High Costs, Lower Demand Plague US Nuclear Industry

In spite of its numerous problems, nuclear power is not dead, or even dying. However, the prognosis varies for U.S. plants in different stages (that is, for those in operation, under construction, and on order) and for different countries.

In the United States, nuclear plants remain controversial. Two plants in the Northeast—Seabrook in New Hampshire and Shoreham on Long Island—are idle primarily because local and state authorities refuse to submit emergency plans. Meanwhile, the Nuclear Regulatory Commission (NRC) has attempted to circumvent these officials by proposing that plans developed by the utilities be considered acceptable.

Cost also continues to be a matter of major concern, with the bill for some newly completed plants now exceeding $3 billion. And locating permanent repositories for the waste generated by nuclear plants caused bitter arguments between the Department of Energy (DOE), the Congress, and the states (see "U.S. Nuclear Waste Program at an Impasse," p. 1 of this issue).

Despite its problems, U.S. nuclear power has made some recent gains. In early 1987—for the first time in almost ten years—a U.S. firm, Combustion Engineering, was awarded a contract to build two new nuclear plants (which will be located in South Korea). In addition, a DOE study published in December 1986 estimated that some regions of the United states will be short of electricity-generating capacity by the late 1990s, a projection that will be used by advocates to argue for more nuclear power.

Thorny issues

Although nuclear power is not at the top of the congressional agenda, several issues in this field may receive legislative action by 1988. They include standardization of reactor and plant design, plant licensing, emergency planning, damages compensation, and NRC reorganization.

Many different designs are used for building nuclear reactors and plants in the United States. For the reactors themselves there are three designs by General Electric, at least three by Westinghouse, several by Babcock and Wilcox, and several by Combustion Engineering. The remainder of a plant, which can account for three-quarters of the cost, is created by architect and engineering firms (A&Es). Seventeen A&Es have worked on nuclear plants in the United States, often developing a design specifically for a particular plant, so that there are now more than sixty plant-reactor designs.

The problems associated with building and reviewing so many designs—and the contrasting simplicity of the apparently successful French program—have led to calls for legislation to standardize nuclear plants built in the future. And, although standardization offers potential advantages, legislative action on this issue will be difficult for three reasons:

1. The NRC does not appear to view standardization as an essential prerequisite to achieving adequate safety, which raises the question of how legislating standardization would be justified.
2. If every manufacturer and major A&E were allowed only one design, the United States could still have about fifty "standard" designs. Clearly, the problems of legislating one or only a few standardizing designs would be great.
3. Agreement on criteria for selecting a design will be difficult to achieve. Questions involved here are whether existing designs or those currently under development would be acceptable, or whether a new set of criteria would be developed and, if so, by whom, using what guidelines.

Plant licensing is another complex issue that may be the subject of legislation by 1988. At present, obtaining a license is a two-step process. First a construction permit is issued, and then much later an operating license hearing is held, after which a decision is made to grant or to withhold the license. It has been proposed that this process be changed to "one-stop"—one-step—licensing, but there are serious questions about whether a single step is adequate and, if so, about when and how it should occur.

Publicity surrounding cases of poor construction and concern about adequate emergency planning make it likely that any legislation passed would require at least a review before operation began. Requiring only a review rather than a hearing after the completion of construction certainly would make life easier for utility executives and would reduce legal fees. However, it would not prevent the NRC from imposing conditions for operation. A "one-stop" hearing after a plant had been built would be financially unacceptable, because people would not be willing to invest construction money if there was no government review until the plant was ready to operate. Pressure from industry and DOE to adopt "one-stop" licensing is lessening, because fewer see it as the panacea needed to bring back nuclear plant construction orders.

Emergency planning is a third issue receiving close scrutiny. Following the Three Mile Island (TMI) accident in 1979, the NRC established a rule requiring each plant to have an approved emergency plan. Because the responsibility for developing this plan rests with state and local governments, the rule has become a mechanism for state or local officials to prevent plants from operating—as in the cases of the Seabrook and Shoreham plants.

The nuclear industry and some NRC officials have been urging relaxation of the emergency plan rule. They say that the source term—the amount of radioactivity that scientists calculate would be released in a nuclear accident—is less than is generally believed and that the emergency planning zone can, therefore, justifiably be reduced in size.

After several years of research on the release of radioactivity in nuclear accidents, the NRC asked the American Physical Society for a review of current knowledge on this subject. The Society's review supported NRC's contention that there had been resolution of substantial uncertainties about the release of radioactivity after a nuclear accident. But it pointed out that other uncertainties remain and concluded that it would be imprudent to reduce the source term at this time.

Efforts to relax the emergency plan rule were further hampered by the Chernobyl accident, which demonstrated that emergency response to a nuclear accident can be critical. More recently, some NRC officials have proposed to change the rule to allow the local utility to plan the off-site emergency response if state and local government officials will not do so. If the commission approves the proposal, the new rule undoubtedly will be challenged in court, and legislation may be introduced to cancel the rule change.

Damages compensation for victims of nuclear accidents is another issue under review. Discussion has centered around Price-Anderson—section 170, entitled "Indemnification and Limitation of Liability," of the Atomic Energy Act—which establishes a damages compensation fund for such victims. The provision that covers accidents at new plants expired on August 1, 1987, which is of particular concern to manufacturers in the nuclear industry. Another provision that applies to plants already in operation is still in effect. Congressional action on this topic is scheduled to take place before the end of 1987.

Meanwhile, understanding of how Price-Anderson works, what is covered, and what the restrictions are remains surprisingly limited. At the 1985 NRC meeting that was held to prepare a Price-Anderson report to the Congress, there was obvious confusion over these matters among NRC staff members and commissioners, representatives from the nuclear industry and nuclear insurance companies, intervenors, and state government officials.

Supporters of Price-Anderson from industry argue that the section is unique in that it requires all members of the industry to assume responsibility for their colleagues. (An accident at any reactor run by a private utility would mean that plant owners throughout the United States would be assessed $5 million for each of their reactors to compensate for damages.)

These industry groups also argue that although the section limits the liability of the utilities, it does not limit the amount of compensation that can be paid to the victims. Implicit in this argument is the belief that, following a catastrophic accident, Congress would use federal funds as needed to make up the difference between the compensation paid by utilities and the total amount of damages.

Despite their concerns about Price-Anderson, intervenor groups have acknowledged that it does offer the advantage of consolidating parties and jurisdiction—that is, all parties who wish to sue are required to do so in a single group and in the same court. This court then addresses only two issues: what damages were caused, and whether or not the accident actually caused those damages. Largely because of consolidation, these issues should be resolvable more quickly than, for example, large numbers of lawsuits filed by individuals after an airplane crash.

When Price-Anderson extension comes up for congressional action, several essential changes should be made. Certainly the current cap of about $750 million on funds for compensation should be raised, particularly in view of the fact that the Atomic Energy Act of 1954 set a $500-million cap on compensation, and this amount was arbitrary in any case. In addition, all parties would benefit from a rewriting of the section to clarify meaning which is so murky that a rapid court process is highly improbable even with the current system of consolidation.

Perhaps one of the most fundamental U.S. nuclear regulatory issues to be resolved is that of NRC reorganization. Every thorough review of the NRC has concluded that it is poorly managed, primarily because of its structure, not its personnel. Each NRC chairman since 1977 has recommended that the commission of five members be replaced by a single administrator.

Others outside the NRC have concurred: John Kemeny, chairman of the presidential review of the TMI accident in 1979, strongly recommended replacing the commission by a single administrator, as did Mitchell Rogovin, director of the independent TMI review for the NRC in 1980. The 1986 National Research Council study on nuclear safety research essentially recommended a single administrator. Also in 1986, Senator Alan K. Simpson (R-Wyo.), then chairman of the Senate NRC oversight subcommittee, introduced a bill to make the change to a single administrator. Representative Morris K. Udall (D-Ariz.), chairman of the House NRC oversight committee, has also said he is in favor of a single NRC administrator.

Reduced demand

Opponents of the U.S. nuclear power program argue that the plants are unsafe or too expensive. And, while costs are the second most serious cause of problems in U.S. nuclear power, the primary cause of the ills of the program is reduced demand for electricity. This is not a dramatic reason—and it has no moral overtones—but its effects are far-reaching.

The federal government predicted in the 1960s and 1970s that demand for electricity would grow continuously at about 7 percent per year. This prediction led the Atomic Energy Commission (AEC) and its successor, the Energy Research and Development Administration (ERDA), to forecast large numbers of nuclear plants for the United States. In 1974 the AEC estimated that about 1,000 large nuclear plants would be built in the United States by the year 2000. In 1977, even after the first oil shock and subsequent economic adjustments, ERDA still predicted that there would be about 380 large nuclear plants by the year 2000. In fact, it is unlikely that the total will be more than 110.

The effects of these overestimates of demand are now being seen as many U.S. utilities are either working off excess capacity or trying to absorb the costs of new large plants that are not yet needed. In 1982 the Tennessee Valley Authority (TVA) cancelled four reactors in which TVA had invested a total of almost $2 billion. At the time the plants were ordered, TVA had forecast an annual growth in demand of about 7 percent, although demand had actually been averaging closer to a 1 percent increase annually. Also in 1982, the Long Island Lighting Company (Lilco) proposed a phase-in plan for its Shoreham nuclear plant that would raise electricity rates by 50 percent in three years.

Large increases in rates such as those proposed by Lilco are leading to federal-state conflicts. For example, in January the D.C. Court of Appeals supported the Federal Energy Regulatory Commission (FERC) against the state public utility commissions of Arkansas, Mississippi, and Louisiana. The issue was FERC's allocation of $3.5 billion for the Grand Gulf nuclear plant to the four operating companies in the Middle South Utilities system. The public utility commissions had opposed this allocation primarily on grounds that those in the state who purchased electricity should not have to pay for unneeded capacity.

Some utility planners and many spokespeople for the nuclear industry are warning that the United States will run short of electricity. The standard measure of adequacy is the reserve margin, that is, the capacity to generate electricity beyond the maximum predicted need. Because of planned and unexpected shutdowns and unforeseen demand, generally approved operating practice is to have a reserve margin somewhat above 20 percent, perhaps as high as 22 or 23 percent. The reserve margin for the entire United States in 1986 was 33 percent.

However, a December 1986 DOE report predicted that the U.S. reserve margin would drop to 22 percent in 1995. Another forecast from a large regional group of utilities, the New England Power Pool, used 2.2 percent annual demand growth as the basis for forecasting a 2,000-megawatt shortfall below a 22 percent margin in the year 2000.

These predictions are not dire, although the DOE does predict a reserve margin of less than 1 percent in some regions by 1995. Furthermore, several factors that could affect the reserve margin remain uncertain, namely, increases in supply from improvements in interconnection between utilities in various regions, additional imports from Canada, more cogeneration, and continued low growth in demand resulting from further conservation measures.

Big bills

Cost is the second most important source of the difficulties besetting the U.S. nuclear program. The costs of power plants (nuclear, coal, wind-powered, and other types) are described in terms of construction costs and operating costs.

The standard method of describing construction costs is in dollars per kilowatt (the total cost to build a plant divided by its electric capacity in kilowatts). All interest charges on the money borrowed to build a plant are included.

Operating costs, in cents per kilowatt hour, are levelized costs; that is, the costs are allocated over the assumed thirty-to-forty-year lifetime of the plant. Operating costs include the construction cost, operating and maintenance costs, fuel costs, and, most recently, decommissioning costs for nuclear plants. The annual income necessary to cover these costs is then used as the amount that must be recovered by electricity sales, and the necessary cents per kilowatt hour is calculated accordingly.

Nuclear power plants in the United States are now known for high costs. Not so well known is the wide variation in costs for individual plants. The construction costs of two recently licensed 1,000-megawatt nuclear plants differed by a factor of 2.5: $4,000 per kilowatt for the Public Service Hope Creek plant in New Jersey and $1,600 per kilowatt for Duke Power's Catawba 2 in South Carolina. It is estimated that the Nine Mile Point plant in New York, not yet operating, will cost at least $5,300 per kilowatt.

Nuclear power has long been defended as having high capital costs offset by low lifetime costs. But 1984 was a bleak year for nuclear power: for the first time the levelized costs for large electricity-generating plants in the United States showed coal to be a cheaper source of power than nuclear energy.

While reduced demand and high costs are the two major causes of the difficulties in the U.S. nuclear power program, three other problems—involving operating practice, management, and public support—also have far-reaching effects.

Overall, operating practice at U.S. nuclear power plants has been poor, although some individual utilities do perform very well. One measure of an electric power plant efficiency is capacity factor—the percent of the time the plant is available to produce power. U.S. plants have low capacity factors; for the two-year period 1983-84, they averaged only 56 percent, compared with 75 percent for overseas plants. By the same capacity-factor measure, all Canadian plants and three-quarters of French plants were in the top one-half worldwide in 1983-84, compared with only one-quarter of U.S. plants. Of the top twenty plants in the world in terms of capacity factor, only three were in the United States, and of the bottom twenty, ten were U.S. plants.

Weakness in U.S. operating practice was clearly seen in the Three Mile Island accident, and, oddly, was reflected in the Chernobyl accident. Although there are great differences in the economic systems of the two countries and in their reactor designs, the causes of the accidents were very similar: complacency on the part of operators, management, and government; operators taking a series of deliberate steps that defeated the safety systems; operators not understanding their plant; inadequate operator training; and weaknesses in approving operating procedures. In the United States, these problems are symptomatic of a fundamental weakness that shows up in low capacity factors, poor construction, and erratic operation—that is, poor management.

Poor management is the most pervasive and insidious problem in the U.S. nuclear industry. The personnel running nuclear utilities should be technically trained. Utilities should treat plant operators as air-lines consider their pilots—as highly paid professionals who hold in their hands not only public lives, but the financial life of the company.

However, the fatal flaw for nuclear power in this country is its inability to gain public support. In the United States, public support is necessary for major programs to be successful. Nuclear power has failed to get that support. Opinion polls from 1975 to 1986 showed a steadily increasing disavowal of nuclear power in the United States in addition to the expected increase in opposition immediately after the Three Mile Island and Chernobyl accidents. A poll taken in mid-1986 showed that nearly 80 percent of the public opposed the building of more nuclear power plants in the United States.

Finally, nuclear power must live with the problem of proliferation. Although there are only indirect links between nuclear power technology and nuclear weapons, the public does associate the two. Consequently, what happens in terms of proliferation affects nuclear power.

Nuclear weapons have not proliferated at the rate predicted twenty years ago. The reasons include forecasts by people who did not understand how difficult it is to make a nuclear weapon; tough U.S. and Soviet stances against proliferation, leading to formal and informal arrangements to restrict the availability of nuclear weapons material and know-how; and the actions of the International Atomic Energy Agency (IAEA).

Although the IAEA has been under strong attack in the Congress for not being able to guarantee nonproliferation, the agency's defenders argue that such an objective is an unreachable goal for an international body. The IAEA does provide the best mechanism, both existing and postulated, for constraining the spread of nuclear weapons.

Nevertheless, criticism of the effectiveness of the IAEA and challenges to non-proliferation continue. Currently they come from Pakistan and Israel, but have been mounted by India, Brazil, Argentina, South Korea, and perhaps Taiwan. The problem of proliferation of nuclear weapons will likely continue to affect the nuclear power industry as one of the many challenges it faces.