Chapter VII:
The Defense Environment of the 1950s
The Secretary of Defense
under the National Security Act of 1947 had little authority over the three
armed services. The first Secretary of Defense, James V. Forrestal, in fact,
had been hoist by his own petard. As Secretary of the Navy he had helped convince
Congress that the new Secretary should have a bare minimum of authority over
the services and only a very small staff. As the first Secretary of Defense,
Forrestal found himself embarrassed and harassed by open interservice rivalries
which he lacked the authority to settle. Twice, in conferences with the joint
Chiefs at Key West in March 1948 and at Newport in September of the same year,
he thought he had negotiated an armistice only to discover that the services
interpreted these agreements in terms of their own parochial interests. Another
discovery was that he had little effective control over defense budgets either.1
Forrestal in 1948
recommended to the President amending the National Security Act of 1947 to provide
the Secretary of Defense with greater authority and control over the military
services. The Commission on the Organization of the Executive Branch of the
Government established by Congress in July 1947 under former President Herbert
C. Hoover agreed with the Secretary in its report. Acting on these recommendations
Congress passed the National Security Act amendments of 1949 (Public Law 216
of 10 August 1949) which redesignated the National Military Establishment as
the Department of Defense, provided the Secretary of Defense with a deputy and
three assistant secretaries, including a Comptroller, and created a nonvoting
chairman of the joint Chiefs of Staff. The service secretaries lost their seats
on the National Security Council,
[271]
their cabinet status,
and their direct access to the President but not to Congress.2
The National Security
Act amendments of 1949 had also granted the Secretary of Defense greater control
over financial management which he used to reorganize the military budgets along
functional lines. The impetus for the reform came from industrialists outside
the military establishment, in particular the Hoover Commission Task Force on
National Security Organization headed by Ferdinand Eberstadt, a New York investment
broker who had assisted Mr. Forrestal earlier in developing the Navy's unification
proposals.3
The Hoover Commission
in its recommendations for reforming federal administrative management to provide
greater executive control over operations had taken-up where President Roosevelt's
Committee on Administrative Management had left off ten years earlier. Before
then the traditional focus of administration generally and of budgets in particular
was honesty, efficiency, and economy epitomized in the Army's doctrine of accountability.
The Roosevelt Committee, opposed at the time by traditionalists, had inaugurated
a new period. of administration where the emphasis was on executive control
over operations through vertical integration along functional lines, management
engineering techniques for work measurement, and functional budgets. The Bureau
of the Budget took the lead in this movement. In the Army the leadership had
come from General Somervell and his Control Division. The demise of Army Service
Forces at the hands of the traditionalists
[272]
stalled the movement
within the Army until the Hoover Commission sparked its renewal, this time under
the leadership of the Comptroller of the Department of Defense.4
The Army's budget
reflected its fragmented organization. There were twenty-five major "projects"
or appropriations classifications based upon the technical services, each with
its own individual budget, which accounted for 80 percent of the funds spent
by the Army. (Table 1) Neither the Secretary of the Army nor the General
Staff possessed any effective control over these funds. Congressionally oriented
procedures for spending and accounting for appropriated funds also made financial
control difficult. The Army's various accounting systems only told Congress
how much of the funds in any appropriation had been committed or obligated,
not how much had been actually spent or when. They contained no information
on what happened to mat6riel or supplies after their purchase. The existence
at all levels of command of unfunded obligations, principally military pay,
and expendable items were added impediments to rational financial control.
Congress emphasized
the independence of the technical services in its traditional restrictions on
transferring funds among major appropriations categories. Technical service
chiefs could and did transfer funds freely among their various activities, functions,
and installations, but neither the Secretary nor the General Staff could legally
transfer funds among the several technical services or other staff agencies
without going to Congress for approval.
Given these conditions
there was no rational means of determining how much the Army's operations cost,
no means of distinguishing between capital and operating expenses in most instances,
and no means of determining inventory supplies on hand. Repeated requests for
deficiency appropriations each year made even control by Congress over spending
difficult.
Finally it was not
possible to correlate budgets and appro-
[273]
LIST OF MAJOR PROJECTS AND SUB-PROJECTS
INCLUDED IN FISCAL YEAR 1949 BUDGET OF THE ARMY MILITARY (ACTIVITIES) FUNCTIONS
Office of the Secretary of the Army
A. Contingencies of the Army
B. Penalty Mail-Military Functions
General Staff Corps
C. Field Exercises
D. National War College
E. Inter-American Relations-Department of the Army
Finance Department
F. Finance Service, Army
1. Pay of the Army
2. Travel of the Army
3. Expenses of Courts-Martial
4. Apprehension of deserters
5. Finance Service-for compensation to clerks and other employees of Finance
Department
6. Claims for damage to or loss or destruction of property, or personal
injury, or death
7. Claims of military and civilian personnel of the Army for destruction
of private property
G. Retired Pay, Army
Quartermaster Corps
H. Quartermaster Service, Army
1. Welfare of enlisted men
2. Subsistence of the Army
3. Regular supplies of the Army
4. Clothing and equipage
5. Incidental expenses of the Army
Transportation Corps
I. Transportation Service, Army
Signal Corps
J. Signal Service of the Army
Medical Department
K. Medical and Hospital Department
Corps of
Engineers
L. Engineer Service, Army
1. Engineer Service
2. Barracks and quarters, Army
3. Military posts
Ordnance Department
M. Ordnance Service and Supplies, Army
Chemical Corps
N. Chemical Service, Army
Army Ground Forces
O. Training and Operation, Army Ground
Forces
P. Command and General Staff College
[274]
TABLE I
LIST OF MAJOR PROJECTS
AND SUB-PROJECTS INCLUDED IN FISCAL YEAR 1949 BUDGET OF THE ARMY MILITARY (ACTIVITIES)
FUNCTIONS-Continued
United States Military Academy
Q. Pay of Military Academy
R. Maintenance and Operation, U.S. Military Academy
National Guard
S. National Guard
Organized Reserves
T. Organized Reserves
Reserve Officers' Training Corps
U. Reserve Officers' Training Corps
National Board for Promotion of Rifle
Practice, Army
V. Promotion of Rifle Practice
Departmental Salaries and Expenses
W. Salaries, Department of the Army
1. Office of Secretary of the Army
2. Office of Chief of Staff
3. Adjutant General's Office
4. Office of Inspector General
5. Office of Judge Advocate General
6. Office of Chief of Finance
7. Office of Quartermaster General
8. Office of Chief of Transportation
9. Office of Chief Signal Officer
10. Office of Chief of Special Services
11. Office of Provost Marshal General
12. Office of Surgeon General
13. Office of Chief of Engineers
14. Office of Chief of Ordnance
15. Office of Chief of Chemical Corps
16. Office of Chief of Chaplains
17. National Guard Bureau
Office of the Secretary
X. Contingent Expenses, Department
of the Army
Y. Printing and Binding, Department of the Army
Source: Cresap,
McCormick and Paget Final Report, 15 Apr 49, p. III-17.
priations with the military plans, missions,
functions, and operations of the Army as a whole.5
After investigating these conditions
Mr. Hoover told Congress that he and his committee thought the military budget
[275]
system had broken
down. The Army and Navy budget structures were antique. "They represent
an accumulation of categories arrived at on an empirical and historical basis.
They do not permit ready comparisons, they impede administration, and interfere
with the efficiency of the Military Establishment. Congress allocates billions
without accurate knowledge as to why they are necessary or what they are being
used for." Both Hoover and Eberstadt agreed that the efficient operation
of the Defense Department required a complete overhaul of the military budget
structure and its procedures and fiscal policies.
Hoover urged reorganizing
the budget "on a functional or performance basis, by which the costs of
a given function can be compared year by year . . . ." Eberstadt recommended
that the Secretary of Defense have full authority and control over the preparation
and expenditure of the defense budget, assuring "clear and direct accountability
to the President...and the Congress through a single official.6
Title IV of the National
Security Act amendments reflected the recommendations of the Hoover Commission
and the Eberstadt Committee on financial management. Section 401 established
the Office of Comptroller in the Department of Defense and delegated broad authority
to him over the financial operations of the department. The Comptroller was
to direct preparation of the department's budget estimates, including the formulation
of uniform terminology, budget classifications, and procedures. He was responsible
for supervising accounting procedures and statistical reporting. Section 402
provided for comptrollers in each of the three services responsible directly
to the service secretaries and acting in accordance with directions from the
Defense Department Comptroller. The fiscal, administrative, and managerial organization
and procedures of the several departments, it declared, should be compatible
with those of the Office of the Defense Department Comptroller.
[276]
Section 403 called
for adoption of "Performance" budgets and new accounting methods which
would "account for and report the cost of performance of readily identifiable
functional programs and activities, with segregation of operating and capital
programs." It also required the service budgets to follow a uniform pattern.
Section 405 provided for working capital funds to finance retail and industrial
activities within each service such as Quartermaster depots and Ordnance arsenals.7
Performance budgets
meant nothing until the new Department of Defense Comptroller identified what
functional budget classifications to adopt. The first Comptroller of the Department
of Defense was Wilfred J. McNeil who remained in this position for a record
ten years. He was intimately familiar with defense budgets and financial practices
as de facto Comptroller of the Navy since 1945, following similar
service in uniform during World War II. He helped Mr. Eberstadt prepare his
report on military financial management and participated in drafting Title IV
of the National Security Act amendments of 1949.
Mr. McNeil later said
"There were entirely too many fiscal masters; fiscal management was divorced
from management responsibility; and there were no clear lines of authority for
responsibility and management; it was all diffused. Money and responsibility
should parallel each other in any business operation-otherwise there can't be
any tight reins on spending. If the man responsible for operating something
has to account for where the money goes, he's naturally going to make sure it
isn't spent for something it should not be." 8
One of McNeil's first
reforms was the inauguration on 17 May 1950 of the new
performance budget with eight broad functional classifications in place of the
traditional technical service oriented budget. For the Army they were military
personnel, operations and maintenance, procurement and produc-
[277]
tion, research and
development, military construction, army national guard, reserve personnel requirements,
national guard military construction, and army civilian components.
With this one directive
McNeil wiped out the independent budgets of the technical services dating back
in some instances to the Revolution. The chiefs no longer would defend their
budgets before Congress. Instead this would be the responsibility of the several
General Staff divisions. Congressional restrictions on transferring funds among
appropriations would hamstring the technical services rather than the General
Staff. Although they would continue to spend the largest amount of the Army's
budget, the technical services would do so under the supervision and control
of several General Staff divisions.9
The General Staff
gained further control over technical service budgets through its membership
on a new Budget Advisory Committee (BAC) (Army Regulation 15-35, 2 October 1951)
. The technical services were not represented on this committee which passed
on their budget requests. Under the old system the General Staff had little
choice but to forward the technical services' requests. Under the new system
the technical services had to justify their budgets in detail before the BAC.
10
The Army's functional
budget was similar to those developed in modern industrial corporations to control
their operations. It told in detail what it cost to support the Army in terms
of men and material resources. It did not reveal the other side of the picture,
the cost of the Army's operations at home or abroad. It did not reveal the gap
which had grown, as General Marshall had warned, between American military commitments
and the resources available to meet them following World War II.
To close the gap both
the Hoover Commission and Cresap, McCormick and Paget had recommended the development
of an "Army Program System" that would translate strategic plans into
functional operating programs which in turn could be translated into the new
functional budget.
The Army accepted these recommendations
in principle
[278]
and on 12 April 1950
announced the inauguration of the new system which would assist in the alignment
of resources and military requirements. Theoretically the Army Programs were
intended to be concrete operational plans designed to translate JCS strategic
plans into action. They were to include a detailed time schedule for meeting
specified program objectives, the resources required in detail, and a means
of reviewing progress. The program cycle would contain three. phases: program
development, when plans would be translated into operating programs; program
execution, when they would be translated into budgets, later into appropriations,
and finally carried out; and program review and analysis. The department assigned
responsibility for program development to the Deputy Chief of Staff for Plans,
for program execution to the Deputy Chief of Staff for Administration, and program
review and analysis to the Comptroller.11
Translating these
theoretical concepts into action required time to adjust and revise operating
programs in practical terms to reflect planning missions at one end and at the
other to relate as closely as possible to budget classifications. It also took
more time to educate all levels of the Army in the mechanics of the new system.
There were interruptions.
The Korean War pushed the new Program System into the background. Budget requests
to support the Korean War were developed in a series of "crash actions"
as deficiency appropriations requests in addition to the normal annual budget
requests. Deadlines imposed by the Korean emergency did not allow time to translate
plans into programs and then into budgets. It was not until after the
[279]
Korean armistice in
July 1958 that any attention could be paid to developing the new Program System.
It took two years
just to develop the "Program Budgets" themselves into sufficient detail
for submission to Congress. The element that caused the greatest difficulty
and required the most indoctrination was the program planning phase. The Army
was accustomed to submitting its budget request six months before the President
submitted his total budget to Congress and a year in advance of the fiscal year
for which the funds were requested. Under the Army Program System the cycle
for translating strategic plans into detailed operating programs began one year
ahead of the budget cycle or two years before the target fiscal year. The programs
in turn were based on JCS mid-range planning projected several years ahead of
the target year. The Joint Strategic Objectives Plan, as it was designated to
distinguish it from long-range planning estimates and short-range contingency
or capabilities planning, set forth concrete military requirements in terms
of major forces, strengths, facilities, and materiel. These became the Army
Strategic Objectives Plan developed for two years before the target fiscal year
that formed the basis for later developing Army Control Program Objectives.
Concrete Program Objectives came next, accompanied by instructions, operating
assumptions, and schedules for completing approved Control Programs in time
to prepare budget requests for carrying out the programs. The co-ordination
of plans and programs is indicated in Chart 24.12
Translating mission-oriented
strategic plans into functionally oriented operating programs and again into
functional budgets on schedule proved impossible. By the end of the decade the
gap between military requirements laid down in strategic plans and the resources
available in military appropriations was still enormous.13
The principal reason
for this gap was the continued divorce at the very top of responsibility between
strategic planning and budget preparation which General Marshall had warned
against. The Joint Chiefs conceived their plans in terms of requirements for
men and materiel without considering the
[280]
JOINT ARMY PLANNING RELATIONSHIPS
* When the initially developed budget undergoes a subsequent
change resulting from decisions by higher authority, these changes are reflected
through a corresponding revision of the control program.
Source: FM 101-51, DA, 10 Oct 57, DA Planning & Programming
Manual, p.16.
[281]
available funds. The Secretary of Defense
continued to make decisions on budgets, particularly budget reductions, without
adequate knowledge of the impact these decisions might have on U.S. military capabilities.
He allocated funds in bulk to each of the three services which made further allocations
within their own departments in accordance with their own priorities. Joint operations
frequently suffered, notably when the Air Force could not provide the Army with
military air transport.
Another reason for
the disparity between Army plans and budgets was that its programs were rarely
completed in time to be incorporated into current budget requests. Since generally
the organizations or agencies responsible for developing programs also prepared
the Army's budget estimates, the pressing requirements of current operations
hampered program planning for the future. This handicap continued despite repeated
revisions of the Army Program System.
Beginning in 1954
the Army developed a different approach, finally designated in 1359 as the Army
Command Management System (ACMS). Under this system installation commanders
received budget guidance in the form of five-year projected estimates, based
on Army mid-range planning, which were revised annually with changes in Congressional
appropriations. These estimates, called "Control Programs," involved
five major areas: Troop, Materiel, Installations, Reserve Components, and Research
and Development. Using these estimates installation commanders were supposed
to prepare detailed requests for funds under twenty-one major functions covering
the Army's nontactical peacetime support activities. (Table 2) The ACMS
was intended to provide both cost and performance data for the amorphous budget
category, Operations and Maintenance, Army (O&MA). However, Mr. McNeil's
continued insistence on archaic obligation and expenditure data remained a major
obstacle. Another problem was the effort to develop a prototype Class I installation
automatic data processing system at Ft. Meade which would integrate supply,
personnel, and financial reporting. In this pioneering research project one
officer involved said the actual programming was "like dropping a ping-pong
ball into a box-full of mouse traps and ping-pong balls. If you've ever tried
it, you know that all hell breaks loose. We hit problems that were far more
complex than
[282]
TITLES AND CODE ZONE DESIGNATIONS OF
MAJOR ACTIVITIES UNDER THE ARMY MANAGEMENT STRUCTURE
Major Activity |
Code Zone |
Military Personnel, Army |
1000.0000-1990.0000 |
Tactical Forces |
2000.0000-2090.0000 |
Training Activities |
2100.0000-2190.0000 |
Central Supply Activities |
2200.0000-2290.0000 |
Major Overhaul and Maintenance
of Materiel |
2300.0000-2390.0000 |
Medical Activities |
2400.0000-2490.0000 |
Army-wide Activities |
2500.0000-2590.0000 |
Army Reserve and ROTC |
2600.0000-2690.0000 |
Joint Projects |
2700.0000-2790.0000 |
Procurement of Equipment and Missiles,
Army |
4000.0000-4990.0000 |
Research, Development, Test, and
Evaluation |
5000.0000-5990.0000 |
Military Construction, Army |
6000.0000-0990.0000 |
National Guard Personnel, Army |
7000.0000-7090.0000 |
Operation and Maintenance, Army
National Guard |
7100.0000-7990.0000 |
Reserve Personnel, Army |
8000.0000-8490.0000 |
Military Construction, Army National
Guard |
8500.0000-8590.0000 |
Military Construction, Army Reserve |
8600.0000-8690.0000 |
Operation and Maintenance of Facilities |
9000.0000-9090.0000 |
Promotion of Rifle Practice |
0500.0000-0590.0000 |
Army Industrial Fund Activities |
3000.0000-3990.0000 |
Other Operational Activities |
0800.0000-0990.0000 |
Source: AR 1-11, 17 Jan 58.
anyone ever dreamed. Every problem we
hit begat a host of new ones, chain-reaction style. And on anything like this,
you have to have all the kinks out." 14
The development of
functional budgets and programs was the principal effort in the 1950s to improve
the Army's financial management. During this period continued criticism from
agencies outside the Department of Defense focused on other problems that made
effective control over the Army's finances difficult. The Flanders Committee
in 1953 investigated the efforts of the military services to carry out the fiscal
reforms
[283]
called for in the
National Security Act amendments of 1949. It complained of the lack of progress
in setting up uniform budget and accounting systems, revolving operating funds,
and the development of adequate financial statistical reports.15
The Davies Committee
report in December 1953 recommended integrating the Army's financial management
systems with its formal command organization. The new functional budget and
programing system ought to reflect the actual cost of Army operations, it said,
on the basis of the "missions" the Army had to perform rather than
the functional means of accomplishing them. The committee also criticized the
existence within the Army of more than thirty separate accounting systems which
could not be correlated rationally. There should be a single, uniform system
of accounting that would adequately measure the "costs of performance."
That meant the adoption of modern, "accrual" cost-accounting systems
and double-entry bookkeeping at all levels of the Army.16
President Eisenhower's
first Secretary of Defense, Charles E. Wilson, appointed a special Advisory
Committee on Fiscal Organization and Procedures within the Department of Defense,
known as the Cooper Committee, which recommended replacing the traditional "obligation-allotment"
form of accounting with modern "cost-of-performance" budgets as a
more rational means of controlling defense costs. One committee member, Wilfred
McNeil, disagreed. As Comptroller of the Department of Defense he was successful
in preventing the adoption of this recommendation.17
The Second Hoover
Commission on the Organization of the. Executive Branch of the Government in
1955 again criticized military budgeting and accounting systems as archaic and
recommended that Congress require budgets and accounting systems based on a
cost-of-performance or "accrual" basis. Congress passed such a law
in 1956 (Public Law 863, 89th Congress,
[284]
1 August 1956) , but
because of Mr. McNeil's opposition it remained largely a dead letter.18
Certainly one of the
original reasons for the whole movement toward unification in the armed forces
was the belief that separate service supply organizations were duplicating and
wasteful. General Somervell's Army Service Forces was the principal champion
of the idea of a completely unified supply and service system for the Army,
Navy, and Air Force, and it was largely because of the ASF influence that the
War Department, in the first phases of the unification hearings, supported the
idea of a unified supply organization. General Lutes, wartime ASF Director of
Operations and in 1947 Director of Supply, Services, and Procurement on the
General Staff, had a study prepared in January 1947 that declared:
Procurement, supply
and service operations will never be as efficient on the basis of voluntary
cooperation as they can be if integration is required . . . . There are large
savings to be had in unified service, supply and procurement for the Armed Forces.
These are not only savings in money, but also savings in the resources that
are scarce in time of war-men, material, facilities, time.19
By 1947, however,
Lutes represented a voice crying in the wilderness as the Army in general backed
away from the concept of a fourth military department handling supplies and
services, even as it fragmented its own functions in this area by restoring
the technical services to their former power and authority. The Navy had consolidated
its own common supply functions in the 1890s under the Bureau of Supplies and
Accounts but championed "voluntary cooperation" among the services
in supply as well as in other fields. The fledgling independent Air Force, irked
with its dependency on the Army supply system for many common supplies and services,
sought to establish its own independent supply system. The Eisenhower-Spaatz
agreement of 15 September 1947 provided that the Army and Air Force would use
each other's services and facilities
[285]
"where economy
consistent with operational efficiency will result." According to some
critics the Air Force interpreted "operational efficiency" as requiring
a completely separate supply system regardless of duplication and overlap.
The Army position
by the early 1950s had changed similarly. A single supply service, the Assistant
Secretary of the Army asserted in 1951, would mean the military would lose "command
control of supply and thus direction of military operations." The necessity
for unity of command over military operations became, in all three services,
the basis of opposition to any revival of the Somervell-Lutes proposals 20
The principal champions
of integrated supply systems were to be found in Congress and the business world,
not in the military services. Under constant prodding from the outsideby Congress
and the two Hoover Commissions-the movement toward increased co-ordination and
integration of service supply systems gained momentum between 1947 and 1960.
The impact on the Army's technical services was considerable.
The National Security
Act of 1947 gave the Secretary of Defense an ill-defined authority to eliminate
duplication and overlap among the services in the supply area and created a
co-ordinating authority in the Munitions Board which, along with other functions,
was to work toward these ends.21
The accomplishments of the Munitions Board along this line were not great, at
least in part because some regarded it as the forerunner of a. single defense
supply service, but it did initiate some important programs that laid the basis
for future developments, and it conducted many studies. Its main work was in
the area of procurement where it originated what was later to be known as the
Coordinated Procurement Program under which one service acted as purchasing
agent for certain categories of items for the others. This arrangement provided
benefits of consolidated purchasing such as lower prices and fewer purchasing
personnel. It did not provide effective control over inventories nor eliminate
the duplication created by the existence of
[286]
different storage
and distribution systems, including the seven different systems of the Army
technical services .22
Meanwhile, in 1948,
Congress passed the Armed Forces Procurement Act, requiring the development
of uniform procurement procedures for all the military departments. The main
purpose of the act was to free the military services from the prewar requirement
that all procurement in peacetime be by open competitive bidding, a requirement
that was impractical in many cases. The act spelled out specific requirements
that must be met to justify negotiated contracts. To carry out the purposes
of the act a special committee drafted a set of Armed Services Procurement Regulations
(ASPR's) that became, with their periodic revisions, the bible of procurement
procedures for the Department of Defense.
Not unexpectedly performance
did not live up to ASPR's promise. The Second Hoover Commission Task Force on
Military Procurement in 1955 stated the ASPR lacked adequate coverage. Consequently
it had spawned a mass of subordinate individual service regulations. There was
also, it said, a wide gap between the ASPR and actual procurement procedures
at the working level in the field which frustrated contractors. The Task Force
recommended rewriting the ASPR to take care of these deficiencies.23
Other important steps
in the 1947-50 period included the establishment of the Military Sea Transportation
Service (MSTS) and the Military Air Transport Service (MATS) with responsibilities
for handling ocean surface transportation and air transportation, respectively,
for the three services. The
[287]
activation of MSTS
removed the Army Transportation Corps from the field of shipping in which it
had been engaged since the Spanish-American War. On the other hand, insistence
on unity of command over operations frustrated efforts to create an integrated
military land transportation service in the United States until 1956.24
The First Hoover Commission
report in 1949 did not, as the second report was to do, recommend a radical
alteration of the DOD logistics structure. However, it did recommend that ".
. . the National Security Act of 1947 be specifically amended so as to strengthen
the authority of the Secretary of Defense in order that he may integrate the
organization and procedures of the various phases of supply in the constituent
departments of the National Military Establishment."25
The provisions of the National Security Act amendments in 1949 concerning the
establishment of performance budgets, stock and industrial funds, and cost-of-performance
accounting made greater integration of defense activities possible. The stock
fund principle introduced in the Army as a result-the Navy had long had one
and the Air Force was never to use stock funding to any considerable degree-greatly
facilitated cross-servicing and provided a mechanism for the later operation
of single managers.
Another outgrowth
of the Hoover Commission report was the Federal Property and Administrative
Services Act of 1949 creating the General Services Administration (GSA) with
government-wide responsibility for central purchasing and management of common
supplies and services. However, it provided that the Secretary of Defense might
use his discretion in exempting the military services from purchasing common
supplies through GSA. For the most part successive Secretaries of Defense did
so, but under a policy that provided for maximum use of GSA facilities where
it would promote efficiency and economy. As the situation developed, GSA assumed
responsibility for providing general office supplies and equipment for the armed
services and for planning, constructing, managing,
[288]
and operating buildings
occupied by the military establishment in the United States. Beyond this the
extent to which each service used GSA as a purchasing agent for common supplies
was a decision to be made by that service 26
Indeed defense policy quite specifically stipulated that there should be separate
service supply establishments. The philosophy under which successive Secretaries
of Defense proceeded, at least until 1955, was set forth in 1949 by Secretary
Louis A. Johnson: "Each of the services is responsible for the logistic
support of its own forces except when logistic support is otherwise provided
for by agreement or assignments as common servicing, joint servicing, or cross
servicing at force, command, department or Department of Defense level."
27
Congress, demanding
greater progress in integrating supply management, became disenchanted with
the Munitions Board. In the Defense Cataloging and Standardization Act of 1952
it transferred the board's functions to a new Defense Supply Management Agency.
The Eisenhower Reorganization Plan No. 6 abolished both this agency and the
Munitions Board, replacing them with a single executive, an Assistant Secretary
of Defense for Supply and Logistics.28
The Korean War led
to several investigations by Congress of military supply management which threatened
to impose a common supply service on the military services from the outside.
The investigation begun in 1951 by the House Committee on Expenditures in the
Executive Departments, known as the Bonner Committee, was the most important.
It charged that, contrary to the Eisenhower-Spaatz agreement, the Air Force
in developing its own supply system had included items commonly used by both
the Army and Air Force. It criticized the lack of co-ordinated supply management
among the Army's technical services, citing conspicuous examples of waste in
competitive buying, overstocking, and duplication in the use of personnel, space,
and facilities. It accused the services of giving only lip service to the principles
of integrated supply manage-
[289]
ment and quietly agreeing
among themselves to emphasize separatism rather than integration.
The Bonner Committee
concluded that supply management within the Department of Defense and the services
lacked adequate centralized control. As a result of its recommendations Congress
adopted the O'Mahoney amendment to the Defense Appropriations Act of 1953 prohibiting
the "obligation of any funds for procurement, production, warehousing,
distribution of supplies or equipment or related supply management functions,
except in accordance with regulations issued by the Secretary of Defense."
Complying with this provision, Secretary of Defense Lovett on 17 November 1951
issued DOD Directive 4000.8, Basic Regulations for the Military Supply System,
ordering the Air Force to abide by the principles of the Eisenhower-Spaatz agreement,
among other things. It listed eleven general principles governing the management
of defense supply and service activities, including cross-servicing, single
procurement, cataloging and standardization, conservation, surplus disposal,
transportation, and traffic management. It expressly prohibited the addition
of new independent or expanded supply functions involving standard, common-use
items without the approval of the Secretary of Defense. One provision required
the services to establish "one single supply and inventory control point
for each specified category of items." By 1960 there were twenty-four such
"National Inventory Control Points" in the Army. 29
The Bonner Committee
became the Riehlman Committee in January 1958 with the change in Congressional
control to the Republican party and continued its investigations. Later that
year it reported that the services were still too slow in improving their management
of supply, asserting that one major reason was that each service was "dedicated
to its own systems and procedures." The new Assistant Secretary of Defense
for Supply and Logistics, Charles E. Thomas, and his successor, Thomas P. Pike,
both disagreed with these findings and opposed further efforts to integrate
common supply activities as a fragmented approach which did not recognize "the
[290]
basic fact that each military supply system
is maintained solely to provide supplies as needed by the tactical forces that
they were called upon to support." 30
While the Riehlman
Committee produced no tangible results, the Commission on Organization of the
Executive Branch of the Government created by Congress on 10 July 1953 and known
as the Second Hoover Commission did. Its Task Force on the Business Organization
of the Department of Defense charged the Department of Defense and the military
services with continued waste, overlapping, and duplication of effort in nearly
all aspects of supply management. Co-ordination was piecemeal and fragmentary.
Substantial economies and greater efficiency could only be achieved by creating
within the Department of Defense a civilian defense supply and service administration,
which would perform common supply and service functions all over the world.31
The military services
opposed a civilian common supply agency even more than a military one. They
charged it would be less responsive to military requirements and so jeopardize
the success of military operations. An Army staff study, The Fourth Service
of Supply and Alternatives, prepared in the Business and Industrial Management
Office of the Office of the Deputy Chief of Staff for Logistics in September
1955 followed this line of argument. The Hoover Commission and its Task Force
on the Business Organization of the Defense Department, it said, did not give
adequate attention to the military aspects of military supply management. They
emphasized peacetime operations. The Task Force's assertions concerning the
inefficiency of military supply management were based on "unsupported assumptions."
A civilian agency,
according to DCSLOG, would produce further duplication of personnel and functions,
increase competition for scarce professional and technical skills, and make
it difficult for the Army to train its own military logistical
[291]
managers and service
troops. Recruiting civilian supply personnel in wartime or for overseas services
was also undesirable. In conclusion the Task Force said that a civilian "Fourth
Service of Supply" would impair the Army's ability to carry out its assigned
military missions.
It admitted the existence
of deficiencies in supply management, but it said the best way to correct them
was to improve management practices within the existing organizational framework
rather than create a separate agency. Among the alternatives it suggested were
to accelerate adoption of uniform inventory management practices, the standardization
of supply documents, and some form of integrated management for subsistence
and for medical supplies.32
Congress did not appear
as impressed with the argument of military necessity as it was with the Hoover
Commission's indictment of waste and inefficiency in the military services.
To avoid having Congress take the matter away from the services entirely, the
Department of Defense did an about face. A prime mover in bringing about a more
favorable attitude toward greater integration in supply management was the new
Deputy Assistant Secretary for Supply and Logistics, Robert C. Lanphier, Jr.,
a Midwestern electric and utility company executive. Beginning in late 1954
a task force in his office spent several months exploring how best to achieve
a maximum degree of integration with a minimum of disruption to the existing
service organizations. The solution proposed and approved by the Secretary of
Defense was to appoint "Single Managers" for a selected group of common
supply and service activities.
The single manager
concept was the most significant advance toward integrated supply management
within the Department of Defense or the armed services since the end of World
War II. Basically it was an expansion of the Single Service Procurement Program
to provide more effective control over inventories at one end of the supply
cycle and greater control over wholesale distribution at the other. Like the
Single Service Procurement Program it superimposed a new organiza-
[292]
tional pattern on
the existing one instead of creating a new organization.
When the concept was
presented to the Secretary of Defense and the services the Navy, adhering to
its traditional opposition to integration in any form, opposed it on principle.
A new Deputy Secretary of Defense for Supply and Logistics, Reuben B. Robertson,
Jr., a paper company executive who had served as vice chairman of the Second
Hoover Commission's Committee on the Business Organization of the Department
of Defense, overruled the Navy's objections and approved a directive outlining
the procedures and principles to be followed in setting up single managerships.
Under this directive
the Secretary of Defense would formally appoint one of the three service secretaries
as single manager for a selected group of commodities or common service activities,
and he, in turn, would select an executive director to operate the program.
Single managers in the Army were established within the existing technical service
organizations. The Secretary of the Army designated a major general as executive
director who served under the chief of the technical service responsible for
the particular type of commodity or service involved. The Single Managers for
Subsistence and for Clothing and Textiles operated under the Quartermaster General,
while the Single Manager for Military Traffic Management was under the Chief
of Transportation.
The responsibility
of the single managers for determining requirements involved common cataloging
and standardization as well as inventory control. They operated under a stock
or consumer revolving fund, buying what they needed, selling to the military
departments and consumers, and using the funds paid to replenish their stocks.
This eliminated the expense and delay in calling for open bids each time supplies
were requested. Through their control over wholesale storage they were able
to direct distribution to consumers from the nearest depot, regardless of the
service operating it, in such a manner as to avoid the needless and expensive
cross-hauling involved when each service maintained its own completely separate
distribution systems.
Under this system
the technical services preserved their organizational integrity. The single
managers operated through
[293]
normal command channels,
reporting to the Secretary of the Army through the chiefs of the technical services
and the Deputy Chief of Staff for Logistics, who was assigned Army staff responsibility
for the single manager programs. The impact of losing control over their inventories
was minimal because this was a function the technical services had never chosen
to exercise effectively. They continued to calculate their own requirements,
and they retained their own wholesale and retail distribution systems. The chief
difference was that they operated their wholesale depots now as agents of the
single managers rather than the several technical services. In sum, the single
manager concept balanced demands for greater integration of supply management
with the military services' insistence that effective military operations required
each service to maintain its own independent supply system.
One factor which delayed
expansion of the single manager system was the requirement for common cataloging,
standardization, accurate inventories, and the necessity to set up cost-ofperformance
accounting systems for each single manager stock fund.33
According to Mr. Lanphier,
in late 1955, the most appropriate categories for initial single manager assignments
were subsistence and clothing and textiles, both of which were assigned to the
Army's Quartermaster Corps, and petroleum and medical and dental supplies, both
of which were assigned to the Navy's Bureau of Supplies and Accounts. These
were areas of largely common-use items, where common cataloging was relatively
complete, where some defense-wide co-ordination existed, and where much of the
wasteful duplication of inventories and cross-hauling existed. The Military
Subsistence Supply Agency was the first to be inaugurated on 4 November 1955,
followed in the next year by the three others. The coordination of transportation
services was also far advanced. Creating single managerships out of the existing
Military Air Transport Service and the Military Sea Transportation Service was
simply a change in designation. The single manager for land traffic management,
the Military Traffic Management
[294]
Agency (MTMA), was
a new agency and assigned to the Transportation Corps. This function had gradually
developed over a decade against considerable Air Force opposition.34
Assignments in 1959
of single managers for general supplies to the Army's Quartermaster Corps as
the Military General Supply Agency (MGSA), for industrial supplies to the Navy's
Bureau of Supplies and Accounts as the Military Industrial Supply Agency (MISA)
, for construction supplies to the Army's Corps of Engineers as the Military
Construction Supply Agency (MCSA), and for automotive supplies to the Ordnance
Department as the Military Automotive Supply Agency (MASA) were the result of
studies undertaken by the Armed Forces Supply Support Center mentioned below.
They required much more work in common cataloging, standardization, accurate
inventorying, and the installation of cost-of-performance accounting. Further
studies by the Armed Forces Supply Support Center in 1960 envisaged an additional
single managership for electrical and electronic supplies.35
The impact of the
single manager system within the Army was greatest on the Quartermaster Corps,
which was, by 1960, responsible for three of them. Three of its four identifiable
procurement systems were single managerships, and the Quartermaster General
asserted that nearly all his supply and procurement personnel were involved
either with the single managerships or other efforts to integrate defense supply
management.36
Another program designed
to eliminate duplication and prevent waste by permitting the transfer of surpluses
among the services was the Interservice Supply Support Program begun in 1955.
It consisted of six area co-ordination groups under a joint council of the services
and governed thirty-three commodity co-ordination groups.37
[295]
Finally, in June 1958,
the Department of Defense created the Armed Forces Supply Support Center (AFSSC)
to administer the common cataloging, standardization, and Interservice Supply
Support programs, the latter redesignated as the Defense Utilization Program.
It was also to study military supply activities continually and recommend improvements
in their management. From such analyses came proposals leading to the creation
of the four additional single managerships in 1959 and 1960, referred to above.38
While all these efforts
were being made within the Department of Defense to improve supply operations,
Congress continued its criticisms of waste, duplication, and overlap and of
the slow progress being made to eliminate them. In 1958 nearly a dozen different
bills were introduced into Congress to establish a fourth service of supply.
The outcome was the addition of the McCormack-Curtis amendment to the Defense
Reorganization Act of 1958 which granted the Secretary of Defense explicit authority
to consolidate or integrate the supply and service functions of the three services,
subject to a Congressional veto where this involved legislative changes. The
amendment stated:
Whenever the Secretary
of Defense determines it will be advantageous to the Government in terms of
effectiveness, economy, or efficiency, he shall provide for the carrying out
of any supply or service activity common to more than one military department
by a single agency or such other organizational entities as he deems appropriate.39
Congress continued
to scrutinize defense supply management and to demand further integration.
At the end of the
fifties the Secretary of Defense and the Joint Chiefs of Staff possessed much
greater authority and control over the three military services than Congress
had provided for or intended in the National Security Act of 1947. The cold
war, the revolution in technology, mounting defense costs, and pressure from
industry and Congress were responsible for this increase in power. It was most
apparent in four
[296]
areas-strategic planning
and the direction of military operations, financial management, logistics, and
research and development.
The first substantial
increase in the authority of the Secretary of Defense and the Joint Chiefs of
Staff came with the passage of the National Security Act amendments of 1949.
The Secretary of Defense's increased authority over defense budgets, calling
for their reorganization along functional lines, was perhaps the most significant
change. At the same time the Secretary's civilian staff and the joint staff
serving the joint Chiefs of Staff were also increased in size.
The Eisenhower Reorganization
Plan No. 6 of 1958 brought about further increases in the authority of the Secretary
of Defense and the Joint Chiefs. The various functional boards set up under
the 1947 act, stymied by interservice rivalry, were replaced by a series of
functional assistant secretaries with authority to act on behalf of the Secretary.
The civilian staffs of the service secretaries were also reorganized on functional
lines, reflecting the changes within the Office of the Secretary of Defense.
The joint staff was again increased in size and the chairman of the joint Chiefs
given greater authority over the joint staff's operations.
Congressional prodding
and Soviet technological achievements led to the Defense Reorganization Act
of 1958 which centralized authority over the services in the Secretary of Defense
and his office even further. The chain of command over military operations was
changed to run from the President and Secretary of Defense through the joint
Chiefs rather than through the service secretaries who had acted as executive
agents since 1953. The JCS, its staff doubled to four hundred, was completely
reorganized along conventional military staff lines, replacing the system of
JCS committees, most of which were abolished. The authority of the chairman
of the joint Chiefs of Staff over the joint staff was increased, and at the
request of President Eisenhower he was given a vote in JCS decisions previously
denied him by Congress.
Authority over the
research and development of new weapons and weapons systems was centralized
under a new Director of Defense Research and Engineering. The McCormack-Curtis
[297]
amendment also gave the Secretary of Defense
greater discretional authority to integrate service supply activities.40
This increasingly
centralized control by the Secretary and the Department of Defense obviously
diminished the role of the services. They had become support commands, responsible
for training, administration, and logistical support of military operations,
limited further by the authority of the new Director of Defense Research and
Engineering over the development of new weapons.
The Department of
the Army became responsible largely for functions performed by the Army Ground
Forces and the Army Service Forces during World War II. There was continual
discussion of whether the Secretary of the Army should act as an independent
spokesman for the Army or as an executive vice president for the Secretary of
Defense. The answer depended somewhat on the personalities involved and their
length of service. The long tenure of Mr. McNeil as Defense Comptroller and
Secretary of the Army Wilber M. Brucker undoubtedly increased their personal
influence, but generally the tour of duty for top-level civilian administrators
was relatively brief as it had been in the past. Other things being equal, under
a strong Secretary of Defense the service secretaries were more likely to act
as his executive agents than under a weak one. 41
[298]
Endnotes
Previous Chapter
Next Chapter
Return to the
Table of Contents