Improving Security at the Department of Energy's Weapons Labs

Report Homeland Security

Improving Security at the Department of Energy's Weapons Labs

September 24, 1999 19 min read Download Report
Ronald Utt
Ronald Utt
Visiting Fellow in Welfare Policy

Ronald Utt is the Herbert and Joyce Morgan Senior Research Fellow.

Recent revelations that the People's Republic of China (PRC) may have illegally acquired advanced nuclear weapons and radar technology from several of the laboratories funded by the U.S. Department of Energy (DOE) provoked the Clinton Administration and Members of Congress to propose ways to enhance security at the labs.

As Representative Douglas Bereuter (R-NE) has stated, this espionage at the DOE's nuclear weapons laboratories facilitated "the most gravely serious thefts by China of sensitive U.S. technology and information that America has ever witnessed."2 In the wake of these revelations, however, debate in Congress was brief and was limited largely to committees that have a compelling interest in maintaining the status quo as well as the regional benefits that derive from the labs.

The many proposals to improve security at the labs range from adding another layer of DOE bureaucracy under a new security chief for the laboratories to fundamentally restructuring the agency in a way that would move the labs from DOE oversight. Proposals first introduced by Senator Rod Grams (R-MN) in S. 896 and Representative Todd Tiahrt (R-KS) in H.R. 1649 would transfer the three key labs that still perform nuclear weapons work to a more secure setting within the Department of Defense (DOD). They also would redistribute many of the DOE's remaining non-weapons programs to other civilian agencies, privatize the power marketing administrations, and effectively shut down the DOE in recognition of its 25 years of ineffective operation and security lapses.

A compromise proposal introduced by Senators Pete Domenici (R-NM) and Jon Kyl (R-AZ) would reorganize the three most important nuclear weapons labs into a semiautonomous entity within the Department of Energy, prospectively called the Agency for Nuclear Stewardship but renamed the National Nuclear Security Administration. This body would have considerable discretion over its own management and operations, and oversight by the Secretary of Energy would be limited. Such a proposal was endorsed in a June 1999 report by a special investigative panel of the President's Foreign Intelligence Advisory Board (PFIAB) headed by former Senator Warren B. Rudman (R-NH).

Energy Secretary Bill Richardson initially opposed the Domenici-Kyl proposal, but after the White House made it clear that it did not necessarily support Richardson's position, he reluctantly endorsed the proposal in early July. Then, in August, when the details of the proposal were included as an amendment to the National Defense Authorization Act for Fiscal Year (FY) 2000, Richardson changed his mind and announced that he would recommend a veto.

The fact that the Domenici-Kyl plan was strongly endorsed by several senior Members of Congress, as well as by a bipartisan, blue-ribbon, independent presidential panel, and opposed by the Cabinet agency whose lax oversight had contributed to the security problems in the first place led many Americans to conclude that it was the sort of reform that would be effective in tightening lab security. Unfortunately, however, by blurring the lines of responsibility and the chain of command even more than they are now, the proposal could very well diminish federal oversight and perpetuate the third-rate security practices common at the labs for nearly two decades.

Two independent experts on government management and reform recently expressed similar concerns in testimony before Congress. Professor Donald F. Kettl of the University of Wisconsin concluded that:

The proposal for a quasi-independent agency for nuclear stewardship focuses on precisely the right issue: improving national security at the nation's nuclear complex. However, it misdiagnoses the problem. It could well make the real problem worse. It fails to strengthen DOE's links to its field operations and misses the critical imperative to redefine DOE's culture.3

Victor Rezendes of the U.S. General Accounting Office (GAO) also expressed skepticism about the proposal:

One approach would create a separate agency within DOE, to be managed by a new Under Secretary for National Security. Another would create a semiautonomous agency whose director would report directly to the Secretary. Another would transfer DOE's nuclear weapons activities to the Department of Defense. While each of these proposals clarifies some lines of authority in the national security area, they are a piecemeal approach to DOE's structural problems and ignore the broader organizational issues.... Reorganization efforts that ignore the broader picture could create new, unintended consequences.4

Notwithstanding the objections of Secretary Richardson, the Domenici-Kyl plan is conceptually similar to the DOE's plan for improving security in that it relies on bureaucratic reshuffling and add-ons conducted within a bureaucratic entity with a decades-long track record of abject failure in addressing its own security problems. Whereas Richardson wants DOE to resolve the labs' notorious security problems, the Senate's compromise proposal would have the labs do it themselves, albeit with the help of a new semiautonomous bureaucracy within DOE.

The problem is that this compromise would preserve the status quo. The new body would have considerable discretion over its own management and would operate independently of DOE. Such a "reform" could very well worsen the perennial security problems at the nuclear weapons labs by removing meaningful federal oversight.

If this congressional compromise meets with a veto, as Secretary Richardson has threatened, Congress will then have an opportunity both to initiate a more thorough review of the role these labs play in meeting America's scientific and defense needs and to develop alternative ways to make them more secure. A better solution might be to transfer the three key labs that have exhibited the gravest security breaches from the Department of Energy to the Department of Defense and restructure the DOE to focus solely on energy issues and environmental concerns emanating from energy production and use.


As the federal agency with direct and primary oversight of America's 17 national labs, including the three nuclear weapons labs, the U.S. Department of Energy received much of the blame for the recently uncovered security breaches in which sensitive missile technology and information may have been leaked to China. As Secretary of Energy Richardson noted, "DOE security at the labs was not given the proper priority that it deserved in the '70s, in the '80s and in the '90s, and there should have been more attention to security at the national laboratories."5

But three decades of documented and admitted failure to protect vital secrets is a stunning record of managerial incompetence that demonstrates why it is time to find another federal home for U.S. nuclear weapons research. As the textbox "GAO's Consistent Scrutiny of DOE Lab Oversight"  shows, the problems with security are long-standing and have been criticized consistently by the General Accounting Office. The Energy Department receives all the blame for security breaches only because that is where the buck currently stops.

The 17 national laboratories are not formally part of the federal government. Although they are managed by the Department of Energy and play an integral role in national security, they are administratively and organizationally independent of DOE, and their employees are not part of the federal civil service.

GAO's Consistent Scrutiny of DOE Lab Oversight

For nearly two decades, the U.S. General Accounting Office has issued critical reports on security problems in the DOE's nuclear weapons labs. The following titles are indicative of the GAO's findings:

  • Safeguards and Security at DOE's Weapons Facilities Are Still Not Adequate (1982)

  • Security Concerns at DOE's Rocky Flats Nuclear Weapons Production Facility (1985)

  • Nuclear Security: DOE's Reinvestigation of Employees Has Not Been Timely (1987)

  • Major Weaknesses in Foreign Visitor Controls at Weapons Laboratories (1988)

  • Potential Security Weaknesses at Los Alamos and Other DOE Facilities (1990)

  • Accountability for Livermore's Secret Classified Documents Is Inadequate (1991)

  • Safeguards and Security Weaknesses at DOE's Weapons Facilities (1991)

  • Poor Management of Nuclear Materials Tracking System Makes Success Unlikely (1995)

  • National Laboratories Need Clearer Missions and Better Management (1995)

  • DOE Needs to Improve Controls Over Foreign Visitors to Weapons Laboratories (1997)

Technically, these 17 labs are organized as government-owned, contractor-operated facilities (GOCOs). In effect, the federal government owns the land, facilities, equipment, and technology, but direct day-to-day management and operation are contracted out to private entities. For example, the Lawrence Livermore National Laboratory in California and Los Alamos National Laboratory in New Mexico are operated and managed by the University of California. Together with the Sandia National Laboratory in New Mexico, which is managed by Lockheed Martin Corporation, these labs and two non-lab GOCOs form America's largely mothballed "atomic bomb factory."

No nuclear bombs have been built in the United States since 1988, yet these weapons facilities are still the repositories of all U.S. nuclear secrets and research geared toward nuclear weapons. Given the grave responsibility of oversight, the managers of these labs should regard secrecy and security as matters of paramount importance. But recent reports from the Cox Committee6 and the PFIAB,7 as well as nearly two decades of critical GAO assessments (see textbox "GAO's Consistent Scrutiny of DOE Lab Oversight"), document the long-standing pattern of serious security lapses at the labs. The independent status of these labs could be an important contributing factor in these pervasive security problems.

After the recent security breaches involving Chinese espionage became known, Secretary Richardson immediately promised that the DOE would find a way to solve the labs' security problems. But the DOE heretofore has failed consistently in its efforts to do so. Its managerial morass also may have contributed to the inconclusive outcome of an investigation by the Federal Bureau of Investigation to evaluate the evidence of internal espionage accumulating over the years.8

As if to underscore these problems, DOE's new security chief, retired General Eugene Habiger, committed a security gaffe shortly after his appointment. On his first visit to the Sandia lab, he inadvertently revealed classified information on nuclear weapons in a speech to an audience of 200.9

How Congress's Compromise Would Diminish Oversight

In contrast to Secretary Richardson's promise to accept responsibility for and act on the labs' security problems, the proposal from Senators Domenici and Kyl included in the Defense authorization bill before Congress would kick this responsibility down to the labs themselves and free them from day-to-day DOE oversight. This change could give greater control to the very institutions, managers, and employees who are most responsible for the scores of security breaches at the labs in the first place. Putting the three key nuclear weapons labs and several related contractor-operated facilities under the oversight of a new semiautonomous body within the DOE will not increase oversight enough to change the status quo (see textbox "Reinventing Government: A New Nuclear Security Bureaucracy").

Reinventing Government: A New Nuclear Security Bureaucracy

Under a compromise congressional proposal attached to the National Defense Authorization Act for Fiscal Year 2000, a new federal entity called the National Nuclear Security Administration (NNSA) would be located within the Department of Energy. This new agency would be responsible for three nuclear weapons labs and two non-lab contractors that have been associated historically with the development and manufacture of nuclear weapons. It would also be responsible for a few other components of these labs.

The new body would be run by an administrator who would be a new Under Secretary in the DOE. Other new management positions would include three new deputy administrators and a general counsel. New "Offices" supervised by deputy administrators would also be created within the agency.

To emphasize the body's separate, semiautonomous status within the DOE, the President would be required to submit a separate NNSA budget to Congress. The NNSA would have its own offices for personnel, legislative, public, procurement, and legal affairs and would maintain staff to conduct liaison with the Department of Energy, other federal entities, state and local governments, and Indian tribes.

In addition to the staff required to perform these administrative functions, the administrator would be authorized to hire up to 300 scientific, technical, and engineering employees to carry out his or her responsibilities. These new employees would be exempt from the provisions of existing federal civil service law and would instead be covered by the more flexible provisions of the Atomic Energy Act of 1954.

The labs are independent entities whose operations are managed by private contractors. Most of the security problems involved the contractors' employees; they reflect years of neglect and/or violations by personnel who are not DOE employees and are not subject to direct day-to-day DOE supervision. The effect of this "arms length" arrangement on the DOE's ability to oversee lab operations was painfully illustrated when Secretary Richardson merely "recommended" disciplinary action against the three Los Alamos employees responsible for some of that lab's security failures. The Secretary was limited to recommending such action because the employees in question worked for the lab's contractor, the University of California, not for DOE; thus, they were beyond his managerial reach.10

The record of security problems at DOE labs and the DOE's systematic failure to rectify long-standing inadequacies indicate that what the labs need most is more and better federal oversight, not less. The record also suggests that they are no more capable of reforming themselves than is the Department of Energy. For this reason, Congress should revisit some of the other lab-security reform proposals that have been offered and not rush to reject one over another simply because it is opposed by a White House appointee.

Other Approaches to Reform

Unlike the administrative changes proposed by Senators Domenici and Kyl, Secretary Richardson, and the President's Foreign Intelligence Advisory Board, elements of the plan in the bills offered by Senator Grams and Representative Tiahrt offer a potentially effective solution to the security problems by shifting the three key weapons labs to the Department of Defense. In addition to helping to solve the weapons security problem, the Grams-Tiahrt proposal would dismantle the DOE. The comprehensiveness of this approach has discouraged support for the overall proposal because it attempts to do too much and would affect many other Energy programs that benefit influential constituencies. It would stir up additional political opposition that likely would kill the bill and end any chance of achieving the important goal of creating a more secure environment for America's nuclear weapons programs.

A potential remedy that combines elements of the Domenici-Kyl proposal with those of the Grams-Tiahrt approach would be to transfer the key weapons labs from the Department of Energy to the Department of Defense, and then to re-engineer what remains of the DOE to focus solely on energy issues and environmental concerns that arise from energy production and use. Specifically, oversight of the three key weapons labs and the two other contractor-run facilities still actively involved in the maintenance of the U.S. nuclear arsenal would be shifted from the DOE to the DOD. The labs would then be under the day-to-day management of a department that has experience in maintaining high levels of security and for which "top secret" status is a day-to-day concern. The DOE's main policy objectives, by contrast, focus on civilian and commercial sectors for which issues of secrecy and security are of incidental concern.

Removing the labs from the DOE is clearly something that Congress and the Administration should have an opportunity to consider more thoroughly, yet some Members of Congress oppose this approach as well as further investigation into its potential viability. Such an opportunity for additional study was squelched when the House-Senate conference committee on the recently passed National Defense Reauthorization Act formally excluded a House amendment that would have required the President to prepare a report on whether the Department of Energy should continue to maintain nuclear weapons responsibility. Given the long-standing security problems at the weapons labs, Congress should not have dismissed so quickly any opportunity to study promising reforms.


How the Labs Originated

America's 17 national labs can trace their origins to World War II, when the federal government created the Manhattan Project to accelerate the development of an operational nuclear bomb. To accomplish this extraordinary objective in a short period of time, the federal government combined the talents of many of the world's leading scientists and technicians, working at universities and test sites throughout the country, with the vast financial and material resources needed to develop and test the weapon. This effort was, of course, successful; but then, rather than close the sites as was done with many other wartime weapons production facilities, the government, facing the advent of the Cold War and Russia's subsequent development of its own atomic bomb, made the lab sites permanent and expanded them to meet the new challenges.

Referring to these facilities as laboratories may have accurately reflected the experimental and speculative nature of the effort when the Manhattan Project was created, but the accelerated production of nuclear weapons in response to the Cold War effectively made them bomb factories. By 1988, they had produced an estimated 30,000 nuclear weapons devices. Since then, after a series of arms treaties with the former Soviet Union and the end of the Cold War, no nuclear bombs have been produced and none are planned for the foreseeable future.

As a result, the labs have experienced a substantial decline in their traditional activities; efforts throughout most of this decade have focused on identifying civilian research activities and additional defense-related work that would be appropriate for the various talents of the labs. However, because these talents are concentrated in nuclear technologies, and because concerns for safety and environmental impact severely limit the application of nuclear technologies to civilian purposes, recent efforts to redirect laboratory talents to other purposes have been challenging, with mixed success.

Until 1947, the labs fell under the management of the U.S. Department of Defense (specifically, the U.S. Army Corps of Engineers); in 1947, the management and oversight of the labs were consolidated under the newly created Atomic Energy Commission, which was tasked with overseeing military and civilian applications of nuclear science.

Nuclear technology was developed first for weapons purposes, but it quickly became apparent that modifications in the technology could have civilian applications, the most important of which would be to produce a substitute fuel for electricity generation and other power needs. The laboratories, as well as responsibility for the nation's nuclear programs, remained with the Atomic Energy Commission and then the Energy Research and Development Administration until 1977, when the labs and several other government programs were combined into a new Department of Energy.

Although much of their work has been directed toward national security objectives, shortly after World War II the labs were kept deliberately outside of and independent of the DOD. It was believed that civilian control over weapons of mass destruction would better ensure that such power would not be misused by the military. Former Secretary of Energy Hazel O'Leary summed up this line of thinking in a 1995 interview: "You need a clear wall between the technical people who design the weapons and certify their safety and reliability and those who would use and deploy and maybe, in their haste to deploy, would not make the careful review of the reliability and safety."11

This reasoning, however, is absurd. The notion that the military would be disinterested in a weapon's safety and reliability makes no sense; nor does the idea that the armed forces may use such a weapon in "their haste." Only the President can launch a nuclear strike, and neither the DOE nor any of the thousands of people working in the laboratories have any say in the matter.

Indeed, recent allegations of sustained and systemic security lapses at the labs suggest that instead of enhancing America's safety, this historic separation of responsibility led to several decades of negligent oversight which severely jeopardized the safety and security of every American. As Representative Bereuter rightly noted after the Cox Committee released an abridged version of its report to the public last May, the alleged espionage at the weapons labs facilitated "the most gravely serious thefts by China of sensitive U.S. technology and information that America has ever witnessed."12

How the Labs Are Organized and Managed

Although the weapons labs play an integral role in national security and fall under DOE oversight, they are not formally part of the federal government or of the DOE, and their employees are not part of the federal civil service. Formally, the labs are government-owned, contractor-operated (GOCOs) facilities. In effect, the government owns the land, facilities, equipment, and technology, but the direct day-to-day management and operation of the labs has been contracted to private entities.

For example, the Lawrence Livermore National Laboratory in California and Los Alamos National Laboratory in New Mexico are managed by the University of California. Together with the Sandia National Laboratory, also in New Mexico but managed by Lockheed Martin Corporation, these three labs, along with two non-lab GOCOs, form the heart of America's "bomb factory." When fully operational in the manufacture of a bomb, the division of labor among the seven GOCOs was as follows:

  1. Los Alamos and Lawrence Livermore designed the nuclear components, which are the explosive material in any nuclear bomb.

  2. Sandia performed the non-nuclear work (for example, electronics, trigger mechanisms, and safety features).

  3. Kansas City Plant manufactured the mechanical and electronic components of the bomb.

  4. Rocky Flats, Colorado, and Y-12 at Oak Ridge, Tennessee, produced the nuclear material.

  5. Pantex in Amarillo, Texas, assembled these components into an operational bomb and shipped it to the military.

Because no nuclear weapons have been built since 1988 and none are planned, the duties of five of these GOCOs changed. Neither Rocky Flats nor Y-12 is now involved in weapons work. Three other labs still conduct research on nuclear devices but have branched out into other scientific fields and weapons research. Lawrence Livermore, for example, conducted the advanced research on submarine-detecting radar devices that became the target of alleged Chinese espionage.13 Pantex, which once assembled the bombs, now does necessary maintenance work on existing nuclear weapons and disassembles other atomic weapons as part of the shrinkage of the U.S. nuclear arsenal. Kansas City still manufactures bomb parts, but only as replacements in existing weapons, which may have deteriorated through age.

How the Labs Are Funded

With no bombs to build, funding for the weapons program decreased throughout the early and mid-1990s. That trend has since been reversed, and lab spending is scheduled to increase each year between FY 1998 and FY 2000.

Of the DOE's estimated spending of $15.5 billion for FY 1999, $12.4 billion (or 80 percent) of the department's budget is for "Atomic energy defense activities." Of this amount, $4.4 billion is for "weapons activities" at the labs, while another $4.3 billion is for the environmental remediation of former weapons production sites, including the labs. The remaining amount, less than $4 billion, is spread among a variety of smaller defense-related activities. Only three-fourths of $1 billion from the Department of Energy's budget, or less than 5 percent of its total budget, will go to programs defined as "energy supply."14

Because the labs are not formally part of the government and are operated by private contractors, neither the President's budget nor the congressional appropriations bills include any specific reference to lab funding. Rather, a lump sum--$4.4 billion in FY 1999--is appropriated by Congress to the Department of Energy for "Weapons Activities." In turn, the DOE, nominally at its discretion, reallocates these funds to a variety of research, development, and weapons-related projects that are performed by the 17 national labs and six related entities that work under contract to the department. Of this lump sum, $3.2 billion, or 73 percent of the total, goes to the five facilities described above as the bomb factory. On a regional basis, about three-quarters of the bomb factory money is spent in facilities located in New Mexico.

A look at two of the DOE's contract labs illustrates their differences in focus. Oak Ridge National Laboratory was established in Tennessee in 1942 to assist in the development of the first atomic bomb and through the 1960s was an important component of the nuclear weapons programs. It currently employs 4,453 workers under a budget of nearly half a billion dollars from all sources, including the DOE. In recent years, Oak Ridge has shifted increasingly toward civilian programs, and defense programs account for just 7 percent of the $385 million it received from the DOE in FY 1998.15

In contrast, the Sandia National Laboratory in Albuquerque, New Mexico, employs a work force of 7,374 and operates with a budget of nearly $1.3 billion from all sources. Of this amount, about $1 billion is from the DOE, and 82 percent of that allocation, or $822 million, is for national defense purposes. At Los Alamos, with 6,900 workers, the percentage of weapons work is 69 percent, while the same share accounted for the defense work of the 7,300 employees at Lawrence Livermore.


Ronald Utt
Ronald Utt

Visiting Fellow in Welfare Policy