Public Policy

Summary
The Energy Information Administration (EIA) will release the Annual Energy Outlook 2005 (AEO 2005) next week, and the Senate Energy and Natural Resources Committee has requested testimony about EIA’s forecasting through 2025. Although EIA’s forecasting of nuclear power’s contribution to U.S. electricity supply has improved in recent years, the Nuclear Energy Institute (NEI)¹ believes that EIA’s outlook, particularly with respect to new nuclear plant construction, is based on erroneous assumptions.

Need for Accurate Analysis and Forecasting
There is increasing evidence that the United States faces serious energy supply and delivery problems. Even assuming successful conservation and efficiency programs, U.S. dependence on imported oil is at a historic high. Natural gas prices across the country have increased dramatically. The transportation infrastructure for delivery and natural gas requires significant expansion. The transmission infrastructure necessary to move electricity within and between states and regions is seriously overloaded, placing reliability at risk.

The imminent threat to reliable supplies of energy at stable, predictable prices is generating interest in new national energy policy legislation. The appropriate authorizing committees in both the Senate and House are holding hearings. At times like these, policymakers in the administration and the Congress must have access to the most accurate analysis and forecasting possible. In the case of nuclear energy, the EIA’s forecasts are not accurate, appear based on hypothetical speculation and do not reflect realistic analysis of the current status of nuclear energy in the United States.

EIA’s Forecast for Nuclear Energy
Each year, EIA’s Office of Integrated Analysis and Forecasting publishes an annual forecast of U.S. energy supply and demand called the Annual Energy Outlook. AEO 2005 provides projections of energy supply and demand in all consuming sectors and for all fuels through 2025. This year’s Outlook projects that total nuclear generation will grow from 764 billion kilowatt-hours in 2003 to 830 billion kilowatt-hours in 2025.

This EIA projection contrasts sharply with EIA’s forecasts several years ago. AEO 1999 predicted that 50,800 megawatts of nuclear capacity would be retired by 2020. The following year, the publication predicted that 42,700 megawatts would be retired.  AEO 2001 forecast shutdown of 28,100 megawatts, and AEO 2002 forecast shutdown of 9,700 megawatts of nuclear generating capacity. These wildly divergent results were produced through a combination of analytical errors, including use of out-of-date data, imposition of arbitrary additional costs and “double-counting” of additional costs.

EIA’s assessment of the outlook for the existing U.S. nuclear power plants nonetheless has improved dramatically. AEO 2005 predicts nuclear generating capacity will increase from 99.2 gigawatts in 2003 to 102.7 gigawatts in 2025 as a result of uprates of existing plants. It projects that all existing nuclear plants will continue to operate.

The Nuclear Energy Institute commends EIA for recognizing that the 103 nuclear operating reactors that supply 20 percent of U.S. electricity will continue to operate to the end of their initial 40-year license terms and, in virtually all cases, will renew their licenses and continue to operate for an additional 20-year period. However, NEI believes there is substantial room for improvement in EIA’s outlook for new nuclear plants in the United States.

The Outlook for New Nuclear Units
The Annual Energy Outlook 2005 assumes no new nuclear power plants will be built before 2025 in the United States. The National Energy Modeling System (NEMS) model reaches this conclusion because EIA analysts have assigned an unrealistically high, and inflated, capital cost to new nuclear generating capacity. The EIA assumes new nuclear plants would have an overnight capital cost of $1,928 per kilowatt of capacity.

NEI commends EIA for its initiative, during 2003, to conduct a series of workshops on the issue of new nuclear plant capital cost. AEO 2004 included a summary of  those workshops, which reflected industry’s view that new nuclear plants in the United States could be built for $1,400 to $1,500 per kilowatt (including first-of-a-kind costs for the initial reactors of a series) and $1,100 to $1,200 per kilowatt (for the nth of a kind).  Unfortunately, these cost estimates, which have a strong factual basis, were not reflected in either AEO 2004 or AEO 2005.

The summary of the 2003 workshops in AEO 2004 acknowledged that “there is reason to believe that new reactors will be less costly to build than those currently in operation in the United States. Over the past 30 years, there have been technological advances in construction techniques that would reduce costs. In addition, the simplified, standardized and pre-approved designs clearly result in cost savings.”

EIA then ignored this finding and assumed that new nuclear plants would experience the same delays, lengthy construction periods and high costs experienced by some of the plants built in the 1980s and 1990s. Consequently, it estimated that a new plant would face overnight capital costs in the range of $1,928 per kilowatt.

The industry believes there is ample evidence to demonstrate that EIA’s approach is flawed, and that there is a reasonably solid basis for industry’s capital cost estimate. Two examples are cited here: the AP1000 design developed by Westinghouse and the Advanced Boiling Water Reactor (ABWR) developed by GE Nuclear Energy.

The Westinghouse AP1000
Westinghouse is currently pursuing Nuclear Regulatory Commission design certification of its AP1000 nuclear power plant. The AP1000 is a 1,117-megawatt advanced light water reactor. It is essentially a higher power version of Westinghouse’s 600-megawatt design, the AP600, which the NRC certified in 1999. More than $400 million was invested in developing and licensing the AP600 design, including an extremely detailed cost database comprising more than 1,900 commodity categories and 25,000 specific items. The cost estimate was verified by Westinghouse, several international architect-engineers, the Electric Power Research Institute and several utilities. A comparably detailed cost estimate was prepared for the AP1000 by modifying the AP600 estimate to reflect the design changes.

In 2002, an industry team—comprised of Westinghouse, seven major U.S. power companies and architect-engineer Bechtel—completed a $1-million re-evaluation of the AP1000 reactor design. As part of that re-evaluation, Bechtel performed a thorough review of the modifications made to the original cost estimate and, after making some minor adjustments, endorsed the AP1000 cost estimate.

Although the specific numbers are proprietary information, the overnight capital cost for building the first two AP1000 reactors at one site is less than $1,400 per kilowatt. This includes all of the first-time costs for completing design, engineering and licensing of the first project. After the first few projects have been completed, the capital cost for later plants will be approximately $1,000 per kilowatt, which is competitive with other sources of baseload electricity. Once those first reactors are built and capital costs reach the $1,000-per-kilowatt range, all future plants would be financed and built without federal government financial assistance.

The Westinghouse-Bechtel estimate of less than $1,400 per kilowatt has a solid analytical basis, has been peer-reviewed, and reflects a rigorous design, engineering and constructability assessment.

The GE Nuclear Energy ABWR
GE Nuclear Energy and its partners have built several ABWRs in Japan and are building two reactors in Taiwan (the Lungmen project). In 2002, GE and Black & Veatch (B&V) completed an independent cost estimate of the ABWR.  This study resulted in volumes of data, including quantities, vendor costs and construction labor rates. The source of information for every piece of data is referenced. Most references for quantities of materials are to the Lungmen project database, and thus accurately reflect what would be required to build a plant.

This cost estimate was reviewed and re-reviewed by GE, B&V and a U.S. utility. As the estimate was based on actual experience from current and previous ABWR projects, it is considered to be solid.

The bottom line: A single-unit ABWR could be built in the U.S. for $1,445 per kilowatt on an overnight basis. Two units on the same site roughly one year apart would have an average cost of $1,300 per kilowatt. These estimates are for a 1,450-megawatt reactor and include owner’s costs, supplier profit and contingency. These costs are slightly higher than the estimates for the AP1000 because the Westinghouse reactor incorporates a number of so-called passive safety features that reduce the total capital cost. GE Nuclear Energy also is developing a boiling water reactor design that incorporates similar advanced passive safety features. The company expects that overnight capital cost for this design will be approximately 20 percent lower than the ABWR.

In addition, EIA ignores real experience from overseas, which demonstrates clearly that new nuclear power plants are the most economic option for new generating capacity and not, as EIA suggests, the least economic. The chart above shows the results of the economic assessment conducted by TVO, the Finnish electric power company, which led to its decision in 2004 to order and build a fifth nuclear power plant. This analysis shows that a new nuclear power plant is markedly more economic than the other alternatives and lends credence to the capital cost estimates developed by the U.S. nuclear industry.

The Nuclear Energy Institute believes EIA would better serve the policy community by using real-life analysis and cost information rather than its own hypothetical assumptions, which prejudice its forecasts against nuclear power.

The continuing prejudice against new nuclear plant construction reflected in EIA’s Annual Energy Outlook has serious negative consequences. Once such example of erroneous EIA data used in the energy policy debate occurred in 2003 when the Senate evaluated whether a federal production tax of $18 per megawatt-hour for the first eight years of operation for the first 6,000 megawatts of new nuclear capacity built would stimulate new nuclear generating capacity in the United States.

The EIA report (SR/OIAF/2004-02) concluded that the production tax credit would, in fact, stimulate construction of 6,000 megawatts of new nuclear power capacity, but that further expansion beyond 6,000 megawatts would not occur because new nuclear plants still would not be economic. The EIA analysis was incorrect because EIA again used inflated assumptions about the capital cost of new nuclear power plants and rejected the logic that, as more plants are built, capital costs would decline making the next units less expensive.

It is not difficult to predict what the EIA’s NEMS model would forecast if EIA staff used more reasonable and realistic cost estimates for new nuclear plants. In 2002, the Electric Power Research Institute used the NEMS model to forecast the amount of new nuclear capacity that would be built using more reasonable capital cost assumptions than EIA. The result: At $1,250 per kilowatt, 23,000 megawatts of new nuclear capacity would be built by 2020. At $1,125 per kilowatt, 62,000 megawatts of new nuclear capacity would be built by 2020.

Conclusion
Given the potential importance attached to the Energy Information Administration’s forecasts, NEI believes it is important that these forecasts have a sound factual and analytical basis. At a minimum, NEI urges that EIA’s forecasting function would benefit from rigorous peer review of all EIA’s nuclear-related assumptions and methodologies, and peer-reviewed development of new economic models better able to simulate the dynamics of competitive electricity markets, the performance of existing nuclear power plants and the timing for construction of new nuclear units.

---

¹ The Nuclear Energy Institute (NEI) is the organization responsible for establishing nuclear industry policy on matters affecting the nuclear energy industry. NEI’s members include all companies licensed to operate commercial nuclear power plants in the United States, nuclear plant designers, major architect-engineering firms, fuel fabrication facilities, materials licensees and other organizations and individuals involved in the nuclear energy industry.