SYSTEMATIC REVIEW FOR DECLASSIFICATION OF NATIONAL SECURITY INFORMATION - DO BENEFITS EXCEED COSTS?

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP93B01194R001100010001-8
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 62
• Document Creation Date: December 12, 2016
• Document Release Date: February 1, 2002
• Sequence Number: 1
• Publication Date: October 15, 1980
• Content Type: REPORT

Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
BY THE COMPTROLLER GENERAL
Report To The Congress
OF THE UNITED STATES
Systematic Review For Declassification
Of National Security Information--
Do Benefits Exceed Costs?
Executive Order 12065 specifies that Federal
agencies review all classified records consid-
ered permanently valuable as they become 20
years old to determine if they can be declas-
sified. However, most of the declassified rec-
ords probably will not be requested by the
public, and it is unlikely that the December
1988 deadline for completing the review will
be met.
This report recommends that the order be
modified to require that only those records be
reviewed which are specifically requested by
the public and those likely to be requested.
The change would improve responsiveness and
save money without changing the policy of
openness in Government.
'O
C1 LCD-81-3
ACCOUXpproved For Release 2002/04/01 : CIA-RDP93B01194R00110 1 1-815, 1980
 Approves) For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
Request for copies of GAO reports should be
sent to:
U.S. General Accounting Office
Document Handling and Information
Services Facility
P.O. Box 6015
Gaithersburg, Md. 20760
Telephone [phone]
The first five copies of individual reports are
free of charge. Additional copies of bound
audit reports are $3.25 each. Additional
copies of unbound report (i.e., letter reports)
and most other publications are $1.00 each.
There will be a 25% discount on all orders for
100 or more copies mailed to a single address.
Sales orders must be prepaid on a cash, check,
or money order basis. Check should be made
out to the "Superintendent of Documents".
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
To the President of the Senate and the
Speaker of the House of Representatives
This report describes the results of our review of th
systematic declassification review program provided for in
Executive Order 12065. The review was performed in the
Washington, D.C., area, and is the third in a series of
reports on the classification of national security infor-
mation. Our first report, "Improved Executive Branch
Oversight Needed for the Government's National Security
Information Classification Program," was issued March 9,
1979. Our second report, "Continuing Problems in DOD's
Classification of National Security Information," was
issued October 26, 1979.
We are sending copies of this report to the Director,
Office of Management and Budget; the Administrator of
General Services; the Assistant to the President for
National Security Affairs; the Attorney General; the
Director, Central Intelligence Agency; and the Secretaries
of Defense, Energy, and State.
Comptroller General
of the United States
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
COMPTROLLER GENERAL'S SYSTEMATIC REVIEW FOR
REPORT TO THE CONGRESS DECLASSIFICATION OF
NATIONAL SECURITY
INFORMATION--DO BENEFITS
EXCEED COSTS?
D I G E S T
To provide greater openness in Government,
Executive Order 12065 requires Federal
agencies, primarily the National Archives
and Records Service (NARS), to review all
classified records, considered to be of
permanent value, as they become 20 years
old to determine if they can be declassi-
fied. One exception is foreign government
information, which must remain classified
until it becomes 30 years old.
The National Security Council is responsible
for providing overall policy direction for
agencies to use in implementing the order.
The order makes the Administrator of General
Services responsible for implementing and
monitoring the classification program and
directs him to delegate that responsibility
to the Information Security Oversight Office
within the General Services Administration.
NARS is also within the General Services
Administration.
GAO evaluated the review program at NARS,
the Central Intelligence Agency, and the
Departments of Defense, Energy, Justice,
and State and found that:
--There is little chance that all 20-year-
old classified records will be reviewed
by the required December 1988 date.
--NARS inaccurately reported the results of
its declassification reviews for several
years.
--A review of only those records requested
by the public and those expected to be
requested would improve responsiveness
to public requests and would reduce costs.
Tear Sheet. t *prawa4,Ftm Ise 2002/04/01 : CIA-RDP93B0l194R001100010001-8
cover date should be noted hereon. 1 LCD-81-3
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
1988 REVIEW DEADLINE
UNLIKELY TO BE MET
NARS has estimated that it will not be
able to review all 20-year-old classified
records in its custody until after the
year 2000. NARS said that agencies have
not scheduled all their records nor sub-
mitted their schedules for approval so
that it can determine which records will
require a declassification review.
Most of the records will require a costly,
time-consuming, page-by-page review
because (1) foreign government information,
which cannot be declassified for 30 years,
is intermingled with domestic information
that can be declassified after 20 years
and (2) information generated since
World War II, containing intelligence
sources and methods, will require a more
thorough review and coordination with the
originating agencies.
NARS has never been able to fill all
the authorized posit: ions in its Records
Declassification Division. During
fiscal years 1977 through 1979, only
about 70 percent of the positions were
filled. 'S'taff turnovers, averaging
41 percent from 1973 through 1979, have
prevented the staff. from developing the
subject matter knowledge needed for
efficient reviews. (See p. 5.)
NARS INACCURATELY REPORTED
DECLASSIFICATION PROGRESS
NARS substantially overstated its declas-
sification activities for fiscal years
1973 to 1977. Instructions on the form
7096, used to accumulate declassification
statistics, allowed unclassified pages to
be included. Records Declassification
Division officials increased surveyors'
estimates of the number of pages declas-
sified and further increased the total
Approved For Release 2002/04/01 CIA-RDP93BOl 194RO01 100010001 -8
ii
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
number of pages declassified in reports
to the Interagency Classification Review
Committee.
Internal reports for fiscal years 1973
through 1977 showed that NARS declassified
161 million pages. However, MARS reported
215 million declassified pages to the
Review Committee, an overstatement of
54 million pages. That amount was included
in the Review Committee's annual report for
1977. (See p. 16.)
The General Services Administration's
Inspector General has never reviewed NARS'
declassification program. Problems dis-
closed in GAO's review might have been
detected sooner if audits had been
performed. (See p. 18.)
REVIEWS ON REQUEST AND ANTICIPATED
DEMAND WOULD BE ADEQUATE
Although GAO fully endorses openness in
Government, it believes that the Executive
order should be modified because most
information being systematically reviewed
and declassified will probably not be
requested by the public. In addition,
responsiveness to the public could be
improved by reviewing only those documents
requested and those expected to be re-
quested. The existing method is ineffi-
cient and costly.
NARS has estimated that over 90 percent
of the 244 million pages reviewed for
declassification during fiscal years
1980 through 1988 will not be requested
by the public.
Since
systematic reviews became a require-
ment
in 1972, the same classified documents
often
have
been reviewed many times before
being
made
available to the public. Agen-
cies
review
the material before sending it
to NARS, the Records Declassification
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
Division makes systematic reviews, and the
MARS custodial divisions review for other
restrictions before releasing material to
the public.
NARS, the Central Intelligence Agency, and
the Departments of Defense, 'Energy, and
State anticipate spending $88 million in
salaries and benefits during fiscal years
1981:through 1988 to systematically review
in.foTmationxfor declassification.
Prior to the establishment of the Records
Declassification Division in 1972, the
custodial divisions were responsible for
all declassification reviews. (See p. 21.)
RECOMMENDATIONS
To ensure that the President, the Congress,
and the public are better informed about
the results of the declassification program,
the Administrator of General Services should
revise the form 7096 to show the number of
(1) pages reviewed, (2) classified pages
reviewed, (3) classified pages declassified,
and (4) classified pages exempt from declas-
sification. The Administrator should also
direct the Inspector General to periodically
conduct comprehensive reviews of the MARS
declassification program.
The Chairman of the National Security
Council should draft, and submit to the
President for approval, a revision to
Executive Order 12065 that would modify
the requirement for the systematic review
of 'information for declassification.
iatic revietas for
.Spec ifically,-`sys kerr-r
declassification should be made of records
'requested by the public and those which
the Archivist of the United States antici-
pates will be requested.
After modification of the systematic review
requirement, the Administrator of General
Services should direct the Archivist of the
Approved For Release 2002/04/0i1vv. CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
United States to transfer responsibility
for all declassification reviews to the
custodial divisions. To ensure uniform
application of declassification policies
and procedures, prescribed by approved
declassification guides, the management
personnel in the Records Declassification
Division should be used in an advisory
and coordinating capacity with the cus-
todial divisions, agencies, and the
Information Security Oversight Office.
AGENCY COMMENTS
NARS agreed with GAO that there was little
chance that the systematic review for de-
classification of all 20-year-old classi-
fied records would be completed by December
1988. NARS said that it was taking steps
to reduce staff turnover and to improve
its reporting of declassification activity.
The Information Security Oversight Office,
speaking for the Administration, strongly
opposed GAO's suggestion that all 20-year-
old classified information not be reviewed.
Among other things, it told GAO that the
report failed to address or evaluate open-
ness in Government and that the recom-
mendation was based primarily on cost
considerations. NARS and the Departments
of Defense and Energy also opposed this
suggestion. As a result of their comments,
GAO revised the report to more fully
address openness in Government, clarified
its conclusions, and revised its sug-
gestion concerning systematic review. GAO
supports openness, but it believes that
selective declassification reviews could
provide information of most interest to
the public, while promoting efficiency
and economy in Government. (See pp. 14,
19, 29, and 31.)
rear hApproved For Release 2002/04/01,: CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
C o n t e n t s
P arc e
DIGEST i
CHAPTER
INTRODUCTION 1
Objectives, scope, and
methodology 3
2 LITTLE CHANCE OF AGENCIES REVIEWING
ALL 20-YEAR OLD CLASSIFIED RECORDS
BY 1988 5
Records scheduling problems delay
reviews for declassification 5
Most 20-year old information will
require page-by-page review 9
NARS declassification staff has
been below authorized levels 11
Conclusions 13
NARS comments and our evaluation 14
3 NARS INACCURATELY REPORTED ITS
PROGRAM RESULTS 16
Increased estimates of pages
declassified 16
Discrepancies in statistics
compiled by NARS and reported
to the oversight office 17
Internal reviews of NARS declas-
sification operations have not
been made 18
Conclusions 18
Recommendations 19
NARS comments and our evaluation 19
4 REVIEWS ON REQUEST AND ANTICIPATED
DEMAND COULD ADEQUATELY PROVIDE
INFORMATION TO THE PUBLIC 21
Most information systematically
reviewed and declassified will
not be requested by the public 22
Responsiveness to the public
could be improved 23
The existing method of systematic
review for declassification
is inefficient and costly 25
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
Page
Custodial divisions should be
responsible for systematic
declassification reviews 26
Conclusions 28
Recommendations 29
ISOO comments and our evaluation 29
NARS, DOD,, CIA, and Department
of Energy comments and our
evaluation 31
I Letter dated July 3, 1980, from the
Acting Archivist of the United States 33
II Letter dated June 26, 1980, from the
Director, Information Security
Oversight Office
ABBREVIATIONS
CIA Central Intelligence Agency
FBI Federal Bureau of Investigation
FOIA Freedom of Information Act
GAO General Accounting Office
GSA General Services Administration
ICRC Interagency Classification Review Committee
ISOO Information Security Oversight Office
NARS National Archives and Records Service
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
INTRODUCTION
Implementation of the program for the classification and
declassification of national, security information is governed
by Executive Order 12065, which took effect December 1, 1978.
It superseded Executive Order 11652, which was in effect from
June 1972 through November 1978. Both orders provide for a
periodic, systematic review for declassification of records.
The new order requires a review for only those records deemed
permanently valuable.
In both orders the President has designated the heads
of certain agencies and officials of those agencies to be
authorized classifiers and declassifiers of information.
The National Security Council is responsible for overall
policy guidance. Permanently valuable classified infor-
mation transferred to the National Archives and Records
Service (MARS), within the General Services Administration
(GSA), is to be declassified by the Archivist of the United
States in accordance with agency declassification guidelines,
the provisions of the Executive order, and directives of the
Information Security Oversight Office (ISOO). ISOO replaced
the Interagency Classification Review Committee (ICRC), which
had oversight responsibility under the previous Executive
order.
The previous order required the use of a general
declassification schedule of 6 to 10 years for the automatic
downgrading and eventual declassification of information,
depending on the level of classification. Information
requiring protection for a longer period could be exempted
from a declassification review for 30 years, and classifi-
cation beyond 30 years could be extended indefinitely at
the discretion of the head of the agency originating the
document.
The new order abolishes the general declassification
schedule and limits the classification of most information
to 6 years. It further provides that information requiring
protection for a longer period can be classified for up to
20 years. Section 3-4 of Executive Order 12065 specifies
that information constituting permanently valuable records
of the Government must be reviewed for declassification at
the end of 20 years, but classification can be extended for
10-year periods, provided the information is reviewed at the
end of each period. Foreign government information may be
Approved For Release 2002/04/011: CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
classified for up to 30 years. The mandatory review provision
of the order also provides that, upon request to either NARS
or the originating agency, any document or section thereof
that no longer requires protection shall be declassified.
The exception would be information less than 10 years old
originated by the President or his representatives.
NARS has until December 1, 1988, to review all classified
information, other than foreign government information, that
is more than 20 years old. In order to meet this deadline,
the order's implementing directive requires heads of agencies
and designated officials, within 60 days of the effective date
of the order, to survey all classified records 20 years old or
older held in agency storage areas and Federal records centers
and to identify those that require scheduling for future dis-
position. Such scheduling is required to be completed by
December 1, 1980. The directive further provides that clas-
sified nonpermanent records that are scheduled to be retained
for more than 20 years need not be systematically reviewed,
but shall be reviewed for declassification upon request.
The agencies given declassification authority under both
orders are to formulate, issue, and maintain systematic review
guidelines. These guidelines shall state specific, limited
categories of information which, because of their national
security sensitivity, should not be declassified automatically.
Such information is to be reviewed item by item to determine
whether continued protection beyond 20 years is needed. Agen-
cies are to cooperate with NARS in drafting guidelines which
will allow NARS to review and declassify most documents on
its own, without having to forward them to the agency for
review.
Declassification responsibilities within NARS are divided
among its divisions. The Records Disposition Division deter-
mines, through agency records schedules, which records are per-
manently valuable and should be accessioned into the Archives
and which are temporary and should be disposed of. The divi-
sion also helps agencies develop schedules and furnishes guid-
ance on the transfer of material to Federal records centers.
When permanently valuable records are actually shipped to
the Archives, it is the responsibility of the custodial divi-
sions to accept these records. Information can be accepted by
any one of seven custodial divisions, depending on the subject
matter of the material. These divisions then inspect the con-
dition of the records, arrange them, and prepare descriptive
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
2
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
guides and other aids that the public can use to request the
information. The custodial divisions are responsible for
reviewing this information for declassification when the
public requests it informally or under the provisions of the
Freedom of Information Act (FOIA), or the mandatory review
provision of Executive Order 12065. The Records Declassifi-
cation Division is responsible for doing systematic declas-
sification reviews of the Government's 20-year-old classified
records that are in the custody of the-Archives.
Implementation of declassification responsibilities
varied among the agencies we visited. Navy and Air Force
systematic declassification groups are located within the
historical offices of each department. FOIA, the Privacy Act:,
and mandatory declassification reviews are organized under
the records management offices. In both the State Department
and the Federal Bureau of Investigation (FBI) declassification
functions are under one office and separate from the records
management offices. Both the Army and the Central Intelligence
Agency (CIA) have all their declassification and records man-,
agement functions within the same office.
OBJECTIVES, SCOPE, AND METHODOLOGY
Because of the size of the Government's national security
information classification program, our review is being under-
taken in phases. This work was initiated at the request of
the Chairmen, Subcommittee on Priorities and Economy in Gov-
ernment, Joint Economic Committee, and the Subcommittee on
Government Information and Individual Rights, House Committee
on Government Operations. It is being issued to the Congress
because of widespread interest in the subject. Our first
report (LCD-78-125, Mar. 9, 1979) discusses the need for im-
proved executive branch oversight of the program. Our second
report (LCD-80-16, Oct. 26, 1979) discusses the continuing
problems in the Department of Defense's (DOD's) classifica-
tion of national security information. This report evaluates
the declassification activities of NARS and several Govern-
ment agencies that handle large quantities of classified
information.
We visited the offices having responsibility for the
declassification program within the following organizations
in the Washington, D.C., area:
--National Archives and Records Service
--Department of the Army
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
--Department of the Navy
--Department of the Air Force
--Department of Energy
--Department of Justice
--Department of State
--Central Intelligence Agency
In addition, DOD's Office of Information Security pro-
vided us with information on the status of declassification
programs of the other major DOD components.
We-reviewed Executive Orders 11652 and 12065, their
implementing directives and instructions, as well as agencies'
declassification guidelines. We discussed implementation of
the program with agency officials. We obtained information
on the cost of the program, results to date, and planned
future declassification efforts of NARS and the agencies.
We started this review with the objective of evaluating
compliance with the systematic review requirements of Execu-
tive Order 12065. Our findings and recommendations on this
evaluation are included in chapters 2 and 3. On the basis of
our observations and discussions with officials from NARS and
various other Government agencies during our review, it became
apparent that consideration should be given to revising the
systematic review requirements. Our findings and recommenda-
tions on this matter are included in chapter 4.
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
CHAPTER 2
LITTLE CHANCE OF AGENCIES REVIEWING
ALL 20-YEAR-OLD CLASSIFIED RECORDS BY 1988
Executive Order 12065, effective December 1, 1978, and
its implementing directive require that (1) all 20-year-old
classified material be on approved schedules by December 1,
1980, (2) agencies issue guidelines showing specific, limited
categories of information which require item-by-item review
to determine if continued protection is needed beyond 20
years, and (3) all permanently valuable 20-year-old classi-
fied material be systematically reviewed for declassification
by December 1, 1988.
Some agencies, which have large amounts of classified
information, will not have all their records on approved
schedules by December 1, 1980.
All agencies, except the National Security Council and
the Office of Science and Technology Policy, have submitted
their systematic review guidelines to NARS.
NARS has estimated that all permanently valuable 20-year--
old classified records in its custody will not be systemati-
cally reviewed for declassification until after the year 2000.,
NARS will not be able to meet the December 1, 1988, deadline
for systematic review because
--agencies have not scheduled all their records and
submitted the schedules to NARS for final approval
in a timely manner,
--an increasing amount of material will require a
page-by-page review,
--the NARS declassification staff has been below
authorized levels, and
--there has been frequent staff turnover in the Records
Declassification Division.
RECORDS SCHEDULING PROBLEMS DELAY REVIEWS
FOR DECLASSIFICATION
All 20-year-old classified material will not be on
approved schedules by December 1, 1980, because some
5
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
agencies have been slow in submitting their schedules to NARS
for approval. NARS will not know the amount of material re-
quiring a declassification review by December 1988 until all
records are on approved schedules.
Since March 1973 NARS has periodically informed the
heads of Federal agencies of their responsibility to schedule
all records to identify which were temporary or nonpermanento
and which were permanent and had archival value.
advised agencies that Federal records centers would no longer
routinely accept records lacking definite retention periods.
In August 1974 NARS set a December 31, 1976, time limit for
development and submission of updated and comprehensive
schedules. A comprehensive records schedule lists the types
of agency records---permanent, or nonpermanent--and contains
precise instructions for their future disposition. Instruc-
tions for nonpermanent records must include minimum retention
periods, while those for permanent records must provide for
future transfer to the Archives of the United States. The
records schedule is also supposed to contain an inventory of
the records on hand, identifying the quantities and types of
permanent and nonpermanent records.
Despite NARS efforts to encourage agencies to schedule
their records for final disposition, about 3.8 million cubic
feet of unscheduled records were stored in Federal records
centers in December 1979. We found that about 247,000 cubic
feet of these records contained some classified material.
Based on a conversion factor of about 2,500 pages to 1 cubic
foot, there were about 617,500,000 unscheduled pages within
that 247,000 cubic feet, an unknown number of which were
classified.
Because most agencies do not separate classified and
unclassified material, few can submit schedules for only
their classified records. Thus, the actual volume of
records needing to be scheduled is considerably more than
617,500,000 pages. For example, the Navy had about 644,000
cubic feet of records, or 1.6 billion pages, at Federal
records centers. The Navy knew that about 102,000 cubic
feet of these records contained classified information
because the boxes were so marked. The Navy, however, must
schedule both its classified and unclassified records because
they are intermingled. The following table shows those
agencies with the largest volume of unscheduled classified
records stored in Federal records centers in December 1979.
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
Cubic feet of
unscheduled
Unscheduled records
containing classified
material
Agency
records
cubic feet
a
es`
p
g
Department of the Navy
644,000
102,000
255,000,000
Department of the Army
591,000
5,000
12,500,000
Department of the
Air Force
204,000
14,000
35,000,000
A- other DOD agencies
117,000
65,000
162,500,000
Agency for International
Development
40,000
23,000
57,500,000
Department of State
22,000
9,000
22,500,000
Department of Justice
128,000
9,000
22,500,000
U.S. Information Agency
7,000
5,000
12,500,000
Department of Commerce
48,000
5,000
12,500,000
General Services Admin-
istration
13,000
2,000
5,000,000
Department of the
Treasury
647,000
2,000
5,000,000
U.S. Arms Control and
Disarmament Agency
1,000
1,000
2,500,000
155 other agencies
1,338,000
a/5,000
12,500,000
Total
3,800,000
247,000
b/617,500,000
a/Only 32 of the 155 agencies have unscheduled records con-
taining classified material.
b/Contains an unknown amount of classified pages.
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
In January 1980 NARS again took steps to get all agencies
to schedule the nearly 3.8 million cubic feet of unscheduled
records in the records centers by requiring agencies to submit
plans for revising inadequate disposition instructions for
their records by February 15, 1980. NARS officials informed
us in March 1980 that agencies had complied with this require-
ment. However, NARS will not be able to review and approve
by December 1, 1980, all agency schedules already submitted
because (1) agencies were slow in submitting their schedules
and (2) it takes time to review and approve them. NARS esti-
mates that it could be early 1982 before all Navy and Army
records schedules are approved.
The FBI's central records system, which includes both
administrative and investigative files, contains most of its
classified material. These and other FBI records are stored
in 121,000 square feet of space in its headquarters building.
The only FBI records contained in the Archives were micro-
filmed investigative files from World War I to 1923.
In May 1977 the FBI submitted a draft schedule for its
central records system to NARS. Before approving the schedule,
which indicated which records would be retained and which
would be destroyed, the Archivist, aware of congressional
interest in this information, referred the schedule to the
two interested committees for review and comment. Before
a decision was made, a class action suit was brought against
NARS and the FBI to stop the destruction of all FBI records.
Historians and others allege that NARS had in the past approved
record schedules allowing the FBI to destroy billions of pages
of information not deemed permanently valuable.
In January 1980 a Federal judge directed the FBI and
MARS, in consultation with historians and other interested
parties, to devise a plan, to be approved by the court, that
specifies instructions for the retention or disposition of
FBI records. Until the court decides which FBI records are
permanently valuable, the FBI will not be able to systemati-
cally review its permanently valuable classified information
as required by Executive Order 12065.
In addition to the 3.8 million cubic feet of unscheduled
records in the records centers, an unknown quantity of 20-year-
old classified material was still being retained by the agen-
cies. The amount of these records needing a declassification
review will not be known until they are either transferred
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
to the Federal records centers or the Archives. Agencies are
permitted to keep older records if they have a continuing
need for them.
MOST 20-YEAR-OLD INFORMATION WILL
REQUIRE PAGE-BY-PAGE REVIEW
Most NARS reviews under Executive Order 12065 will have
to be done by a costly and time consuming page-by-page review
rather than by bulk declassification, because (1) foreign
government information, which does not have to be reviewed
for declassification until it is 30 years old, is inter-
mingled with domestic information that can be declassified
after 20 years, and (2) increased amounts of information
containing intelligence sources and methods will require a
more thorough review and coordination with the agencies that
originated the information.
Under Executive Order 11652, all 30-year-old information,
whether it was foreign government or domestic information,
had to be reviewed by NARS for declassification. Under Exec-
utive Order 12065, NARS must review 20-year-old Government
information for declassification, while foreign government
data still remains classified for 30 years. According to NARS
officials, this means that they must review the same type
of information page by page that was once bulk declassified,
in order to identify and segregate foreign government infor-
mation not yet 30 years old.
Bulk declassification consists of NARS using agency
declassification guidance to survey a sample of records. If
no records are identified as exempt from declassification
based on the guidance, the entire group of records is immedi-
ately declassified. However, if MARS finds even one document
that could be exempt, based on the declassification guidance,
the entire group of records is reviewed page by page. Obvi-
ously, bulk review is the most efficient way to declassify
information. From fiscal years 1973 through 1979, NARS used
21 staff-years to bulk review 247,000,000 pages, while for
the same period 256 staff years were used to review 137,000,000
pages for declassification by the page-by-page method.
NARS officials explained that since World War II, the
volume of intelligence information and the amount of such
information exchanged among agencies has increased substan-
tially. Executive Order 12065 also requires a declassifica-
tion review of 20-year-old intelligence information, much of
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
which contains sources of information and methods of operation
that the originating agencies believe require extended pro-
tection. Because of this, the amount of information that is
reviewed page by page will increase, while the amount that is
bulk declassified will decrease.
NARS officials have estimated that under the previous
Executive order an individual could cover approximately 300
pages an hour in a page-by-page review of 30-year-old mate-
rial. Under the current order an individual will only be able
to review between 100 and 125 pages an hour because 20-year-
old material is being reviewed. Much of this information may
reveal sources and m(,thods still in use. As a result, this
information is reviewed very carefully and at a slower rate.
For example, a recent 80,000 page project was reviewed under
the new Executive order at a rate of about 108 pages an hour.
Reviewing 20-year-old records for declassification that
contain information about intelligence sources and methods
will also require more frequent coordination with the agencies
that originated the information. In June 1979 the CIA issued
systematic review guidelines which allow NARS to automatically
declassify all 20-year-old information, unless it falls into
any of 29 specified categories. The guidelines instruct NARS
not to declassify any information falling into these catego-
ries. Such information is to be referred to the CIA for re-
view. These categories refer to intelligence sources and
methods, yet are all inclusive and do not specifically define
what information can and cannot be declassified within the
categories. As a result, NARS reviews and withholds all in-
formation falling into these categories for the CIA's review.
A similar situation occurs when NARS reviews an agency's
records that contain intelligence information originated by
other agencies. We reviewed three projects completed by NARS
in January 1980, which required extensive coordination with the
National Security Agency, the State Department, CIA, Army,
Navy, Air Force, and the FBI. Of the 31,800 pages withheld
for coordination, only 2,800 were subsequently released.
State Department guidelines specifically require NARS to
refer information falling under intelligence sources and
methods to State for further review and possible referral
to other agencies.
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
NARS DECLASSIFICATION STAFF HAS
BEEN BELOW AUTHORIZED LEVELS
NARS officials told us that a primary reason that the
systematic review requirement of the order will not be met
by December 1988 is the lack of an adequate number of people
in the Records Declassification Division to perform the re-
views. NARS officials have estimated that from 1980 to 2000,
they will review about 466 million pages of information for
declassification and that the systematic review of all 20-year-
old classified material will still not be accomplished. A
backlog of 24 million pages will still exist.
From 1973 through 1979 the staff was substantially below
authorized levels. For example, during fiscal years 1977
through 1979 only about 70 percent of the authorized positions
were filled. The primary reasons were the constant and high
turnover and the time required to obtain security clearances
for incoming personnel.
Frequent staff turnover in the
Records Declassification Division
A Records Declassification Division was formed within
NARS to perform systematic reviews. This requirement was
established by Executive Order 11652 in June 1972. Since
the division was created, personnel turnover has been high.
From 1973 through 1979 the staff turnover rate averaged 41
percent, ranging from a low of 19 percent in 1978 to a high
of 71 percent in 1973. Efficient review of information for
declassification requires subject matter expertise, which is
acquired with experience. However, if the high rate of turn-
over experienced by the declassification division from 1973
through 1979 continues, the staff will not develop the subject
matter knowledge to efficiently review all 20-year-old perma-
nently valuable classified information. The following tabu-
lation shows the authorized staff level, average staff level,
and staff turnover from 1973 through 1979.
11
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
Fiscal
Authorized
l
Average
staff
el
l
Percent of
authorized
staff
Staff
turnover
Percent of
yearly
turnover
year
staff leve
ev
1973
109
65
59
46
71
1974
102
66
65
24
36
1975
120
82
68
31
38
1976
105
85
81
46
54
1977
105
71
68
22
31
1978
103
72
70
14
19
1979
100
68
68
28
41
NARS officials agree that high staff turnover has been a
problem since the division was ortadvancement opportunity.of
the staff leave for better pay
NARS officials told us that it takes about 2 years to
become proficient in doing declassification reviews and that
most of the turnover occurred within that 2-year period.
Thus, a large part of the staff had to be constantly trained
and its work closely reviewed to ensure accuracy. As a
result, the staff was not as proficient as it should have
been to perform declassification reviews, which in turn,
meant that less information was reviewed for declassification
than would otherwise have been the case with a more proficient
staff. We were told, for example, that because inexperienced
staff had to be closely supervised and their work constantly
reviewed, it took over 2 years to review a project with only
900 cubic feet (or about 2.25 million pages) of records.
Staff turnover, in addition to delaying the review
process, is costly. New staffinembers cannot review any
classified material until a top secret security clearance
is obtained, which takes approximately 4 months and now
costs about $950.
We identified 230 individuals granted top secret clear-
ances from 1973 through 1979, of which 211 either resigned
shortly after beginning work or transferred to another NARS
12
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
division. It appears that the remaining 19 individuals
declined the position offered to them after they had been
cleared. About a third of the 211 individuals transferred
to other divisions in NARS which may have required them to
have a security clearance. It cost almost $27,000 to clear
the individuals who left the declassification division in
1979. If the high turnover continues, future costs could
be even higher.
NARS is aware of the turnover problem and plans to in-
crease the career ladder for 19 of its 108 technical declassi-
fication positions by one grade. However, we do not believe
this action is enough to correct the problem because it will
only affect a small percentage of the staff. Moreover, it is
not known whether the increase of one grade will be enough
to retain qualified personnel.
CONCLUSIONS
There is little chance that NARS will be able to sys-
tematically review for declassification all 20-year-old
permanently valuable material by December 1, 1988. NARS has
estimated that such reviews will not be completed until after
the year 2000.
One reason NARS will not meet the December 1988 deadline
is because some agencies did not have all their 20-year-old
records on approved schedules that indicated which are perma-
nently valuable and require a declassification review. The
records of all agencies are required to be on approved sched-
ules by December 1, 1980. Some organizations, like the Army
and Navy, will not have their records on approved schedules
at least until 1982. NARS did not know the total volume of
unscheduled 20-year-old classified records of all Government
agencies. Consequently, NARS did not know how many permanently
valuable records will have to be reviewed for declassification.
However, as of December 1979 there were nearly 3.8 million
cubic feet of material in Federal records centers that needed
to be scheduled, and it is unlikely that the 20-year-old per-
manently valuable classified material can be scheduled and
reviewed for declassification by 1988.
Most information will also require a costly and time
consuming page-by-page review to identify and segregate
foreign government information that must remain classified
Approved For Release 2002/04/01 : CI-RDP93B01194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
for 30 years. Increased amounts of information containing
intelligence sources and methods will require a more thorough
review and coordination with the agencies that originated
the information.
Between 1973 and 1979 MARS was not able to fill all the
authorized positions in its Records Declassification Division.
Frequent turnover in the declassification division does not
permit the staff continuity needed to gain the subject matter
expertise to declassify information. Staff turnover averaged
41 percent from 1973 through 1.979. Although it takes about
2 years to become proficient in doing declassification reviews,
most of the staff that has left has done so within 2 years
after being hired.
In our opinion, if the recommendations made in chapter 4
are adopted, the problems discussed above should be resolved.
NARS COMMENTS AND OUR EVALUATION
On July 3, 1980, the Acting Archivist of the United
States, in commenting on our report (see app. I), agreed
with our finding that there was little chance that all 20-
year-old classified records would be reviewed by December
1988 with the present level of resources. NARS will, he
said, be able to review and declassify the greater part of
historically significant documents by then. NARS also agreed
with our identification of the causes for its inability to
meet the 1988 goal.
NARS was not sure whether the failure of agencies to
schedule their records was the fault of poor records manage-
ment or the lack of personnel, but it suggested that we should
have "forcefully recommended greater agency attention to
records scheduling."
Greater agency attention to records scheduling may be
needed. However, it was not within the scope of our work
to examine and identify the causes of the scheduling problems
at the nine agencies that each had 1,000 cubic feet or more
of unscheduled records containing classified material stored
in Federal records centers. Without the identification of
causes, we have no basis for such a recommendation.
There is agreement that an increasing amount of material
will require page-by-page review because foreign government
information in the files is not 30 years old, thus, not per-
mitting bulk declassification of files that are 20 years old.
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
14
 Approved For Release 2002/04/01: CIA-RDP93B01194RO01100010001-8
However, NARS said that "if records have not been requested
by a researcher, we can and do defer declassification action
until the passage of time eliminates this impediment to bulk
declassification."
With respect to the problem of staff turnover, NARS said
that it was reexamining its hiring policies and was consid-
ering the possibility of hiring, on a part-time.basis, retired
experts from those agencies that classify large quantities of
information.
Approved For Release 2002/04/011x CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
CHAPTER 3
NARS INACCURATELY REPORTED ITS PROGRAM RESULTS
Since 1973 both ICRC and ISOO have stressed the
importance of accurately reporting the results of classifi-
cation decisions. Inaccurate reporting of program results
hampers ISOO in effectively carrying out its oversight re-
sponsibility. It also misinforms the President, the Congress,
and the public, and results in a loss of credibility and
confidence in the entire classification program.
We found that declassification statistics compiled and
reported by NARS were inaccurate because instructions on the
form allowed unclassified material to be included and NARS
arbitrarily increased estimates of the number of pages it
declassified. We also found discrepancies in the statistics
compiled by NARS and reported to the oversight office.
INCREASED ESTIMATES OF
PAGES DECLASSIFIED
GSA form 7096 was used to report a summary of statistics
of various archives programs and activities. One section
was devoted to statistics about the records declassification
program and was used by NARS to compile the declassification
information reported to ISOO. The instructions for completing
this form provided for each page of a document to be counted
as a page declassified even when the document contained unclas-
sified pages. As a result of these instructions, the number
of pages reported as declassified was higher than what it
should have been.
When NARS reviews information for declassification, all
material in a record group is first surveyed to determine
whether it can be bulk declassified or will require a page-
by-page review. The surveyor samples a number of documents
in order to make this determination. The number of documents
examined depends on the surveyor's knowledge of the subject
:natter, his judgment, and the specificity of the guidance
used.
In addition to recommending either a bulk or a page-by-
page review, the surveyor also estimates the percentage of
classified material in the record group. This estimate is
used to compute the amount of material declassified when it
is compared to the total amount of information in the record
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
16
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
group. At the time of our review, there were no controls
to ensure that the surveyor's estimates were accurately
reflected in NARS internal reports. We found that declas-
sification division officials increased the amount of
declassified material that surveyors reported and used
inflated figures in their periodic activity reports.
We identified 20 projects surveyed between January 1978
and August 1979 in which surveyors' estimates of classified
material differed from the figures included in NARS' internal
reports. In each case the report showed a higher percentage
of classified material than the surveyor estimated. For
example, a project containing 156 cubic feet of material was
reviewed in August 1978. The surveyor estimated that 3 per-
cent, or 11,700 pages, were classified. The MARS report,
issued in January 1979, showed 30 percent, or 117,000 pages,
were classified. Another project, surveyed in August 1978,
contained 2,865 cubic feet of records. According to the
surveyor's estimate,'3 percent, or 214,875 pages were clas-
sified. The January 1979 report showed that 30 percent of
the project contained classified pages. In both cases, a
zero had been added to the surveyor's percentage estimates.
Our analysis of the 20 projects, including a review of
the surveyors' workpapers and draft reports, indicated that
the surveyors' estimates were increased in the internal re-
ports in order to show a higher number of pages declassified.
The surveyors' estimates of classified pages were increased
by almost 3 million pages.
DISCREPANCIES IN STATISTICS COMPILED BY NARS
AND REPORTED TO THE OVERSIGHT OFFICE
Although inflated statistics were used to compile the
number of pages declassified, MARS further increased the
statistics that it reported to ICRC.
We examined NARS quarterly reports and project logs for
fiscal years 1973 through 1977 and compared them to the fig-
ures published by ICRC. ICRC received higher declassification.
figures and published them in its annual report for 1977.
ICRC was the oversight body responsible for monitoring the
Government's classification program. NARS internal reports
showed that 161 million pages were declassified from 1973 to
1977. However, NARS reported to ICRC that 215 million pages
had been declassified, and ICRC included that amount, an over-
statement of 54 million pages, in its 1977 annual report.
Approved For Release 2002/04/d17: CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
We informed NARS officials of these matters in November
1979. In February 1980 we were told that GSA form 7096 was
being revised to show the number of pages reviewed and the
number of pages declassified. In February 1980 the Director
of the Records Declassification Division notified all branches
that any adjustment to the surveyor's initial estimate must be
documented and authorized in writing by the appropriate branch
chief, including an explanation of changes made.
INTERNAL REVIEWS OF NARS DECLASSIFICATION
OPERATIONS HAVE NOT BEEN MADE
Prior to October 1978, GSA's internal reviews were con-
ducted by the Office of Audits, which reported to the Adminis-
trator of GSA. These audits were financial reviews and
management studies of the efficiency and effectiveness of
various operations. The Inspector General's office was es-
tablished in October 1978 by Public Law 95-452 and the Office
of Audits was placed under it. The Inspector General is
appointed by the President and confirmed by the Senate and
reports to both the GSA Administrator and the Congress.
Since October 1978 the Inspector General has conducted
four audits of NARS, including reviews of various NARS opera-
tions and two Presidential libraries. The Inspector General
has never performed a review of the operations of the NARS
records declassification prociram; however, NARS conducted an
internal study in May 1977 at. the request of the Archivist.
We were told that the study addressed declassification pro-
cedures, staff morale, and turnover in the Records Declassi-
fication Division. The Archivist was informed that problems
existed in these areas, but a. report was never issued.
Certain problems disclosed in our review, like the high
personnel turnover discussed on page 11, were also identified
in that study. Our review also disclosed deficiencies in the
way MARS has reported its program results. We believe that
the inaccurate reporting of program results might have been
detected and corrected sooner if the Inspector General had
performed periodic audits of the NARS declassification program.
CONCLUSIONS
NARS records declassification program had never been the
subject of an audit by GSA's Inspector General. The past
overstatement of declassifica.tion'statistics compiled and
reported to ICRC might have been disclosed earlier if program
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
18
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
performance had been reviewed. We believe that the
overstatement of pages declassified from 1973 to 1977
misinformed the ICRC, the President, the Congress, and
the public.
RECOMMENDATIONS
We recommend that the Administrator of General Services
--Revise the section of GSA form 7096 that deals with
declassification statistics to show the number of
(1) pages reviewed, (2) classified pages reviewed,
(3) classified pages declassified, and (4) classified
pages exempt from declassification.
--Direct the Inspector General to periodically conduct
comprehensive reviews of the NARS declassification
program.
NARS COMMENTS AND OUR EVALUATION
NARS told us that GSA form 7096, which was revised in 1.972
to account for systematic review work performed was inadequate
as a means of reporting work completed at the end of a calendar
year because incomplete projects were not included. Conse-
quently, the Records Declassification Division Director
estimated the accomplishments of the entire systematic review
program in his division, the custodial divisions, and the
Presidential libraries, and reported those figures to ICRC.
A revised form 7096, along the lines we suggested, has been
approved for use in fiscal year 1981. Because of this revision
and other internal procedural decisions, NARS expects to be
able to report the total number of pages examined, the number
of classified pages examined, the number of pages declassified,
and the number of pages remaining classified.
Since we were able to reconstruct the number of pages
declassified during fiscal years 1973 through 1977 from
quarterly reports and project logs, including provision for
incomplete projects, we believe that NARS could have done
likewise.
NARS also told us that it was taking steps to ensure
that any changes in surveyors' original estimates were fully
documented and justified and cited the Records Declassifi-
cation Division Director's February 1980 memorandum.
19
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
NARS agreed that internal reviews had not been made and
said that the situation would be rectified and that problems
addressed by us would be addressed in a review of the entire
program.
20
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
CHAPTER 4
REVIEWS ON REQUEST AND
ANTICIPATED DEMAND COULD ADEQUATELY
PROVIDE INFORMATION TO THE PUBLIC
The primary objective of the Executive order's
declassification provision is to make information no longer
requiring protection available to the public, by requiring a
systematic declassification review of 20-year-old information.
According to the statement by the President when he signed the
order, limiting classification and accelerating declassifica-
tion would increase openness in Government.
We have always been a strong advocate of openness in
Government, and we believe that such openness would not be
diminished if the systematic review part of the program were
modified to be more responsive to the public. We believe
that systematic reviews of certain groups of records should
be made (1) when requested by the public and (2) in anticipa-
tion of requests.
A major but intangible benefit of systematic review of
all 20-year-old records is the public's perception that it
provides for increased openness in Government. Although we
recognize the importance of this perception to public con-
fidence in the classification program, we believe that its
benefits and costs should be compared to the benefits that
could be derived from certain modifications to the system-
atic review requirements of the program. We believe that
these modifications are needed because
--most information systematically reviewed and declas-
sified will probably not be requested by the public,
--responsiveness to the public could be improved, and
--the existing method of systematic review for declas-
sification is inefficient and costly.
The Records Declassification Division was established in
1972 to perform systematic reviews for declassification that
were previously done by the custodial divisions. We believe
that the custodial divisions should again assume the respon-
sibility for such reviews.
21
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
MOST INFORMATION SYSTEMATICALLY
REVIEWED AND DECLASSIFIED WILL
NOT.BE REQUESTED BY THE PUBLIC
Based on MARS' estimates,, over 90 percent of the pages
that will be reviewed for declassification during fiscal years
1980 through 1988 will not be requested by the public.
Estimates of the types of information that the public
wants are based on requests made to NARS by individuals,
indicating the type of information they will be coming into
review during the year. Future years' estimates are based
on what the Director of the Records Declassification Division
believes the public will request. Estimates of pages to be
systematically reviewed through 1988 are based on the number
of people reviewing information in the declassification divi-
sion and the rate at which they can review information.
The following tabulation is taken from NARS' estimates.
Total Pages Pages reviewed
Fiscal pages systematically by public request Percent
year reviewed reviewed (note a) of total-
------------------ (000 omitted)----------------
1980
33,100
31,175
1,925
6
1981
26,800
24,375
2,425
9
1982
26,500
24,225
2,275
9
1983
26,500
24,225
2,275
9
1984
26,500
24,225
2,275
9
1985
26,200
24,025
2,175
8
1986
26,000
24,000
2,000
8
1987
26,000
24,000
2,000
8
1988
26,000
24,000
2,000
8
Total
243,600
224,250
19,350
8
a/Includes informal requests and requests made under the
mandatory review provision of Executive Order 12065 and
the FOIA.
Approved For Release 2002/04/uu : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
The Director of the Records Declassification Division told
us that specific requests for information would increase by
about 5 million pages a year if there were no requirement for
systematically reviewing all 20-year-old material. Officials
of the custodial divisions thought that the estimate was too
high and that the increase in specific requests would be about
1 million pages or more a year.
Even if the number of specific requests were to increase
by as much as 5 million pages a year, there would still be a
substantial reduction in the efforts needed to review records
that probably will not be requested.
The public may not be aware of the fact that, even under
the systematic review provision of the Executive order, most
classified information will never be reviewed for declassi-
fication because it is not considered permanently valuable.
NARS has estimated that only about 5 percent of all Government
records are considered permanently valuable and subject to a
systematic declassification review. Much of the other 95 per-
cent, while not considered permanently valuable, could still
be of interest to the public; yet these records are not sys-
tematically reviewed. These records are made up of classified
information that is originated and used by Government agencies
in their daily operations. Some of the nonpermanent records
are maintained by the agencies for many years. Most of this
type of information is destroyed when it is no longer needed.
Nevertheless, most classified information is not systematically
reviewed for declassification and is not made available to the
public. And as noted above, it is probable that only a small
portion of the 5 percent of the Government's records that are
systematically reviewed will ever be requested by the public.
RESPONSIVENESS TO THE PUBLIC
COULD BE IMPROVED
Most classified and otherwise restricted information is
not requested by the public under the provisions of the
Executive order, FOIA, the Privacy Act, or other statutes.
Most of NARS' requests for information are informal and made
by individuals in writing, by phone, or in person. NARS
estimates that between fiscal years 1981 and 1988, the public
will request about 17 million pages of classified or otherwise
restricted information. Of this amount, about 16 million
pages, or 95 percent, will be requested informally.
Approved For Release 2002/04/01 : iiA-RDP93B01194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
We believe that responsiveness to the public could be
improved if the efforts now being directed to systematically
reviewing all 20-year-old material were z.edirected to re-
viewing only that information (1) requested informally,
(2) requested formally under provisions of the Executive
order, FOIA, and the Privacy Act, and (3) which NARS antic-
ipates will be requested.
When information has been systematically reviewed but
not declassified, an individual may request, under the
mandatory review pro'.;ision of the Executive order or FOIA,
that the classified portions of the information be declassi-
fied and released. According to NARS officials, individuals
may rely even more heavily on. FOIA and mandatory reviews in
the coming years, because the amount of information exempted
under systematic review is growing due to the increase in
intelligence information requiring review as it becomes 20
years old. Fiore of this information will remain classified
beyond 20 years in order to protect sources and methods.
From 1972 through 1979, when 30-year-old material was
being reviewed, 143 million classified pages were system-
atically reviewed for declassification by one NARS group and
only 2 million pages, or about 1 percent, remained classified.
However, NARS officials estimated that the amount of material
that is reviewed and not declassified could increase by 20
percent or more in the future as more 20-year-old material is
reviewed. MARS recently reviewed a project which contained
intelligence information. That project contained approximately
15,000 documents. About one-third of the documents were not
declassified. Individuals wanting access to information not
systematically declassified may turn to mandatory requests
under the Executive order or.FOIA to obtain the information.
The mandatory review provision of Executive Order 12065
provides that, upon request to either NARS or the originating
agency, any document or section thereof that no longer requires
protection shall be declassified, except information less
than 10 years old originated by the President or his repre-
sentatives. The originating agency has 60 days to decide
whether to release the requested information in total or in
part or whether to deny the request.. The decision can be
appealed to the agency within 60 days. The agency has 30
days to act on an appeal. The mandatory review provision
also allows individuals to specifically request information
previously exempted from declassification under systematic
review, thus requiring a decision on a case-by-case basis.
Approved For Release 2002/04/91: CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
The ICRC, which had oversight responsibility under the
previous order, reported 12,800 mandatory review requests
from 1973 through 1977. About 8,000, or 62 percent, of the
requests were granted in total or in part.
FOIA provides that, upon request, an agency shall
declassify any document or portion thereof that no longer
requires protection under the provisions of an Executive
order. The agency has 10 working days to release the docu-
ment in total or in part, to deny the request, or to provide
notification that the request is being processed. Decisions
can be appealed to the head of the agency or to the specific
office where the request was made. If the appeal is rejected,
the requestor can take the case to court.
Federal agencies were not required to report the total
number of FOIA requests received each year. Consequently,
statistics on the total number of requests, Government-wide,
were not readily available. However, we were able to obtain
information on FOIA requests and denials from the military
departments, the CIA, and the State Department. During 1978
and 1979, 111,700 requests for information were made to these
organizations; 96,800, or 87 percent, were granted in total
or in part. As of December 1979 the State Department and the
CIA had a backlog of over 4,100 FOIA requests.
THE EXISTING METHOD OF SYSTEMATIC
REVIEW FOR DECLASSIFICATION IS
INEFFICIENT AND COSTLY
Inefficient declassification procedures could be elimi-
nated and costs could be reduced if the systematic review
requirements were modified.
Since systematic declassification reviews became a re-
quirement in 1972, the same classified documents often have
been reviewed a number of times before being made available
to the public. Those reviews have included agency reviews
before material was accessioned to NARS, systematic reviews
by the Records Declassification Division, and reviews by
NARS custodial divisions for other general and specific
restrictions.
Most agencies only review and declassify information that
they have originated, which they believe is no longer sensi-
tive. They are not authorized to declassify material in their
25
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
files that was originated by another agency. Prior to acces-
sioning material to NARK, the CIA, Navy, and the Department
of Energy review their files and recommend to NARS which
information can be declassified. The recommendation, however,
only applies to information that the agency believes it can
declassify using its own declassification guidelines. Because
one agency's files may contain information from other agencies,
the NARS declassification staff must again review the files
using the other agencies' declassification guidance. NARS
identifies that material requiring further review and the
agencies' declassification personnel actually performs another
review of that information before a final declassification
determination is made.
Before the public can have access to the material, the
NARS custodial staff again reviews the files to identify and
remove material restricted under provisions of FOIA and the
Privacy Act. If the public later requests material specifi-
cally under the FOIA or the mandatory review provision of the
Executive order, that material has to again be reviewed by
either a WARS custodial division or the originating agency,,
even if the material was exempted from declassification in
any of the previous reviews.
These multiple reviews for declassification were time
consuming and costly.
NARS, the CIA, and the Departments of Defense, Energy,
and State have estimated that., at current levels of effort,
their salary and benefit costs for systematic reviews for
fiscal years 1981 through 1988 will total about $88 million.
The CIA told us that, at its current level of effort, it will
spend about $16 million during the 8-year period, but it will
only be able to review about 22 percent of its 20-year-old
classified information by December 1, 1988. The CIA also told
us that it would need an additional $64 million to meet the
review requirement of the Executive order by December 1988,
and that the additional funds would have to be taken from
other CIA programs.
CUSTODIAL DIVISIONS SHOULD BE RESPONSIBLE
FOR SYSTEMATIC DECLASSIFICATION REVIEWS
Prior to Executive Order 11652 in 1972, the NARS custodial
divisions were responsible for performing all declassification
and other reviews prior to the release of records. Since then,
they have continued to perform declassification reviews under
Approved For Release 2002/04/0216: CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
FOIA and other programs. The only exception is the systematic
review program which was assigned to the newly formed Records
Declassification Division.
Custody and control of the Federal Government's records
in the National Archives are divided among seven custodial
divisions, depending on the subject matter of the records.
Each division is responsible for preserving, arranging,
describing, and providing a reference service to the public
for the records that it holds. Virtually all classified
records fall within the Military Archives Division, the Civil
Archives Division, and the General Archives Division. As
part of arranging and describing the records, staff in these
divisions must review and remove those records which meet
certain restrictions to access, such as those prescribed by
FOIA and the Privacy Act. In addition, there are about 20
other general restrictions that the staff must identify.
These include restrictions on records originated by the
Joint and Combined Chiefs of Staff, copyrighted material,
and DOD personnel and inspector general reports.
Prior to Executive Order 11652, when systematic reviews
were initiated, the custodial divisions reviewed classified
material for possible declassification as part of their normal
workload. They concentrated primarily on information specif-
ically requested by the public, with secondary emphasis on
information that they anticipated the public would request.
In 1972, in order to implement the systematic review require-
ment of Executive Order 11652, NARS created the Records
Declassification Division and gave it the primary responsi-
bility for systematically reviewing information for declassi-
fication. However, the custodial divisions continued declas-
sifying some information specifically requested by the public
informally and under FOIA and the Privacy Act. Since 1972 the
NARS custodial divisions have declassified over 2.6 million
pages in response to specific public requests for information.
The custodial staffs believe they could again incorporate
a declassification review as part of their responsibilities.
NARS officials agree that the custodial staffs have better
subject matter knowledge than the declassification staff and
could apply declassification guidance with less agency assis-
tance, thus reducing the time and cost of declassifying
information.
NARS officials said that when agencies accession perma-,
nently valuable 20-year-old records to NARS, certain groups,
Approved For Release 2002/04/01 :201A-RDP93B01194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
for which a public demand is anticipated, could be selected
for systematic review. Reviews of this type have been done
in the past.
Unlike the high staff turnover in the declassification
division, the custodial divisions' turnover has been low.
From 1972 through 1979 the annual staff turnover in the three
custodial divisions, that had virtually all the classified
information, averaged only 9 percent, as compared to an
average turnover rate of 41 percent in the declassification
division for the same period. Continuity in personnel has
permitted the staff to gain experience, develop an indepth
and thorough knowledge of information in agency files, and
establish close working relationships with agency personnel.
The custodial staffs, unlike the declassification division
staff, deal with the public on a daily basis and assist them
in obtaining information in which they are interested.
LIARS was unable to estimate the exact number of people
that would be needed in the custodial divisions if they
assumed the responsibility for all declassification reviews.
However, we believe that some administrative economies could
be effected by consolidating the personnel in other existing
divisions. Management personnel of the declassification
division could be used in an advisory and coordinating capac-
ity with the custodial divisions, agencies, and ISOO to ensure
the uniform application of declassification policies and
procedures.
CONCLUSIOIIS
A major objective and benefit of Executive Order 12065's
requirement for the systematic review for declassification of
all 20-year-old, permanently valuable classified information is
the public's perception that it provides for increased openness
in Government. We agree that such a perception is important
to public confidence in the classification program and in
the Government in general. We fully endorse the policy of
openness in Government. However, we believe that modification
of the systematic review requirement of the order would make
the declassification part of the program more responsive to
the public. Such modification might have some effect on
public perception, but we believe that if the public were
made fully aware of the costs and benefits of the existing
system and proposed modification as described in this chapter,
general acceptance of the change might be forthcoming.
28
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
Since most of the 20-year-old classified information
that is systematically reviewed and declassified may never be
requested by the public, the effort required for such reviews
could be used more efficiently if it were redirected to areas
of greatest concern to the public. Instead of being used to
review all 20-year-old material, that effort could be used to
review those groups of records specifically requested and
those likely to be requested. Some effort could also be used
to respond to FOIA and mandatory review requests. In addi-
tion to being more responsive to the public, we believe that
the Government could save a substantial part of the estimated
$88 million that will be spent during fiscal years 1981
through 1988 to review all 20-year-old classified information.
RECOMMENDATIONS
We recommend that the Chairman of the National Security
Council draft, and submit to the President for approval, a
revision to section 3-4 of Executive Order 12065 that would
modify the requirement for the systematic review for declas-
sification of all 20-year-old permanently valuable classified
information. Specifically, systematic reviews for declassi-
fication should be made of those records or groups of clas-
sified records requested by the public and those records
which the Archivist of the United States anticipates will be
requested.
We recommend, after modification of the systematic
review requirement, that the Administrator of General Services
direct the Archivist of the United States to transfer respon-
sibility for all declassification reviews to the custodial
divisions. To ensure uniform application of declassification
policies and procedures, as prescribed by approved declassi-
fication guides, the management personnel in the Records
Declassification Division should be used in an advisory and
coordinating capacity with the custodial divisions, agencies,
and ISOO.
ISOO COMMENTS AND OUR EVALUATION
On June 26, 1980, the Director of ISOO, on behalf of
the Administration, strongly opposed our suggestion for
eliminating the requirement that all 20-year-old, permanently
valuable classified information be systematically reviewed.
(See app. II.) ISOO also expressed the view that our report
had three shortcomings:
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
29
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
1. It failed to address or evaluate the public policy
of open Government as a product of systematic
review.
2. It concluded that FOIA and mandatory review
procedures of the Executive order were adequate
substitutes for systematic review.
3. It failed to consider alternative and less drastic
solutions to the problems identified in the report
before recommending abolishment of the systematic
review.
According to ISOO, the report "largely ignores any
criteria other than cost in evaluating the program" and that
its cost is substantially less than that of FOIA. ISOO further
stated that the existing system helps to offset the perception
held by many that Government officials hide their mistakes
behind the cloak of secrecy, and that our recommendation would
lend credence to that perception by dooming vast quantities of
records to permanent closure.
As a result of the comments, this chapter has been revised
to reflect some of the concerns expressed by ISOO and other
agencies (see below). We have more fully addressed the ques-
tion of the public's perception of openness in Government and
clarified our conclusions to show that we do not consider the
FOIA and mandatory review procedures substitutes for the sys-
tematic review of all 20-year-old classified material. Sys-
tematic reviews would be made, but only of those records that
are requested and those that the Archivist expects to be re-
quested in the future. Our suggestion has been revised ac-
cordingly.
We agree with ISOO that cost is not the most important
reason for changing the systematic review requirement of the
Executive order. Responsiveness to the public is certainly
more important. While the costs incurred under the FOIA may
be higher than the costs associated with systematic review,
the FOIA costs are the direct result of specific requests,
whereas the systematic review costs are for reviewing rec-
ords--most of which will probably not be requested by the
public. We believe that it is incumbent upon the Government
to be responsive to its citizens, but it is equally incumbent
for the Government to do so in an efficient and economical
manner.
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
30
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
Executive order provisions and the perceptions of openness
in Government that they often are intended to convey must be
examined and evaluated from time to time. Another provision
of Executive Order 12065, related to systematic review, deals
with the duration of classification. The order provides for
most information to be classified for periods not in excess of
6 years. It also authorizes some information to be classified
for more than 6 years, but states that such authority "shall
be used sparingly." That provision of the order, besides
conveying the impression that there will be greater openness
in Government, if followed, would have a major impact on the
efforts required for systematic reviews. However, whether
justified or not, the authority to extend classification beyond
6 years has not been used sparingly. An ongoing review of
classified documents. held by contractors shows that 97 percent
were classified for more than 6 years. An earlier review at
Government installations showed that over 50 percent were
similarly classified. (See p. 3.)
NARS, DOD, CIA, AND DEPARTMENT OF
ENERGY COMMENTS AND OUR EVALUATION 1/
MARS believes that access to classified information would
be curtailed if systematic reviews were eliminated, because
sound research methodology by serious scholars requires general
access to the records potentially concerning their subject of
research. It does not believe that research conducted within
the confines of theFOIA and mandatory review provision would
be acceptable to researchers.
As revised, we believe our recommendation would not
hinder research, because NARS already informally reviews
and declassifies records requested by researchers. As
noted on page 15, NARS said that if records have not been
requested by a researcher, it can and does defer declassifi-
cation action. Furthermore, section 4-3 of the order, "Access
by Historical Researchers and Former Presidential Appointees,"
provides for access to classified information by persons
engaged in historical research projects, when certain condi-
tions are met.
1/Since agency comments were lengthy we have included only
those of two principal groups involved--NARS and ISOO.
Other comments are also summarized.
Approved For Release 2002/04/0131CIA-RDP931301194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
DOD believes that the progressively more stringent
policies that the executive branch has evolved during the
past 27 years for assuring orderly, earliest possible declas-
sification have well served the goal of maintaining public
confidence in the integrity of the classification system.
DOD also believes that we should have considered cost reduc-
tions that might be realized by a less radical revision of
the Executive order, such as reverting to the systematic
review of all 30-year-old classified material.
We have already addressed the issue of public confidence
in the classification system. With respect to the suggestion
that we should have considered a less radical revision, we
believe a less radical revision would not resolve the problem.
The Department of Energy told us that while it supports
the systematic review program, it does not have adequate
resources to properly conduct the required review.
The CIA did not comment on our recommendation. It said,
however, that the multiple layers of review mentioned in the
report were necessary because records custodians, such as
NARS, regardless of their familiarity with the records, do not
possess the experience, background, and knowledge needed to
make declassification determinations with respect to such
things as intelligence, national defense, and foreign
relations.
Our recommendations, as modified, would eliminate the
need for some of the multiple layers of review, but we agree
with the CIA that such reviews may be necessary in some cases.
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
32
 APPEApftovad For Release 2002/04/01 : CIA-RDP93BOl194RO0110001YL( ADIX I
General National Archives
Services and
Administration Records Service Washington, DC 20408
JUL 3 1980
Mr. R. W. Gutmann
Director
Logistics and Communications Division.
United States General Accounting Office
Washington, DC 20548
Enclosed is a paper presenting the position of the National Archives and
Records Service (NARS) to the proposed draft report entitled,
"Systematic Review for Declassification of National Security Information?
-
Do Benefits Equal Costs? It This agency cannot accept the conclusions
and recommendations of the draft report that the systematic review
program in operation in the National Archives since 1972 is not suffi-
ciently cost effective to warrant its continuation.
We oppose this conclusion on several bases. First, this government's
avowed public policy position of openness would be seriously undercut
if the systematic review program were abolished. This point is cogently
argued in the response of the Director of the Information Security Over-
sight Office to the proposed draft report and has our wholehearted
endorsement. Second, the Freedom of Information Act and the mandatory
review provisions of EO 12065 are inadequate substitutions for systematic
review. Third, many of the findings in the report are not pertinent to
the conclusion to end systematic review. Fourth, several of the report's
findings are inaccurate and do not support the conclusion. Finally, the
systematic review program's problems identified in the GAO report are
susceptible to solutions within NARS or in cooperation with other agencies.
Thank you for giving us this opportunity to express our views for the
proposed draft report. If you would like more information from us on
the matter, Edwin A. Thompson, Director of the Records Declassifi-
cation Division of NARS is available to discuss the report in further
detail with Mr. Boker of your office.
Alc
LAMES E. O'NEILL
Acting Archivist
of the United States
33
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 APPENDI `pproved For Release 2002/04/01 : CIA-RDP93B0l194R001148 i.$ I
National Archives and Records Service
RESPONSE TO THE DRAFT GAO REPORT, "SYSTEMATIC REVIEW FOR
DECLASSIFICATION OF NATIONAL SECURITY INFORMATION-DO BENEFITS EQUAL
COST"
1. THE FREEDOM OF. INFORMATION ACT AND/OR MANDATORY REVIEW
PROVISIONS OF E.O. 12065 ARE INADEQUATE SUBSTITUTIONS FOR SYSTEMATIC
REVIEW.
The draft report assumes that materials of interest to historical researchers and others
can be made available more cost effectively through reliance on mandatory review
procedures and/or the Freedom of Information Act than through the systematic review
program. The National Archives and Records Service does not accept the premise that
the perceived cost-effectiveness should be the sole or even the primary measure of the
program's worth. However, if we were to accept cost-effectiveness as the basis for
evaluating the program, we remain unconvinced that mandatory review and Freedom of
Information Act (FOIA) requests provide acceptable alternatives.
If systematic review were eliminated, we believe that access to classified information
would be curtailed. Sound research methodology by serious scholars requires general
access to the records potentially concerning their subject of research. To fully explore a
research topic, a researcher requires comprehensive access to the records, i.e., to
examine the records with minimal constraints. The researcher frequently does not know
specifically what is available relating to his/her topic and would, therefore, find it
extremely difficult, if not impossible, to initiate specific FOIA requests. When told that
the files are closed because they are classified and that he must specifically identify the
documents he wants released, the researcher becomes frustrated and angry at a
Government that preaches openness but does nothing positive to make that happen.
Researchers are also frustrated by the time delays associated with the FOIA and
mandatory review processes. Only the most persistent researcher without publication or
34
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 APPEI fouled For Release 2002/04/01 : CIA-RDP93B01194RO0110001)QQ i DIX I:
other deadlines can consider researching a post-1950 topic involving classified records
using the Freedom of Information Act and the mandatory review provisions of E.O.
12065.
As the agency primarily concerned with scholarly access to the Government's
permanently valuable records, we are keenly aware of the negative impact of closed
records. It was this experience which led us in 1971-72 to propose the systematic review
of the vast quantity of World War II records in our custody. We championed the inclusion
of the principle of systematic review, which became an important part of E.O. 11652, in
1972 because we knew that thorough historical research was not possible as long as the
records remained classified.
Since 1972 about 250,000,000 pages of pre- 1950 records, formerly closed because they
contained classified information, have been systematically reviewed and declassified.
Research and scholarship in the history of this period have developed in directions which
were impossible a decade ago. To cite just one example, during the past 9 months nearly
200 requests were submitted by researchers for over 1,000 cubic feet of records created
by SCAP and OMGUS, the U.S. occupying authorities in postwar Japan and Germany.
Systematic declassification review of the 18,000 cubic feet of these records was
completed just a few years ago. We are confident that this flourishing research would
not have been undertaken if the records had not been systematically reviewed and made
available for research.
We firmly believe that the encumbrances surrounding access to classified records has a
very real chilling effect on serious research. Articles and books on the post-1950 period
are not being written because access to the necessary records is so difficult. We also
know that research based on documents released through the Freedom of Information Act
and mandatory review process is frequently flawed because it is based only on fragments
of information, rather than the complete record. Researchers are unable to synthesize
35
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8
 APPENDIXpRroved For Release 2002/04/01 : CIA-RDP93B01194R0011QF& I
and place in historical perspective the bits of information made available to them. The
inevitable result is an uninformed or misinformed public.
While systematic review does not result in declassification of everything examined, it
does ensure that 80%-95% of most records are released and allows the researcher to
better appraise for himself the significance of items being withheld. The National
Archives' experience has been that most researchers are fully satisfied with the records
released under the systematic review programs and only infrequently request another
review of doc\ ments initially denied. We believe that researchers will strongly oppose
conducting their research entirely within the confines of FOIA or mandatory review.
II. THE DRAFT REPORTS' CONCLUSION TO END SYSTEMATIC REVIEW IS NOT
PERTINENT TO THE FINDINGS.
A. Systematic review for declassification of all 20-year-old permanently
valuable material will not be accomplished by December I, 1988. It is true that NARS
cannot reach this objective with the present level of resources. However, this is
irrelevant to the question of whether the systematic review program should be continued
or terminated. Even though present resources are inadequate to reach the goal of
reviewing all permanently valuable classified records by 1988, NARS will be able to
systematically review and declassify the greater part of approximately 250 million pages
of historically significant documents by 1988. It seems illogical to us to conclude that
because all permanently valuable classified records will not be systematically reviewed
by 1988 that the program should be terminated, thereby ensuring that none of the records
are systematically reviewed.
The problem of ensuring that records schedulles are developed for all agencies' records is
one which we are well aware. While we do not argue with the fact that some agencies
may not be adhering to good records management practices,the fault does not lie with
the systematic review program. This is surely a case of putting the cart before the
36
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 APPi r vE.d For Release 2002/04/01 : CIA-RDP93B0l194R001100010WP DIX I
horse. But the blame for this inadequacy rests on the failure of agencies to schedule
their records. Whether it is the fault of poor records management or the lack of agency
personnel to do the job, the fact remains that the report would have better addressed this
facet.of declassification if it had forcefully recommended greater agency attention to
records scheduling. Recommending the abolition of systematic review instead is
completely illogical.
NARS will continue to work with the agencies to accelerate the development of records
schedules, to appraise records, and to accession them into the National Archives. NARS
will also reappraise previously accessioned classified records to ensure their continuing
value before they are reviewed for declassification. While we recognize that these steps
must be taken to guarantee that only permanently valuable records are systematically
reviewed, they are not germane to the question of whether systematic review should be
continued.
B. Most, post-1950 classified information will require costly and time consuming
page-by-page review. We agree. Only a small portion of the post-1950 record files thus
far surveyed have proven susceptible to quick "bulk" declassification action. But,
however small that portion, this procedure is the most cost-effective method for
declassification and we would lose this capability if the systematic review program is
discontinued.
In a number of cases researcher requests have caused us to examine on a page-by-page
basis records we might have been able to "bulk" declassify if we had been able to delay
declassification review until the foreign. government information in the files was 30 years
old. If records have not been requested by a researcher, we can and do defer
declassification action until the passage of time eliminates this impediment to bulk
declassification.
37
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194R0011go'19B1 I
APPENDIX I
Of greater concern ore agency guidelines which require NARS reviewers to identify and
withhold for agency action a significantly larger percentage of the documents examined
than was true for pre-1950 classified information. This is the direct result of the
continuing sensitivity today of many issues and programs which first arose during this
period. The undeniable consequence is that it takes longer and that it is, therefore, more
costly to systematically review more recent records. But we can, by better management
of the NARS review program, by better coordination of potentially sensitive information
with agency specialists, and by joint efforts to improve the guidelines, ensure a more
efficient and consistent systematic review effort. These are real problem areas
warranting closer and continuing attention by NARS and all of the agencies with whom
we work. We do not agree, however, that because the declassification review of more
recently dated records is slower and more costly, that the entire systematic review
program should be discontinued.
C. The present review program is costly and results in the systematic review of
much information not requested by the public. NARS does not accept the GAO's
conclusion that systematic review is not necessary because researchers have to date only
examined approximately 10% of the documents declassified. The purpose of archival
appraisal is to designate a universe of documents from which researchers will select
portions for examination. If researchers were already examining all of the documents
appraised as permanently valuable, then clearly the archivists involved have failed to
identify a large enough universe.
Our experience has shown that research interests change with time. Trends in
scholarship change over time, resulting in the examination of records which were not in
demand 20 or even 10 years ago. Systematic declassification review results in rapid and
cost-effective release of information in advance of the changing demand. Surely these
anticipatory reviews do not justify the demise of the program.
38
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 APPMJ pi4Nei1 For Release 2002/04/01 : CIA-RDP93B0l194R001100010R NDIX
NARS also cannot accept the GAO's interpretation of the figures used in the table on
page 24 of the draft report. The estimate of the number of pages reviewed in a given
year as a consequence of direct public requests is part of the ZBB presentation. It
represents the minimum effort involved if no systematic review program were conducted
that particular year. Our estimated resource requirement to meet these public demands
represents about 15% of the Records Declassification Division's present budget. But if
systematic review were discontinued indefinitely and the backlog of classified records
grew, the demand by researchers for release of classified information will easily double
within 5 years. Within a decade we might well be spending as much to respond to public
requests as we are now devoting to the entire systematic review program.
But the immediate real cost of the proposed termination of systematic review would be
the chilling effect on research into our recent past, and the growth of a large
accumulation of permanently valuable records which are not available to the public.
Sooner or later this mass of classified records would have to be acted upon or the
National Archives would soon become a storage vault of largely unavailable
information. The Government and the researcher community found such a situation
intolerable in 1972. The situation should not be allowed to develop again.
D. Some classified documents have been reviewed a number of times before
being made available to the public. Re-reviews of classified records are helpful and in
some cases absolutely essential. Some agencies have reviewed records prior to their
accessioning into the National Archives or prior to their systematic review by NARS
personnel. These agency reviews, which largely concentrate on intelligence files,
identify documents which are still sensitive, which enables NARS to proceed rapidly
through the review of the classified information originated by other agencies.
We agree that some of the early efforts by agencies to review their records were
unsatisfactory and required wasteful total re-review. The early problems encountered
39
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 APPEND IX-pfroved For Release 2002/04/01 : CIA-RDP93B0l194R00110Di- I
(including review of records later determined to be non-permanent) have been overcome
through closer inter-agency coordination. The ongoing agency reviews are proving to be
helpful because they greatly increase the speed with which NARS can act on certain
files, and they eliminate the need for NARS to identify potentially sensitive documents
and refer them to agency specialists.
A total re-review of the documents withheld between 1972 and 1978 has resulted in the
declassification and refiling of about half of the records originally withdrawn. This is
largely the result of the development of new guidelines and the completion of agency
coordination. The overall result has been the release of several million pages of records
once thought to be sensitive and is a strong argument for periodic re-reviews.
Finally, some records withdrawn from the files under the systematic review procedures
are subjected to a further review when specifically requested by a researcher. In such
instances the requester has determined from the description available that the specific
document is pertinent to his research. Requests of this type represents less than 2% of
the records presently being reviewed at the specific request of researchers. Clearly,
most researchers are well satisfied with the product of systematic review.
E. Internal review of the NARS declassification operation have not been made.
This is a valid criticism and one which will be rectified. NARS is committed to an
efficient and effective program aimed at the rapid declassification review of the
Government's permanently valuable records. The problems addressed by the GAO
examiners will be among those matters covered in a further review of the entire
program.
Ill. SEVERAL OF THE DRAFT REPORT'S FINDINGS ARE INACCURATE OR
MISINTERPRETATIONS OF THE FACTS.
A. NARS custodial divisions should again do declassification reviews on
40
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 APPEN oiled For Release 2002/04/01 : CIA-RDP93B0l194R00110001 tt$DIX
request. The GAO report correctly notes that prior to the issuance of E.O. 11652 in 1972
the NARS custodial units performed such declassification reviews as were done. In truth,
however, the declassification effort in NABS was severely limited. A few agencies made
declassification decisions and transmitted that information to NARS. NARS' role was
merely to mark the documents declassified. Most records, no matter how old, were
unaffected and the custodial divisions routinely transmitted the request to the
responsible agency for action. Declassification guidelines as we know them today were
unheard of before 1972.
Declassification review has become a specialized activity requiring training and
experience for its successful accomplishment. The most effective training is to work
with those already experienced in applying the guidelines, and particularly to work
alongside the agency declassification review specialists. New problems, new
interpretations of guidelines, and agency instructions are a constant matter for the
attention of all reviewers. Expertise in declassification of the records of a single agency
is not sufficient, as classified information from a large number of agencies is regularly
encountered in most reviews. NARS' experience over the past 8 years convinces us that
the develoment of declassification review expertise is the key to proper and consistent
declassification decision making. We are certain that this can best be obtained by
concentrating this specialized function in a single organization working on a regular basis
with agency declassification review specialists.
Since 1972 the Congress has appropriated funds to NARS for the specific purpose of
systematically reviewing classified records. These funds were not intended for the
review of unclassified or declassified records for public release (a normal part of the
NARS custodial units' work). For these reasons and for good management of the
especially appropriated funds, we do not believe that this activity should be disbursed
among the NARS custodial units.
41
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R00110%0)R99 X
APPENDIX I
B. NARS survey estimates of pages declassified were increased. NARS is taking
steps to ensure that any changes in original estimates are fully documented and
justified. The Records Declassification Division Director issued a memorandum to all
Branches on February 29, 1980, immediately after this matter was brought to his
attention by the GAO auditors. In the future, any change of the percentage estimate
must be explained, recorded, and signed by the Branch Chief on the document involved.
C. Discrepancies in statistics compiled by NARS and reported to the Ovesight
Office. The,NARS program statistical reporting form (GSA Form 7096) was revised in
1972 to account for systematic review work performed. The key statistic is
"Declassification Completed" which counts the number of classified pages in the material
examined-which was declassified as a result of "bulk declassification" action or page-by-
page review. However, this figure was available only when the NARS review of a block
of records was completed following agency examination and determination by an agency
head that the information withheld required continued classification beyond 30 years. In
many instances, review projects were begun in one year and not completed until the next
year.
The inadequacy of this figure as a basis for reporting work completed at the end of a
calendar year for the Interagency Classification Review Committee's annual report was
readily apparent. The Director of the NAR S Records Declassification Division,
therefore, made the best estimate he could of the accomplishments of the entire
systematic review program in his unit, in custodial units, and in the Presidential
libraries. The figures reported to the ICRC were total pages reviewed and the percent
declassified. Because of its inadequacy for ICRC reporting purposes, the figures shown
on the GSA Form 7096 were never used as the basis of the NARS report to the ICRC.
Revision of the statistical reporting form along the lines suggested by the GAO has
already been approved and will be used beginning in FY 1981. But the most significant
42
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
APPENDIX I APPENDIX I
changes have been brought about by internal procedural decisions which will assure that
review projects are completed in a timely manner and that reports input into the
computer will be used for statistical analysis and monthly compilation. Beginning in FY
1981, NARS expects to be able to report the total number of pages examined, how many
of these were classified pages, how many were withheld, and consequently how many
classified pages were declassified.
IV. THE DRAFT REPORT FAILS TO CONSIDER REMEDIES LESS DRASTIC THAN
ABOLISHING SYSTEMATIC REVIEW IN ITS ANALYSIS OF THE PROGRAM'S
PROBLEMS.
Although we do not agree with a number of the report's findings per se, we do believe
that the report does raise some valid questions. We will cite some of the specific
problems with systematic review referred to in the report and suggest some specific
remedies, or at least avenues to pursue, which we believe the report should have
considered, rather than simply calling for its abolition.
The draft report states that only a small percentage of records made available to
researchers are actually used by researchers and, therefore, the cost of making records
available does not equal perceived benefits. Even if we were to assume the fact that not
all systematically reviewed material is actually used by researchers, that fact is not the
fault of systematic review. Rather, if there are too many records in the universe under
review, the fault rests with the appraisal criteria or the application of these criteria used
to determine what in fact is historically valuable. To be sure, the era of United States
history currently undergoing review was by its very nature conducive to the creation of
materials considered to be permanently valuable. It should also be recognized that
information in the National Archives is preserved for use by present and future
researchers and that over time, historical interests and emphasis changes. Records that
are being researched now may not be as heavily used in the future, while records which
are little used currently may be primary sources for future research. It seems for more
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
43
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
reasonable that if a problem does exist it is remedied more effectively and for less
drastically by a systematic reappraisal of the pertinent records series before systematic
review is conducted- NARS, working closely with the agencies, is undetaking further
close examination or reexamination of records prior to their accessioning or
declassification review.
Another example pertains to the report's emphasis on NARS declassification staff
turnover as one of the major problems of operating an efficient systematic review
program. While staff turnover is an operational problem, it is surely not a proper index
for determining the validity of the program itself. To a large extent, the personnel
turnover is the result of present hiring practices. First, for many of these employees,
this is their first full-time job. Second, the constant review of records on a regular basis
is tedious and often boring work. Finally, promotion potential is limited. The solution to
employee turnover is surely not to abolish these positions entirely.
The management of NARS is reexamining its hiring policies for personnel engaged in this
type of systematic review. Under consideration is the possibility of hiring on a part-time
basis retired experts from those agencies who classify large quantities of information.
This would result in an employee mix that would reduce turnover while simultaneously
blending youthful energy and enthusiasm with the wisdom of experience.
CONCLUSION
The systematic review of security classified records of historical value is, in our view, an
essential part of this Government's commitment to openness. This commitment cannot
be adequately met by piecemeal releases resulting from FOIA and mandatory review
requests. Further, processing these individual requests is many times more expensive
than the costs incurred in routine systematic review.
We are opposed to reverting to the conditions which prevailed in the National Archives
Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
44
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
APPENDIX I APPENDIX I
before 1972 as recommended in this draft report. Several of the problems identified by
the GAO examiners will receive immediate management attention in NARS. However,
we cannot accept the draft report's conclusion that this important and publicly valuable
program ought to be terminated because of a number of relatively minor problems which
are susceptible to solutions within NARS and the other Federal agencies.
Approved For Release 2002/04/01 ;491A-RDP93B01194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
General Information Security
Services Oversight Washington, DC 20405
Administration Office
Mr. R. W. Gutmann
Director
Logistics and Communications Division
United States General Accounting Office
Washington, D.C. 20548
Dear Mr. Gutmann:
Thank you for the opportunity to review and reply to the draft of a proposed
General Accounting Office report entitled, "Systematic Review for Declassifi-
cation of National Security Information -- Do Benefits Equal Cost?" In my
capacity as Director of the Information Security Oversight Office (ISOO), I
am pleased to enclose a response on behalf of the Administration. I under-
stand that several of the agencies to which you referred the draft report
may be responding directly to you with their individual comments.
Because of our significant problems with the draft report, I would be more
than happy to meet and discuss it with you or members of your staff at any
mutually convenient time or place. I am convinced that our objectives are
basically the same, and that we can ultimately agree to the approaches we
should pursue to attack the problems that exist within the systematic review
program. I would also appreciate the opportunity to review and comment upon
any future draft of the GAO report. I can be reached at 633-6880.
STEVEN GARFINKEL
Director
Approved For Release 2002/04/01 4 JIA-RDP93B01194RO01100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93BOl194RO01100010001-8
APPENDIX II APPENDIX II
ISOO'S RESPONSE TO THE DRAFT
OF A PROPOSED GAO REPORT ENTITLED,
"SYSTEMATIC REVIEW FOR DECLASSIFICATION OF NATIONAL
SECURITY INFORMATION -- DO BENEFITS EQUAL COST?"
The General Accounting Office (GAO) has submitted to several agencies
of the executive branch for their review and comment a draft of a proposed
GAO report entitled, "Systematic Review for Declassification of National
Security Information -- Do Benefits Equal Cost?" The draft report concludes
that the systematic review program established in Section 3-4 of Executive
Order 12065 is not cost-effective, and, therefore, should be abolished by an
amendment to the Order. The Administration strongly opposes the conclusions
and recommendations of the draft report.
GAO's examination of the systematic declassification review program is
potentially of great assistance to the executive branch in its efforts to
implement this program effectively and efficiently. To be sure, a number of
problems identified by GAO in its present draft are real and must be remedied,
if the program is to achieve its objectives in a more meaningful way. The
Administration fully supports such exploration. For example, there is no doubt
that systematic review is costly and burdensome to the agencies that must
implement its requirements. To improve the productivity of the system would
benefit both the government and the public. To this end the draft report makes
a number of observations about the costs associated with the present system
which demand and will receive our attention.
Nevertheless, the present draft falls far short of its potential for
constructive criticism. Despite a lengthy examination of systematic review,
the draft report largely ignores any criteria other than cost in evaluating the
program. Far more disturbing, however, are the report's recommendations,'which
we contend are unnecessarily drastic and largely insupportable. We are convinced
that the effectuation of these recommendations would result in an irreversible
public disservice. Above all other concerns, we urge that the final report seek
remedies which fit the legitimate problems associated with systematic review
rather than simply calling for its abolition.
In our view, the shortcomings of the draft report are threefold: First,
its failure to address or evaluate the public policy of open government as a
product of systematic declassification review; second, its conclusion that the
Freedom of Information Act (FOIA) and/or agency mandatory review procedures are
adequate substitutes for systematic review; and, finally, as alluded to above,
the recommendation that systematic review be abolished without any consideration
of alternative and less drastic solutions to the problems identified in the report.
Approved For Release 2002/04/01 : CIA-RDP93BO1I 94RO01100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93BOl194RO01100010001-8
I. One of the greatest shortcomings of the draft report is its one-
dimensional scope. It focuses only on the cost of conducting the systematic
review program. It ignores both the tangible and intangible benefits that
accrue from a program that is one of the cornerstones of the public policy
frequently referred to as "open government." The importance of the systematic
review program cannot be appreciated without taking into consideration its
philosophical and even psychological purposes.
In its laws and policies regarding public knowledge of governmental
activities, the United States is unique among nations. No other nation promotes
the public availability of information through statutes such as the Freedom of
Information Act, the Privacy Act, the Government in the Sunshine Act, and the
Federal Advisory Committee Act. The value of these laws is not limited to the
knowledge gleaned from their usage. The right to know is at least as valuable,
since this principle stands as one of the hallmarks of our democratic institutions.
In the same manner, systematic review establishes the people's right to know for
one of the most sensitive of areas, that to which we sometimes refer as "state
secrets."
Systematic review establishes the principle that an open society
cannot tolerate the permanent or indefinite closure of historically valuable
records. Moreover, by establishing a finite period of restriction, systematic
review helps to offset the perception held by many of a security classification
system abused by government officials intent on hiding their mistakes behind
the cloak of secrecy. The abolition of systematic review would lend further
credence to this perception, severely damaging the public credibility that is
necessary for the system to remain viable.
As a practical matter the abolition of systematic review at this
time will doom vast quantities of records adjudged by records appraisers to be
permanently valuable to the opposite fate -- permanent closure. GAO's own
statistics would indicate that substitute declassification procedures are able
to handle the review of approximately 2 million pages of classified materials
annually. Even understaffed, systematic review results in the annual review of
approximately 24 million pages. Assuming reasonably constant staffing levels,
the absence of systematic review will leave over 20 million pages of additional
unreviewed records each year. Within a few years the stack of unreviewed
documents would become so vast as to be outside the government's ability to
process them.
This is not a program whose price tag will ever be measurable in
the same terms as one which deals in goods or services. The Freedom of Infor-
mation Act is enormously expensive to implement, dwarfing the cost of systematic
review. Its primary beneficiaries are most often vested commercial interests,
not the research community or the general public. Do these factors warrant its
repeal? We daresay not, just as we contend that systematic review must also be
retained for the principles it espouses.
Approved For Release 2002/04/01 461A-RDP93BO1194RO01100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
APPENDIX II APPENDIX II
11. The draft report assumes that materials of interest to historical
researchers and others can be made available more cost effectively through
reliance on mandatory review procedures and/or the Freedom of Information Act
rather than through the systematic review program. We do not believe that these
procedures are adequate substitutes.
First, the absence of systematic review would effectively eliminate
bulk declassification. No matter how much the percentage of bulk declassifi-
cation may decline, it will always remain the most cost-effective means of
declassification. Further, without a systematic review system, it will be
virtually impossible to explore other avenues of promoting the most expeditious
declassification of large quantities of records.
Second, compared to systematic. review the costs of processing FOIA
or mandatory review requests for similar quantities of materials are enormous.
The report proposes to shift the burden of gaining access to the government's
older classified documents to the mandatory review and FOIA process without
adequately investigating the costs associated with those programs. We believe
that a more thorough investigation of program costs would show little if any
savings to the government by substituting mandatory review and the FOIA for
systematic review. To the contrary, it is our belief that a more in-depth
analysis of the alternative approach proposed in the report will show it to be
a more costly approach. Very many, if not most, FOIA and mandatory review
requests require the input of high level agency officials. Intricate denial
and appeal procedures frequently extend the processing period over several months
or even years. Although we have no way to predict, nor does the draft report
examine, how great an increase in the number of mandatory review and FOIA
requests would result from the elimination of systematic review, we believe it
would be substantial. Most researchers of formerly classified records rely
almost exclusively on systematic review to provide a universe of records from
which to select portions for examination. The demise of systematic review as
recommended by the draft report will inevitably result in an ever-expanding
backlog of classified records, thereby forcing significantly greater numbers of
researchers to turn to the more costly mandatory review and FOIA processes.
Moreover, the personal costs to the researcher will be significantly higher in
terms of expenses and delays if forced to rely on the highly regulated procedures
of FOIA or mandatory review rather than systematic review.
Finally, in dismissing the singular significance of systematic review
in providing a more complete universe of materials for examination, the draft
report fails to show any understanding of research methodology. A researcher
begins by selecting a rather broad research topic, reads all of the secondary
sources related to the topic, determines which records may offer fruitful areas
of investigation, and begins the laborious task of sifting through thousands of
documents to identify those pertinent to his research. A researcher seldom, if
ever, begins the research knowing what documents exist that will prove essential
or useful to the project. Because of this basic research methodology, many
researchers working on the more recent historical period (post-World War II)
have come to rely on the product of systematic review to enable them to identify
theparameters of their research.
Approved For Release 2002/04/01 : CM-RDP93B01194RO01100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93BOl194RO01100010001-8
Access to records declassified under systematic review often
provides researchers with leads to other documents which may not yet be
declassified. Often records can be requested under mandatory review or FOIA
because the researcher has discovered enough information in the records opened
through systematic review to permit the identification of other related docu-
ments. Without the interplay of systematic review and mandatory review or
FOIA, researchers will find the conduct of their research much more difficult.
There will he far less assurance that the researcher will have access to even
a significant portion of pertinent records. Scholarship will necessarily suffer.
III. The most disturbing feature of the draft report is its extremely
simplistic solution to theproblems associated with systematic review; namely,
the abolition of the program. Overlooked are far less drastic solutions that
are compatible with the identified problems.
For example, the draft report cites as a failing of systematic
review the fact that it is unlikely that the agencies will catch up with the
20-year review date by 1988, as mandated in the Order. While this may be true,
it is not logical to suggest that systematic review is both the cause and
effect of a failure to meet the time requirements. Rather, the cause is a
combination of three factors for which far less drastic remedies might be
appropriate: the lack of sufficient resources; the inefficient use of existing
resources; and, an overoptimistic projection of the time requirement.
Similarly illogical are those portions of the report which call for
the abolition of the program in response to problems of records appraisal,
records scheduling, duplications of review efforts, and personnel turnover.
In each case, there appear to us to be reasonable solutions or mollifiers
which the draft report ignores in favor of destroying the program.
CONCLUSION
In signing Executive Order 12065, President Carter stated:
While some material must be classified, the
government classifies'too much information,
classifies it too highly, and for too long.
These practices violate the public's right
to know, impose unnecessary costs, and
weaken protection for truly sensitive infor-
mation.by undermining. respect for all
classification.
Two years later, these words are just as valid as they were then.
Yet, the proposed draft CAO report calls for the abolition of one of the primary
programs designed to minimize the abuses of the classification system. Most
significantly, it calls for the abolition of a system that only two years ago
received the endorsement of the President, the Congress and the public.
Approved For Release 2002/04/01 : CIA-RDP93BOl194RO01100010001-8
50
 Approved For Release 2002/04/01 : CIA-RDP93B0l194R001100010001-8
APPENDIX II APPENDIX It
The Administration continues to endorse systematic review. We
suspect we are not alone in this endorsement. At the same time, however, we
are aware that much needs to be done to make the program work more economically
and more productively. The GAO inquiry into systematic review can be a posi-
tive step in this direction. The abolition of systematic review is not, however,
a positive step -- it is a step backward that jeopardizes the credibility of
the security classification system.
(941187)
Approved For Release 2002/04/01 : 6dh-RDP93B01194R001100010001-8
 Approved For Release 2002/04/01 : CIA-RDP93B01194RO01100010001-8
AN EQUAL OPPORTUNITY EMPLOYER
UNITED STATES
GENERAL ACCOUNTING OFFICE
WASHINGTON, D.C. 20548
OFFICIAL BUSINESS
PENALTY FOR PRIVATE U3E,$30O
PO$:Act AND Fits PAID
U. S. GXNtRAL ACCOUNTING OFFICE
THIRD CLASS
Approved For Release 2002/04/01 : CIA-RDP93BOl 194RO01 100010001 -8