Chapter VIII: 
The McNamara Revolution
One of the major issues of the 1960 Presidential campaign was the alleged inadequacy of the Eisenhower administration's direction and management of the nation's security. Two of the principal critics were retired Army Chief of Staff General Maxwell D. Taylor and the former Army Chief of Research and Development Lt. Gen. James M. Gavin. The Subcommittee on National Policy Machinery of the Senate's Committee on Government Operations, under Senator Henry M. Jackson of Washington, began a series of hearings and investigations in January 1960 which also concentrated on the inadequacy of this country's national security organization. Senator John F. Kennedy, when running for President, appointed Senator Stuart E. Symington of Missouri, a former Secretary of the Air Force under President Truman, chairman of an advisory committee to investigate the organization and operations of the Department of Defense. Finally two RAND Corporation officials, Charles J. Hitch and Roland N. McKean, criticized the financial management of the Department of Defense in The Economics of Defense in the Nuclear Age.
General Gavin charged that the roles of the joint Chiefs as heads of separate military services were incompatible with their functions as the nation's top military planners because they could not in practice divorce themselves from the particular interests of their individual services. There were "interminable delays" in reaching decisions caused by disagreement and deadlock among the services. He suggested abolishing the joint Chiefs of Staff and substituting a Senior Military Advisory Group to the Secretary of Defense. Its members would be senior officers who had just completed a tour of duty as their service's chief of staff, and a functional joint staff would support them.1

General Taylor had become the principal military spokesman of the Marshall tradition of tight executive control over the armed services before and after his retirement as Chief of Staff of the Army. In The Uncertain Trumpet, he, like General Gavin, was critical of current military strategy because it neglected the Army in favor of the massive deterrent of the Strategic Air Command. Concentration on total nuclear war similarly neglected the requirements of conventional and limited warfare, the principal type of conflict that had developed during the cold war.
Like General Gavin, Taylor also criticized the procedures by which the joint Chiefs of Staff reached their decisions. Repeating General Marshall's dictum, he told the Jackson Committee that "you cannot fight wars by committee." A single armed services chief of staff should run the Secretary of Defense's "command post" for him, assisted by an advisory council. In summary effective control over operations required more efficient planning as well as a more efficient planning organization.
The current role of the Defense Department Comptroller disturbed General Taylor. Given the fact that the joint Chiefs of Staff were often in deadlocked disagreement, he asserted that "strategy has become a more or less incidental by-product of the administrative processes of the defense budget." To avoid this situation he would restructure defense budgets on the basis of the strategic missions to be performed rather than on the resources or functions required to perform them. What was needed was a strategy of "flexible response" capable of meeting all levels of conflict from "cold" through "limited" to "total" war; "atomic" deterrent forces based on intercontinental missiles rather-than manned bombers; "counterattrition forces" capable of fighting "brush fire wars;" guerrilla and other "limited" conflicts; mobile reserve forces, including mobilization stockpiles; air lift and sea lift forces; antisubmarine warfare forces; continental air defense based on the development of antimissile missiles; plus whatever resources were required to support general mobilization and civil defense programs. The three military services would be reorganized similarly as operational commands while the three service departments would be organized to mobilize, train, and support

them. In this manner American military commitments could be balanced effectively with the resources required to fulfill them, another objective which General Marshall had posited at the end of World War II.2
Outside the military services a special Preparedness Investigating Subcommittee of the Senate Armed Services Committee under the chairmanship of Senator Lyndon B. Johnson, Democrat of Texas, in 1957 began a continuing series of inquiries into satellite and missile programs, into the role of the Bureau of the Budget in formulating and executing defense budgets, and into other major issues.
Senator Jackson's Subcommittee on National Policy Machinery investigated "whether our Government is now properly organized to meet successfully the challenge of the cold war." 3
Former Secretary of Defense Robert A. Lovett, a leading civilian disciple of General Marshall, was the first witness to testify before this committee. Echoing his predecessor, he said bluntly that the "committee system" under which the Department of Defense and, indeed, the entire federal government operated traditionally was the principal obstacle to effective decision-making. He admitted that the committee system had developed out of the federal form of government as part of "a series of checks and balances" to prevent any one group within the government from becoming too powerful.
The often forgotten fact is that our form of government, and its machinery, has had built into it a series of clashes of group needs...This device of inviting argument between conflicting interests-which we can call the "foulup factor" in our equation of performance-was obviously the result of a deliberate decision to give up the doubtful efficiency of a dictatorship in return for a method of protection of individual freedom, rights, privileges, and immunities.
Mr. Lovett feared that within the executive branch alone there was an observable trend to expand the committee system

to the point where mere curiosity on the part of someone or some agency and not a "need to know" can be used as a ticket of admission to the merry-go-round of "concurrences." This doctrine, unless carefully and boldly policed, can become so fertile as spawner of committees as to blanket the whole executive branch with an embalmed atmosphere . . . . The derogation of the authority of the individual in government, and the exaltation of the anonymous mass, has resulted in a noticeable lack of decisiveness. Committees cannot effectively replace the decision-making power of the individual who takes the oath of office; nor can committees provide the essential qualities of leadership.4
Thus did Mr. Lovett compare the Marshall tradition concept of tight executive control with the traditional procedures of completed staff actions.
Senator Stuart Symington represented Air Force critics of the JCS committee system. As chairman of a task force on defense organization and management appointed by Senator Kennedy during his 1960 campaign for President, Symington heavily weighted his committee with Air Force spokesmen. One was Thomas K. Finletter, the first Secretary of the Air Force. Another was former Assistant Secretary and later Under Secretary of the Air Force Roswell L. Gilpatric.
Not surprising, the criticisms and recommendations made by the Symington Committee reflected policies advanced by the Air Staff in 1959 in its "Black Book on Defense Reorganization" favoring "total unification." Interservice rivalry, the committee said, prevented the JCS from functioning effectively. To eliminate this rivalry it recommended abolishing the joint Chiefs of Staff in favor of a single armed forces Chief of Staff, called the "Chairman of the Joint Staff," who would be chief military adviser to the Secretary of Defense and the President and direct the activities of the joint staff. He would also preside over a Military Advisory Council composed of those senior officers who had just completed tours of duty as chiefs of staff. Divorced from their services they would no longer feel required to place service interests above everything else.
Second, the Symington Committee proposed to abolish the three "separately administered" services and reorganize them as "organic units within a single Department of Defense." The Secretary of Defense would be assisted by two Under Secretaries, one for Weapons Systems and another for Administration. The former would be responsible for all logistical support

activities, including research and development, production, procurement, and military construction and installations. The latter would be responsible primarily for personnel and financial management. A series of functional directorates similar to the existing Assistant Secretaries of Defense would act as the department's staff.
Finally, to integrate the services completely the committee recommended adopting uniform recruitment policies, uniform pay scales, unified direction of all service schools, and a more flexible policy of transferring personnel among the services. The military services would retain their individual chiefs of staff who would have direct access to the Secretary of Defense. The services would also retain such vestiges of their former separate identities as their distinctive uniforms.5
Spokesmen for the Army's Marshall tradition and the Air Force were the major critics of the Eisenhower defense policies and organization. Representatives of the Navy, which remained the principal supporter of the JCS committee system, were conspicuous by their absence. Supporting the critics was the

Picture - Mr. VANCE
observable trend of the previous decade in the direction of greater authority and control over the services by the Secretary of Defense. As one student of the defense organization put it: "Gradually, and with a finesse which demands respect, the services are being dismembered and disembowelled, so that the question of their utility is decided continually in decrements. Since we cannot reasonably expect to turn the clock back, the only relevant question is whether the process is too fast or too slow." 6
The trend toward centralized authority in the Secretary of Defense seemed likely to continue, but future developments were partly contingent on the man President Kennedy selected as his Secretary of Defense, Robert S. McNamara. McNamara was a highly successful industrial manager, a "comptroller" in the broadest sense of that much-abused and misunderstood term. Most of the reforms he instituted as Secretary of Defense and the techniques he employed were ones which management experts since the days of General Somervell's Control Division had repeatedly recommended. What was unique was the rapidity with which he absorbed information and made decisions. What had disturbed him most at the outset was the long time it took to get decisions out of the Department of

Picture - Mr. HITCH
Defense. In the General Marshall tradition he placed the blame for delay on the committee system with its endless bargaining and compromises. He intended to replace committees where possible by asserting greater executive authority, responsibility, and control over the department and its operations. As he said, "The individual in the position of responsibility must make the decision and take the responsibility for it." 7
Secretary McNamara was surprised to find that there was no management engineering agency within his office responsible for reviewing organization and procedures. He promptly assigned this function to the department's new General Counsel, Cyrus R. Vance, a veteran of the Johnson Defense Preparedness Subcommittee. Another Johnson subcommittee veteran, Solis Horwitz, became Director of the Office of Organizational and Management Planning under Mr. Vance. This agency was responsible for directing or supervising studies requested by Secretary McNamara in its assigned area and for monitoring major organizational changes in the Department of Defense stemming from such projects.
One study led to regrouping the functions of the Assistant

Secretaries of Defense. The two Assistant Secretaries for Manpower, Personnel, and Reserve and for Health and Medical services were combined under one Assistant Secretary for Manpower. The Assistant Secretaries for Supply and Logistics and for Property and Installations were also combined under one Assistant Secretary for Installations and Logistics. An Assistant Secretary for Civil Defense was added because this function had been transferred to the Defense Department. Other studies resulted in abolition of more than five hundred superannuated departmental committees and in a major reorganization of the Air Force's field establishment into a research and development or Systems Command and a Logistics Command.8
Mission or Program Budgets
Secretary McNamara's first major reform was to revise the Defense Department's budget to reflect the military missions for which it was responsible. The person most directly responsible for this project was the new Defense Comptroller, Charles J. Hitch. The Office of the Comptroller in the Army for several years had advocated such a budget. When McNamara became Secretary of Defense the Army's Chief of Staff was General George H. Decker, a former Comptroller, who sought to develop some means of presenting the Army's costs of operation in mission terms. In the fall of 1960 shortly after he became Chief of Staff, Decker had initiated additional investigations of this concept.9
Mr. Hitch believed that the combination of functional budget categories and the rigid budget reductions of the Eisenhower administration had created unmanageable problems, with each service favoring its own projects at the expense of joint ones, concentrating on new weapons systems at the expense of conventional ones, and neglecting maintenance.
The Army's own modernization program emphasized the development of missiles and Army aviation at the expense of conventional weapons and equipment, Mr. Hitch charged. In

an era of financial austerity the Army's major overhead operating costs, the operations and maintenance program, suffered most. More and more equipment was useless for lack of spare parts. Deferred maintenance seriously impaired the Army's combat readiness. Local commanders often had to transfer operations and maintenance funds intended for repairs and utilities for more urgent missions, an illegal transaction made possible by the thin dividing line that existed in practice between procurement activities and overhead operations.10
Another major weakness of the existing budget was the failure to relate functional appropriations to major military missions or objectives. Mr. Hitch proposed a series of nine "Program Packages" designed to solve this problem. (Table 3)
Major Programs   FY 1961
Actual 2
FY 1962
FY 1962
FY 1963
Strategic Retaliatory Forces         7.6     9.1     8.5
Continental Air and Missile Defense Forces          2.2     2.1     1.9
General Purpose Forces          14.5     17.5     18.1
Airlift/Sealift Forces          .9     12     1.4
Reserve and Guard Forces          1.7     1.8     2.0
Research and Development          3.9     4.3     5.5
General Support         12.3     12.7     13.7
Civil Defense              .3     .2
Military Assistance          1.8     1.8     1.6
Total Obligational Authority      46.1     44.9     51.0     52.8
1 Total obligational authority represents the total financial requirements for the program approved for initiation in a given fiscal year, regardless of the year in which the funds were authorized or appropriated.
2 Breakdown not available for fiscal year 1961.
Source: Annual Report of the Department of Defense, FY 1962, p. 367.

Only three of the new categories referred to major military missions: strategic retaliatory forces, continental air and missile defense forces, and general purpose forces for conventional or limited war. Four categories, air lift and sea lift forces, research and development, general support, and reserve forces were supporting activities. Military assistance and civil defense, the latter soon replaced as a separate category by retired pay, were separate categories for political reasons as much as anything else because Congress insisted on treating these areas separately from regular defense appropriations.11
Congress did not accept these program packages as a substitute for the service-oriented, functional appropriations structure developed in the previous decade. As a consequence, Mr. Hitch and the services with the aid of computers developed a means, known as a torque converter, of translating program packages into appropriations categories and vice versa, both for the current fiscal year and projected several years into the future.
Applying appropriations categories to major military missions or to the research and development of major new weapons systems was not too difficult. The problem was how to apportion overhead operating costs like operations and maintenance among the major missions and similarly to break down the general support package into standard appropriations.12
Since the major purposes of Mr. Hitch's reforms were to enable Congress, the President, and the Secretary of Defense to assert greater control over defense budgets and operations and to balance military requirements with the resources available to carry them out, much depended on the accuracy and uni-

formity of the statistical information contained in budget requests. If inaccurate information were fed into computers, the answers would be inaccurate. The lack of reliable cost data, particularly for the Army's operations and maintenance program with which the department and the Army had been struggling for more than a decade, remained a major unsolved problem complicated by the continuing shortage of funds available for this category of appropriations.13
The analysis of resource requirements and their allocation among competing military programs on a rational basis was the responsibility of a new Office of Programming within the Department of Defense Comptroller's Office under Hugh McCullough, a veteran with twenty years' experience in military financial management including the research and development of the Navy's Polaris missile system. Within this new office a Systems Planning Directorate developed means by which to measure and translate into financial terms the mat6riel, manpower, and other resources required by the military services, a function currently known as force planning analysis.
The most difficult assignment was that of the Weapons Systems Analysis Directorate under one of Secretary McNamara's famous "whiz kids," Dr. Alain C. Enthoven, a young RAND Corporation alumnus. The failure to relate appropriations to new weapons systems from their conception to their operational deployment and ultimate obsolescence was, Hitch asserted, another great weakness of the existing budget structure. What was needed, and what Dr. Enthoven's office attempted to supply, was a rational means of estimating the costs of new weapons systems, including not only the costs of research and development and of procurement and production but their annual operating costs. Military officers neglected the latter in their estimates because they were not accountable for these costs. In evaluating alternative weapons systems and strategies Enthoven and his staff employed cost-effectiveness analysis developed by economists and systems analysis developed by operations research analysts. Their evaluation included analysis of the objectives of competing strategies and

their often unstated underlying basic assumptions. It sought wherever possible to substitute rational judgment for guesswork in reaching decisions. As Mr. Hitch said:
In no case . . . is systems analysis a substitute for sound and experienced military judgment. It is simply a method to get before the decision-maker the relevant data, organized in a way most useful to him . . . . What we are seeking to achieve through systems analysis is to minimize the areas where unsupported judgment must govern in the decision-making process.14
Cost effectiveness and systems analysis introduced the jargon of statistics and computer technology into military planning. When "the standard economic model of efficient allocation" employed in cost effectiveness studies was defined as "the maximization of a quasi-concave ordinal function of variables constrained to lie within a convex region," a communications gap opened between the systems analysts and those combat veteran officers unfamiliar with the language. Within the Army it was several years before similar agencies for Force Planning Analysis (21 February 1966) and Weapons Systems Analysis (20 February 1967) were established on the Army staff to match the organization in the Department of Defense Comptroller's Office. By that time the urgent requirements of the Vietnam War had displaced cost effectiveness in priority within the Department of Defense.15
Centralized Defense Functions
When McNamara became Secretary of Defense the centralization of authority in the Office of the Secretary of Defense was apparent in the number of agencies operating directly under the Secretary or the joint Chiefs rather than under the

service departments. One of the earliest of these was the Armed Forces Special Weapons Project (AFSWP), an ad hoc interdepartmental, triservice organization, set up on 1 January 1947 by joint directive of the Secretaries of the Army and Navy as the successor to the Manhattan District when the new Atomic Energy Commission took over most of the latter's functions and facilities. AFSWP was a combined logistical support, training, and combat developments agency for the military application of atomic energy. Serving the Army, Navy, and later the Air Force it was never a joint agency as such. It reported to the Secretaries of War and Navy and later to the Secretary of Defense through the service chiefs.
Following the Department of Defense reorganization of 1958, the Armed Forces Special Weapons Project was redesignated as the Defense Atomic Support Agency (DASA) and placed under the JCS. The National Security Agency (NSA), created in 1952, continued to perform highly specialized technical and coordinating functions in the intelligence area under the direction of the Secretary of Defense. The Advanced Research Projects Agency (ARPA) was created in February 1958 as a separately organized research and development agency of the Department of Defense.
The Defense Communications Agency (DCA) was created on 12 May 1960 as an agency of the Department of Defense responsible to the Secretary through the JCS for the "operational and management direction" of the Defense Communications System, including all Department of Defense "world-wide, long-haul, Government-owned and leased, point-to-point circuits, terminals, and other facilities," to provide secure communications among the President, the Secretary of Defense, the JCS, and other government agencies, the military services and departments, the unified and specified commands, and their major subordinate headquarters.
The first joint defense agency Secretary McNamara established was the Defense Intelligence Agency (DIA), established under the JCS by a directive on 1 August 1961 to "organize, direct, and manage the Department's intelligence resources and

to coordinate and supervise such functions still retained by the three military departments."16
Nearly all these agencies transferred some functions or activities of the Army, Navy, and Air Force to the Department of Defense under the JCS. Another function Secretary McNamara wanted to investigate, common supplies and services, affected the Army more directly. The issue was whether the existing single manager system provided the most effective means of integrating these activities. As outlined earlier this system had been adopted as a means of avoiding complete integration under a fourth service of supply and against considerable opposition from the military services. They continued resistance to further integration, disagreeing on what items should be classified as common supplies and services and on the development of more uniform supply distribution procedures. Congress continued to exert strong pressure for further if not complete integration through a separate defense common supply and service agency.17
On 23 March 1961 Secretary McNamara asked Mr. Vance and the several Assistant Secretaries for Installations and Logistics to study this question, which he labeled Project 100. They were to investigate and list the advantages and disadvantages of (1) continuing the existing single manager system operating under the several service secretaries, (2) assigning responsibility for operating a consolidated supply and service agency under one secretary, or (3) operating such a service under the Secretary of Defense.18
The Project 100 Committee submitted its report on 11 July 1961. The principal weaknesses, it thought, of continuing the existing system of multiple single managers were that the numerous channels of command and staff layers required de-

layed decisions and impeded effective control over operations. Any increase in the numbers of single manager assignments would further complicate this problem, producing duplication and greater diversity of procedures. Finally the single managers had to compete for limited manpower and operating funds with other service functions.
The principal disadvantages of consolidating these functions under one department were that the service selected might tend to favor its own programs and at the same time interfere in the supply management of the other two services. It would also call for a major reorganization with all the attendant confusion, disruption, and temporary loss of efficiency. Interference in the supply management of the services and the disruptive effect of a major reorganization were also disadvantages of setting up a separate consolidated common supply and service agency. It might also be less responsive to combat support requirements.19
The committee recommended that whatever organizational pattern was selected common supply and service functions should remain a military responsibility because their sole purpose was to support military operating forces. Such an integrated system should also be adaptable to wartime use immediately. Each service should retain full control over the development and management of its assigned weapons systems. All of them would continue to require military personnel trained in supply and service management. Common supply and services activities should be restricted to wholesale distribution within CONUS, and the services should retain their own retail distribution systems and facilities as under the existing single manager systems.20
The service chiefs and secretaries split in their choice of alternatives. Secretary McNamara publicly announced his decision on 31 August 1961 that a separate common supply and service agency to be known as the Defense Supply Agency (DSA) would be established. The Department of Defense directive issued on 6 November 1961 establishing DSA, effective 1 January 1962, differed from the Project 100 Committee's concept in two important respects. The committee thought

there should be a Defense Supply Council composed of the Deputy Secretary of Defense, the service secretaries, the chairman of the joint Chiefs of Staff, and the Assistant Secretary of Defense for Installations and Logistics. This council would actively supervise DSA's operations. Secretary McNamara made the council a purely advisory agency and granted the director broad executive authority to run the Defense Supply Agency. Second, he did not limit the choice of the director specifically to a military officer as recommended by the committee. The man he chose, however, was a former Quartermaster General of the Army, Lt. Gen. Andrew T. McNamara. Finally, at the request of the JCS which did not want the responsibility for DSA, Mr. McNamara ordered the director to report directly to him instead of through the JCS as was the case with nearly all the other joint defense agencies.21
When the Defense Supply Agency was set up, it took over the eight commodity single managers, the Military Traffic Management Agency, the Armed Services Supply Support Center, the thirty-four Consolidated Surplus Sales offices, the National Surplus Property Bidders Registration and Information Office, the Army and Marine Corps clothing factories, and the management of a proposed electronics supply center. DSA was to administer the Federal Catalog Program, the Defense Standardization Program, the Defense Utilization Program, the Coordinated Procurement Programs, and the Surplus Personal Property Disposal Program.
The Defense Supply Agency staff included both military and civilian personnel from all services on a joint basis, but 95 percent of its staff were civilians. Originally nearly 60 percent of its staff came from the Army, including most of the Quartermaster's supply management personnel. By the end of June 1963, DSA was managing over a million different items in nine supply centers with an estimated inventory value of about $2.5 billion.
In general DSA was to act as a wholesale distributor of supplies to the services within the continental United States. The military services would decide what they wanted, where they wanted it, and when. DSA would decide how much to buy, how much to stock, and how to distribute it to meet the

needs of the services. The services retained responsibility for selecting those items which should be placed under integrated management.22


Previous Chapter     Next Chapter

Return to the Table of Contents