HOUSE OF REPRESENTATIVES REPORT NO. 94-1746
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP83T00573R000600200010-6
Release Decision:
RIPPUB
Original Classification:
K
Document Page Count:
24
Document Creation Date:
December 12, 2016
Document Release Date:
April 24, 2002
Sequence Number:
10
Case Number:
Publication Date:
November 16, 1976
Content Type:
MF
File:
Attachment | Size |
---|---|
CIA-RDP83T00573R000600200010-6.pdf | 1.29 MB |
Body:
Osw OP C%.
Approved For Rrfease 8005/' :
r^ns 00 00
1976 November 16
UNITED STATES-DEPARTMENT OF COMMERCE
National Bureau of Standards
Washington, D.C. 20234
Jf~
MEMORANDUM FOR FIPS Points of Contact
FIPSCAC Membership
State Information Systems Coordinators
From: Harry S. White, Jr. ,~ ~v`~?~
Associate Director for ADP Sta dards
Subject: House of Representatives Report No. 94-1746
Attached for your information and use is House Report No. 94-1746,
entitled "Administration of Public Law 89-306, Procurement of ADP
Resources by the Federal Government". Currently, each of the
organizations cited in the Recommendations Section (pages 15 and 16)
are preparing responses to Congress on these items.
Attachment
NBS Approved For Release 2002/05/17: CIA-RDP83T00573R000600200010-6 `J
1901-1976
b1T16 _~g~6^
Approved For Release 2002y05/17 : CIA-RDP83T00573R0006002004~0-6
Union Calendar No, 870
94th Congress, 2d Session - - - - - - - House Report No. 94-1746
ADMINISTRATION OF PUBLIC LAW 89-306,
PROCUREMENT OF ADP RESOURCES
BY THE FEDERAL GOVERNMENT
THIRTY-EIGHTH REPORT
BY THE
COMMITTEE ON GOVERNMENT
OPERATIONS
together with
ADDITIONAL VIEWS
OCTOBER 1, 1976.--Committed to the Committee of the Whole House on the
State of the Union and ordered to be printed
U.S. GOVERNMENT PRINTING OFFICE
WASHINGTON : 1976
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17 : CIA-FtDP83T00573R000600200010-6
COMMITTEE ON GOVERNMENT OPERATIONS
JACK BROOKS, Texas, Chairman
L. H. FOUNTAIN, North Carolina
JOHN E. MOSS, California
DANTE B. FASCELL, Florida
WILLIAM S. MOORHEAD, Pennsylvania
WM. J. RANDALL, Missouri
BENJAMIN S. ROSENTHAL, New York
JIM WRIGHT, Texas
FERNAND J. ST GERMAIN, Rhode Island
FLOYD V. HICKS, Washington
DON FUQUA, Florida
JOHN CONYERS, JR., Michigan
BELLA S..ABZUG, New York
JAMES V. STANTON, Ohio
LEO J. RYAN, California
CARDISS COLLINS, Illinois
JOHN L. BURTON, California
RICHARDSON PREYER, North Carolina
MICHAEL HARRINGTON, Massachusetts
ROBERT F. DRINAN, Massachusetts
EDWARD MEZVINSKY, Iowa
BARBARA JORDAN, Texas
GLENN ENGLISH, Oklahoma
ELLIOTT H. LEVITAS, Georgia
DAVID W. EVANS, Indiana
ANTHONY MOFFETT, Connecticut
ANDREW MAGUI:RE, New Jersey
LES ASPIN, Wisconsin
FRANK HORTON, New York
JOHN N. ?ERLENBORN, Illinois
JOHN W. WYDLER, New York
CLARENCE J. BROWN, Ohio
GILBERT GLIDE, Maryland
PAUL N. McCLOSKEY, JR., California
SAM STEIGER, Arizona
GARRY BROWN, Michigan
CHARLES THONE, Nebraska
ALAN STEELMAN, Texas
JOEL PRITCHARD, Washington
EDWIN B. FORSYTHE, New Jersey
ROBERT W. KASTEN, JR., Wisconsin
WILLIS D. GRADISON, JR., Ohio
WILLIAM M. JONES, General Counsel
JOHN E. MooRE, Staff Administrator
WILLIAM H. COPENHAVER, Associate Counsel
LYNNE HIGGINBOTHAM, Clerk
J. P. CARLSON, Minority Counsel
LEGISLATION AND NATIONAL SECURITY SUBCOMMITTEE
JACK BROOKS, Texas, Chairman
JOHN E. MOSS, California FRANK HORTON, New York
BENJAMIN S. ROSENTHAL, New York JOHN N. ERLENBORN, Illinois
JIM WRIGHT, Texas JOEL PRITCHARD, Washington
DON FUQUA, Florida
WILLIAM: S. MOORHEAD, Pennsylvania
JAMES V. STANTON, Ohio
MICHAEL HARRINGTON, Massachusetts
ELMER W. HENDERSON, Staff Director
THOMAS J. PAYNE, Professional Staff Member
RICHARD C. BARNES, Professional Staff Member
DAVID E. HOLT, Professional Staff Member
GUADALUPE R. FLORES, Professional Staff Member
CRAIG J. GEHRING, Professional Staff Member
DOROTHY A. MCGEE, Clerk
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 20OZ05/17 : CIA-RDP83T00573R00060020Q 0-6
Hon. CARL ALBERT,
HOUSE OF REPRESENTATIVES,
Washington, D.C., October 1',1976.
Speaker of the (louse of Representatives,
Washington, D.C.
DEAR MR. SPEAKER : By direction of the Committee on Government
Operations, I submit herewith the committee's thirty-eighth report
to the 04th Congress. The committee's report is based on a study made
by its Legislation and National Security Subcommittee.
JACK BROOKS, Chairman.
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 200205/17 : CIA-RDP83TOO573R00060020QWO-6
CONTENTS
Page
1. Introduction --------------------------------------------------- 1
II. Hearings ------------------------------------------------------ 3
III. Surveys of ADP procurements ---------------------------------- 4
IV. Basic causes of noncompetitive procurements--------------------- 5
Lack of adequate justification------------------------------- 5
Lack of long-range planning--------------------------------- 6
Lack of standards------------------------------------------ 7
Lack of high level languages-------------------------------- 8
Lack of utilization reviews---------------------------------- 9
Lack of use of functional specifications---------------------- 10
V. Interim upgrades----------------------------------------------- 11
VI. Procurement of smaller dollar value items------------------------ 12
VII. ADP fund------------------------------------------------------ 13
VIII. Conclusion ---------------------------------------------------- 14
IX. Recommendations --------------------------------------------- 15
Additional views of Hon. Sam Steiger--------------------------------- 17
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 20OZ05/17 :.CIA- P83T 057 00060020QWO-6
union ales ar o. 870
94TH CONGRESS
HOUSE OF REPRESENTATIVES
REronr
1
2d Session
No. 94-1746
ADMINISTRATION OF PUBLIC LAW 89-306, PROCURE-
MENT OF ADP RESOURCES BY THE FEDERAL GOV-
ERNMENT
OCTOBER 1, 1976.-Committed to the Committee on the Whole House on the
State of the Union and ordered to be printed
Mr. BRooKs, from the Committee on Government Operations,
submitted the following
THIRTY-EIGHTH REPORT
together with
ADDITIONAL VIEWS
BASED ON A STUDY BY THE LEGISLATION AND NATIONAL SECURITY
SUBCOMMITTEE
On September 28, 1976, the Committee on Government Operations
approved and adopted a report entitled "Administration of Public
Law 89-306, Procurement of ADP Resources by the Federal Govern-
ment." The chairman was directed to transmit a copy to the Speaker
of the House.
The annual Automated Data Processing (ADP) costs in the Fed-
eral Government are estimated to range between $4 and $10 billion.
These costs include the procurement of ADP hardware, maintenance,
and services, as well as personnel and other operational costs. These
estimates, however, do not generally include Federally financed ADP
expenditures by Federal grantees and contractors. Also, they may well
not include all Federal expenditures on "special purpose" ADP
equipment which is generally designed as an integral part of non-
ADP operations. If the costs of these additional functions were in-
cluded, Federal annual ADP costs could easily exceed $15 billion. This
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17 : EIA-A83T00573R000600200010-6
means that current Federal ADP expenditures may amount to approxi-
mately 4 percent of the entire Federal budget.
The magnitude of these costs compels the Congress to reevaluate the
manner in which we perceive and manage Federal ADP operations in
the Federal Government, and, more importantly, how these operations
impact upon the budgets and management of executive agencies.
Traditionally, computers have been embraced by government man-
agers as a revolutionary tool for improving efficiency and conserving
resources. As a result, these manager's expectations have all too often
exceeded the reality of the computer's capability to solve their prob-
lems. The level of current Federal expenditures on ADP resources must
compel us to reexamine the true impact of computers on government
operations. Since we are now devoting such a large share of our Fed-
eral budget to ADP resources, we need to scrutinize more carefully
whether these expenditures are justified from a cost/benefit stand-
point. Equally important, however, is the question whether acquisition
of ADP resources is necessary to carry on an essential program or
whether such acquisition will foster non-essential activity simply be-
cause the computer is available.
Both of these issues can have such an impact on government agencies'
operations that they require a more careful scrutiny of ADP procure-
ment and operations than is now being conducted within the Govern-
ment. Automatic data processing equipment can no longer be viewed
as a mere tool but must now be treated in the same way another major
programs and, as such, receive the attention of top management.
The primary means for providing this direction by top management
lies in, the effective administration of Public Law 89-306 (The Brooks
Act). This Act, while directed primarily toward the procurement of
ADP resources, in effect, compels agencies to examine their needs and
uses of ADP if they are to achieve the economies and efficiencies in-
tended by the Act:
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002&5/17 : CIA-RDP83T00573R000600200QkO-6
II. HEARINGS
The Government Operations Committee, through its Subcommittee
on Legislation and National Security, conducted three days of hear-
ings on June 28 and 29; and July 1, 1976, in which procurement and
utilization under the Brooks Act* were reviewed. Preparatory to these
hearings, the Subcommittee devoted several months to reviewing the
administration of this Act by the Office of Management and Budget
(OMB), the General Services Administration (GSA), and the Na-
tional Bureau of Standards (NBS). Considerable information was ob-
tained from these agencies regarding their role in administering the
Act, as well as information relating to Federal user agencies' ADP pro-
curement and use. In addition, data was obtained directly from the
user agencies on their ADP inventory and its acquisition.
At the hearings, testimony by representatives of OMB, GSA, and
NBS provided the Subcommittee with additional information con-
cerning their activities under the Act. The General Accounting Office
(GAO), in turn, presented a ten-year summary covering the imple-
mentation of the Act. Finally, the Computer and Business Equipment
Manufacturers' Association and the Computer Industry Association
testified as to the ADP industry's position regarding the implementa-
tion of the Act.
A review of this information and testimony has disclosed that the
Brooks Act has been neither administered nor implemented in accord-
ance with the intentions of Congress:
GSA has repeatedly authorized noncompetitive procurements
which were not adequately justified.
GSA has failed to enforce regulations and restrictions in ADP
procurement authority delegated user agencies.
GSA has not provided adequate managem- ent'guidance to user
agencies:
OMB has failed to establish concise, clearcut policy.
OMB has not provided adequate direction in the enforcement
of those policies it has established.
NBS has failed to provide necessary hardware and software
standards.
Federal user agencies have consistently failed to cooperate with
GSA.
Federal user agencies have shown a general reluctance to ad-
here to the purpose and intent of the Brooks Act.
Under the Brooks Act, GSA has the authority for procuring ADP
resources required by Federal user agencies. GSA may either procure
those ADP requirements which a user agency specifies or delegate
to an agency authority to procure under such restrictions and condi-
tions as GSA specifies. The Act does not permit GSA to impair or
interfere with an agency's determination of its requirement. This can-
not be interpreted, however, as preventing GSA from determining
the best means for an agency to fulfill its ADP requirements. A dis-
pute between GSA and a user agency is to be resolved by OMB.
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
approved For Release 2002/05/17 : CIA-83T00573R000600200010-6
III. SunVEYs OF ADP PROCUREMENTS
In a survey conducted by the Subcommittee of delegations of ADP
procurement authority granted by GSA, it was found that in Fis-
cal Year 1975 only 36 percent of the systems required were procured
in a fully competitive manner.' The balance were noncompetitive-
sole source, make and model, brand name or equal. This low percentage
of competitive procurements is clearly inconsistent with the goals and
objectives of the Brooks Act and reflects a decline in the level of com-
petition from historic percentages developed by the Subcommittee of
user agency procurements over an eight year period. This latter sur-
vey showed that an average of 60 percent of all systems procure-
ments were competitive during that eight year period. What is par-
ticularly disturbing about this sharp decline in the level of competition
is the fact that, as GSA admitted to the Subcommittee, noncompeti-
tive procurements are more costly to the Federal Government than
competitive procurements. According to the GSA, since, enactment of
Public Law 89-306, over $681 million. in cost avoidance has been
achieved in 302 competitive ADP contracts.
It was pointed out at the hearing that. the Brooks Act has stimulated
a greater participation in the Government market by ADP manufac-
turers than in the commercial market. Whereas, 70 percent of the com-
mercial computer market is dominated by one firm, that firm controls
only approximately 30 percent of the Government market. While the
Brooks. Act can. justifiably take credit for this, these statistics do not
represent true competition in the sense that they do not reflect whether
competition occurred in any specific procurement.
1 This percentage does not include the large number of noncompetitive delegations
grantee! by GSA under Temporary Regulation, E-32, relating to renewals of leased equip
ment rented under the ADP Schedule. If these procurements had been included, the per-
centage of competitive procurements would have been even lower.
(4)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 20024+05/17 : CIA-RDP83T00573R000600200Qy0-6
IV. BASIC CAUSES OF NONCOMPETITIVE PROCUREMENTS
The basic causes for noncompetitive procurements are lack of (1)
adequate justifications for ADP acquisition, (2) long-range planning,
(8) standards, (4) high level languages, (5) utilization reviews, and
(6) use of functional specifications.
GAO has documented several cases in which GSA granted procure-
ment delegations which had not been fully justified by user agencies.
Among these were procurement requests by the Department of Agri-
culture, Veterans' Administration, and the' Social Security Adminis-
tration. Several reasons appear to exist for GSA's actions. Foremost,
in the Committee's view, is GSA's lack of resources to examine in
depth an agency's justification for a noncompetitive procurement. To
conduct an adequate in-depth review of a justification for a major
ADP system would require the expenditure of hundreds of man-hours
by technically qualified personnel. Generally, GSA lacks sufficient
manpower to devote that amount of time to such reviews. The
Federal Computer Performance Evaluation and Simulation Center
(FEDSIM) does possess the capability to perform such services but
its resources are so limited that it is unable to provide a significant
degree of timely assistance in this area.
In addition, GSA, on occasion, appears to grant a user. agency a
delegation of procurement authority because of the latter's prestige or
dominance in the bureaucratic structure of government. In essence,
this prestige or dominance permits an agency to exercise influence
over another agency without taking any apparent overt action. This
influence stems from the mission of the agency and the support it can
garner from the higher levels of government. GSA, unfortunately,
ranks below many other agencies in the world of bureaucracy. As a
consequence, it is unable, standing alone, to resist pressures from a
more dominant source if such source signals GSA that it intends to
exert pressure concerning a particular request.
Even worse, there have been occasions where user agencies have re-
fused to cooperate with GSA when the latter sought fuller justifica-
tion for noncompetitive procurement requests. In recent months,
Departments of Interior, Commerce, Transportation, and HEW have,
in one way or another, challenged GSA's authority to obtain addi-
tional documentation for procurement requests. These challenges have
been based on the requirements section of the Brooks Act which pro-
hibits GSA from questioning an agency's requirements. The validity
of these challenges are clearly without legal support under the law.
If an agency disagrees with GSA regarding justification for a partic-
ular procurement, the Act provides that the dispute be brought to
OMB for resolution. In the past neither GSA nor the agencies have
exercised this right to appeal to any extent.
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17 :OCIA-RDP83T00573R000600200010-6
The Cornmitt, e is very disturbed over an apparently increasing
trend of user agencies to interpret the law to suit their own purposes.
Whether intentional or not, this type of behavior will have the effect
of subverting the effectiveness of Public Law 89-306. If GSA were to
be denied the right to. require full documentation for ADP procure-
ments, the-objectives of fully competitive procurements under the Act
would be effectively destroyed. Federal agencies strongly resisted en-
actinenjt of the Prooks Act. Since passage, they have shown little will-
ingness to comply voluntarily with the law. The low level of fully
competitive procurements further illustrates their non-cooperative
state of mind. Either user agencies must learn to comply with the full
thrust of the Act or OMB must take all necessary action to make
them do so. In particular, OMB must make it clear to every user
agency that its right to determine its own ADP requirements under
the Act does not include the right to dictate a specific brand name
.of equipment as its requirements or refuse to supply GSA with appro-
priate feasibility studies. In addition, GSA must be given the re-
sources to perform adequate reviews of user agencies' procurement
justifications.
GSA, in turn, must take a more aggressive position when dealing
with user agencies. Being fully cognizant of the bureaucratic facts of
life, however, it has to be realized that GSA will not be able to assume
this new role unless OMB fully supports it.
The manner in which the Brooks Act is being administered is unac-
ceptable. If this continues, the Act's effectiveness will become seriously
jeopardized at a cost of millions of dollars to the taxpayers.
The second major cause for noncompetitive procurements is the fail-
ure to prepare effective long-range plans. These plans must be based
upon a user agency's projected missions and programs for a period of
five or , more years, and not merely a guesstimate of future ADP needs
divorced from the agency's missions and programs. Once these plans
have been prepared, an agency should develop the ADP requirements
necessary to support its mission and program needs. At this stage, the
ADP resources should be procured on a fully competitive basis through
the use of functional specifications. It is noted that several witnesses
at the hearings supported functional specifications for major systems
procurements. GSA is in the process of issuing regulations which,
while still containing many shortcomings, encourage the use of func-
tional. specifications. OMB, if it effectively enforces its action, has gone
a step further by issuing Circular A-109, which requires all major sys-
tems, including ADP systems, to be procured by means of functional
specifications.
Assuming that an agency has planned effectively, the ADP resources
to be acquired should fully support the agency's missions and programs
for the duration of the plan and should obviate the necessity for in-
terim upgrades, add-ons, and replacement systems. The only apparent
exception should be in those instances where new or increased respon-
sibilities have arisen which an agency could not have reasonably pro-
di cted at the tinie the original system. was procured. The establishment
of ADP requirements at the time a plan is developed is not meant to
imply, however, that all hardware and software must be procured at
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 200 (85/17 CIA-RDP83TOO573R000600200 O-6
the beginning of the plan. Instead of acquiring unneeded capacity in
the short run, contracts can be awarded which call for phased installa-
tion of equipment over a period of years.
In effectively implementing this, planning system, OMB must assume
primary responsibility for its success. Not only must it issue clear and
concise policy guidance, but it must also insure that such policy is ad-
hered to by user agencies. The success of this planning concept is
equally dependent upon Congress' approving funds for ADP procure-
ments which it supports and which are consistent with agency plans.
Further, in order for this planning and procurement role to be effec-
tive, a need exists to develop line item appropriation authority for
ADP.
At the present time, it is difficult for Congress and OMB to monitor
effectively agency expenditures or requests for expenditures for ADP
because they are buried in other funding requests. There are those who
argue that ADP is merely a tool designated to assist agencies in ful-
filling mission needs. As a consequence, it is maintained that singling
out ADP for separate consideration would distort the budgetary proc-
ess. In actuality, however, current and prospective usage of ADP is so
interrelated with agency policies, planning and programs that fund-
ings of ADP requirements can become as important as funding of pro-
grams themselves. In addition, ADP expenditures represent such a
significant percentage of the overall budget that effective management
of budgetary priorities requires that such expenditures be singled out
for separate consideration.
Only if this planning concept is faithfully pursued will the spirit
and letter of the Brooks Act be fully realized. The Act requires the
economic and efficient procurement of ADP resources. This can
only be realized if, in the long run, procurements are conducted in a
fully competitive manner.
It is recognized, however, that under the most favorable circum-
stance it will take time to implement effectively botha workable plan-
ning process and the functional specification concept. Conceivably, its
objectives may never be fully realized. Pending successful implemen-
tation, many steps can be initiated immediately which will contribute
significantly to the achievement of fully competitive ADP systems
procurements.
LACK Or STANDARDS
First and foremost is the development of meaningful hardware and
software standards. Under the Brooks Act and Executive Order 11717,
NBS is charged with the responsibility for developing ADP stand-
ards. At the hearings, GAO expressed serious concern about the lack
of progress being made by NBS in the development of standards. This
concern of GAO was prompted, in large part, by its-recognition that
standards are essential to the achievement of full competition and to
the saving of large sums of money by the Government. To date, NBS
has only developed to a limited extent standards necessary to fully
implement the Act., even though it acknowledged at the hearings that
lack of standards seriously impedes effective competition. As the Com-
puter Industry Association testified at the hearings, NBS has devel-
oped no meaningful hardware standards and only a relatively few
software standards for ADPE. This sentiment- was echoed by GSA.
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17: CIA-RDP83T00573R000600200010-6
Perhaps the most dramatic example of this failure is NBS' ten-year,
effort without success to develop Input/Output Interface standards.
These standards, which would enable peripheral devices of different
manufacturers to be connected to the same central processing unit,
would, as GSA has informed the Committee, greatly enhance the ef-
fectiveuess in achieving competition in both systems and peripheral
procurements. Yet, NBS has not aggressively pursued the develop-
ment of these standards because it apparently has been committed to
the adoption of voluntary standards developed under American Na-
tional Standards Institute (ANSI) procedures. This voluntary proc-
ess under ANSI is generally sound so long as it does not unduly impede
the development of worthwhile standards. When, however, as in the
case of the Input/Output Interface standards, conflicting interests
serve to preclude the timely adoption of a standard, NBS has the obli-
gation to develop Federal Standards. In response to a Subcommittee
question, NBS stated that development of Input/Output standards is
essential in order to provide more effective competition. Furthermore,
at the hearings NBS testified that additional delay in the adoption of
voluntary Input/Output standards will compel it unilaterally to adopt
Federal standards.
It has been alleged that the development of ADP standards will
preclude the Government from taking advantage of technological ad-
vances and perhaps even stifle the development of such advances. If,
in fact, there is any validity to thisallegation, the pitfalls suggested
therein can be avoided by NBS maintaining constant vigilence to
assure that standards are modified in accordance with advances in
technology.
Even in those case in which standards have been adopted, their
implementation has not been effectively enforced. NBS has informed
the Subcommittee that it has no data on agency compliance with stand-
ards and, furthermore, maintains that enforcement of standards pres-
ently resides in each individual agency. GSA has incorporated most
Federal standards into its procurement regulations, but it has made
little or no effort to see that they are complied with. OMB, in turn,
last exercised its policy guidance role in this area in 1966 when it
issued Policy Guidelines to NBS. Since that time, OMB has neither
updated these guidelines nor taken action to see that they are fol-
lowed. This means that even if standards are developed in the future,
such development will be of little value unless an effective means of
enforcement is developed. While NBS has indicated that centralizing
enforcement responsibility will improve compliance, it maintains that
it is not an appropriate agency to perform this function. In conse-
quence, OMB must establish procedures for the effective enforcement
of ADP standards, and designate GSA as the agency responsible for
enforcing compliance with such standards.
LACK OF HIGH LEVEL LANGUAGES
Another fundamental course of action which must be adopted is the
requirement that Federal user agencies write their ADP programs in
higher level languages. Writing programs in higher level languages
contributes to effective hardware competition by eliminating most of
the costs associated with converting software from one brand of equip-
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 200?105/17j CIA-RDP83TOO573R00060020QQ40-6
ment to another brand. Increasingly, software conversion costs. are
dictating that specific ADPE be procured, thereby preventing full
competition. The survey conducted by the Subcommittee of delega-
tions of procurement authority for FY 1975 revealed that a large num-
ber of noncompetitive procurement requests were justified on the
grounds of avoiding excess conversion costs. This clearly suggests
that a sizable share of existing programs have been written in lower
level languages and that there is little evidence of effective central
direction in the Government encouraging conversion to higher level
languages.
The Government's primary effort to meet this problem has been to
concentrate on the development of higher level languages. Even here;
however, success has been limited. Cobol was adopted as a federal
standard in 1968; yet, the continuing large number of noncompetitive
procurement requests justified on the basis of avoiding conversion costs
suggests that the standard has not been widely complied with. More-
over, even if user agencies have written programs in Cobol, it is doubt-
ful how successful this will be in reducing the conversion problem. As
noted earlier, there is no centralized enforcement of standards by NBS
or any other agency. This means that a user agency may adopt Cobol
but employ unique features which will impede conversion. Further-
more, even though NBS has been developing a Fortran standard for
several years, none has yet been adopted. Only when standard high
level languages are developed and their use enforced will a barrier
to effective competition be eliminated.
Standardization of Cobol and Fortran alone will not be enough. As
GSA informed the Committee, the establishment of large data base
applications, which include considerable expense for software, storage,
and training has created a new problem. There is no standard data
base management system or approach. Each manufacturer and service
company has its own system. Without a standard for data base man-
agement systems, conversion problems will continue. In the absence
of standards for high level languages and data base management sys-
tems, the policy should be established that, when a user agency acquires
a new ADP system, it will generally be required to absorb software
conversion costs and not include those costs in determining what
equipment will be acquired.
Another factor contributing to the high level of noncompetitive
ADP procurements is the failure of user agencies to effectively and
efficiently utilize their ADP resources. The economic benefits of com-
petitive procurements of ADP resources under the Brooks Act are
dependent upon the efficient utilization of such resources. By means
of such utilization, noncompetitive interim upgrades, add-ons and
replacements can be minimized. And, even in those cases where in-
creased capacity for an interim period is required in spite of efficient
utilization, such can frequently be obtained through the competitive
procurement of peripherals, minicomputers or service contracts.
At the hearings, the GAO cited many examples of agencies' failures
to properly utilize their equipment. Perhaps the most flagrant ex-
ample was that of the Social Security Administration which was
operating equipment at 50 percent of capacity and running second
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
AM.
Approved For Release 2002/05/17: CIA-RDP83T00573R000600200010-6
10
eneration programs on third generation equipment: In addition, the
GAO also described procurements by the U.S. Department of Agri-
culture and the Treasury Department which were not supported ade=
quately by utilization reviews. In each of these, costly equipment was
acquired through noncompetitive procurements.
The GAO reported that where inadequate utilization was un-
covered, as in the above examples, such inadequacy was found ? to be
caused by one or more of the following: improper design, inefficient
application, or operational deficiencies. In each case, of course, the
user agency must be held primarily accountable for these inefficiencies.
However, OMB, GSA, and NBS must share a portion of the responsi-
bility for this state of affairs : OMB for failing to provide policy
guidance and overall leadership, GSA for failing to challenge ques-
tionable justifications, and NBS for failing to develop criteria for
measuring systems performances. Yet, even taking into account the
deficiencies of these three agencies, prudent management would
dictate that user agencies themselves should strive to optimize their
ADP utilization in order to achieve a high level of economies and
efficiencies in their programs. Their failure to do so is another example
of their lack of interest in supporting the objectives of the Brooks Act
or in making the best use of the taxpayers' dollars.
LACai OF USE OF FUNCTIONAL SPECIFICATIONS
Finally, another step that can be taken to increase competitive sys-
tems procurements in the short run is the use of functional specifica-
tions. At present, virtually all ADP procurements are made on the
basis of restrictive specifications. Earlier, functional specifications
were discussed in terms of their use in conjunction with long-range
systems plans. However, even in the absence of long-range plans,
agencies should be required to use functional specifications when
acquiring a new system. Clearly, the promulgation of OMB Circular
A-109 and GSA's Pending Federal Procurement Regulation both call
for the use of functional specifications in ADP systems proceurements
and are, therefore, consistent with this proposal.
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 20GM5/17 : CIA-RDP83T00573R000600200 0-6
V. INTERIM UPGRADES
The adverse impact of the above six causes (lack of adequate justi-
fication, long-range planning, standards, high-level languages, utili-
zation reviews, functional specifications are most readily seen in the
case of the significant number of noncompetitive interim upgrades,
add-ons, and replacement systems that are authorized each year. Until
the above causes for noncompetitive procurements are eliminated
user agencies will undoubtedly continue to request large numbers of
delegations for such noncompetitive procurements. Before GSA grants
such a delegation, however, it should require the requesting agency
to document that its existing system is only operating essential pro-
grams and that all steps have been taken to optimize the efficiency of
the programs and systems. In addition, GSA should not grant a dele-
gation for a noncompetitive procurement unless the user agency docu-
ments that it cannot meet its needs through the competitive procure-
ment of minicomputers or service contracts.
At the present, GSA is in the process of establishing a new Tele-
processing Services Program (TSP) which is intended to permit
user agencies to contract competitively for ADP services. Under this
program, GSA will contract with a large number of ADP service firms
throughout the country. Once operational, it will be mandatory upon
agencies to acquire ADP services under this program unless GSA
grants a waiver. If properly implemented, the TSP program could
reduce significantly the number of noncompetitive interim upgrades,
add-ons, and replacement systems. However, the Committee is con-
cerned that these hoped for benefits will not be fully realized because
of GSA's failure to date to effectively manage the program. Among
these failures have been a lengthy delay in issuing the RFP, the
development of inadequate procedures governing its use, and an undue
delay in awarding contracts. Unless GSA immediately moves to cor-
rect these failures, the success of the TSP program will be jeopardized.
In those cases where GSA determines that a noncompetitive interim
upgrade, add-on, or replacement system is justified, it should grant a
delegation only on the condition that the user agency has given a firm
commitment that it will go competitive within a time period not ex-
ceeding two years. Needless to say, GSA must strictly enforce this
time limit, a role it has not assumed to date.
(11)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17 : CIA-RD'83T00573R000600200010-6
VI. PROCUREMENT OF SMALLER DOLLAR VALUE ITEMS
The discussion to date has been directed primarily to the procure-
ment of major ADP systems. Implicit in such discussion has been the
realization that procurement of major systems by means of functional
specifications can be more time consuming and costly than a noncom-
petitive procurement or an equipment specification procurement. How-
ever, the benefits to be realized through such fully competitive procure-
ments will in the long run far outweigh short-term disadvantages.
By way of contrast, immediate cost benefits and time savings can
be realized through the institution of simplified procedures for pro-
curerr ent of smaller dollar value items.
The Subcommittee survey of fiscal year 1975 procurement delega-
tions revealed that 56 percent of all delegations were for procurements
valued at $250,000 or less. Most items procured in this category were
minicomputers, peripherals, software, and maintenance where highly
competitive markets exist. The relatively low-dollar volume involved
in such procurements, as well as the existence of competition, has meant
that GSA has had to devote a disproportionate amount of its limited
resources to small dollar item procurements which are least vulnerable
to noncorapetition. By the same token, the resources of user agencies
have been unfairly taxed by their being required to follow the same
procedures for small dollar items as they must for major procurements.
To conserve the resources of both GSA and user agencies, new proce-
dures should be established whereby user agencies could procure ADP
items below $250,000 without the need to obtain a delegation of author-
ity from GSA. These procedures should only be implemented, how-
ever, if certain safeguards are established. Full competition must be
preserved and clearly documented in user agency files. CPU's and
systems may not be procured, except for minicomputers. Most impor-
tantly, GSA must maintain active oversight over such procurements,
including periodic audits, to insure compliance with the terms of the
program. Serious breaches of these procedures by a user agency must
lead to GSA's suspension of such agency's further authority to use
these procedures.
(12)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 200112A95/17 : CIA-RDP83T00573R000600200Q, 0-6
VII. ADP FUND
The emphasis of these hearings has been upon the procurement of
ADP resources under the Brooks Act. It is only fitting, then, that
reference be made to the operation of the ADP Fund which was cre-
ated by the Act and was to be administered by GSA.
When the Fund was initially established, it was contemplated that
its role in ADP procurement would be more prominent than it is
today. In a report issued in October, 1975, the GAO indicated that a
primary cause for the restricted use of the Fund has been the limited
financial resources provided for it. This would appear to be short-
sighted since many worthwhile programs could be administered effec-
tively through the Fund. with potential economic savings to the
Government.
Among such programs would be the Opportunity Buy Program
and Requirements Contracts. The former program enables GSA to
purchase equipment at a significant economic saving for utilization by
a user agency when the latter is financially unable to make the pur-
chase itself. In the case of requirements contracts, the Fund could
guarantee the acquisition of minimum quantities of ADPE at large
potential cost savings.
The Fund can also be used effectively to support the TSP Program,
discussed earlier, Data Processing Centers, and Multiyear Leasing.
With regard to the latter, many are advocating enactment of Multi-
year Leasing legislation which would permit the leasing of equip-
ment for several years although only obligating payment on a year-by-
year basis. GSA already has legal authority. to enter into multiyear
leases. Therefore, if the Fund were fully capitalized, GSA would have
the financial capability to take advantage of multiyear leasing without
the need to enact such legislation.
Finally, it is anticipated that the Fund could be used as a central
procurer for the acquisition of software products for Government-
wide use. Through such purchase significant savings can be achieved.
There are those who advocate that the Fund be the sole procurer of
all Government ADP resources, including large systems. Given GSA's
limited resources and the knowledge and expertise residing in user
agencies, relating to ADP, the Committee does not believe that it.
would be beneficial at this time to centralize all procurements in the
Fund. Particularly, in the case of large systems procurements, it would
seem far more reasonable to have the user agencies procure their needs
under delegation from GSA rather than for GSA to shoulder the entire
burden for such procurements through the Fund. In turn, this will
free GSA to devote a greater share of its limited resources to increas-
ing its oversight of major procurements to the end that they be fully
justified and competitive.
(13)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
VIII. CONCLUSION
Over ten ,years ago, the Committee, -anticipating the key role that
computers would play in the operations of Government, recommended
and Congress enacted Public Law 89-306 (The: Brooks Act) which
established a framework for the economic and efficient procurement
and utilization of ADP resources. The growth in the expenditures of
and use made of computers have far exceeded the levels anticipated
by the Committee.
The magnitude of this growth prompted the Committee to conduct
an investigation and to hold hearings into the administration of the
Act. The Committee found that, overall, the Act while providing sig-
nificant benefits has neither been satisfactorily administered nor effec-
tively implemented to its fullest potential. The Act has -opened the
Government market more widely to the ADP industry than to the
commercial market. Millions of dollars have been saved in ADP pro-
curements. However, the low level of fully competitive ADP procure-
ments coupled with the apparent lack of effective utilization of ADP
resources will, i f allowed to continue, ultimately result in seriously
jeopardizing the, effectiveness of the Brooks Act at a cost of millions
of dollars annually to the taxpayers.
GSA, OMB, NBS, and the user agencies must all share responsi-
bility for this situation : GSA for failing to administer the Act effec-
tively, OMB for failing to provide adequate policy direction, NBS for
failing to establish standards, and most significantly the user agencies
who, having the most to gain from the Act, has resorted to every possi-
ble means to avoid full compliance with it. If the full benefits of the
Act are to be achieved, GSA, OMB, NBS, and the user agencies must
join together in. a commitment to fully support and adhere to the
provisions of the Act.
(14)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2001 6'5/17 :. CIA-RDP83TOO573R000600200000-6
IX. RECOMMENDATIONS
1. While a user agency has the right under the law to determine that
it has a need for ADP requirements, OMB must issue clear policy
guidelines upholding GSA's authority to determine the array of ADP
resources which would best meet such requirements.
2. Disputes between GSA and user agencies over. their respective
rights should be referred to OMB for immediate resolution; as pro-
vided for in Public Law 89-306.
3. OMB must establish concise, clearcut policies providing direc-
tion in the administration and enforcement of Public Law 89-306 by
GSA, NBS, and, in particular, the user agencies who, to date, have
demonstrated a general reluctance to comply with the law.
4. GSA must provide concise and clearcut guidance in its adminis-
tration of Public Law 89-306 through the issuance and enforcement
of regulations and restrictions imposed in the delegation of procure-
ment authority.
5. Federal user agencies must be required to develop long-range
plans governing ADP needs and equipment expectations based upon
realistic utilization reviews.
6. ADP systems must be procured competitively in accordance with
functional specifications except in unusual instances when considera-
tion of national interest may require reliance upon equipment
specifications.
7. Interim upgrades, add-ons, and replacement systems should be
avoided except in those cases where new responsibilities have arisen
which a user agency could not have reasonably predicted at the time
the original system was procured.
8. Where GSA authorizes a non-competitive interim upgrade, add-
on, or replacement system, it should do so only upon the condition that
the user agency give: a firm commitment that it will go competitive
within a period not exceeding two years-such commitment being
strictly enforced by GSA.
9. Interim upgrades, add-ons, or replacement systems should only
be authorized in those cases where additional requirements cannot be
met through the procurement of minicomputers or service contracts.
10. NBS must develop necessary hardware and software standards
to insure maximum economies and efficiencies in the procurement and
utilization of ADP resources.
11. OMB must establish procedures for the effective enforcement of
ADP standards and designate GSA as the agency to enforce compli-
ance with such standards.
12. All ADP programs should be converted to higher level lan-
guages except in those cases where OMB specifically determines that
the national interest requires otherwise.
(15)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/171clA-RDP83T00573R000600200010-6
13. Software conversion costs should not be considered in the evalua-
tion of bids for the procurement of ADP systems except to the extent
that such costs involve direct out-of-pocket costs for program
conversion.
14. Research should be undertaken, pursuant to OMB's direction,
aimed at the development and use of management information sys-
tems which contribute to competitive ADP procurements and
utilization.
15. Federal user agencies should be authorized to procure ADP
resources, excluding CPU's, below $250,000 without specific delega-
tion of authority from GSA as long as the agencies document that the
procurements are fully competitive. Procurement of -CPU's should
always require a delegation from GSA.
16. GSA should be granted additional resources to enable it to more
effectively administer Public Law 89-306.
17. FEDSIM should be granted additional resources to enable it to
increase its level of assistance to GSA and user agencies.
18. The ADP Fund should be adequately funded to support the
TSP program, Data Processing Centers, Multiyear Leasing, Oppor-
tunity Buys, Requirements Contracts, and Government-wide software
products.
19. Congress and OMB should direct that each agency request ap-
propriations for all ADP expenditures (hardware, software, mainte-
nance operations, grantees, and contractors) under a single line-item.
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002&5/17: CIA-RDP83TOO573R000600200QO-6
ADDITIONAL VIEWS OF IION. SAM STEIGER
Our Committee, over the years, has played an important role by
insuring the efficient and economic procurement and utilization of
ADP,resources by the Federal Government. This report most cer-
tainly will give needed new direction in this area. Because of the im-
portance of this report, it is important that some additional views be
presented.
The report primarily focuses on the apparent statistical decline in
the number of competitive ADP procurements. While tlii.s is signif-
icant as a trend, the causes, not just the symptoms must be attacked.
A management system must be developed to capitalize on and har-
ness for greater productivity the basic strengths of the competitive
free enterprise system to bring the vast benefits of computer tech-
nology to bear in solving the many problems of government. If the
Subcommittee fails to broaden its focus it will provide the Execu-
tive Branch a directional bias which, if not clarified, could prevent
the government from realizing the maximum benefits from computer
technology and competition.
The specific sections for which additional views are offered are
Section IV, Basic Causes of Noncompetitive Procurements; and Sec-
tion IX, Recommendations.
SECTION IV, BASIC CAUSES Or NONCOMPETITIVE PROCUREMENTS
Section IV identifies the basic causes for noncompetitive procure-
ments as the lack of (1) adequate justification for ADP acquisition,
3) standards, (4) high-level languages,
(2) long-range planning, (33,
(5) utilization reviews, and (6) use of functional specifications. In
addition to these causes identified in the report, there are several addi-
tional factors which would bring us even closer to the basic problems
that exist in the procurement and use of computers in the government
today. These to some extent have been previously highlighted by the
Blue Ribbon Panel on Defense Procurement and the Commission on
Government Procurement and are as follows :
The excessive cost, time, and complexity of today's au-
thorization and acquisition process for computers are well
documented and understood throughout both the Executive
and Legislative Branch. We have seen that the current pro-
curement process can take from three to six years and can
incur procurement costs that are greater than the cost of the
equipment. Failure to acknowledge and address this adverse
factor impacts not only the goal of maximizing competition
but substantially reduces the effectiveness of the use of com-
puter technology in government. Major effort is required to
improve and streamline the procurement process. Failure to
(17)
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2002/05/17 1clA-RDP83T00573R000600200010-6
do so will continue to deny the government and the taxpayer
the full benefits of the competitive free enterprise system.
The second issue is the growing age of the government's
computers and its effect on governmental productivity and re-
spons~veness. Again we can look at the Defense Blue Ribbon
Panel'Report which concluded that personnel cost alone to
do a job on some older computers were greater than total costs
to do the job on newer, more efficient systems. A management
system is needed to assure that the productivity potential of
computers is realized by the government.
Third, the Report does not consider the main factor in
the apparent statistical reduction in the level of "fully com-
petitive procurements". Most automated government func-
tions today must accommodate two or three complete changes
in. computer technology during the automated system life.
During the past ten years, since the passage of the Brooks
Act, most operational government functions have been auto-
mated to some degree and represent a substantial economic
investment by the government. Each of these systems gen-
erally has a total life of somewhere between ten and twenty
years. Asa result, most computer equipment acquisitions to-
day are for the purpose of improving existing systems to
achieve increased productivity and responsiveness or accom-
modate additional workloads rather than the acquisition of
wholly new systems. Due to the rapid technological advance-
ment, the steady decline in the cost of computer technology
and the requirement to minimize the disruption of assimilat-
ing new generations of technology during the total systems
life; new procurement processes must be developed which will
take into account the need for system enhancement without
total redesign that will accommodate both the goals and maxi-
mizing competition and 'the benefit of technology.
SECTION IX, RECOMMr!:NDATIONS
Here again, the Report's focus is on providing recommendations
that a;re designed to achieve immediate symptomatic fixes for an ap-
parent statistical increase in "full competition" without acknowledg-
ing the need for more broad, indepth changes that will improve the
acquisition, management, and utilization of computer technology
throughout the government.
Recommendations Nos. 1 through 5 partially address the need for
it more effective computer technology management system in govern-
ment. One that focuses not only,on acquisition but on management
and utilization as well; as was envisioned by the Brooks Act., How-
ever, the recommendations stop short of recommending that OMB
fully implement such it system. If the government is to maximize the
benefits of technology and maximize competition, it is imperative that
OMB exert policy leadership through the establishment of an orga-
nizational function that will serve as the focal point for the develop-
ment, implementation, and execution of a, comprehensive system to
manage the government's vast computer resources. In discharging
this responsibility, OMB should : '
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6
Approved For Release 2005/1719CIA-RDP83T00573R000600200-6
Develop a management system for the government's computer
resource; including, long-range planning, capital investment, au-
thorization, acquisition, utilization, and inventory management/
retirement.
Develop basic policies for the procurement, management, and
utilization of computers to maximize the benefits of technology
and competition.
Monitor and audit the function of all elements to insure the
enforcement of policy and proper management of the resource to
achieve the desired objectives.
Continuously refine and modify the management system to
adapt to future ADP environments.
Periodically report back to the Subcommittee on progress, find-
ings and problems.
Only through the indepth, continuous involvement and planning
by a central agency such as OMB, will the government ever be able to
maximize the benefits of technology, its vast investment in this tech-
nology, and competition.
Recommendations Nos. 7 through 9 attempt to reduce the number of
noncompetitive procurements. This is a laudable objective which
should be fully supported. However, in doing so, the Report recom-
mends constraining the current methods of improving or enhancing
existing systems without acknowledging the need for a complete re-
vamping of the ADP procurement process. As a result, the agencies
have no reasonable alternatives with which to accommodate workload
or productivity increases short of complete system replacement. In
addition, the recommended two year limit imposed on interim up-
grades in Recommendation No. 8 fails to recognize that it is virtually
impossible to complete the fully competitive process in less than three
to six years. Thus, this recommendation must be accompanied by
major efforts to streamline and improve the system.
Decisions relative to interim improvement to system capacity and
productivity should be based on cost/ benefit analysis rather than on
preferential bias for particular sources of supply as contained in
Recommendation No. 9.
Finally, Recommendation No. 19 directs that each. agency request
appropriations for all ADP expenditures under a single line item.
While there is merit with the underlying desire to insure that Con-
gress is aware of the actual level of ADP expenditures within each
agency, this recommendation would eliminate the essential relation-
ship of ADP expenditures to mission or program accomplishment
which is the real purpose of the expenditure. The recommendation
should therefore call for a requirement that each agency expand its
budget justification to provide a consolidated report and explanation
of all direct ADP expenditures.
SAM STEIGER.
0
Approved For Release 2002/05/17 : CIA-RDP83T00573R000600200010-6