Congressional Budget Office - March 1997


Annual Savings							Five-Year
Savings from			(Millions of dollars)		Cumulative
the 1997 Plan		1998	1999	2000	2001	2002	Total

Budget Authority	267	387	429	681	964	2,728

Outlays			200	357	419	618	893	2,487

NOTE: Savings relative to the Administration's 1998 plan appear in Appendix A.

For the first four decades of the nuclear age, the United States developed, tested, and produced nuclear weapons for its arsenal. The Department of Energy (DOE) and its predecessors have been responsible for that task. During much of the Cold War, the arsenal held over 25,000 warheads of more than a dozen different types. The weapons were designed and developed at the three weapons laboratories (Los Alamos, Lawrence Liver-more, and Sandia) and tested at the Nevada Test Site; materials and components for the weapons were produced at more than a dozen facilities across the country.

The end of the Cold War has changed the requirements for the arsenal. In response to the second Strategic Arms Reduction Treaty (START II), the United States plans to keep roughly 5,000 warheads of seven different types in its active inventory beyond 2003. DOE has started to consolidate its production facilities as it adjusts to its declining workload.

The United States, along with all other declared nuclear powers except China, has also unilaterally halted all underground testing. To establish a permanent worldwide moratorium, the United Nations' Conference on Disarmament negotiated the Comprehensive Test Ban Treaty (CTBT), which will make it difficult for any country to develop new weapons. President Clinton signed the treaty in September 1996 but has not yet submitted it to the Senate for ratification.

To preserve its ability to ensure, over the long run, the reliability and safety of the weapons that remain in the nuclear stockpile under a CTBT, the Department of Energy has developed a stockpile stewardship and management program. One goal of that program is to increase funding for activities such as computer simulations, nonexplosive nuclear testing, and fusion research that will become increasingly important for ensuring the reliability of the stockpile in the absence of underground testing. Another goal is to ensure that the weapons labs continue to attract talented scientists by providing challenging work and state-of-the-art facilities. A third goal is to develop facilities that will produce the necessary nuclear and nonnuclear components to replace parts, thus ensuring reliability.

To carry out this plan, DOE will continue to operate both of its weapons design labs (Los Alamos and Lawrence Livermore) and its engineering lab (Sandia). It will also construct several new facilities to provide data on the reliability and safety of weapons as they age. Those facilities include the Dual-Axis Radiographic Hydrotest (DARHT) facility at Los Alamos for hydrodynamic tests and the National Ignition Facility (NIF) at Lawrence Livermore for research on the fusion portions of the weapons. In addition, DOE will conduct "zero-yield" tests at the Nevada Test Site so that it can retain enough skilled technicians to resume testing--as directed by the President--if the United States withdraws from the CTBT for reasons of supreme national interests.

According to the 1997 plan for stewardship, DOE will spend $1.7 billion in 1998 for what has been known historically as weapons research, development, and testing (RD&T), or about $600 million less (after adjusting for inflation) than it spent in 1988 when the laboratories were still operating at a Cold War pace. However, the annual expenditures for RD&T under the Administration's plan, after adjusting for inflation, will still be about the same as in 1980 when the United States was both designing new warheads and maintaining an arsenal of some 25,000 warheads. Further reductions in spending may therefore be possible.

DOE's 1997 plan called for spending about $2 bil-lion in 1998 to manage the stockpile and $2 billion or more each year thereafter. That spending includes an average of nearly $500 million a year through 2002 to develop a new source of tritium, a radioactive gas that is used in all U.S. nuclear weapons and decays at the rate of 5.5 percent a year. Tritium is produced by bombarding special targets with neutrons. The neutrons could come from an accelerator or from the fissioning of uranium atoms within a commercial nuclear reactor. DOE recently decided to work on both technologies through 1998, at which point it will make a decision about which one to develop fully.

This option would reduce the scope of the stewardship program by consolidating the two design laboratories and forgoing all testing activities at the Nevada Test Site. It would also reduce the cost of managing the stockpile by canceling the development of a tritium production accelerator and relying instead on less costly commercial reactors. Taken together, the changes in this option would save $200 million in outlays in 1998 and $2.5 billion through 2002 compared with the Administration's 1997 plan. Measured against the 1998 plan, five-year savings would be about $730 million lower. That plan excludes much of the funding that will eventually be required to develop the tritium accelerator. Savings are actually greater in 1998 and 1999 because the 1998 plan fully funds early design activities.

For illustrative purposes, the above savings assume that weapons design activities would be consolidated at Los Alamos over a period of five years; Lawrence Livermore would no longer have the designing of nuclear weapons as its primary focus. Los Alamos designed the majority of nuclear weapons that are likely to remain in the stockpile. To ensure that the other warhead types could be reliably maintained, some designers from Livermore would have to move to Los Alamos. This option would also maintain a cadre of weapons scientists at Livermore to provide peer review for Los Alamos's efforts. To provide those scientists with challenging work, Livermore would retain substantial computational facilities for modeling the complex processes inside nuclear weapons and would proceed with DOE's plans to build the National Ignition Facility. (The savings would be lower if stewardship activities were consolidated at Lawrence Livermore because that would involve moving more facilities and relocating more weapons designers. Also, the environmental issues raised by introducing new nuclear facilities into the populous area surrounding Livermore could prove difficult to overcome.)

Finally, by canceling the program to develop an accelerator to produce tritium and instead producing tritium in commercial reactors, this option would save $190 million in 1998 and about $2 billion through 2002 relative to the 1997 plan. Eventually, operating savings could total more than $100 million a year.

The central question underlying this option is, What is required to ensure the reliability and safety of the stockpile in the future if the current moratorium on underground nuclear testing is made permanent? DOE's stewardship and management program is the Administration's answer. This option preserves much of what the stewardship plan calls for, including DARHT and NIF, but does not preserve readiness at the Nevada Test Site or fund two full design labs. It also opts for an inexpensive source of tritium.

Some people may feel that this option cuts the program too deeply. They believe that DOE's stewardship program is the minimum effort necessary to maintain the stockpile without underground testing. Cuts would not be prudent, they argue, because scientists will need new facilities to obtain data on reliability that was formerly provided directly by underground nuclear testing.

Supporters of DOE's stewardship program also object to the consolidation proposed here. In their view, two design laboratories are essential for providing a robust stewardship program: competition and peer review will be even more important in the absence of underground testing. Furthermore, they argue, refocusing the efforts of one lab away from weapons research will eliminate its central unifying mission (and thus its motivation for excellence) without replacing that focus with an equally important mission. Consolidation will also result in the loss of some facilities that cannot easily be transferred to the other lab. For many of these reasons, the President recently directed DOE to retain both labs. Advocates of the stewardship program also disagree with this option's proposal to close the Nevada Test Site because doing so would increase the time required to resume underground testing if Russia started a new arms race or the United States discovered a serious problem with its stockpile that could only be corrected by testing. Perhaps equally important to them, closing the Nevada Test Site would restrict the ability of weapons scientists to conduct "subcritical" experiments to learn more about the effects of aging on plutonium.

Other people argue that the stewardship program should be cut further than suggested in this option. Some believe that keeping part of a second lab, increasing money for basic stewardship, and building DARHT and the $1.2 billion National Ignition Facility are unnecessary to support the stockpile. In their view, those facilities may allow DOE scientists to continue designing and testing weapons and to circumvent the test ban treaty. Even if DOE has no intention of designing new weapons, they argue, the perception of such a capability may make it difficult to convince nonnuclear countries--from whom the United States would like continuing support for the Nuclear Non-Proliferation Treaty--that the United States has really given up testing. Other critics contend that the nation cannot afford to keep a portion of a second design lab or NIF; they argue that if NIF can help scientists understand how to harness fusion for civilian energy, as supporters claim, it should be funded outside the nuclear weapons program.

There are several reasons to continue developing an accelerator for producing tritium. Although DOE has explored the idea of buying services from commercial reactors, and utilities that operate the reactors seem enthusiastic, forgoing the accelerator may be premature until DOE is certain that bureaucratic and political hurdles can be addressed and that commercial services will be available. Moreover, some groups argue that relying on commercial reactors to produce tritium will complicate efforts to control the spread of nuclear weapons because it blurs the distinction between military and civil nuclear programs. An accelerator is also appealing because it will not produce the radioactive waste that a reactor generates.