Why is INPO a far better regulator than the NRC?

After Three Mile Island, it belatedly hit all US nuclear plants that, under Price-Anderson, they were the effective insurers of each other. Price-Anderson created a suicide pact.

The industry reacted immediately and set up the Institute of Nuclear Plant Operations (INPO). No legislation, no Executive Order produced INPO. INPO is funded by the plants themselves. INPO is a completely self-regulated inspection service. It is in theory voluntary; but it would be nearly impossible to obtain insurance for a US plant that was not a member of INPO and subject to its inspections.

INPO sent out a team of inspectors to every US plant. Unlike the inept, paperwork pushers at the NRC, these guys were smart experienced plant operators.¹ They knew where the problems lie, and what to look for. Most importantly, they had skin in the game.

By sharing information on problems and best practices, they raised standards and avoided nonsensical situations such as led to TMI.² All the metrics improved, some drastically. The worst plants were pulled up closer to the best plants. Much of the increase in capacity factors in the 1980's was due to demand catching up with an over-built situation, but much of it, including the shortening of refueling outages, was due to INPO.

INPO was and is far more effective than the NRC at regulating nuclear power plants. The NRC could disappear tomorrow, and the real operations at the plant would not deteriorate at all. INPO would see to that. If INPO replaced the NRC in certifying the design and construction of US nuclear plants, the quality and safety of those plants would improve.

Unfortunately, INPO has three features, which result in cripplingly inefficient regulation.

1) INPO is motivated not by the cost of actual harm, but by the cost of preposterously unrealistic perceptions of radiation harm combined with an out-of-control American tort system. A Fukushima sized release in the US could lead to a trillion dollars in claims, despite the absence of any detectable harm top the public.

2) INPO is a creature of the plant owners, not the insurers. As a practical matter, INPO is controlled by the big nuclear utilities. They are not inclined to welcome new competition.

3) INPO is a monopoly. It does not have to compete with other INPO's for customers. One result is that it has become a bloated, paperwork obsessed apparat. A plant's INPO dues are about as large as its liability insurance premium.

(1) and (3) means there is no balancing mechanism between economy and preventing a release. Tasked with the job of preventing a trillion dollar claim, INPO has a monomaniacal focus on safety culture, which it defines as ``an organization's values and behaviors that serve to make nuclear safety the overriding priority."[emphasis mine] In 36 pages of testimony to Congress in 2019, the INPO CEO never mentions cost or economics or the harm associated with alternate sources of electricity once.\cite{inpo-2019}

INPO is based on the twin premises that a release is intolerable (on dollar grounds, not actual harm) and a proper safety culture can prevent that release. INPO is the latest embodiment of the Two Lies. Under an NRC-less, INPO based regulation, nuclear power would remain auto-genocidally expensive.

(2) and (3) are easily correctable. Just remove the middle man, and have the inspectors report directly to the insurers. This will require repealing Price-Anderson. Then enforce competition in the inspection function by allowing the plants to shop for inspection services. but the inspection services will have to be acceptable to the insurers.

This may be the least original idea ever proposed. It's the same system by which we regulate just about all hazardous industries. In ocean transportation, it's called the Classification Society system. In other industries, it's called the TUV system. The replacement system already exists.

(1) is a little harder. Congress must by law replace an out-of-control tort system with a radiation exposure compensation plan that depends on each person's dose rate profile, and nothing else. The exposure payment would be automatic, no fault, not subject to dispute, and exclusive. There would be no requirement to demonstrate any harm. Only government entities could prosecute or sue a plant for negligence.

These straightforward changes, which we call Underwriter Certification, would do just as good a job at regulating nuclear power as INPO but far, far more efficiently. If at the same time, we forced competition on the suppliers of nuclear power, nuclear would be pushed down to its should-cost of 3 cents/kWh or less. Nuclear power would automatically push fossil fuel out of the electricity generation market, except for a bit of peaking. Electrifying substantial portions of currently non-grid markets would become something more than unworkable schemes to extract taxpayer money and further impoverish the poor. The planet would be richer and cooler.

1. One manager of a university research reactor told me this story. Before an NRC inspection, he would put an obvious but inconsequential error in his NRC paperwork. The NRC inspector would catch the error, and write him up. He would apologize and promise that that mistake would never happen again. The NRC inspector could show his boss how well he was doing his job. Everybody went home happy. This guy told me he was far more concerned about the reactor's insurance inspectors than the NRC inspectors. His reactor was not part of INPO.

2. 18 months prior to TMI, the Davis-Besse plant experienced essentially the same failure as at Three Mile Island. However, the Shift Supervisor Mike Derivan went against his NRC-approved training, figured out what was happening, and prevented any damage to the reactor. This information was not distributed to the rest of the industry. So the TMI operators did not realize all they had to do was isolate a stuck open safety valve, and TMI would have been a non-event.