Read the white paper below or download the pdf here.
Presented at the Western Energy Policy Research Conference
Sponsored by the Center for Advanced Energy Studies (CAES) and the Energy Policy Institute (EPI)
Nuclear program director
Snake River Alliance
Title XVII of the Energy Policy Act of 2005 authorized the US Department of Energy to support innovative clean energy technologies that are typically unable to obtain conventional private financing because of high technology risks. The government support comes in the form of loan guarantees.
Before a closer consideration of the nuclear loan guarantee program and its goals and effects, it is useful to consider some historical context.
Nuclear construction has always been daunting. According to “Federal Loan Guarantees for the Construction of Nuclear Power Plants,” a 2011 study by the Congressional Budget Office, “cost overruns averaged more than 200 percent for plants on which construction started between 1966 and 1977. … In total, 40 projects were canceled after construction had started.” Even today, nuclear construction poses tremendous financial uncertainty. Costs will rise because of myriad factors – material costs, limited suppliers for key parts, and the complexity of the technology itself (eg, huge vessels with exceedingly fine tolerances). Construction delays and cost overruns are mutually reinforcing and virtually certain.
Balancing the high historic and projected construction costs is the fact that nuclear power has benefited from extensive public and ratepayer subsidies. The industry has always argued these are temporary bridges. In fact, a 1954 General Electric advertisement said that “the atom will produce power for homes in 5 to 10 years…without government subsidy.” That prediction proved hollow, and subsidies have been a steady and crucial factor in the industry’s survival.
In 2011 the Union of Concerned Scientists published an analysis entitled Nuclear Power: Still Not Viable without Subsidies. It attempted to take into account all the costs and risks shifted from the nuclear industry to government and ratepayers over the past fifty years. UCS first evaluated “legacy” subsidies, which supported nuclear construction through the 1980s but no longer significantly affect pricing.
UCS then looked at ongoing subsidies to existing reactors. One of the most notable subsidies is the Price-Anderson Act, which significantly shifts accident and security risks from industry to taxpayers. Price-Anderson limits the primary insurance a nuclear plant owner is required to carry to $375 million. In the event of an accident, each nuclear plant owner is assessed a prorated amount for a pool totaling $11.6 billion. A sense of how much Price-Anderson support might be worth comes from post-Fukushima damage estimates. German policy makers projected that an accident on the scale of Fukushima could cost $11 trillion. Other ongoing subsidies underwrite uranium mining and enrichment. UCS also looked at the underpricing of the use and degradation of water. Though about $31 billion has been collected from ratepayers for the Nuclear Waste Fund, a repository is expected to cost $100 billion over its operating life. In addition, industry will probably not shoulder the full financial burden of decommissioning its reactors. All these subsidies do affect the current cost of electricity.
Finally, new nuclear construction can be subsidized today through federal loan guarantees, accelerated depreciation, and subsidized borrowing through municipal bonds and Build America Bonds that are available to publicly owned utilities. New reactor construction also can be supported by local property tax abatements and, in many states, by construction work in progress (CWIP) surcharges to customers long before a plant is completed and producing electricity. The 2005 Energy Policy Act added an important new public support and grants production tax credits for each kilowatt-hour generated during a reactor’s first eight years of operation. All the new subsidies will add to the cost of nuclear power.
Based on its full accounting, the Union of Concerned Scientists reached a fairly startling conclusion: “Subsidies to the nuclear fuel cycle have often exceeded the value of the power produced. This means that buying power on the open market and giving it away for free would have been less costly than subsidizing the construction and operation of nuclear power plants.”
Without question, nuclear power more than qualifies for loan guarantees under the Energy Policy Act of 2005 if the only requirement is that the technology is “typically unable to obtain conventional private financing due to high technology risks.” But a further, important requirement from an energy policy perspective is that the technologies targeted for support must avoid, reduce, or sequester air pollutants or anthropogenic greenhouse gas emissions. Nuclear power is covered under Section 1703 of the 2005 Energy Policy Act. Section 1705 covers biomass, hydrogen, solar, wind/hydropower, nuclear, advanced fossil energy coal, carbon sequestration practices/technologies, electricity delivery and energy reliability, alternative fuel vehicles, industrial energy efficiency projects, and pollution control equipment.
In addition, to underline the importance of innovation, the act excludes technologies with more than three implementations that have been active for more than five years.
In 2008, Congress appropriated $18.5 billion for nuclear reactor construction and a further $2 billion for construction of nuclear front end facilities – uranium enrichment plants. The Department of Energy added to the front end funding pot with an additional $2 billion allocation. The Administration asked Congress for more funding in the following years, but those requests were not successful. No additional loan guarantees were included in the budget request for fiscal year 2013.
No new applicants for nuclear loan guarantees were accepted after September 2008. By then, the DOE had received 19 applications from 17 different corporations for the construction of 14 nuclear power plants. Each power plant can have more than one reactor. The dollar value of those requested loan guarantees was $188 billion. To date, no nuclear loan guarantees have been awarded, though one has been offered and rejected and two, worth $10.3 billion, are conditional.
Congress ultimately appropriated $18.5 billion for Section 1705 energy efficiency and renewable energy projects. Despite some headline difficulties such as the Solyndra default, that part of the loan guarantee program has been far more successful. The Department of Energy has granted 17 loan guarantees (and 6 partial guarantees) totaling $9.6 billion to projects across the country in solar manufacturing and generation, biofuel, energy storage, wind generation, transmission, and geothermal. A key distinction of the Section 1705 program is that the money it has distributed has already led to the production of low carbon electricity in seven of its generation projects.
The mechanics of the DOE’s nuclear loan guarantee are fairly straightforward with a few distinctive elements. The loan guarantee can cover up to 80 percent of the cost of the project. The other 20 percent must come either from raising equity or borrowing elsewhere. The equity can come from shareholders, but an increasingly attractive mechanism for regulated utilities is a construction work in progress (CWIP) surcharge on customer bills. Progress Energy is a case study of how valuable CWIPs can be to utilities contemplating nuclear construction. The price tag for Progress Energy’s reactor along Florida’s northwest Gulf Coast has ballooned from $5 billion to upwards of $24 billion, and a Tampa Bay Times investigation concluded the new reactor project could add $50 a month to electric bills because of the state’s CWIP surcharge.
In bankruptcy proceedings, debt holders get paid first, and those with the “right of first lien” are the very first in line. But another distinctive element in the loan guarantee program is that the DOE no longer requires the US government to hold the right of first lien. Simply put, US taxpayers will have to share any proceeds from a liquidation in the event of a loan default. That’s assuming, of course, that there’s much of salvageable value in a half finished nuclear reactor.
A third unusual characteristic of the nuclear loan guarantee program is that the US DOE is required to charge and collect a credit subsidy fee from the borrower. This upfront requirement, which is not in effect for the 1705 loans, has been a fundamental point of contention.
The credit subsidy cost is projected using an Office of Management and Budget model that is not made public. But the projection is essentially based on the likelihood of default and the amount a lender could recover afterwards. The guarantee is that the government will make up the difference between the amount lent and the amount recovered. The credit subsidy cost is usually quoted as a percentage of the loan guarantee.
In order to fully protect US taxpayers, the credit subsidy cost must be covered by the credit subsidy fee. Borrowers, of course, want the fee to be as low as possible. The projected fees for nuclear construction loans range from 1 percent or less (from the industry) to 30 percent. The high end reflects an expected 50 percent chance of default on a nuclear loan (which is a consensus of the Congressional Budget Office, the Government Accountability Office, and Standard and Poor’s). It also reflects an assumed recovery rate of 50 percent, which might be unreasonably optimistic if default occurs early in construction and might not fully account for the fact that US taxpayers no longer have the right of first lien.
The debate surrounding the credit subsidy fee is no small matter. Looking at the $22.5 billion available for nuclear loan guarantees, a one percent shift in the credit subsidy fee represents $225 million of risk to US taxpayers or cost to nuclear developers. Another problem from a policy point of view is that the modeling and negotiating of the credit subsidy fee, and indeed of everything else about the loan, is not made public even though it is public money that is ultimately at risk.
According to 2010 Congressional testimony by the Center for American Progress Action Fund, the Congressional Budget Office expects the DOE to set credit subsidy fees one percent lower than the cost. It further posited that political pressure on the DOE to issue nuclear loan guarantees would likely force it to set the credit subsidy fee “significantly lower than the true costs.”
In its 2011 study, the Congressional Budget Office deemed four reactor loan guarantee applicants to have advanced the most, but noted that “it is not clear that any of them will reach a final agreement.”
In October 2010, the DOE offered a loan guarantee to Unistar, a joint venture of Constellation Energy and France’s Electricite de France, for a new reactor at Calvert Cliffs, Maryland. The project’s price tag was $10 billion. The DOE offer was based on a credit subsidy rate of 11.6 per cent, which would have required Constellation to put up a $880 million credit subsidy fee. Constellation rejected the offer on the grounds that the fee was too high, but did not try to renegotiate. CBO, which posited the guarantee might have been offered at a below-market rate, agreed with some press accounts that Constellation’s decision might also have been influenced by low electricity prices in the region.
Two new reactors at the VC Summer plant in South Carolina have been licensed by the Nuclear Regulatory Commission, and Scana Corporation has applied for a loan guarantee for the $9 billion expansion. The project is already $300 million over budget. In addition, two new reactors are slated for construction at the South Texas Project by NRG Energy and Tokyo Electric Power Company (TEPCO), which operated the Fukushima reactors in Japan. The two partners applied for a $7 billion loan guarantee, but NRG Energy has now withdrawn.
The nuclear reactor construction plan that is the farthest along is at Plant Vogtle in Georgia, where Southern Company’s subsidiary Georgia Power and regional cooperatives and municipal utilities are building two new reactors. The DOE announced a conditional loan guarantee for $8.3 billion in early 2010. Though the plant has received its license from the Nuclear Regulatory Commission and early construction activities have begun, the loan guarantee has yet to be finalized.
The Southern Alliance for Clean Energy sought to learn the terms of the credit subsidy fee negotiations for Plant Vogtle under the Freedom of Information Act. After two years of litigation, recently released documents reveal that Southern Company’s credit subsidy fee may be as low as .5 to 1.5 percent, though its partners seem to have been offered fees ranging from 2.5 percent to 11.1 percent. Statements from both Southern Company and the Nuclear Energy Institute, the nuclear power trade group, indicate that the terms the DOE is able to offer might not be acceptable to Southern Company. Though Southern Company has said it might proceed without a federal loan guarantee, it revealed in a May 2012 filing with the Securities and Exchange Commission that construction may cost $900 million more than the current $14 billion projection.
Two companies applied for loan guarantees to back new uranium enrichment plants. Each asked for $2 billion, which was the full amount available at the time. The Department of Energy then added another $2 billion to the uranium enrichment pot. One application came from the United States Enrichment Corporation for a plant in Ohio. The DOE determined the plant proposed by USEC had serious technical problems, and USEC withdrew its loan application. Since then, the company has benefited from a series of appropriations for a wide range of activities as it tries to improve its technical approach.
Areva bills itself as the largest nuclear company in the world and is owned by the French government. In May 2010 it received a conditional loan guarantee for a uranium enrichment plant in eastern Idaho. Since then the plant has been licensed by the NRC, but Areva has faced serious financial problems after its aggressive nuclear expansion efforts met a global slowdown and then the Fukushima disaster. In 2011, its net loss was €2.42 billion, and the Idaho plant is on hold with no indication its loan guarantee is near finalization. It’s worth noting that the technology Areva proposes for its Idaho plant is nearly identical to that of a new enrichment plant built in New Mexico by Urenco without public funds.
No nuclear loan guarantees have been finalized. Federal loan guarantees have not led to any new nuclear build. On the other hand, no nuclear project has been stopped solely because the DOE refused to guarantee a loan. Though nuclear power’s economic survival continues because of a whole host of subsidies, the policy goal of the 2005 Energy Policy Act – innovative technologies producing low-carbon electricity – has not been met. But the program hasn’t stalled because of the difficulties reaching agreement on loan terms, though those have clearly been substantial. Low natural gas prices and lower electricity demand have challenged all new generating capacity.
Nuclear power faces additional obstacles, and the nuclear road might just be too steep. We’ve spent decades trying to surmount some of nuclear power’s endemic flaws: it has long project gestation time and high capital costs, its technology can lead to nuclear weapons proliferation, and nuclear pollution is very difficult to remediate. So far, no country on earth has fully resolved how to dispose of nuclear power’s most dangerous waste. At the start of what nuclear proponents hoped would be a renaissance, a price on carbon seemed to be a realistic expectation. The Fukushima disaster shook confidence in nuclear power, as it should have. In the short term, at least, a recent court decision about waste disposal put new and renewing licenses on hold.
But the answer is not for the federal government to take a hands-off approach. What’s at stake is too important: energy is a fundamental public good. But the answer isn’t to make it easier to get nuclear loan guarantees, either. Though we’ve focused more here on the nuclear loan guarantee program, it’s undeniable that the technology itself is the largest single challenge to renewed deployment of nuclear power. The answer may therefore be to back winning horses instead, technologies that reduce carbon and enhance US energy production.
General Electric promised subsidy-free nuclear power half a century ago. In July, its CEO, Jeff Immelt, acknowledged, “It’s just hard to justify nuclear, really hard. Gas is so cheap and at some point, really, economics rule. So I think some combination of gas, and either wind or solar…that’s where we see most countries around the world going.”