The Clear Case for CWIP – A Rebuttal to Mark Cooper’s Analysis on “Advanced Cost Recovery”

Uprate at St. Lucie impossible without CWIP.

Two nuclear critics, Peter Bradford and Mark Cooper, recently published a report (pdf) explaining how “advanced cost recovery” for nuclear plants in Florida and South Carolina “creates another nuclear fiasco.” Cooper’s main argument seems to be that Construction Work in Progress (CWIP) shifts to ratepayers all of the risks of building nuclear plants. This is either a deliberate distortion or a misunderstanding of how the cost recovery mechanism works.

How “advanced cost recovery” (aka CWIP) works

When a utility builds any type of project, it uses a mix of debt and equity to pay for the construction. The debt comes from banks and other investors and, of course, the utility must pay interest to use the debt. The equity comes from the utility’s shareholders and also requires a return for its use. The CWIP financing mechanism, which is also allowed by the federal government for interstate transmission projects, allows a company building a project to include the interest on debt and a return on investor’s equity (known as carrying costs) in its electric rates while the project is under construction.

SCANA’s latest filing to the South Carolina’s public utility commission (pdf), Appendix 3 (page 27) explains how CWIP works for the two-reactor expansion at its Summer station. These carrying costs are classified in accounting and electric-utility parlance as “allowance for funds used during construction” (AFUDC). When CWIP is allowed, AFUDC is basically what the power company adds to its rates, and collects from the ratepayers during construction. The AFUDC in SCANA’s case is $238 million.

What’s also shown on page 27 is the yearly and cumulative amount of cash flow SCANA needs to build the two reactors. This is currently $5.5 billion and is funded by debt and equity. Cooper neglects to explain that SCANA’s customers are paying only about five percent of the construction costs of the projects (the AFUDC rate) during construction, while the owners and investors are contributing 95 percent.

Cooper also neglects to explain another significant feature of CWIP. Because the interest on debt and return on equity are being paid each year, these carrying costs are not accumulating as they would in the absence of CWIP, to be recovered when the plant begins operation. Because these costs, like all debt, compound over time, the quicker they are paid, the less they build up. When ratepayers pay these carrying costs while the plants are under construction, customers will save substantial sums over the life of the units. From Georgia Power:

“Approximately $0.5 billion of these customer benefits are associated with the inclusion of CWIP in rate base.” – p. 36 from Vogtle’s 8th Construction Monitoring Report (pdf)

From SCANA:

“Paying financing costs while construction is ongoing, as opposed to waiting until the project has been completed, will lower the cost of the new units by about $1 billion. We estimate this reduction will save our customers approximately $4 billion in electric rates over the life of the new units.” – SCANA Press Release

Peter Bradford asserted that these cost savings are false and the plants will cost customers more money in the end, but (typically) failed to provide any explanation of or factual support for these statements. A total of ten states – and the federal government – allow CWIP today because the legislators and public service commissions in these states know that such a cost recovery mechanism saves money for its customers.

“Ratepayers are on the hook”

Cooper’s paper claims that “ratepayers are on the hook” and taking all the risk of building a nuclear plant because of CWIP. Apparently, Cooper does not understand how utilities build and recover the costs of power plants in regulated states.

In regulated states, when a power plant is completed, the utility requests approval from the Public Service Commission to place the cost of construction in its rate base, which means that the utility begins charging its electric customers for the construction costs. State public service commissions analyze the costs to determine if they were spent “prudently.” If the PSC deems the costs to have been prudent, then they can be passed on to ratepayers. This occurs with or without CWIP.

When CWIP is allowed, public service commissions exercise even greater oversight and cost control. Costs are audited, reviewed and subject to a prudence determination by the PSC, typically on a quarterly or semi-annual basis. The utility must also demonstrate the economic viability of the project before it even begins construction. 

Cooper and his fellow anti-nuclear critics contend that CWIP allows utilities to collect costs without having to build the plant. Cooper’s paper claims that they can get halfway through the project, cancel it, and then collect the incurred costs of construction from ratepayers without having to pay anything. This is of course an erroneous view of how states govern projects.

Cooper’s paper cites an Iowa Utility Board analysis to make its case against CWIP. Last year, the Iowa legislature debated a bill that would allow CWIP which may have provided MidAmerican Energy the opportunity to build a nuclear plant in the state. The Iowa House passed the bill but the Senate didn’t act on it. There was much discussion about the issue and MidAmerican provided a short explanation of the bill (pdf). The explanation clearly states that the utility does not have the unilateral ability to cancel a project. State public service commissions must approve both the decision to build a plant, the decision to cancel one, and how much utilities can collect in rates based on the prudency of spending.

Cooper’s paper claims that CWIP puts all the risk on ratepayers. CWIP actually lowers the risks to ratepayers because it makes the projects more affordable to build, increasing the chance of successfully completing the project.

Most Economical

The five reactors under construction in the Southeast U.S. (Summer 2&3, Vogtle 3&4, and Watts Bar 2) are being built because they’re the most economical projects for those utilities. Cooper’s paper claims that there are many more economical alternatives than those plants. The utilities and state regulators in that region disagree.

Renewable resources in the Southeast are thin and natural gas plants (despite today’s low costs) are still not cheaper than the nuclear plants in the long-run. Utilities are also looking to diversify away from coal, and improved efficiency only gets a utility so far before new capacity is needed. Here’s what Georgia Power says on page 45 (pdf):

“The weighted average expected value of the relative savings for completion of the [Vogtle] Facility as compared to the gas-fired [combined cycle] alternative is $4.0 billion ….”

Paying it Forward

CWIP provides clear cost benefits for customers when building capital-intensive projects like nuclear plants. The two nuclear critics in this case simply do not like nuclear power and continue to misinform and skew the debate about the benefits of CWIP. Here’s their paper’s foregone conclusion on page viii:

“Nuclear power is inherently uneconomic because it relies on a catastrophically dangerous resource that is vulnerable to human frailties and the vicissitudes of Mother Nature.”

When one begins with such a conclusion, it’s easy to guess the rest of the paper.

Many generating plants and transmission lines operating today – not to mention the interstate highways, hundreds of bridges, and dams – have largely been paid for by customers who are no longer on the system. All customers benefit today from cost-effective capital that was deployed years ago. CWIP is not new and, as shown above, allows customers the benefit of lower electricity costs from a reliable, affordable, emission-free source of power that can last 60 to 80 to 100 years. Very few, if any, other technologies can provide such benefits.