The form of the "policy" for promoting innovation in rector design should have taken the form of instructing NRC to come up with a cost benefit methodology, subject to review by Congress.
I forwarded your column to my good friend, Joe Williams, a nuclear engineer who retired in the last few years from his employment at the NRC. Pasted below is what he wrote in response. I submit this with his permission, and he also volunteered to talk to you, Matt, if you have more questions or want to discuss this with. him further. So, here's Joe's take on your column-
Matthew Yglesias essay regarding the Nuclear Regulatory Commission’s ability to license new reactors does not present a full story. He suggests that NRC isn’t serving the public well by failing to license advanced reactor designs, but does not appear to consider the particular technical and regulatory circumstances, or the wounds the nuclear industry has inflicted upon itself.
From my perspective as a retired nuclear engineer with over 40 years of experience, the essay is one of the latest in a very long line of similar items blaming NRC for the failings of the industry. I worked for NRC for nearly 30 years, spending most of the last 20 years of my career working on new reactor licensing issues, and I am familiar with many of the advanced reactor concepts, as well as the technical and regulatory issues involved. The NRC has some considerable flaws, but that it is far from the only factor in the industry’s struggles.
Yglesias takes NRC to task for not accepting the Oklo reactor license application for review. I should add a disclaimer at this point that I briefly worked as a consultant for Oklo after my retirement, though I had no part in preparing their license application. While at NRC, I was part of the branch responsible for management of advanced reactor projects, including Oklo. I did not participate in preapplication activities with Oklo, and left NRC before their application was submitted. I believe that the Aurora design concept is very innovative, and I hope that the Oklo team can effectively address issues with the application and bring the project to fruition.
It is not apparent from Yglesias’ essay that he is familiar with the specific issues that led NRC to make deny the application, described in a publicly-available NRC letter to Oklo. The gist is that "Oklo has repeatedly failed to provide substantive information in response to NRC staff requests for additional information (RAIs) on the maximum credible accident (MCA) for the Aurora design, the safety classification of structures, systems, and components (SSCs), and other issues needed for the NRC staff to establish a schedule and complete its technical review.” NRC’s letter notes that Oklo can demand a formal hearing if they believe the application actually fulfills requirements, but you can be confident there won’t be a hearing.
Yglesias also notes that NRC has not approved any advanced reactor applications. While this is true, it is also misleading. To date, only one advanced reactor application (that is, a reactor that does not use water as a coolant) has been submitted – Oklo’s. In the 20 years I worked on new reactor licensing, I saw more than a few claims that a design would be submitted by a certain date that did not come to pass, not because of regulation, but because the claims were based on hope or hype, rather than reality (see, for example, Transatomic or Hyperion).
His statement that NRC has never approved a design from start to finish since 1975 lacks important context. For one thing, there were no new license applications for about 30 years from before the TMI-2 accident in 1979 until 2008. In the meantime, NRC created a new regulatory process with extensive input from industry stakeholders that provided for prior approval of designs and sites before construction. Starting in the 1990s, this process was used to approve several new designs, and then to approve licenses to allow construction and operation at several sites, including the one being built in Georgia. Many of the other projects were not pursued due to reduced projections for electrical demand after the 2009 recession, while another project in South Carolina failed because of untimely completion of design details by the reactor vendor and architect engineer and poor management of the huge construction project; these issues are well-documented in reporting by the Charleston Post and Courier newspaper. These problems have little, if anything to do with regulation. NRC is far from perfect (I could go on for awhile about that, too), but they aren’t the people that put a bunch of parts in a warehouse without adequate indexing so they could actually be found when they were needed, for example. And this far from an isolated example, as very similar issues have repeatedly come up in other projects dating back to the ‘70s and ‘80s. The industry gripes about NRC’s heavy hand, but the reason they suffer under that weight can almost always be traced to where they screwed up and NRC had to react. See, for example, the rather colorful regulatory history of the Tennessee Valley Authority’s nuclear program.
Yglesias makes what I think is a valid comparison between the hazards of nuclear power and other forms of electrical generation, suggesting that such comparisons should be considered in NRC’s licensing. While they may not be addressed to the extent they should be, the NRC’s review process explicitly requires addressing alternatives in the environmental review for a license.
I put a lot of time and effort towards the goal of building new reactors that can provide the country with carbon-free electricity. Nuclear technology can be a tremendous asset for mankind, but it is not the panacea some of its more fervent advocates claim. For many years, I kept a quote from Richard Feynman over my desk: “For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.” If the nuclear industry wants to be successful, they each need to be honest with themselves about their shortcomings. Otherwise, they will never get to the true source of their problems.
We can talk about NRC’s shortcomings another time.
This seems to miss the point of 90% of Matt's article and focuses on basically 1 paragraph. Clearly this was the paragraph that triggered your friend but it his reply probably would have been better received (certainly by me) had he focused less on his hot-button issue and either conceded or objected to the remainder of the piece.
While my response didn’t address every point Yglesias made, I believe it was more comprehensive than you suggest. I attempted to point out that many of the statements in the essay regarding the things NRC hasn’t done or should be doing don’t withstand scrutiny. The fundamental point is that many nuclear advocates have scapegoated NRC for the industry’s problems for decades. It is not NRC’s fault Oklo’s work did not meet the standards required by regulations to accept the application. Sorry you didn’t find my perspectives useful.
Separately, I see several comments regarding the Navy’s nuclear program. It should be no surprise that many NRC staff, including many senior managers, have served in the Navy, and are quite familiar with the standards established by Naval Reactors and have brought those principles to bear in their decision making.
Skimming through his article, a full 2/3 of it seems unrelated to the NRC. I know nothing of the NRC. I will take you at your word that he was unnecessarily unkind to them. My point was merely that your response might have found a more receptive audience had you posed it for an audience who read Matt's piece as something other than a scathing critique of NRC.
"... build a few for people who have eccentric use cases (remote communities dependent on diesel, companies that for PR reasons want to say they’re zero-carbon ...."
Or the Navy, that wants to power submarines in places where there isn't enough oxygen to burn stuff, and doesn't want to have to re-fuel or re-charge frequently ... which we already have. Nuclear-powered submarines are working out fine, aren't they?
But Matt Y, having listened to you on Bulwark today, I think it might be a good idea to do a post on why you are still a Democrat.
I agreed with a lot of your points today.
The one point I would maybe push back on is when you said the GOP has moved left on some economic issues. I would say rhetorically they are not using the Ayn Rand talking rhetoric. However, have they changed on a policy level? I think no. Youngkin is fighting min. wage increases and is pursing business tax cuts.
The next GOP President and Congress will do one thing. Pass an enormous inflationary upper income tax cut. None of the rhetoric matters. Do you doubt that?
Given that our country has basically never given nuclear a real shot, it isn’t particularly instructive to look at historical deaths stemming from reactor malfunctioning. Instead, we must estimate expected deaths based on proposed reactor location(s), probability of malfunction, magnitude of nuclear event given malfunction, etc.
Matt sort of gets at this by alluding to the “high-end estimate of the worst nuclear accident ever”. But like once-in-a-lifetime Global Financial System Collapses, or once-in-a-lifetime Pandemics, we still run into the tricky problem of making decisions based on very low probability events that carry potential for catastrophic outcomes. Using Matt’s cost-benefit logic, we could just as easily argue against Dodd-Frank Wall Street Reform or legislation for future Pandemic preparedness. My point being that this topic seems a bit cherry-picked for this type of analysis.
I also disagree with the approach of treating premature deaths due to pollution (which in this case disproportionately impact an older population) the same as workplace accident deaths (which by definition happen to the working age population) or deaths resulting from exposure to nuclear radiation (which can kill infants). Such an approach ignores the fact that, to quote Chuck Palahniuk, “On a long enough timeline, the survival rate for everyone drops to zero.”
I think a better way to approach such cost-benefit analysis (to the extent that, per my first point, it’s even applicable in this case) is to evaluate expected life years lost/saved. I’m not saying it changes the conclusion vis a vis whether to expand nuclear energy programs, but I [potentially, didn’t exactly run the numbers] disagree with Matt’s suggestion of where the marginal shift occurs between nuclear and fossil fuels.
Not really. The micro reactors have a lifetime fuel load pre installed. The solids handling of coal does not cost scale down well. Still need to build a railroad. The fossil "micro-reactor" you are thinking of is an industrial scale diesel generator. Fuel is expensive for them
It's not as if these things won't be built. They just won't be built in the US. Fortunately there is a whole world out there capable of moving this technology forward. My money would be on the French and probably the Chinese. Germany has taken themselves out of the game. That's a shame but technology moves on.
There's an interesting wrinkle to the Oklo decision. Rod Adams made this point on the Nuclear Barians podcast, but his theory is that the NRC is now bound by law to complete their review in 3 years. By rejecting the application the NRC reset the three year clock, which is a quite nifty way to side step the law.
You certainly can learn a lot from the past. As an engineer reaching the end of a decades long career I would say that the chief lesson from Chernobyl is the same one you could learn from a wood stove. Never build a wood stove out of wood.
Speaking of economics, one of the things that makes me wonder about the future of nuke plants delivered and operated by private business is that they are essentially impossible to insure.
If a private provider had to 100% fully carry their own liability risk, no project would ever be profitable. The entire enterprise pre-supposes an enormous implicit federal insurance subsidy.
It seems like the big issue with nuclear power is that we still have no good story for how the nuclear waste it produces is managed. Having a larger number of smaller reactors in service would appear to make this problem worse, not to mention increasing the risk that somebody might succeed in stealing the spent fuel in order to do (or threaten to do) something awful with it.
In the South West we have huge swaths of dirt ( just fly over S Cal) that could be covered with solar at no environmental cost. 5 GW installations are not unreasonable. You do not mention Deep Bore Hole storage of nuclear waste. This will solve the waste issue and place it where it does not need monitoring. Current drilling technology (see the Chilean mine disaster relief) is sufficient.
As a resident of the southwest (Southern AZ), while I would agree that there won't be a problem finding land to lay down solar panels, I'm hurt that your would describe our beautiful landscape as "huge swaths of dirt" :p
This issue with nuclear technology seems to me related to this issue with PPP, https://www.nber.org/papers/w29669#fromrss, and to issues Matt has raised before with the costs of building transit infrastructure, and to issues Matt has raised with the FDA and CDC. It seems many of our government bureaucracies are not well suited for contemporary problems. Are the solutions all local to each agency, the slow boring of hard boards? Or are there bigger things Congress or the White House ought to do?
Always great to see a rationalist perspective on the cost/benefits of nuclear.
Here in Washington the same people who believe climate change will wipe out humanity are fighting tooth and nail to take out all Hydro power. Go figure.
I wonder what they are actually checking here. I would assume nearly 80% of college grad age people have tried it (I actually see a link saying 80% of 55 year olds have tried it). Are they actually looking for non-users? Are they screening out people who are inadequately conservative to lie about it?
It seems to me it would make more sense to only hire the people that say yes to screen out the liars.
The form of the "policy" for promoting innovation in rector design should have taken the form of instructing NRC to come up with a cost benefit methodology, subject to review by Congress.
I forwarded your column to my good friend, Joe Williams, a nuclear engineer who retired in the last few years from his employment at the NRC. Pasted below is what he wrote in response. I submit this with his permission, and he also volunteered to talk to you, Matt, if you have more questions or want to discuss this with. him further. So, here's Joe's take on your column-
Matthew Yglesias essay regarding the Nuclear Regulatory Commission’s ability to license new reactors does not present a full story. He suggests that NRC isn’t serving the public well by failing to license advanced reactor designs, but does not appear to consider the particular technical and regulatory circumstances, or the wounds the nuclear industry has inflicted upon itself.
From my perspective as a retired nuclear engineer with over 40 years of experience, the essay is one of the latest in a very long line of similar items blaming NRC for the failings of the industry. I worked for NRC for nearly 30 years, spending most of the last 20 years of my career working on new reactor licensing issues, and I am familiar with many of the advanced reactor concepts, as well as the technical and regulatory issues involved. The NRC has some considerable flaws, but that it is far from the only factor in the industry’s struggles.
Yglesias takes NRC to task for not accepting the Oklo reactor license application for review. I should add a disclaimer at this point that I briefly worked as a consultant for Oklo after my retirement, though I had no part in preparing their license application. While at NRC, I was part of the branch responsible for management of advanced reactor projects, including Oklo. I did not participate in preapplication activities with Oklo, and left NRC before their application was submitted. I believe that the Aurora design concept is very innovative, and I hope that the Oklo team can effectively address issues with the application and bring the project to fruition.
It is not apparent from Yglesias’ essay that he is familiar with the specific issues that led NRC to make deny the application, described in a publicly-available NRC letter to Oklo. The gist is that "Oklo has repeatedly failed to provide substantive information in response to NRC staff requests for additional information (RAIs) on the maximum credible accident (MCA) for the Aurora design, the safety classification of structures, systems, and components (SSCs), and other issues needed for the NRC staff to establish a schedule and complete its technical review.” NRC’s letter notes that Oklo can demand a formal hearing if they believe the application actually fulfills requirements, but you can be confident there won’t be a hearing.
Yglesias also notes that NRC has not approved any advanced reactor applications. While this is true, it is also misleading. To date, only one advanced reactor application (that is, a reactor that does not use water as a coolant) has been submitted – Oklo’s. In the 20 years I worked on new reactor licensing, I saw more than a few claims that a design would be submitted by a certain date that did not come to pass, not because of regulation, but because the claims were based on hope or hype, rather than reality (see, for example, Transatomic or Hyperion).
His statement that NRC has never approved a design from start to finish since 1975 lacks important context. For one thing, there were no new license applications for about 30 years from before the TMI-2 accident in 1979 until 2008. In the meantime, NRC created a new regulatory process with extensive input from industry stakeholders that provided for prior approval of designs and sites before construction. Starting in the 1990s, this process was used to approve several new designs, and then to approve licenses to allow construction and operation at several sites, including the one being built in Georgia. Many of the other projects were not pursued due to reduced projections for electrical demand after the 2009 recession, while another project in South Carolina failed because of untimely completion of design details by the reactor vendor and architect engineer and poor management of the huge construction project; these issues are well-documented in reporting by the Charleston Post and Courier newspaper. These problems have little, if anything to do with regulation. NRC is far from perfect (I could go on for awhile about that, too), but they aren’t the people that put a bunch of parts in a warehouse without adequate indexing so they could actually be found when they were needed, for example. And this far from an isolated example, as very similar issues have repeatedly come up in other projects dating back to the ‘70s and ‘80s. The industry gripes about NRC’s heavy hand, but the reason they suffer under that weight can almost always be traced to where they screwed up and NRC had to react. See, for example, the rather colorful regulatory history of the Tennessee Valley Authority’s nuclear program.
Yglesias makes what I think is a valid comparison between the hazards of nuclear power and other forms of electrical generation, suggesting that such comparisons should be considered in NRC’s licensing. While they may not be addressed to the extent they should be, the NRC’s review process explicitly requires addressing alternatives in the environmental review for a license.
I put a lot of time and effort towards the goal of building new reactors that can provide the country with carbon-free electricity. Nuclear technology can be a tremendous asset for mankind, but it is not the panacea some of its more fervent advocates claim. For many years, I kept a quote from Richard Feynman over my desk: “For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.” If the nuclear industry wants to be successful, they each need to be honest with themselves about their shortcomings. Otherwise, they will never get to the true source of their problems.
We can talk about NRC’s shortcomings another time.
This seems to miss the point of 90% of Matt's article and focuses on basically 1 paragraph. Clearly this was the paragraph that triggered your friend but it his reply probably would have been better received (certainly by me) had he focused less on his hot-button issue and either conceded or objected to the remainder of the piece.
While my response didn’t address every point Yglesias made, I believe it was more comprehensive than you suggest. I attempted to point out that many of the statements in the essay regarding the things NRC hasn’t done or should be doing don’t withstand scrutiny. The fundamental point is that many nuclear advocates have scapegoated NRC for the industry’s problems for decades. It is not NRC’s fault Oklo’s work did not meet the standards required by regulations to accept the application. Sorry you didn’t find my perspectives useful.
Separately, I see several comments regarding the Navy’s nuclear program. It should be no surprise that many NRC staff, including many senior managers, have served in the Navy, and are quite familiar with the standards established by Naval Reactors and have brought those principles to bear in their decision making.
Skimming through his article, a full 2/3 of it seems unrelated to the NRC. I know nothing of the NRC. I will take you at your word that he was unnecessarily unkind to them. My point was merely that your response might have found a more receptive audience had you posed it for an audience who read Matt's piece as something other than a scathing critique of NRC.
"... build a few for people who have eccentric use cases (remote communities dependent on diesel, companies that for PR reasons want to say they’re zero-carbon ...."
Or the Navy, that wants to power submarines in places where there isn't enough oxygen to burn stuff, and doesn't want to have to re-fuel or re-charge frequently ... which we already have. Nuclear-powered submarines are working out fine, aren't they?
We should probably just get rid of the NRC and give this to the Navy or some other entity with a track record of not being complete garbage.
OT
But Matt Y, having listened to you on Bulwark today, I think it might be a good idea to do a post on why you are still a Democrat.
I agreed with a lot of your points today.
The one point I would maybe push back on is when you said the GOP has moved left on some economic issues. I would say rhetorically they are not using the Ayn Rand talking rhetoric. However, have they changed on a policy level? I think no. Youngkin is fighting min. wage increases and is pursing business tax cuts.
The next GOP President and Congress will do one thing. Pass an enormous inflationary upper income tax cut. None of the rhetoric matters. Do you doubt that?
Given that our country has basically never given nuclear a real shot, it isn’t particularly instructive to look at historical deaths stemming from reactor malfunctioning. Instead, we must estimate expected deaths based on proposed reactor location(s), probability of malfunction, magnitude of nuclear event given malfunction, etc.
Matt sort of gets at this by alluding to the “high-end estimate of the worst nuclear accident ever”. But like once-in-a-lifetime Global Financial System Collapses, or once-in-a-lifetime Pandemics, we still run into the tricky problem of making decisions based on very low probability events that carry potential for catastrophic outcomes. Using Matt’s cost-benefit logic, we could just as easily argue against Dodd-Frank Wall Street Reform or legislation for future Pandemic preparedness. My point being that this topic seems a bit cherry-picked for this type of analysis.
I also disagree with the approach of treating premature deaths due to pollution (which in this case disproportionately impact an older population) the same as workplace accident deaths (which by definition happen to the working age population) or deaths resulting from exposure to nuclear radiation (which can kill infants). Such an approach ignores the fact that, to quote Chuck Palahniuk, “On a long enough timeline, the survival rate for everyone drops to zero.”
I think a better way to approach such cost-benefit analysis (to the extent that, per my first point, it’s even applicable in this case) is to evaluate expected life years lost/saved. I’m not saying it changes the conclusion vis a vis whether to expand nuclear energy programs, but I [potentially, didn’t exactly run the numbers] disagree with Matt’s suggestion of where the marginal shift occurs between nuclear and fossil fuels.
Here's a question to ponder: the advantages of factory-built nuclear micro-reactors should also apply to factory-built coal-fired "micro-turbines".
Where are they?
Not really. The micro reactors have a lifetime fuel load pre installed. The solids handling of coal does not cost scale down well. Still need to build a railroad. The fossil "micro-reactor" you are thinking of is an industrial scale diesel generator. Fuel is expensive for them
It's not as if these things won't be built. They just won't be built in the US. Fortunately there is a whole world out there capable of moving this technology forward. My money would be on the French and probably the Chinese. Germany has taken themselves out of the game. That's a shame but technology moves on.
There's an interesting wrinkle to the Oklo decision. Rod Adams made this point on the Nuclear Barians podcast, but his theory is that the NRC is now bound by law to complete their review in 3 years. By rejecting the application the NRC reset the three year clock, which is a quite nifty way to side step the law.
You can find this at the the 48:30 mark.
https://nuclearbarians.substack.com/p/advancing-nuclear-ft-rod-adams
You certainly can learn a lot from the past. As an engineer reaching the end of a decades long career I would say that the chief lesson from Chernobyl is the same one you could learn from a wood stove. Never build a wood stove out of wood.
Speaking of economics, one of the things that makes me wonder about the future of nuke plants delivered and operated by private business is that they are essentially impossible to insure.
If a private provider had to 100% fully carry their own liability risk, no project would ever be profitable. The entire enterprise pre-supposes an enormous implicit federal insurance subsidy.
It seems like the big issue with nuclear power is that we still have no good story for how the nuclear waste it produces is managed. Having a larger number of smaller reactors in service would appear to make this problem worse, not to mention increasing the risk that somebody might succeed in stealing the spent fuel in order to do (or threaten to do) something awful with it.
Yes we do. But it is an article of faith among the anti-nuclear crowd that we don't.
In the South West we have huge swaths of dirt ( just fly over S Cal) that could be covered with solar at no environmental cost. 5 GW installations are not unreasonable. You do not mention Deep Bore Hole storage of nuclear waste. This will solve the waste issue and place it where it does not need monitoring. Current drilling technology (see the Chilean mine disaster relief) is sufficient.
As a resident of the southwest (Southern AZ), while I would agree that there won't be a problem finding land to lay down solar panels, I'm hurt that your would describe our beautiful landscape as "huge swaths of dirt" :p
The southwest desert is gorgeous. But would I sacrifice a few hundred square miles for clean air and solving climate change? Yes.
This issue with nuclear technology seems to me related to this issue with PPP, https://www.nber.org/papers/w29669#fromrss, and to issues Matt has raised before with the costs of building transit infrastructure, and to issues Matt has raised with the FDA and CDC. It seems many of our government bureaucracies are not well suited for contemporary problems. Are the solutions all local to each agency, the slow boring of hard boards? Or are there bigger things Congress or the White House ought to do?
Always great to see a rationalist perspective on the cost/benefits of nuclear.
Here in Washington the same people who believe climate change will wipe out humanity are fighting tooth and nail to take out all Hydro power. Go figure.
The NRC offered me a job in 2000. I didn’t actually start the job because, on a background check, I admitted that I had smoked pot before. Just sayin.
I wonder what they are actually checking here. I would assume nearly 80% of college grad age people have tried it (I actually see a link saying 80% of 55 year olds have tried it). Are they actually looking for non-users? Are they screening out people who are inadequately conservative to lie about it?
It seems to me it would make more sense to only hire the people that say yes to screen out the liars.