Underwriter Certification and the Banks

The cost of nuclear power is critically dependent on three factors:

1) Overnight Capital Cost or CAPEX

2) Build Time

3) Cost of Capital

The GKN has expounded at length on nuclear's overnight cost: why it should be less than $2000/kW and why in the west it is five to ten times that number. We have spent \whole posts on build times, showing that nuclear construction periods should be no more than 4 years, a number that the Americans bettered in the late 1960's, and the Japanese consistently achieved in the 1980's and 1990's.¹ But the cost of capital is nearly as important as Figure 1 shows As a rough rule of thumb, a 1% increase in interest increases the naive LCOE by about 20\%. And that's only if you can build it in 4 years.

Figure 1. Naive APR1400 LCOE as a function of overnight CAPEX and discount rate. Naive LCOE assumes the plant operates at full power 100% of the time it is available. Baseload LCOE might be another description.

Power plants should be an extremely attractive financing opportunity, especially if the plant has a solid Power Purchase Agreement. For low risk projects, banks are happy to lend up to 90% at a slight premium to SOFR (Secured Overnight Financing Rate) because they can borrow at SOFR and pocket the difference. The SOFR rate is typically 1 or 2% above inflation or 1 to 2% real.

But the banks need three things in order to offer financing at near SOFR interest rates:

1) Assurance that the technology can generate electricity per its design specification. For proven technologies that's not a problem.

2. Absolute protection from liability risk. If third part liability can wipe out a utility, there will be nobody left to pay the loan.

3. Protection from regulatory risk. There is no way a bank can be exposed to an omnipotent regulator who can impose new requirements and force costly backfits whenever it feels like it. A 1974 study by the General Accountability Office of the Sequoyah plant documented 23 changes ``where a structure or component had to be torn out and rebuilt or added because of required changes."\cite{ford-1982}[p 208] The Sequoyah plant began construction in 1968, with a scheduled completion date of 1973 at a cost of $300 million. It actually went into operation in 1981 and cost $1700 million. This was a typical experience.

There is no way a bank can be exposed to an omnipotent regulator who can shut down a plant whenever it feels like it, for example, a problem at some other plant. The NRC has shut down whole fleets of nuclear plants for months on multiple occasions in response to relatively minor problems at a single plant. In 1979, the NRC shut down five Atlantic Coast plants at the height of the Second Oil Crisis for a software bug that came into play only if the plants were hit with the strongest earthquake in the historic record before the next planned shutdown. The Japanese shut down all their nuclear plants for over a decade in response to the Fukushima release. Many of these plants were nearly new. Many were not exposed to tsunamis.

This can be tolerable to the banks only if they know the borrower can push off his additional costs on to the rate payer. But in a deregulated market this is a show-stopper.

The key to (2) is that the banks have assurance that the insurance will cover any third party liability. This can only be the case if the compensation is very well defined.

Unless the plant can transfer the cost of any regulatory changes onto the rate payer as is the case in a regulated monopoly, only Underwriter Certification can satisfy the banks with respect to (3). They are familiar with the system. They've seen how it balances benefit and potential harm. They are comfortable with the system. UCert is the only way of obtaining private financing of a power plant for anything but a regulated monopoly.

It's really quite simple. You cannot have an omnipotent regulator, a deregulated market, and private financing of nuclear power. End of story. No amount of NRC "reform" can change that.

1. Moreover, nuclear plants can located where they are needed, obviating the need for major new transmission lines.