NUCLEAR SAFETY

Potential Security

Weaknesses at Los

Alamos and Other

DOE Facilities

 United States

General Accounting Office

Washington, D.C. 20548

Resources, Community, and

Economic Development Division

B-240972

October l&l990

The Honorable John D. Dingell

Chairman, Subcommittee on Oversight

and Investigations

Committee on Energy and Commerce

House of Representatives

The Honorable Pete V. Domenici

United States Senate

The Honorable Jeff Bingaman

United States Senate

At your request, we examined issues related to the adequacy of security at the Department

of Energy’s Los Alamos National Laboratory, New Mexico; the Department’s security

inspection process; and the feasibility of federalizing the Department’s security forces. This

report presents the results of our efforts.

Unless you publicly announce its contents earlier, we plan no further distribution of this

report until 30 days from the date of this letter. At that time, we will send copies to the

appropriate congressional committees; the Secretary of Energy; and the Director, Office of

Management and Budget. We will also make copies available to others upon request.

This work was performed under the direction of Victor S. Rezendes, Director, Energy Issues,

who can be reached at (202) 275-1441. Other major contributors are listed in appendix III.

J. Dexter Peach

Assistant Comptroller General

Executive Summary

Purpose

In March 1989 the contract security force at the Department of Energy’s

(DOE) Los Alamos National Laboratory began a lo-week strike, primarily

because of quality of life issues. During the strike, DOE used temporary

replacements from other facilities. Los Alamos carries out nuclear

weapons research, development, design, and testing activities. Therefore,

continuous, effective security is essential to protect nuclear materials,

weapons, and information.

Concerned about the effect of the strike on site security, the Chairman,

Subcommittee on Oversight and Investigations, House Committee on

Energy and Commerce, and Senators Domenici and Bingaman asked GAO

to evaluate (1) the adequacy of security at Los Alamos and other DoE

facilities, (2) DOE'S oversight of protective forces, and (3) the feasibility

of establishing federal security forces at DOE facilities.

Background

DOE is responsible for the nation’s nuclear weapons program and owns a

broad spectrum of facilities to carry out research, development, and production

activities. Contractors provide security services at all but one

DOE facility-the Albuquerque Operations Office. For all facilities, DOE

established 12 minimum skills that security force members must meet

and the annual training they should receive. Periodically, DOE conducts

inspections and/or performance tests to assess the effectiveness of the

security forces. (See ch. 1.)

Results in Brief

GAO raises concerns about the adequacy of security at Los Alamos

before, during, and after the strike. Before the strike, DOE could not

demonstrate that the security force was properly trained to protect the

facility because training records for some of the force were missing,

incomplete, or inaccurate. During the strike, many replacements did not

meet the 12 required skills. After the strike, an unannounced exercise

showed that as late as April 1990 more than 75 percent of the regular

force did not meet one or more of nine required skills.

GAO also raises concerns about security at some other DOE facilities that

it reviewed. DOE inspections identified recurring and similar weaknesses;

yet, DOE rated only one security program as unsatisfactory. GAO believes

that this occurred because DOE lacks criteria specifying the severity and

frequency of inspection findings that would result in a satisfactory or

unsatisfactory rating. Also, DOE does not have an effective mechanism to

ensure that corrective actions are taken on inspection findings found that some inspection findings went uncorrected for as much as 5 years.

DOE believes that federal and contract security forces are equally

capable of protecting its facilities, and the costs for both are similar.

However, DOE does not have current cost data, and GAO estimates that

annual labor and benefit costs could be about $15 million less if DOE federalized

the security forces at the nine facilities GAO reviewed.

Principal Findings

Potential Security Weaknesses at Los Alamos

DOE and Los Alamos officials believe that security before, during, and

since the 1989 strike was adequate. GAO was unable to verify this assertion

but did find indications that potential security weaknesses exist.

Before the strike, training records for some security force members were

missing, undated, incomplete, or inaccurate. Therefore, DOE could not

demonstrate that Los Alamos’ force was properly trained to protect the

facility.

During the strike, DOE waived physical fitness and medical requirements

for about half of the replacement force, and many were not certified in 1

or more of 12 skills required of the regular Los Alamos force. These situations

may not have occurred if DOE had established skill requirements

for replacements and required contractors to develop contingency plans

specifying the methods to be used to meet the requirements during a

strike. Further, although a facility is most vulnerable during the early

stages of a strike, DOE did not conduct an inspection at Los Alamos until

2 weeks, nor test the replacements’ proficiencies until 6 weeks, after the

strike began. Also, DOE never conducted a force-on-force simulated

attack test over the strike’s duration even though such a test is the best

measure of a security force’s overall ability to protect life and property.

Therefore, DOE had little assurance that the replacements could adequately

protect Los Alamos.

Since the strike, GAO found that most of the regular security force lacked

one or more of nine skills that DOE officials say are needed to ensure the

minimum level of protection for the site. Over 75 percent of the regular

security force lacked such skills during an unannounced April 1990

exercise that DOE conducted at GAO'S request. (See ch. 2.)

Other Facilities Have

Security Program

Weaknesses

DOE periodically inspects its facilities to assess the effectiveness of

security policies, procedures, operations, and force proficiencies. DOE

inspections of Los Alamos and eight other facilities since 1985 found

some weaknesses that were similar and recurring. For example, DOE

found that some security force members at Los Alamos, Argonne,

Sandia, and Savannah River could not appropriately handcuff, search,

or arrest intruders and shoot accurately. DOE also found weaknesses in

the training programs related to those programs.

Despite finding similar problems at the nine facilities, DOE rated only Argonne as unsatisfactory over the 5-year period. GAO believes that the differences occurred

because DOE does not have criteria specifying the severity and frequency

of inspection findings that would result in a satisfactory or unsatisfactory

rating. In the highly important area of security at sensitive nuclear

weapons facilities, DOE should be conservative and consistent-if one

situation warrants an unsatisfactory rating, then other facilities with

similar weaknesses should be similarly rated.

Also, DOE has no systematic method to track or confirm the corrective

actions taken on inspection findings, GAO found that deficiencies identified

as early as October 1985 at six facilities had not been corrected as

of May 1990. DOE'S allowing this situation to occur could send a message

to contractors that security is not important and could perpetuate an

environment in which contractors have little incentive to take corrective

actions. DOE has a mechanism to improve this situation-the awards fee

process. For contractors with repeat security inspection weaknesses, DOE

could vary the amount of fees awarded depending upon the timing and

effectiveness of corrective actions taken. (See ch. 3.)

Some Contract Forces May No Longer Be Cost Effective –

DOE believes that the abilities of, and costs for, a federal and contract

force are similar, but a critical factor is the force’s ability to provide

uninterrupted service. A major advantage of a federal force is that it

cannot legally strike, whereas a major disadvantage of a contract force

is that generally it can strike. The Los Alamos strike cost about $1.6

million over and above the almost $17 million contract cost.

According to a DOE Office of General Counsel official, no legal obstacles exist to

DOE'S negotiating a never-strike provision in its security force contracts

but estimated that it would be costly to do so. Also, turnover may be

lower with a federal force. During the 26 months before the strike, Los

Alamos experienced between 1 l- and 15-percent turnover; the Albuquerque

Operations Office federal force experienced no turnover.

In contrast, a contract force, according to DOE and Los Alamos officials, can

more quickly be reduced or increased to meet changing work demands.

Recommendations

Generally, the advantages and disadvantages of both types of forces

offset each other, and the primary issue becomes cost. LWE conducted

cost studies in the early 1980s for four facilities but has not updated the

studies or conducted additional analyses to determine whether it is still

cost-effective to have contract forces at all its facilities. DOE officials

said they have not done so because privatization was emphasized

throughout the 198Os, and they could not obtain positions for federal

forces. Since that time, contract employee costs have increased faster

than federal employee costs. GAO estimates that federal labor and benefit

costs could be at least $15 million less each year than similar contract

costs at 9 facilities, representing more than 60 percent of DOE'S

5,600 security force members. (See ch. 4.)

To help ensure that security forces have the maximum capability to protect

sensitive nuclear weapons facilities, GAO has made a number of recommendations to the Secretary of Energy to standardize skill requirements for all security force members including strike replacements,

ensure that security force members receive all required training,

withhold a portion of award fees when contractors do not take timely

corrective actions on security inspection weaknesses, and

evaluate the relative costs of federal and contract security services

across the nuclear weapons complex and convert to federal forces at

locations where it is cost-effective to do so.

Agency Comments GAO discussed the facts presented in this report with DOE, Los Alamos,and the security force contractor. The officials generally agreed with the facts but offered some clarifications that were incorporated where appropriate. As requested, GAO did not ask DOE, Los Alamos, or the contractor to comment officially on this report.

Contents

Executive Summary 2

Chapter 1 8

Introduction Causes of the Strike 8

Actions Taken Since the Strike 9

Organization for Overseeing Security 9

Objectives, Scope, and Methodology 10

Chapter 2 14

Concerns About the Was Security Adequate During the Strike? 14

Adequacy of Security at Los Alamos

 DOE Sites Were Not Prepared for Strikes 18

Security Force May Not Have Been Properly Trained 19

Los Alamos’ Security Force Did Not Perform Well During 21

a Surprise Test

Conclusions 22

Recommendations to the Secretary of Energy 23

Chapter 3 24

DOE’s Security

DOE Lacks Specific Criteria for Rating Facilities 24

Inspection Process Can Be Improved

DOE Does Not Have an Effective System to Track 27

Corrective Actions Taken

Other Options to Ensure Corrective Actions 28

Conclusions 29

Recommendations to the Secretary of Energy 30

Chapter 4 31

Some Contract Forces May No Longer Be Cost-Effective

Federal Security Force May Be More Cost-Effective at 32

Some Locations

DOE Has Not IJpdated Its Cost Comparisons 33

Some Aspects of Contract and Federal Security Forces 34

Offset Each Other

Conclusions

Recommendation to the Secretary of Energy

35

36

Appendixes Appendix I: Views of Los Alamos Security Force Members

Appendix II: Comparison of Six Facilities’ Compliance

With Albuquerque’s Draft Contingency Plan

Requirements

38

41

Appendix III: Major Contributors to This Report 43

Tables Table 2.1: Number of Replacement Force That Were Not

Certified in Certain Skills

16

Table 2.2: Incomplete, Missing, or Deficient Training

Records

20

Table 2.3: Results of an Unannounced Test

Table 3.1: Security Force Weaknesses Cited in DOE

Inspection Reports, 1985-89

21

26

Table 4.1: Contract Versus Federal Labor and Benefit

Costs at Nine DOE Facilities

32

Abbreviations

DOE Department of Energy

GAO General Accounting Office

OMH Office of Management and Budget

OSE Office of Security Evaluations

OS&S Office of Safeguards and Security

SSIIS Safeguards and Security Issues Information System

Chapter 1

Introduction

In March 1989 the security force at the Department of Energy’s (DOE)

Los Alamos National Laboratory, New Mexico, began a strike that lasted

10 weeks. Los Alamos conducts both unclassified and classified activities

related to all phases of nuclear weapons research, development,

design, and testing. Therefore, a security force possessing the necessary

skills is the first line of human defense against terrorist or other attacks,

theft or misuse of classified information and materials, and sabotage at

sensitive nuclear facilities and is a key factor in WE'S physical security

program. Security force members who cannot individually or as a team

successfully perform all assigned tasks raises serious questions about

the adequacy of security at these facilities.

To provide security force protection, the University of California, which

operates Los Alamos for DOE, has contracted with Mason and Hanger-

Silas Mason Company, Inc., since 1981. The company e\mploys several

hundred security inspectors, including officers and a rapid response

team, who are authorized to detain, arrest, and use force if necessary to

protect the facility. Security inspectors must meet minimum competency

levels in 12 basic skills and be physically fit to perform their duties.

Causes of the Strike

Mason and Hanger has a labor agreement with the International Guards

Union of America, Local 69. In February 1989 the labor agreement

expired, and on March 13,1989, the security force began a lo-week

strike, which ended on May 21, 1989. During the strike, DOE used temporary

replacements from its other sites as well as the Department of

Defense facilities. DOE and Los Alamos officials told us that the causes of

the strike included longstanding, unresolved labor-management relations

problems-primarily Mason and Hanger’s overtime, disciplinary,

and sick leave policies.

According to Mason and Hanger officials, the overtime occurred because

they never had enough job applicants with DOE security clearances to fill

positions left vacant by security force members who resigned, retired, or

were fired. Applicants, they said, became discouraged by the long wait

for clearances (up to 18 months), and many were no longer interested in,

or available for, employment by the time DOE granted their clearances.

About one out of every four applicants were not available to accept job

vacancies once the clearances had been received.

Some security inspectors told us that the overtime would have been

more bearable, and the attrition rate lower, if Mason and Hanger had

instituted fairer and more sympathetic policies. Fourteen security

inspectors said that the mandatory overtime was excessive and that the

company’s disciplinary policy was harsh. For example, some said that

employees had been disciplined for frivolous reasons (eating while on

duty). Similarly, according to some of these individuals, the sick leave

policy was arbitrary, and in some cases, the company overruled doctors’

opinions and forced employees to use vacation in lieu of sick leave.

According to these individuals, these policies caused security force

members to quit or be fired.

Actions Taken Since the Strike

To minimize future labor problems, a Mason and Hanger official said

that in May 1989 the company changed its policies to require less mandatory

overtime. At the same time, the company removed all demerits

that most security inspectors had accrued and modified its sick leave

policy by allowi’ng security inspectors to charge sick leave starting with

the first day of absence, provided they obtain a note from their doctor.

In addition, the company hired a human relations manager to serve as a

liaison with the security force, and officials believe that labor relations

have improved. According to a DOE industrial relations specialist, the

agency suggested that Los Alamos use the Federal Mediation and Conciliation

Service to facilitate meetings between Mason Hanger and the

union. Los Alamos did so through December 31,1989. As of May 1990,

according to several security inspectors, Mason and Hanger was not

always responsive to inspectors’ grievances, and the possibility existed

that a wildcat strike might occur because labor relations had deteriorated.

The industrial relations specialist told us that a wildcat strike has

never occurred at a DOE facility.

Organization for

Overseeing Security

The Assistant Secretary for Defense Programs is responsible for

directing the activities conducted by DOE'S nuclear weapons facilities.

Within Defense Programs, the Office of Safeguards and Security (0s~)

establishes policies pertaining to the skills and qualifications that

security force applicants must meet, the type and amount of annual

training they should receive, and the content of plans for emergencies

and other contingencies. Also, the Office of Security Evaluations (OSE)

under the Assistant Secretary for Environment, Safety, and Health periodically

assesses the effectiveness of DOE safeguards and security policies,

procedures, systems; and operations. In making these assessments,

OSE is required to periodically conduct performance tests: a simulated

attack on specific DOE targets (force-on-force) or a limited-scope assessment

of, for example, a force’s response to a simulated crisis, emergency,

or unplanned events, such as activation of security alarms. The

limited-scope test can either be announced or unannounced.

DOE headquarters has delegated responsibility for significant aspects of

the security program to eight field offices called operations offices that

oversee the facilities. DOE’S Albuquerque Operations Office, New Mexico,

oversees Los Alamos and eight other nuclear weapons laboratories and

production facilities. To carry out its responsibilities, Albuquerque conducts

various types of security surveys to ensure that the facilities

maintain effective safeguards and security programs. In an unusual

event, such as a strike, Albuquerque may also conduct special surveys

or performance tests.

In turn, the operations offices have delegated certain oversight responsibilities

to the contractors that operate the facilities. For example, the

University of California is responsible for overseeing Mason and

Hanger’s operations to ensure that they comply with DOE’S policies and

procedures. Finally, Mason and Hanger establishes additional skill

requirements for the Los Alamos security force, provides training, and

tests to ensure that the force meets DOE’S and its requirements.

Objectives, Scope, and Methodology

Concerned about the security implications of the strike, the Chairman,

Subcommittee on Oversight and Investigations, House Committee on

Energy and Commerce, and Senators Domenici and Bingaman, asked us

in May 1989 to evaluate (1) the adequacy of security at Los Alamos and

eight other DOE facilities,’ (2) DOE’S oversight of the protective security

forces at the nine nuclear weapons facilities, and (3) the feasibility of

federalizing DOE security forces currently under contract. In subsequent

discussions with congressional staff, we agreed to concentrate on Los

Alamos and gather information on the other eight sites from DOE headquarters

and operations offices. We did not conduct work at the eight

other sites. Further, we did not assess other aspects of DOE’S physical

security requirements, such as adequacy of fences, barriers, and alarms.

Since we agreed to issue an unclassified report, some of the information

cannot be presented in its entirety.

Argonne National Laboratory, Illinois; Lawrence Livermore National Laboratory, California; Nevada Test Site, Nevada; Pantex, Texas; Rocky Flats, Colorado; Sandia National Laboratories, New Mexico;  Savannah River Plant, South Carolina; and the Oak Ridge Y-12 plant, Tennessee.

To obtain an overall perspective on the three issues, we reviewed relevant

provisions of the Atomic Energy Act of 1954, as amended, and

DOE’S security policies and procedures. Also, we met with the head of

WE’S Safeguard and Security Task Force that was reviewing a broad

range of safeguards and security issues at DOE facilities. A classified

report of the task force’s results is expected to be available during the

fall of 1990. We also met with Nuclear Regulatory Commission staff

about that agency’s security program requirements.

Adequacy of Security

We met with DOE headquarters, Albuquerque Operations Office, Los

Alamos, and Mason and Hanger security and procurement officials as

well as 14 Los Alamos security force members about the causes of the

strike; adequacy of security before, during, and after the strike; and

adequacy of training. We judgmentally selected 11 security force members

from Mason and Hanger’s staff roster and invited others to meet

with us. Three accepted the invitation, for a total of 14 (app. I contains

the views expressed). The 11 individuals that we selected worked on the

day, swing, and midnight shifts in all possible job categories and

included union and nonunion members, females and males, and security

inspectors and supervisors. Because of time constraints, we did not take

a valid statistical sample; therefore, the results cannot be projected to

all the Los Alamos security force.

In addition, we reviewed DOE’S, Los Alamos’, and Mason and Hanger’s

security force skill and qualification requirements and examined 1989

training records for about 330 regular security force members and the

39 1 strike replacements. To validate security force performance, we

asked DOE to conduct a “no-notice” limited-scope performance test of the

Los Alamos security force in 9 of 12 required basic skills-security

operations, use of deadly force and limited arrest authority, communication

procedures, firearms, tactics, physical conditioning, self-defense,

nonlethal weapons, and site protection. The security force members

were asked to shoot their handgun (firearms), demonstrate eight basic

moves with a baton (nonlethal weapon similar to a billy club), run at

least one-half mile (physical conditioning), and apprehend suspects demonstrating the six other required skills. We did not test the remaining

three skills: vehicle safety, standards of conduct, and first aid/fire

fighting.

For the test, we selected a statistical sample of security force members

from those working the three primary shifts over a 24-hour period. Staff

from our Offices of Security and Special Investigations who are knowledgeable

about apprehension, arrest, and baton procedures supplemented

the audit team to observe and critique the tests. With the

exception of the running exercise, the tests were conducted on April 3

and 4, 1990. Prior to a running test, DOE requires participants to receive

a medical examination. To comply with this requirement, the running

exercises were conducted on April 11 and 12, 1990. Our results can be

projected with a 96-percent confidence level to the security force members

from which the sample was taken.

We also reviewed contracts between DOE and the University of California,

the University’s subcontract with Mason and Hanger, and the

company’s agreement with the union. We limited our examination to

provisions that pertain to the length of the contract, work stoppages,

contingency requirements, oversight responsibilities, and termination.

DOE Oversight

To determine the actions that DOE takes to oversee the protection of its

facilities, we reviewed the process used to inspect and rate facilities and

the methods employed to ensure that the contractors take corrective

actions on the deficiencies identified. In this regard, DOE headquarters

and Albuquerque provided us with inspection reports for the period

1986439 for the nine sites in our scope. We compared DOE’S findings in

these reports to identify trends or patterns, such as repeat deficiencies

at a particular site. We also obtained DOE’S contingency plan criteria,

draft criteria that Albuquerque had developed, and six plans from contractors

under Albuquerque’s purview.2 We compared the six plans with

Albuquerque’s draft criteria but did not evaluate the adequacy of the

criteria. Using this and other information discussed above, we assessed

DOE’s internal controls for ensuring security at its facilities and work

stoppage preparedness.

Federalization

To determine the advantages and disadvantages of federalizing the Los

Alamos security force, we interviewed DOE headquarters, Albuquerque,

Los Alamos, and the Office of Management and Budget (OMB) officials as

well as Mason and Hanger security force members. We reviewed four

cost analyses that DOE had prepared in the early 1980s. Also, we developed

wage and benefit cost data for the nine facilities in our scope and a

hypothetical federal force. We analyzed wages and benefits because (1)

2Pantex, Texas; Los Alamos, New Mexico; Pinellaa, Florida; Kansas City plant, Missouri; Mound, Ohio;

and Rocky Flats, Colorado.

they represented 60 to 76 percent of the costs in DOE'S four analyses and

(2) an OMB official said that labor costs normally represent about 86 to

90 percent of a contract guard force costs. The costs that we did not

analyze included overtime and shift differential pay, operations overhead,

general and administrative expenses, and contract administration

that would normally be part of a full cost study.

We also reviewed the Law Enforcement Pay Commission report. We determined that the Commission’s findings were not applicable to the scope and nature of this review because the report addressed only specific law enforcement

officers and jobs, such as the Federal Bureau of Investigation, and did

not include federal security inspectors and guards.

We discussed the facts presented in this report with DOE, Albuquerque,

Los Alamos, and Mason and Hanger officials. They generally agreed

with the information but offered some clarifications, which we incorporated

where appropriate. As requested, we did not ask these officials to

comment, officially on this report. Our work was conducted between

July 1989 and May 1990 in accordance with generally accepted government

auditing standards.

Concerns About the Adequacy of Security at

Los Alamos

Although DOE and Los Alamos officials believe that security was adequate

during the strike, we were unable to verify this assertion. However,

we did find indications that potential weaknesses in overall

security exist. During the strike, DOE did not conduct a force-on-force

simulated attack to verify the replacement force’s proficiencies. Also, 2

weeks elapsed before DOE conducted an inspection, and 6 weeks elapsed

before DOE conducted a limited-scope test of the replacements’ performance.

In addition, DOE waived physical and medical requirements for

almost 60 percent of the 391 replacements, and most were not certified

in the minimum job skills required of the regular security force.

Throughout the 198Os, three strikes occurred at other DOE facilities; yet,

neither DOE nor Los Alamos was prepared for the strike. For instance,

DOE had not provided its contractors guidance on how to prepare for or

deal with a strike. As a result, Los Alamos’ contingency plan did not

specify all the actions that should be taken during a strike, Los Alamos

is not unique in this regard; we found similar weaknesses in the contingency

plans of five other DOE facilities that we reviewed. As a result of

the Los Alamos strike, in February 1990 LIOE headquarters sent criteria

to its operations offices and contractors to use for preparing strike contingency

plans.

Aside from the strike situation, the regular Los Alamos security force

may not be properly trained or proficient in protecting other employees,

laboratory assets, or themselves. First, Mason and Hanger’s training and

certification records for 1989 were incomplete, inaccurate, or missing.

Second, the results of the unannounced test that DOE conducted at our

request in April 1990 showed that about 75 percent of the Los Alamos

security force were not proficient in 1 or more of the 12 minimum

required skills.

Was Security Adequate During the Strike?

DOE and Los Alamos officials contend that security was adequate during

the strike. However, we identified several issues that raise questions

about their position. For example, DOE did not conduct inspections

during the early days of the strike, when the facility was the most vulnerable

because not all replacements were on board and those that were

on board were not familiar with their duties, weapons, or the uniqueness

of the site and terrain. In addition, under its policies, DOE is required to

periodically test security force performance, particularly when changes

occur in procedures, measures, or practices.’ However, DOE did not test

the proficiencies of the replacement force until 6 weeks after the strike

began and then only on a limited basis and never conducted a force-onforce

simulated attack-the best measure of a security force’s overall

ability to effectively protect life and property at a nuclear weapons

facility. According to WE officials, a force-on-force performance test

would have been impractical during the strike because the extra staff

needed to simulate an attack were not available.

The strike began on March 13, 1989, but 2 weeks passed before DOE conducted

its first inspection to determine whether the replacements were

competent and capable of protecting Los Alamos. DOE found no problems

during the inspection, which involved visits to guard posts and interviewing

personnel, but the inspection did not include performance tests

of any of the 12 minimum security force skills.

Then, in mid-April 1989 (about 4 weeks into the strike), staff from the

Subcommittee on Oversight and Investigations, House Committee on

Energy and Commerce, notified DOE that they had received allegations of

many instances of firearms “horseplay” by the replacement force. The

following day, two Albuquerque officials conducted an inspection.

Although they found no misuse of firearms, the officials did find that 12

of 30 replacements interviewed lacked proficiency with various

weapons that they were required to use. In response, Mason and Hanger

provided weapons training to the 12 individuals.

Even with this information, DOE did not conduct performance tests. Only

in late April 1989, after another inquiry from the Subcommittee staff

about the replacement force’s training and competency, did DOE conduct

a limited-scope test. DOE conducted the test at two sensitive areas at Los

Alamos using three scenarios: entry with (1) a pipe bomb, (2) an incorrect

badge, and (3) illegal drugs. DOE found that the replacement force

responded correctly during the test.

Later, DOE conducted two additional inspections and identified some

security problems, such as failure to find drug equipment during a simulated

entry attempt, but the reports stated that security was adequate,

at a high state of readiness, or satisfactory. Albuquerque officials told

us that about 5 weeks into the strike they assigned a security official to

monitor, not test, the replacement force until May 16, 1990-l week

before the strike ended. The official’s reports stated that no problems

relating to excessive fatigue, incompetencies, or other security-related

problems existed.

Many Replacements Did

Not Meet Critical Skills

Many of the replacement force did not meet 1 or more of the 12 minimum

skills required of the regular Los Alamos security force. This situation

occurred because DOE'S policies do not specify that temporary

replacements should possess all such skills.

The several hundred replacements consisted of auxiliary and augmentee

personnel. Mason and Hanger, the Department of Defense, and DOE'S

Pantex facility in Texas provided most of the auxiliary personnel. Auxiliary

replacements normally work in scientific or engineering rather

than security-related jobs and, in accordance with DOE'S policies, fill

security inspector positions on an “as-needed” basis upon request (usually

in an emergency), On the other hand, the augmentees were full-time

security inspectors from other DOE facilities, such as the Nevada Test

Site, Savannah River plant, and the Oak Ridge Y-12 plant. As shown in

table 2.1, many replacements were not certified in the minimum skills

required of the regular Los Alamos force.

Table 2.1: Number of Replacement Force

That Were Not Certified In Certain Skill8 Replacement force

Skill Auxiliary Augmentees’

Physical fitness 7

Arrest 38 35

Baton

Weapons:

191b 66

Pistol (day) 7 7

Pistol (night) 190 8

Shotgun (day) 7 14

Shotgun (night)

Rifle day \ 7

Rifle (night 190 20

‘Some individuals lacked more than one skill

No auxiliary personnel were issued batons.

Also, 34 auxiliary personnel did not have the required medical certifications

to show that they had the necessary mental, sensory, and motor

skills to perform their assigned duties safely and effectively. Albuquerque

realized this about 2 weeks into the strike. Albuquerque also

realized that none of the auxiliary personnel were certified in the physical

fitness requirements of the regular force-run a 40-yard dash and

at least a one-half mile distance run. MIE considers these requirements

important because protective forces must perform normal and emergency

duties without undue hazard to themselves, fellow employees, the

site, or the public. Nevertheless, Albuquerque requested, and oss

granted, a waiver of the medical and physical fitness requirements for

the auxiliary personnel.

In addition, DOE did not have qualification requirements for the auxiliary

force even though sufficient time had elapsed since three prior

strikes at other DoE nuclear weapons facilities.2 Although the University

of California required the replacement force to be qualified in weapons

(pistol, shotgun, or rifle) needed to protect Los Alamos and capable of

performing all Los Alamos protective force duties, some auxiliary personnel

were not certified to use a pistol, shotgun, or rifle. Also, the auxiliary

force were not trained on night use of weapons, even though some

were assigned to the night shift during the strike. Finally, none of the

auxiliaries were issued batons because they were not qualified to use

them.

Although a scientist or engineer (auxiliary personnel) may not have all

the required security force skills, the more perplexing issue is: Why

would a regular security inspector from another DOE nuclear weapons

facility (an augmentee) lack all the skills needed to protect Los Alamos?

The simple answer, according to the Director, oss, and other DOE officials

is that each security force contractor establishes different competency

requirements. For example, some contractors require proficiency in

using a shotgun, rifle, or baton, while others do not. DOE officials also

noted that each site is unique; therefore, if a contractor’s security

inspector augments another contractor’s security force, the security

inspector may not possess all the needed skills for a particular location.

Also, the Director, OSS, told us that between 12 and 18 months are

needed for new security staff to learn the tactics, geography, and

targets at a particular facility.

Because of the lessons learned from the strike, DOE issued several memoranda

concerning the use of replacements from other DOE sites. In January

1990 ass issued a memorandum stating that allowing unqualified

Oak Ridge in 1080, Pantex in 1981, and Oak Ridge in 1983.

individuals to use weapons raises questions about their ability to carry

out routine and emergency duties and exposes DOE to unnecessary liability.

A February 1990 memorandum stated that the Director, DOE'S

Central Training Academy, would collect and maintain a computerized

data base of qualified security inspectors and special response team personnel

throughout the DOE complex for use in extreme emergencies.

DOE Sites Were Not

Prepared for Strikes

Neither DOE nor Los Alamos was prepared for the strike because DOE'S

policies pertaining to the content of plans for foreseeable contingencies

do not require contractors to specifically address the actions that should

be taken if a strike occurs.3 We found that contingency plans prepared

for Los Alamos and five other DOE nuclear weapons facilities varied in

detail and coverage. For example, four plans identified posts that could

be shut down, curtailed, and/or consolidated while two did not.

Following the strike, both DOE headquarters and Albuquerque initiated

actions to improve contingency planning for strikes. In June 1989 Albuquerque

asked seven facilities, including Los Alamos, to submit their

plans for review.4 Albuquerque wanted to determine whether the contractors

could effectively deal with emergencies, especially security

force strikes. Six contractors submitted their plans; the seventh (Sandia)

did not have a plan because contractor officials believed that productive

bargaining would prevent a strike.

On the basis of its review of the plans and lessons learned from the

strike, Albuquerque’s Security and Nuclear Safeguards Division drafted

contingency plan criteria that included 18 elements. Our comparison of

the six plans with Albuquerque’s draft criteria showed that none met

even 60 percent of the criteria (app. II shows our comparison of the six

facilities compliance with Albuquerque’s draft contingency plan

criteria).

In August 1989 ass sent a memorandum to the operations offices that

included minimal contingency plan criteria. For example, the memorandum

stated that the plans should identify those security functions

that could be performed by other staff. The memorandum also

requested each facility contractor to submit its contingency plan to OSS.

DOE Order 6632.7, Protective Forces, Feb. 9,lSS.

Los Alamos, New Mexico; Kansas City, Missouri, Mound, Ohio; Pantex, Texas; Pinellas, Florida; Rocky Flats, Colorado; and Sandia, New Mexico.

None did so, and the office did not follow up on its request to get the

plans submitted.

In February 1990 OSS sent another memorandum to the operations

offices requesting them to submit a summary of their facility contractors’

contingency plans by May 1990. The memorandum also instructed

each operations office to incorporate a contingency plan section in its

future Master Safeguards and Security Agreement6 According to an oss

official, the memorandum reminded the operations offices that contingency

plans were now required for strikes and was intended to serve as

guidance, rather than requirements, on preparing the plans.

 We noted that the memorandum identified only 4 of the 18 elements in Albuquerque’s draft contingency plan criteria. Albuquerque officials told us

that they required contractors under their purview to revise their plans

by April 1990 in accordance with the memorandum. All did so. Because

Albuquerque received these plans near the end of our work, we could

not assess the information provided.

Security Force May Not Have Been Properly Trained

We found that many training and certification documents for the

security force before the strike were missing, incomplete, undated,

changed, or unsigned. Without accurate and complete documentation,

neither Mason and Hanger nor DOE can demonstrate that the regular Los

Alamos security force is properly trained to protect the facility.

DOE’S policies specify the training and physical requirements that a

security force must meet and require contractors to maintain records

showing that they have complied with the policies.6 Although Mason

and Hanger officials told us that they retain training records indefinitely,

we found the opposite- records for training provided in 1989 to

about 330 security inspectors were missing or had such deficiencies as

being incomplete, inaccurate, unsigned, or undated. Table 2.2 shows our

findings for arrest, baton, and physical fitness.

“An agreement between the Assistant Secretary for Defense Programs and an operations office concerning

the types of security measures to be taken, risks to be considered, and ways to increasesecurity effectiveness.“

DOE Order 5632.7, Protective Forces, Feb. 9, 1988.

Table 2.2: Incomplete, Missing, or

Deficient Training Records

Skill

Number of Number of

Number of records.

Number with employees available missing deficiencies

Arrest 333 244 89 121

Baton 333 306 27 a

Physical Fitness 333 312 21 0

Although all records indicated a perfect score, we verified 21 randomly selected records and found that  only 5 were signed, dated, and completely filled out. Further, 50 percent of the security inspectors that we interviewed said they needed additional training in this area.

Concerning weapons proficiencies, we cross-checked 14 randomly

selected source documents to Mason and Hanger’s computerized information

and found one missing record. The remainder had no data entry

errors. However, we noted three problems with the source documents:

(1) the range master (firearms instructor) certified his own qualification

record, (2) some records had been altered by erasures or correction

fluid, and (3) all entries were in pencil instead of ink. In addition,

security inspectors must be medically fit to perform their assigned

duties. We found that Mason and Hanger’s computerized medical records

were not up to date, and 47 of 333 inspectors (about 14 percent) had

missed part of their annual medical examination.

According to Mason and Hanger officials, the documentation problems

occurred because they did not have sufficient resources (staff and

funds) to properly maintain the documents. To correct these problems,

Mason and Hanger officials said that the company (1) has sent three

employees to record keeping classes, (2) is obtaining new computer hardware

to provide greater record keeping capability, (3) is updating

training records, and (4) is identifying and scheduling training for some

of the security force. Mason and Hanger officials said they took these

actions not only because of our findings but also because they were concerned

that security inspectors might bring injury suits or otherwise

hold the company liable for failing to train them as happened in a recent

city government case.7 In such a situation, accurate and timely training

documentation would be vital. A company official also said that, if

employees won the suit, the costs would be passed on to the government

because their subcontract with the University of California is cost reimbursable

as is the University’s contract with DOE.

According to the Director, ass, DOE recognizes that these problems exist

not only at Los Alamos but throughout the nuclear weapons complex. As

a result, DOE will require the Central Training Academy to assess the

status and quality of training provided to all security forces. DOE also

plans to ask the Central Training Academy to develop a standardized

course for force members and certify the qualifications of those individuals

who will provide the training,

Los Alamos’ Security Force Did Not Perform Well During a Surprise Test

According to some DOE officials and Nuclear Regulatory Commission

staff, a no-notice test is the best way to assess a security force’s ability

at any given time. DOE'S policies allow for such tests, but DOE has not

conducted unannounced tests in the past because, according to officials,

they raise safety concerns, are difficult to plan, disrupt the work force,

and create stress for all participants. Because of the training record

problems that we found, OSS in April 1990 conducted an unannounced

test at our request of the regular Los Alamos security force in 9 of 12

skills, including weapons, baton, running, and arrest or defense tactics.

According to the Director, OM, these skills are similar to those required

of a police officer and are applicable to the majority of DOE'S security

force members. Table 2.3 shows the number of security force members

that participated in the tests and those that did not meet the skill

requirements.

Table 2.3: Results of an Unannounced Test

Skills

Weapons54

Number Failed

passed Number Percent

52 2 4

Baton 54 51 3 6

Running 50b 50 0 0

Apprehension 54 15 39 72

Test summary 54 12 42c 78

The apprehension test combines six skills: force and arrest, security operations, communications, tactics, self-defense, and site protection.

Three participants were not medically fit to run, and one was on sick leave the day of the test.

The total does not add because some participants failed more than one test.

Although the security force had a basic understanding of selected skills

and generally met the handgun, baton, and running requirements, only

12 (22 percent) of the 64 force members tested passed all 9 skills. The

remaining 78 percent lacked one or more of the skills needed to arrest,

apprehend, communicate, and survive in an adversarial situation; protect

laboratory resources or staff; or defend themselves. For example,

when the participants encountered an adversarial situation, many failed

to stay behind cover and assess the situation. Instead, they left their

cover and walked up to the potential adversary to ask what they were

doing. As a result, in many instances the adversary took a visible

weapon, “killed” the participant or hostage, and left with the classified

documents or government property. In total, 24 participants and hostages

were “killed” during this testing.

If we project the test results to the 191 security force members sampled,

149 would lack one or more of the above skills. As discussed in chapter

3, security force performance weaknesses have been a longstanding

problem at Los Alamos. DOE inspections conducted in 1986 and 1988

identified problems similar to those found in our unannounced test.

Conclusions

A security force must be able to protect DOE'S sensitive nuclear weapons

facilities from such threats as terrorist attacks, unauthorized entry, and

theft of classified documents. To minimize the potential for a security

breach, the security force must meet certain employment qualifications

and continually maintain its skills. When a security force-the first line

of human defense-cannot perform its duties, is ineffective, or improperly

trained, little assurance exists that sensitive facilities are appropriately

safeguarded. Yet, this is the situation that we found at Los Alamos

before, during, and since the 1989 strike.

First, the available evidence does not show that either the regular or

replacement force was properly trained to protect Los Alamos. Training

and proficiency records for the regular security force were missing,

incomplete, or inaccurate. Second, many of the strike replacements did

not meet all the skills required of the regular force. Finally, the results

of our unannounced exercise showed that about 75 percent of the

security force- as late as April 1990-did not have all the skills needed

to protect the facility or defend themselves.

Also, about 60 percent of the replacements used during the strike normally

worked in jobs other than protective services, and DOE waived

some requirements for these replacements, thereby increasing the vulnerability

of the site. Also, DOE did not test the proficiencies of the

replacements until 6 weeks after the strike began-and then only on a

limited basis-and never conducted a simulated, force-on-force, adversarial

test over its duration. Instead, DOE relied on limited security

inspections and the on-site observations of an Albuquerque official to

ensure that Los Alamos was appropriately protected.

We believe that these situations occurred because neither DOE nor Los

Alamos was prepared for the strike. DOE had no skill requirements for a

temporary replacement force and did not require its contractors to

develop contingency plans for strikes. Although three other strikes had

occurred at DOE facilities throughout the 198Os, DOE did not take such

threats serious enough and require all contractors to appropriately plan

for work stoppages that could pose a security risk. In a February 1990

memorandum, DOE attempted to correct this oversight. The memorandum,

however, provided only limited specifics on the plans contents,

thereby increasing the potential that contractors will not address significant

issues for dealing with strikes.

Recommendations to

the Secretary of

Energy

.

.

.

.

To continuously and completely protect sensitive and valuable documents,

personnel, and government property, DOE needs trained and proficient

security forces at its facilities. Therefore, we recommend that the

Secretary of Energy

expeditiously develop specific contingency plan criteria for strikes and

require all contractors to prepare plans that meet the criteria;

establish standardized qualification and skill requirements for all protective

forces and ensure that strike replacements meet the

requirements;

ensure that security force members receive all required training and

institute a mechanism to ensure that contractors document and retain

this information; and

conduct unannounced inspections and performance tests, particularly

immediately upon the initiation of an unusual event, such as a strike, to

obtain more realistic indications of security force competencies.

Chapter 3

DOE’s Security Inspection Process Can

Be Improved

,

DOE Lacks Specific

Criteria for Rating

Facilities

Weaknesses exist in DOE’s security inspection and evaluation process.

DOE lacks specific criteria for the types of deficiencies that would result

in either a satisfactory or unsatisfactory rating for a facility’s security

force. As a result, DOE inspections between 1985 and 1989 identified similar

and recurring problems at Los Alamos and eight other facilities, yet

DOE rated six facilities satisfactory, two marginally satisfactory, and

only the Argonne National Laboratory as unsatisfactory. Despite these

longstanding inconsistencies, DOE has not developed specific rating criteria

or performance incentives for security force contractors to minimize

poor performance.

Further, DOE does not have appropriate controls to ensure that contractors

take corrective actions on security program weaknesses identified

during inspections. We found that some deficiencies went uncorrected

for as much as 5 years even though DOE’S computerized system showed

that contractors’ had taken corrective actions. This situation may not

have occurred if DOE had confirmed that contractors took the needed

actions. In April 1990, Albuquerque took steps to ensure that corrective

actions had actually been taken on inspection findings.

DOE’S policies do not specify the severity and frequency of inspection

findings that would result in a satisfactory, marginal, or unsatisfactory

performance rating. Therefore, inspection ratings can vary even though

the results are similar and recurring. Further, DOE’S reports did not indicate

that the uniqueness of a site would cause variances in the inspection

ratings assigned. Between 1985 and 1989, DOE inspections found

some similar and recurring weaknesses at nine facilities including Los

Alamos but only one received an unsatisfactory rating. Under DOE’S policies,

OSE, ass, and operations offices, such as Albuquerque, are required

to periodically inspect the security activities conducted by the contractors

that operate nuclear weapons facilities.’ On the basis of the inspection

results, both headquarters and the operations offices assign a rating

of satisfactory, marginal, or unsatisfactory to the contractor’s program.

Within headquarters, ass and OSE are responsible for periodically

inspecting the weapons facilities. According to an OSE official, the

inspection team provides an unrated draft of its findings to the operations

offices for comments and then the report and the team’s proposed

rating are reviewed by inspection branch chiefs and division director

and OSE’S Director. These individuals either concur with the suggested

rating or, with the team leader, adjust the rating, considering technical,

management, and other issues, such as the contractor’s performance in

each topical area reviewed.

According to an official, OSE does not have clear-cut criteria specifying

the values assigned to deficiencies that result in a facility being rated

satisfactory or unsatisfactory. The official also noted that, although OSE

strives for objectivity by involving more than one individual in the process

and examining a number of parameters, the ratings will always be

highly subjective. Also, an Albuquerque official, involved with inspections

at OSE and the operations office, confirmed that no criteria exist

for inspectors to consider the severity of findings when making rating

recommendations.

Thus, we found that DOE assigned satisfactory ratings to facilities even

though the inspection reports identified numerous deficiencies. Between

1986 and 1989, DOE identified similar and recurring inspection problems

at nine facilities but rated six as satisfactory, two as marginally satisfactory,

and one-Argonne National Laboratory-as unsatisfactory.

Table 3.1 shows the types of findings at the nine facilities.

Table 3.1: Security Force Weaknesses

Cited in DOE Inspection Reports, 1985- Number of inspections*

99 -Type of finding A B C D E F G H lb

Not skilled on metal detector

Cannot find concealed weapon

1

1

Cannot find drug equipment 1

Unfamiliar with weapons 2 1

Command/control weaknesses 1 1 1 1

Handcuff/search/arrest weaknesses 3 3 1 3 1 1

Cannot appropriately apprehend suspects 1 2 1

Failed to keep weapon from adversary 1

Failed to search following handcuffing 1

Not skilled in Ml 6 1

Unfamiliar with night vision devices 1

Radio/communication weaknesses 1 2 3 1

Patrol not reporting security condition I

Failed to control security situation 1

Undocumented performance tests 1 1 1 1

Lacked firearm skills 2 1 1 1 1 1

No firearms training 1

First aid program deficiencies 1

Training program deficiencies 2 2 I 2 1 1 1

Security inspector missing from post 1

Past-due medical examinations l

Hostile aircraft training weaknesses 1 1

Patrols not conducted to procedures 1

Lacked theft/diversion requirements 1

Failed to assign weapons to post  1

Supervisors lack knowledge

Lack emergency warning signal knowledge

Poor weapons inventory/maintenance

Number of inspections that identify the finding noted

“A = Los Alamos National Laboratory, New Mexico.

B = Argonne National Laboratory, Illinois.

C = Rocky Flats, Colorado.

0 = Oak Ridge Y-12, Tennessee.

E = Savannah River, South Carolina.

F = Lawrence Livermore National Laboratory, California

G = Nevada Test Site, Nevada.

H = Sandia National Laboratories, New Mexico.

I = Pantex, Texas.

As shown in table 3.1, DOE found that the security forces at Argonne

(three inspections), Savannah River (three inspections), Sandia (one

inspection), and Los Alamos (three inspections) could not appropriately

handcuff, search, or arrest intruders and lacked weapons skills and

accuracy. DOE also found weaknesses in the training programs at the

four facilities. Despite finding similar problems, DOE only rated the

security program at Argonne as unsatisfactory. Because DOE lacks criteria

concerning the extent to which severity is considered in rating a

security force program, we could not determine-and DOE officials could

not effectively explain -the rationale for assigning different ratings.

According to the Director, ass, training and headquarters oversight are

more important than developing severity criteria. He said that DOE has

attempted to develop severity models to overcome dissimilarities with

ratings, but the attempts had not been successful. However, during the

spring of 1990, DOE reorganized oss to help improve rating consistency

by, in part, increasing the training provided to inspection teams.

DOE Does Not Have an Effective System to Track Corrective Actions Taken

 DOE does not have an effective system to monitor and follow up on

actions taken as a result of its inspection findings. As a result, we found

that security force performance weaknesses identified as much as 5

years ago had not been corrected as of May 1990. For example, in 1986

OSE found that the Los Alamos security force could not effectively

detain and/or arrest intruders. In 1988, an Albuquerque inspection identified

the same deficiencies. The University of California reported to

Albuquerque that corrective actions had been taken to resolve these

problems. We found that these problems still existed as of April 1990.

According to ass officials, they developed a classified computerized

information system, the Safeguards and Security Issues Information

System (SSIIS) in June 1985 to help monitor inspection weaknesses

found at all facilities and ensure that corrective actions were taken. In

1989 DOE found that the SSIIS data were not reliable or complete and took

actions to update and correct the data. As of February 1990, the system

contained information on DOE headquarters inspection findings from

November 1988 and operations offices’ findings from January 1989.

06s officials say that SSIIS was used primarily to monitor the status of

corrective actions and ensure that headquarters received complete,

timely, and reliable data on operations offices’ findings. An auxiliary

benefit was to encourage operations offices to develop data bases to

monitor inspection findings and corrective actions taken by the contractors

under their jurisdiction. According to Albuquerque security officials,

they do not use SSIIS because the data are not complete or reliable,

and their staff could not access the data directly from their computers.

To correct these situations, oss officials told us that they are developing

a new data base to replace SSIIS. The officials estimated that the system

would be available to the operations offices after about a year.

In the interim, Albuquerque uses three data bases, to track inspection

findings and corrective actions taken. One provides a facility-specific

summary of the number of findings, those requiring corrective actions,

and their status. The second data base provides more detailed information

on inspection findings and the status of corrective actions.

According to an Albuquerque official, the office strives to keep the

information current, but some backlogs have occurred because staff

were not available to input the data. Albuquerque has also developed a

data base to track OSE’S inspection findings for contractors under its

purview.

Despite having three systems to track their own and DOE headquarters

inspection results, as late as March 1990, Albuquerque did not know the

status of corrective actions taken on inspection weaknesses identified as

early as 5 years ago. According to an official who tracks this data, Albuquerque

did not have up-to-date information on DOE headquarters findings

until 1989. The official subsequently found that six contractors

under Albuquerque’s jurisdiction had not corrected security program

deficiencies identified as early as 1985.

As a result, Albuquerque plans to more closely monitor the contractors’

activities. In April 1990 Albuquerque sent a memorandum to its area

offices requiring officials to validate corrective actions taken as a result

of internal-board headquarters and Office of Inspector General-and

external-General Accounting Office-reviews and/or inspections. The

area offices must certify in writing that the actions have been taken;

Albuquerque security personnel will take a sample during subsequent

inspections to validate the area offices’ certification.

Other Options to

Ensure Corrective

Actions

that contractors institute effective programs to protect and secure

nuclear weapons facilities; the operations offices can curtail or suspend

facility operations if an immediate and unacceptable national security or

public health and safety risk exists.2 Also, DOE’S contracts with the University

of California and others that operate nuclear weapons facilities

state that DOE can terminate a contract when DOE determines that termination

is in the best interest of the government, such as for unsatisfactory

performance.

Between 1985 and 1989 DOE had identified numerous weaknesses in contractors’

security programs. Also, in three evaluations (one in 1986 and

two in 1989), DOE had identified numerous, repeat security program

deficiencies at Argonne that eventually resulted in one marginally satisfactory

and two unsatisfactory ratings. In 1989 DOE curtailed the operations

at Argonne until corrective actions could be taken and replaced the

contractor in June 1990. According to officials, DOE prefers to work with

contractors to correct security force weaknesses rather than terminate a

contract for poor performance.

However, DOE has another mechanism that it could use to encourage contractors

to take timely and effective corrective actions on security

inspection deficiencies- the awards fee process. DoE uses award fees,

over and above reimbursing normal costs, to encourage effective contractor

performance. We noted that in fiscal years 1987 and 1988 DOE

had delineated security-including protective forces and systems-as a

functional area for the award fee determination at Pantex and the Oak

Ridge Y-12 plant but not at Rocky Flats. However, the weight-or

importance given to security-varied between the facilities. At Oak

Ridge, DOE consistently considered security as 10 percent of the total fee;

at Pantex, the weight ranged from 10 to 20 percent.

In November 1989 DOE published in the Federal Register a final rule for

withholding award fees if a contractor fails in one major functional area.

For facilities with repeat security inspection deficiencies, DOE could

include a functional area in the award fees process to specifically measure

contractor performance, including the timing, and effectiveness of,

corrective actions taken for inspection findings.

Conclusions

The security at other DOE facilities may be as questionable as the situation

that we found at Los Alamos. DOE inspections at other facilities

since 1986 have identified recurring and similar security force weaknesses

that seem to justify a less than satisfactory rating. Yet, in only

one instance-Argonne- did DOE determine that the security program

was unsatisfactory. We believe that in the highly important areas of

security and adequacy of protective services at facilities involved in the

research, development, or production of nuclear weapons, DOE should

take a conservative approach and should not allow security weaknesses

to persist. If DOE finds the same problems at more than one facility, then

DOE should similarly assess the level of security. If one situation warrants

an unsatisfactory rating, it seems reasonable that other facilities

would be similarly rated.

Also, DOE'S actions can send a message to contractors that security is not

important and perpetuates an environment in which corrective actions

are not taken on the problems identified. In this regard, DOE does not

have an effective mechanism to ensure that actions are taken to correct

inspection weaknesses. The ineffectiveness of DOE'S process is best illustrated

by the fact that inspections of the nine facilities over 5 years

identified some of the same problems; therefore, many deficiencies did

not get corrected. If DOE used the award fee or some other mechanism to

affect the contractors’ profits or instituted other punitive measures,

such as terminating security force contracts for poor performance, the

contractors may pay greater attention to quickly and effectively correcting

weaknesses identified in security inspections.

Recommendations to the Secretary of Energy

To ensure consistency among inspection ratings and provide an incen-

tive for security forces’ contractors to correct inspection deficiencies, we

recommend that the Secretary of Energy develop specific criteria to eliminate any inconsistency for rating facility’s security as either satisfactory, marginal, or unsatisfactory and withhold a portion of award fees when contractors do not take timely corrective actions on security inspection weaknesses,

Some Contract Forces May No Longer Be Cost-

Effective

We estimate that federal labor and benefit costs would be at least $15

million less each year than similar contract costs at the nine M)E facilities

we reviewed. Nearly all 5,600 security personnel that protect DOE'S

nuclear weapons facilities are contractor employees. In the early 1980s

DOE assessed the costs of contracting for security functions at four facilities

and found that contract costs were less than federal costs. DOE has

not updated the analyses or conducted additional ones to determine

whether contract costs are still less than federal costs.

According to DOE officials, both contract and federal forces are equally

capable of protecting sensitive nuclear facilities, but a critical factor is

the force’s ability to provide uninterrupted service. A major advantage

of a federal force is that it cannot legally strike, whereas a major disadvantage

of a contract force is that it generally can. The Los Alamos

strike cost about $1.6 million over and above the almost $17 million contract

cost. According to a DOE Office of General Counsel official, no legal

obstacles exist to DOE'S negotiating a never-strike provision in its

security force contracts but estimated that it would be costly to do so.

Also, turnover may be lower with a federal force. During the 26 months

before the strike, Los Alamos experienced between 1 I- and 15-percent

turnover; the Albuquerque Operations Office federal force experienced

no turnover. In contrast, a contract force, according to DOE and Los

Alamos officials, can more quickly be reduced or increased to meet

changing work demands. Although both types of security forces offer

advantages and disadvantages, they generally offset each other, and the

primary issue becomes cost.

DOE believes that federalizing security force services may be more

acceptable today than under previous administrations, which emphasized

privatization of such activities. Also, OMB officials told us that they

would work with M)E to prevent a recurrence of past budgetary

problems that hindered the hiring of federal employees for security

force positions.

Federal Security Force May Be More Cost- Effective at Some Locations

For nine DOE facilities we reviewed, we found that DOE could save about

$16 million annually in labor and benefit costs by converting to federal

forces,’ We estimated that contract labor and benefit costs ranged from

6 to 38 percent higher than similar federal costs at the facilities. Table

4.1 shows the contract and federal security force costs at the nine

facilities.

Table 4.1: Contract Versus Federal Labor

and Benefit Costs at Nine DOE Facilities Security force costs Difference

Facility Contract Federal Dollars Percent

Argonne $1,538,950 $1,336,500 $202,450 15

Lawrence Livermore 7559,378 5,986,156 1,573,222 26

Los Alamos 11,782,093 9,556,950 2,225,143 E

Nevada Test Site 13.547,681 10,154.262 3,393,419 33

Oak Ridge(Y-12)

Pantex -

Rocky Flats

Sandia

16,588,826 14,760,534 1,828,292 12

10,933,211 10,040,871 892,340 9

9,867,181 7,146,206 2,720,975 38

5,495,996 49585,083 910,913 20

Savannah River 19,983,746 18,951,143 1,032,603 5

Total $97,297,062 $82,517,705 $14,779,357

Total number of employees at the nine facilities was over 3,000; total contract security force costs were $193.1 million.

However, the estimated annual savings shown in table 4.1 could be even

higher. First, we did not include overtime costs in this comparison. Contractor

salaries are more than federal salaries; therefore, overtime,

which is calculated as a multiple of base pay, would be higher for a contract

than federal force. Second, the estimated federal salaries were

developed using the salaries of DOE’S federal nuclear materials courier

force.2 The courier force salaries would be higher than federal security

inspector’s because the couriers have higher skill and training requirements.

For these reasons, our estimated annual labor and benefit savings

of about $15 million may be conservative.

Also, DOE’S Inspector General Office is currently evaluating the cost effectiveness

of selected support service contracts. According to Inspector General officials, their analyses will include all costs and their tentative findings indicate that significant cost savings could result from

federalizing the services performed by these contractors.

‘The nine facilities represented more than 60 percent of DOE’s 5,600 total security force members for1989.

‘DOE has three federal courier forces. They are comparable to the Los Alamos security force because both are responsible for protecting sensitive material. However, specific emergency reactions are different  because the security force protects a fixed site while the courier force protects nuclear material

transported by truck between DOE facilities.

DOE Has Not Updated Its Cost Comparisons

DOE officials believe that the costs for a contract or federal force would

be similar. However, DOE has not updated four cost comparisons prepared

in the early 1980s nor has it conducted additional analyses. On

the basis of the earlier analyses, DOE converted three federal forces to

contract forces (including Los Alamos); the Albuquerque Operations

Office force remained as the only federal force. The 1980 Los Alamos

study concluded that DOE could save about $2 million over a 3-year

period by replacing the federal force with a contract force.

In addition, a 1981 DOE cost study for Oak Ridge showed that the agency

could save $274,000 over a 3-year period by replacing the federal force

with a contractor force. In 1982, we found that DOE’S study was deficient

and concluded that DOE’S converting the Oak Ridge force may increase

government costs by as much as $1.2 million over the 3-year periodq3 At

that time, we recommended that the Secretary of Energy reassess the

decision to contract for the Oak Ridge guard services. DOE did not terminate

the contract. The cost results presented in table 4.1 show that the

Oak Ridge contractor’s 1989 wage and benefit costs exceeded federal

force costs by more than $1.8 million.

According to DOE officials, they have not updated or conducted additional

cost comparisons because (1) the prior administration emphasized

privatization throughout the 198Os, (2) the government’s policy is to

contract for security forces unless some “overwhelming” reason precludes

doing so, and (3) DOE could not obtain OMB’S approval for the

number of federal positions needed. An OMB official told us that agencies

should periodically conduct a cost analysis if the difference in contract

and federal labor and benefits costs are close to, or exceed, 20 percent.

According to OMB officials, a difference of less than 20 percent would

likely be obliterated through (1) a process that allows contractors to

resubmit lower bids and (2) the lo-percent conversion penalty for

unforeseen costs that federal agencies must add to their estimated in house

costs.

For five of the nine facilities shown in table 4.1, the estimated costs of a

contract force were at least 20 percent higher than a federal force; four

were less than 20 percent. Yet, DOE has not reanalyzed the cost differences

between contract and federal forces since the early 1980s.

According to the Chief, Management Systems Development/Evaluation

Branch, DOE has not done so because the operations offices have more

pressing problems, such as the billions of dollars of environmental

cleanup, and do not have the staff to initiate multiple cost comparison

studies.

Some Aspects of Contract and Federal Security Forces Offset Each Other

According to DOE officials and others with whom we met, either contract

or federal security forces can protect DOE facilities. According to a 1976

Nuclear Regulatory Commission report, neither option offers an overall

advantage over the other; therefore, the deciding issue is cost. One of

the major disadvantages of a contract force is that it can strike, which

generates security concerns and increases costs. The Los Alamos strike

cost about $1.6 million over and above the almost $17 million fiscal year

1989 contract cost. On the other hand, a federal force would eliminate

contract administration costs, which vary according to the force size.

Although several strike prevention mechanisms exist, they also carry a

price tag. For example, a never-strike provision that extends beyond the

contract expiration date and through renewal negotiations would provide

DOE a mechanism to eliminate work stoppages. According to an

Albuquerque Office of General Counsel official, no legal obstacles exist

to DOE's negotiating a never-strike provision in its security force contracts,

but the official believed that such a provision would be costly

and estimated that the union might require a 26-percent increase in

wages before agreeing not to strike.

A number of other differences exist between federal and contract forces.

For example:

l Federal protective forces may experience less turnover. During the 26

months before the strike, Mason and Hanger experienced turnover rates

of between 11 and 15 percent. Albuquerque’s federal guard force experienced

no turnover during the same period. Also, DOE'S nuclear materials

couriers experienced turnover rates as low as 8 and 10 percent during

1987 and 1988 even though the employees worked more overtime

(about 76 percent) than the Los Alamos force (about 40 percent).

. A contract force can more rapidly respond to changes in workload

demands. According to officials, one reason that DOE contracted for

security services at Los Alamos was that OMB did not authorize additional

personnel as security demands increased. Although the federal

personnel register included individuals wanting to be on the Los Alamos

security force, DOE could not get additional positions approved in the

early 1980s.

. A contract force can more easily terminate an employee who does not

meet the required qualifications and skills. A member of DOE'S security

task force study said that a federal force has built-in protection and terminating

an employee is a slow and cumbersome process.

Although costs and turnover rates may be lower for a federal force, DOE

has never converted a contract force to a federal function. Some DOE and

Los Alamos officials believe that federalizing the Los Alamos security

force would traumatize employees and cause political turmoil in the

local community but cited no specific examples to support their opinions.

In addition, DOE noted that its operations offices do not have personnel

slots to manage federal security forces, and OMB would require

DOE to staff such positions from existing slots. OMB officials told us that

they would work with DOE through the budget review process to help get

the necessary positions if DOE can demonstrate that conversion is cost effective.

However, DOE officials said that the budget process is very

time consuming and cumbersome and has not been responsive to their

prior requests for additional federal personnel positions.

Conclusions

Contractors provide security services for all but one DOE facility. At nine

facilities, we found that selected federal force costs may be at least $15

million less costly than contract forces. Believing that little or no cost

difference exists, DOE has not reexamined analyses conducted in the

early 1980s nor conducted additional ones to determine whether using

contract forces today is cost-effective and, therefore, still warranted.

DOE says that both a federal and contract force are equally capable of

protecting its facilities. However, a federal force may offer financial and

other unquantifiable benefits that are not, nor are they required to be,

considered in weighing the costs and benefits of obtaining security services.

For example, a federal force cannot legally strike-the Los

Alamos strike cost about $1.6 million over and above the almost $17

million contract costs. Also, a federal force may be more stable. High

turnover rates, such as the 11 to 15 percent that occurred at Los Alamos

before the strike, increase hiring, training, and security clearance costs.

DOE would have to weigh these benefits against other factors, such as its

ability to terminate employees. Nevertheless, taken together, a federal

force may be able to offer numerous advantages across the board for

DOE.

Recommendation to

the Secretary of

Energy

In this era of scarce budget resources, DOE needs to obtain protective

services in the most cost-effective manner. Because significant savings

may be realized by having federal rather than contract employees provide

security services, we recommend that the Secretary of Energy conduct

an in-depth analysis of the relative costs of federal and contract

security services across the nuclear weapons complex and convert to

federal forces at locations where it is cost-effective to do so.

Appendix I

Views of Los Alamos Security Force Members

The following summarizes the views expressed by 14 Los Alamos

security force members concerning the (1) causes of the strike; (2) adequacy

of security before, during and after the strike; (3) actions that

could improve security; (4) quality and quantity of training; (5) usefulness

of DOE'S inspections, and (6) advantages and disadvantages of federalizing

the security force. The views of these individuals cannot be

projected to all the Los Alamos security force.

Strike Causes

All said the strike was caused by a combination of excessive, mandatory

overtime, a restrictive sick leave policy, and a punitive disciplinary

policy that increased stress because demerits could result in job loss. A

common complaint was that the company gave demerits for minor reasons

(eating at a desk, picking up a newspaper, and minor vehicular

damage). Some also said that sometimes the company required

employees to take annual leave in lieu of sick leave, even if they had a

doctor’s statement and sick leave time was available. Some acknowledged

that Mason and Hanger instituted such a policy because of

reported past sick leave abuses by the Los Alamos force. All but one

said that the strike could have been prevented if Mason and Hanger had

been more flexible and understanding with its employees and exhibited

better human relations skills.

Adequacy of Security Before, During, and After the Strike

l About half said that security was poor before the strike because of

fatigue and low morale caused by excessive overtime; the other half said

security was good to adequate.

Twelve said that security was poor or marginal during the strike

because the replacements did not meet the physical fitness and medical

requirements and were not trained in using the same weapons as the

regular Los Alamos force. Also, several said that the replacements were

unfamiliar with the site, and some posts were not staffed because not

enough replacements were available.

. Thirteen said that security has improved since the strike. Some security

inspectors also noted that an “open campus” atmosphere prevents them

from fully implementing DOE'S requirements because laboratory personnel

complain when security inspectors do their jobs correctly. Also,

some claimed that Los Alamos allows its personnel to circumvent the

policies established.

Actions That Could

Improve Security

The security force members offered various suggestions to improve

security. Some examples follow.

Increase training, especially to qualify with new weapons, increase

firing range time, and provide more communications training.

Obtain better equipment, such as 4-wheel drive vehicles, clothing, and

holsters.

Improve safety and health conditions that will decrease sick leave and

injuries. For example, when a security inspector fell from a guard tower,

Los Alamos made changes to make it safer but did not take the same

actions to improve other towers. Guard posts should be improved: concrete

floors are cold and some traffic islands are “falling apart,” which

allows exhaust fumes to enter the buildings and affect their health.

Allow Mason and Hanger to manage security instead of Los Alamos and

encourage labor relations training for Los Alamos, Mason and Hanger,

and security force members.

Quality/Quantity of

Training

Some security force members said that they were adequately trained to

perform their jobs. One said that some members need more training on

using weapons, but the firing range is seldom open. About 50 percent

said that more training is needed on using the baton and arrest procedures.

Some suggested that training classes should be smaller so the

security force could receive more individual attention, and Mason and

Hanger should use trained instructors rather than field supervisors.

They also expressed concerns that training is often canceled, usually

because of lack of funds.

Usefulness of DOE’s

Inspections

Eight said that DOE'S inspections are useful because they help identify

problems. On the other hand, some noted that the limited-scope performance

tests do not represent real-life situations because DOE announces

when the exercises will be held, training increases right before the tests,

and some force members are preselected to participate. For these reasons,

some force members did not believe that the tests appropriately

assessed their competencies.

Federalizing

Almost all said that they would like to be in the federal system because

the benefits (sick and annual leave, health plan, and retirement) are

better, and military service counts toward retirement. They also said

that a federal force would take security more seriously, receive better

training, and employee problems would be resolved in a timely manner.

Many said that the major disadvantage would be a reduction in the

salary received. However, some noted that the benefits were more

important than salary.

Comparison of Six Facilities’ Compliance With

Albuquerque’s Draft Contingency

Plan Requirements

Requirements 1 2 3 4 5 6’

1. Identify number of posts needed

during emergencies. N Y Y Y Y Yb

2. Identify number of non-striking

protective force individuals that

would be available during strike on

all shifts. N Y Y Y Y Y

3. identify posts that could be shut

down, curtailed, and/or

consolidated. Y Y Y N Y N

4. Determine the priority of site

operations that can be shut down

or curtailed. N N N N Y N

5. Identify functions that could be

accomplished by other staff. N N Y Y Y N

6. Describe how auxiliary guard

uniforms will be identified. N Y N N N N

7. Describe how non-striking security

inspectors would be identified. N N N Y N N

8. Identify inventories and

inspections to be made upon

departure of striking employees

a. weapons N N N N N Y

b. keys N N N N N N

c. emergency response vehicles/

equipment N N N N N Y

d. perimeter posts, radio, telephone,

and alarms. N N N N N Y

9. identify a plan to notify

management and auxiliary guards

in the event of a strike. N Y Y Y N N

10. Estimated amount of time

required to implement the site

contingency plan. Y N N N N N

11, Identify how, and for how long,

local law enforcement agencies

will provide assistance during a

strike. N N N N Y N

12. Identify training to be held upon

the arrival of augmentees. N N N N N N

Page 41 GAO/RCED-@l-12 Security at Loa Alamos

Appendix II

Comparison of Six Facilities’ Compliance

With Albuquerque’s Draft Contingency

Plan Requirements

Requirement8 1 2 3 4 5 6’

13. Identify training for auxiliary guards to be he d prior to duty

assignment.  Y N Y N Y

14. Identify a point of contact to

coordinate logistics, and explain

how you will meet logistical needs. N Y N N N Y

i5. Identify and discuss transition

plans when the strike is

terminated. N Y N N N N

a1 -Pantex, TX.

LOS Alamos, NM.

S=Pinellas, FL.

4=Kansas City, MO

5=Mound, OH.

G-Rocky Flats, CO.

Y=yes; N-no.

Counting each of these items as separate topical areas results in a total of 16 requirements.

Major Contributors to This Report

Resources,

Community, and

Economic

Development Division,

Washington, D.C.

Denver Regional

Office

Judy England-Joseph, Associate Director, Energy Issues

Mary Ann Kruslicky, Assistant Director

Philip A. Olson, Assignment Manager

John D. Gentry, Regional Manager’s Representative

Craig D. Richards, Evaluator-in-Charge

Lois J. Curtis, Site Senior

Sarah A, Narvaez, Evaluator

Pamela K. Tumler, Report Analyst