did it involve a tangible increase in release risk at all? More generally, I personally doubt that NQA-1, a unique and more strict fabrication QA program that applies solely for the nuclear industry, has benefits that are worth the cost. I'd love to see a cost/benefit analysis of using NQA-1 vs. using the standard QA programs that are used for most heavy construction/industry.

In terms of policy, all energy sources should be held to the same standard, in terms of dollars spent per life saved (where impacts other than public deaths could also be considered). Perhaps more specifically, all regulations, for all sources, should be subject to cost/benefit analyses that use the standard government public safety criterion (which is about $10 million dollars per life saved at the moment). Under such a standard, nuclear regulations would likely become much less strict, whereas restrictions on fossil (mainly coal) plant pollution would become far more strict.

Needed action

I'm considering writing to Alexander to express the above thoughts, but past experience suggests that my letter will almost certainly go straight into the circular file, since I'm not a resident of Tennessee. I have never written to a congressman or senator from another state and gotten a response. My view is that the chairman of a house or senate committee is a national office, with respect to the issues covered by the committee. Congressional practice, however, is that even committee chairs (let alone committee members) ignore all correspondence from people who do not live within their state or (house) district.

Thus, I'm hoping that anyone reading this who hails from the state of Tennessee will write Alexander about this issue. Feel free to use any, all, or none of the ideas expressed above. In my view, the main points would be:

  • NRC regulations should be subject to cost/benefit analysis. Not only should any new regulation be subject to such analyses, but a bottoms-up review of all existing NRC regulations and other practices and procedures should be performed, with each requirement or practice being subject to cost/benefit analysis.
  • The regulations and requirements governing all energy sources should be subject to cost/benefit analyses, with the same criteria (dollars per life saved, etc.) applied to all sources. The standard government public safety criterion (now about $10 million per life saved) should be considered as the basis for all analyses.
  • With respect to fabrication/construction QA, the value of having a unique nuclear industry QA program (NQA-1) should be investigated, and subject to cost/benefit analysis. That is, the possibility of having the nuclear industry be subject to the same fabrication/construction QA programs that apply for other large construction projects (bridges, buildings, industrial sites, etc..) should be evaluated, with respect to cost vs. benefit.