SP-4102 Managing NASA in the Apollo Era


Chapter 3

Headquarters Organization, or the View From the Seventh Floor*


[27] The internal administrative history of NASA is much more difficult to write about than its external affairs. Bureaus rose and fell; functions performed in one were parceled out among several and vice versa. Bureaus with imposing titles sometimes had little real power, while others with little formal authority had much say in making and carrying out management decisions. Furthermore, the larger the organization and the greater the rate of change, the more complex internal relations became, with more opportunity for jurisdictional conflict and a greater need for specialized units to prepare and enforce agency-wide rules. A final reason for the complexity of the relations (and the difficulty in explaining them) is that many agency practices were worked out through informal unwritten understandings. Such agreements might appear because of the tendency for institutional reality to outrun its formal documentation, or perhaps because officials preferred that certain relations not be formalized, for fear that it might prove impossible to draft a management statement acceptable to everyone. 1

These features-the complexity of jurisdictions, the network of informal relations, the shifting of functions between offices, the blurring of the lines of authority-highlight the problem of understanding and explaining NASA management. A logical way to begin is with an account of headquarters organization in general and of "top management" in particular-that is, the three or four officials at the top of the hierarchy. This approach is attractive because top management was the only group who could represent the entire agency to the outside world. The peculiar decentralized structure of NASA is best understood as a series of delegations of authority from top management to the program offices and from [28] headquarters to the centers, and the major reorganizations of the 1960s, while they affected all of NASA, began as changes at headquarters in its relations with the field installations. In this chapter headquarters administration is analyzed as it changed from 1961 to the resignation of NASA Administrator James E. Webb in October 1968. The next year, ending with the presentation of the report of the President's Space Task Group in September 1969, is discussed in chapter 9. The discussion begins with a summary account of the problems confronting management in organizing headquarters functions. Then the four cycles of reorganization of the 1960s are examined-why they occurred and what they effected--and the discussion concludes with an analysis of the functions top management performed.




When NASA began its official existence on 1 October 1958, Administrator T. Keith Glennan and his deputy, former NACA Director Hugh L. Dryden, faced the problem of pulling together the programs inherited from NACA and those to be transferred from DOD. Dryden had at first seemed the logical choice to head the agency. Before coming to NACA in 1947, he had done notable work in aeronautical research at the National Bureau of Standards, where he rose to the position of Deputy Director. At NACA, as later at NASA, Dryden had earned almost as much respect for his ability as an administrator as he had earlier for his research on boundary-layer flow. But he was no manager in the sense that Glennan and Webb were. As one colleague remarked, "he really wasn't the guy to bang away morning, noon, and night on keeping programs and projects going on schedule and within funds and that kind of thing. What he was, rather, was a man with very good judgment on what objectives we ought to have in NASA, on what relationships were really important.... He had a very good understanding of how things got done or might get snarled up in the Government." 2 Yet the same firmness, methodical approach, and caution that were respected by those who knew him well were not likely to appeal to Congress-still less were his remarks about setting the pace of the space program that seemed to betoken a lack of aggressiveness. What made Glennan acceptable was his combination of experience and political loyalties. He was an engineer, not a scientist; before World War II he had worked as a sound system engineer in the motion picture industry, and during the war he had been in charge of the Navy's Underwater Sound Laboratories. His Republican politics made him acceptable to the White House, while his manifest success in making the Case Institute of Technology, of which he was president, a leading technical institute? and his experience as a member of the Atomic Energy Commission, the National Science Foundation, and the Institute for Defense Analyses implied that he understood the problems involved in managing a science-based agency.

[29] Between them, Glennan and Dryden had the mixture of scientific and administrative experience to preside over the transition from an agency doing most of its work in-house to one expected to contract out for most of its research and development Nevertheless, the transition proved so difficult that it was not completely effected until the end of 1963. In truth, the problems they faced in 1958-1959 were manifold and interrelated: to plan for an extended period of growth in funding and manpower, not as though these were grafted onto a preexisting agency but as the foundation for a national space program; to maintain a proper balance between in-house and contractor capabilities, since in-house work would be necessary both to attract and retain the most talented scientists and keep them available to furnish management with objective technical advice, or at least not advice biased in favor of the corporations with which NASA might do business. They had to (1) develop and update a long-range plan that could justify the agency program internally and before Congress, (2) run the agency on a day-to-day basis, (3) refine the systems management reporting techniques used in the development of Polaris and Atlas and transfer them to the fabrication of launch vehicles and spacecraft, and (4) strike a balance between headquarters and centers-neither too rigidly structured nor too decentralized. Their work was further complicated by the way in which one problem impinged on all the others. For example, the perfecting of management reporting systems depended on the authorization of projects ambitious enough and complex enough to justify their use.

What was inseparable in fact may profitably be dissociated for analysis. The problem of headquarters organization logically preceded the resolution of the other major problems. This was so for three reasons: first, because Glennan and Dryden had to decide which functions to reserve for themselves and which to delegate to the centers; second, because they needed expert staff work to assist them in formulating policy; and third, because agency-wide functions like procurement and financial management might best be handled by central offices. While not a blank check, the Space Act had nonetheless given the Administrator considerable freedom to decide NASA's structure. Starting from scratch, Glennan had more reason than his successor to turn to outside consultants and committees to study management problems. It was not that he actually hired more consultants than Webb, who sometimes seemed to have signed up half the country to consult and advise. Rather, Glennan used them specifically to set up the entire headquarters structure. At a time when NASA was tending toward something very different from what its predecessor had been, the last thing he needed was an internal task force that would recommend the establishment of an agency modeled on NACA. For this reason, Glennan called in outside experts. He hired a consulting firm, McKinsey and Company, as early as October 1958 to assist in organizing headquarters functions and again in 1960 to study contracting systems and to provide staff for the Advisory Committee on Organization. Glennan's choice of outside consultants was also influenced by the lack of NASA personnel skilled in management analysis and by his knowledge of the potential advantages of forming ad hoc study Committees whenever substantial changes had to be made. A committee [30] legitimized open discussion of the fact that a problem did exist, created an arena for bargaining, and served as a conduit for ideas and policy suggestions. The specific advantage of a committee external to the agency was that members could raise issues and suggest improvements without fear of jeopardizing their positions. On the other hand, the danger existed that such committees would simply rubber stamp what had been decided; that, in searching for consensus, the members would rub away the sharp edges of their differences, leaving only smooth rounded surfaces of agreement; and that, by being outside the organization, they had no real stake in whatever conclusions they might reach. The committees appointed by Glennan certainly did justify decisions already taken; that is, they rationalized Glennan's intuitive ideas. This may well have been their principal contribution and the reason why Glennan used them.3

Early in his tenure, Glennan started to tackle a problem inherent in the administration of an organization such as NASA. To make the best use of his time as head of an agency that was about to expand greatly, Glennan had to subdivide the duties of management. To Dryden fell the responsibility for dealing with the universities and the national and international scientific communities; to Glennan, attending to those matters of planning and budgeting and interagency coordination that affected the totality of NASA. In his first year as Administrator, Glennan put forward the notion of a general manager (somewhat along the lines of the Atomic Energy Commission) to handle agency routine, freeing himself and Dryden to formulate policy and conduct the external relations of the agency. Having carried his point in the teeth of considerable opposition, Glennan named Richard Homer, Assistant Air Force Secretary (Research and Development), as Associate Administrator in April 1959. This office was of fundamental importance in the development of NASA policy; the organizational history of the next decade can be charted to a degree by the extent of the Associate Administrator's responsibility for program planning, monitoring the centers, and, through his management of daily operations, freeing the Administrator for long-range planning.

The essential problem in establishing the position of Associate Administrator was that it set up a layer between the line organization and the two top officials. A conflict of roles was unavoidable given that the organization antedated the office. "The top program and administrative directors . . . were powerful individuals in their own right. All were on the scene several months before the position of Associate Administrator was filled . . . the former NACA laboratories had had a history of partial autonomy and resistance to central controls."4 To prevent the laboratories and program offices from breaking off and becoming semiautonomous, certain changes were called for: a staff located in the Associate Administrator's office to enable him to pull the agency's programs together, especially in submitting the annual budget requirements; a standard procedure for authorizing specific programs, updating these authorizations, and integrating them with the long-range plan; and a procedure whereby program managers could be brought face to face with top management to discuss areas that concerned troth groups. None of these had been fully achieved prior to 1961. There were, to be sure, [31] formal meetings for bringing agency-wide problems to light: a semiannual management conference usually held at one of the centers, biweekly meetings chaired by the Associate Administrator and attended by officials from the major program development offices, and a Space Exploration Program Council that, attended by high-level officials, served as a kind of agency supercouncil. All these reviews and Committees represented steps in the right direction, but none went very far. Their effectiveness was directly proportional to the frequency with which they met, the precision with which the agenda could be prepared, and the authority that those attending could exert in enforcing whatever decisions were reached. Where meetings were infrequent and where problems were so general and broad that nothing emerged beyond a vague sense of the meeting, little of consequence resulted. This was largely the case with the semiannual conferences and the Space Exploration Program Council.

Program authorization was yet another problem that took several years to work out. Before 1961 "some projects had been approved verbally only, some had been in the budget but no formal approval action could be found, some were contracts implying some type of formal approval, some were commitments made in letters to outside organizations, some had appeared in a variety of places . . . and could be presumed to have been approved." 5 There is no mystery about the lack of a standard project approval procedure, attributable in part to the number and variety of programs inherited or started up, ranging in size from the development of the F-1 engine for the Saturn rocket to small-scale Explorer spacecraft. No one procedure covered them all. Developing a standard authorization procedure presupposed agreement on official definitions of "program," "project," "system," "systems integration," and so on. NASA also had to recruit a staff capable of matching programs against budgets and extrapolating future programs from present resources; indeed, it was essential that NASA adopt or develop data processing systems to track resources from their appropriation by Congress to completion of the programs.

There had to be standardized procedures for promulgating management directives; few procedures were more useful for the establishment of a headquarters organization than the development of a NASA Management Manual that would distinguish between ad hoc circulars, directives that stated policy in a general way, handbooks of detailed procedure, and instructions intended as standard descriptions of powers and authorities for the time being. (The extent to which the Manual actually made policy or lagged behind is discussed separately.) Ultimately, the success with which NASA created a uniform authorization procedure would depend on having a single official, a final point of contact, responsible for approving project plans. That official, as will be seen, was the Associate Administrator, the key figure around whom all agency programs would ultimately pivot.

Glennan and Dryden worked quickly and effectively to build an agency capable of handling large, complex programs. By the summer of 1960 NASA had acquired many of the people and facilities that would prove indispensable to a greatly expanded mission one year later. In the Jet Propulsion Laboratory NASA....



Figure 3-1. NASA organization chart, 17 January 1961.

Figure 3-1. NASA organization chart, 17 January 1961.


[33] ....had a research installation superbly equipped to design and track deep-space and lunar probes such as the Ranger and Surveyor series; in Marshall, a team with a long in-house tradition and competence in the development of large launch vehicles; in the Naval Research Laboratory, scientists who formed the nucleus of the space science program, a pool of capabilities with few rivals in other Government laboratories; finally, in the Space Task Group, located at Langley but an autonomous subdivision of Goddard, the core of what would become in 1962 the Manned Spacecraft Center in Houston. In fact, by the end of 1960 almost all of NASA's programs for the next decade had either begun or were under study.

The failure of Dryden and Glennan to go still further in establishing a stronger headquarters organization was only partly administrative; it was political as well. Glennan, appointed well into Eisenhower's second term, had taken a leave of absence from the Case Institute of Technology. He had neither the authority nor the inclination to lock his successor into organizational arrangements established in NASA's formative period. Moreover, he had to contend with the uncertainties dogging the very concept of a national space program at a time when there were no manned spacefiight projects approved beyond Mercury, the overlapping spheres of NASA and the Air Force had yet to be defined, and the mechanism for placing and supervising large research and development (R&D) contracts still had to be worked out. Until NASA's mission in space had been enunciated, there was only so much that Glennan could do to mold the organization at headquarters (figure 3-1).

Nevertheless, Glennan established the Associate Administrator as the agency's general manager for day-to-day operations. He set up an Office of Plans and Program Evaluation to prepare and revise an official Ten-Year Plan, issued, as one of his last official acts, a management instruction that provided a uniform mode of project authorization, and established an Office of Launch Vehicle Programs that separated the fabrication of launch vehicles from their ultimate use. Finally, in March 1960 he appointed an Advisory Committee on Organization to evaluate NASA on the assumption that "the opportunity to make comprehensive changes in NASA's organization and procedures would not exist too much longer. " 6

The committee, chaired by Lawrence Kimpton, former chancellor of the University of Chicago, consisted of corporate and foundation executives, men of Glennan's type; this fact almost automatically deprived the committee of any influence in NASA below the highest level. That the Kimpton committee's conclusions were bland, that its report had few original ideas, should not seem too surprising. Perhaps the lesson of the Kimpton committee was that the problems it was invited to address-the proportion of work that should be done in-house, the functions of the Associate Administrator, the responsibilities of headquarters staff for monitoring the centers-went too deep to be disposed of on the basis of an ad hoc committee's recommendations. Significant structural changes would be determined by the content of NASA programs, which in turn would be determined from the outside, by the executive branch and by Congress.7


[34] TURN AND COUNTERTURN, 1961-1963


When Glennan left Washington on the day President Kennedy was inaugurated, the nature, the scope, and even the future of the NASA mission were in doubt. Nor did Kennedy's belated appointment on 7 February 1961 of James E. Webb to succeed Glennan resolve these questions. While the President's advisors had recommended that he name a scientist to head NASA, the President had wanted someone with experience in the political rough-and-tumble of Washington, someone who could handle Congress, the Bureau of the Budget, and those elements in DOD that wished NASA to be essentially a supporting agency for the Air Force. As former Director of the Bureau of the Budget and as Under Secretary of State, Webb had the experience and the energy-if not at first the desire-for the job. He demanded, on being offered the job, the authority to run the agency as he saw fit, without altering everything at once. He insisted that Dryden remain as Deputy Administrator and that both Dryden and Associate Administrator Robert C. Seamans, Jr., who had succeeded Richard Homer in September 1960, serve with Webb as a multiple executive, making the most important decisions together and otherwise working in those spheres for which each was best qualified. This arrangement lasted until Dryden's death in December 1965. As Administrator, Webb represented NASA to other agencies, to Congress, and to the White House; Dryden kept open the lines of communication to the international scientific community and to the National Academy of Sciences, of which he was home secretary; while Seamans, as general manager, prescribed internal policy, served as cochairman of the Aeronautics and Astronautics Coordinating Board, and until 1967 was the official to whom most of the so-called "functional" offices reported.8

Very soon after Webb's appointment was confirmed, it became apparent within NASA that headquarters structure, especially as it pertained to the centers, would have to be changed. As an outsider, Webb had no stake in any existing arrangements. Like his predecessor, he might have decided to call in an outside group to evaluate NASA as a whole. Instead, he began by working with a very small group of insiders, none of whom had been part of NACA: the Director of Business Administration, Albert Siepert, who had come to NASA from the National Institutes of Health; Alfred Hodgson, Siepert's special assistant; and Jack Young, who had earlier been involved in the McKinsey studies of NASA before becoming Deputy Director of the Office of Administration. Without any management instruction or directive, the settled practice became, in Siepert's words, to reserve to NASA officials rather than to outside consultants "the continuous study of our management policies." Appearing before a House committee in 1962, he explained that one of the conclusions of the Kimpton report had been that "NASA has reached a period of maturity in its young life where it should definitely strengthen its management analysis staff. It ought to be able to appraise with its own in-house competence most of the operating problems as they develop."9 This shift, prior to the major changes that it helped to effect, remained standard procedure during the next seven years.

[35] Between February and August 1961 these men carefully laid the ground work for the October reorganization in a series of staff papers that set forth the available options.10 The campaign of self-evaluation, which began with a "Summary Look" in February, went on to a 14 April staff study (on the eve of the lunar landing decision) and culminated in a 12 June paper on "Reappraising NASA" Organizational Structure to Achieve the Objectives of an Accelerated Program.' Each paper represented an advance toward the eventual solution. Whereas the February paper analyzed problems and listed alternatives without recommendations, the subsequent papers began by recommending procedural changes [the] way in which decisions were implemented-as a means of making substantive decisions.

Undoubtedly the most important procedural change was the establishment- proposed in April and made in June-of a "programming" office as a staff arm of the Associate Administrator, which would be charged with responsibility for budget preparation, management reporting, and the preparation of budgetary guidelines for the program offices. "Its loyalty would be to the Associate Administrator rather than to any program or group of programs, or to any combination of field installations or program offices. " 11 The new office's responsibilities would be agency-wide; its staff, several of whom were recruited from Abe Silverstein's Office of Space Flight Programs, would check the tendency of any single headquarters bureau to become semiautonomous or to grow at the expense of all the others. The head of the new Office of Programs,** DeMarquis Wyatt, who had worked for many years at Lewis before coming to Washington, had most recently been in charge of program planning and coordination in Silverstein's office. Wyatt was quick to grasp the possibilities of the new office, namely, that there had to be some single person or office to ensure that "on-going actions, new actions, and future plans were fitted into an identifiable work plan." The basic premise of the Office of Programs was that "a simultaneous technical and fiscal evaluation must be made of plans and actions in order to best accomplish the objectives of [the] program." 12

Wyatt's office became the bureau where all NASA planning converged, and its four divisions-facilities coordination, which oversaw all NASA construction; management reporting, which supplied data for scheduling and reporting to all levels of management; resources programming, which prepared the data for NASA's budget; and project review, which would evaluate requests for new projects or changes in old ones-provided the staff services that Seamans needed as the agency's general manager.13 A danger was that the new office might become something more than a staff arm, interposing itself between the program directors and the Administrator. This seldom happened. For several years the Office of Programs served principally to expedite decisions reached at the program office ever; following the organizational changes of 1967 the Office of Programs, under [36] a different name, surrendered its budgetary functions to prepare those studies on which (it was hoped) long-range planning was intended to be based.

The full-scale reorganization of headquarters announced on I November 1961 was both procedural and substantive (figure 3-2). That is, it prescribed what the program offices were to do as well as how they were to do it. The principal change, under which the center directors were to report to the Associate Administrator, was the most difficult to carry out; it was dropped two years later, and its importance may easily be overestimated. Under the interim arrangement, the program directors at headquarters remained responsible for program budgeting and funding. They set up the technical guidelines, established milestones for reporting progress on a monthly basis, and retained the authority to reprogram funds to the centers. The 1961 reorganization purported to provide Seamans with direct observation of the centers, to bring the center directors into the planning process, and to guarantee control of the $2-3 billion in capital investment anticipated for center facilities. It was not aimed primarily at bringing about a mechanical smoothness of functioning. Rather, this change was made in frank recognition that the lunar landing decision had made manned spaceflight the dominant activity within the agency; that the lunar landing mission was something to which all the centers would have to contribute directly; and that-because all the centers had responsibilities in space science, applications, and advanced research-there ought to be one official with agency-wide responsibilities to whom they should report.

The second major change was the realignment of headquarters program offices. The previous four were abolished,*** and four new ones were created: Space Science under Homer Newell, Advanced Research and Technology (DART) under Ira Abbott, Manned Space Flight (OMSF) under D. Brainerd Holmes, and Applications under Morton J. Stoller.14 An Office of Tracking and Data Acquisition (OTDA) headed by Edmund C. Buckley was established for agencywide support in telemetry and automatic data processing. The responsibilities for developing and using hardware were once again united. The tendency of these changes was to separate the broad areas of the space program that required greater autonomy. In particular, by stressing the importance of advanced research and technology, the 1961 reorganization marked a recrudescence of the NACA concept of research in fundamental aspects of aeronautical and space vehicle design and systematic testing "to obtain data for aeronautical and space vehicles of the future." 15 But OART was now required to go beyond this, and, by reducing theory to design, it was to produce real prototypes of advanced subsystems, whether or not that hardware would ever be used. OART would stand in relation to NASA very much as NACA had stood in relation to the services. OART would have to anticipate problems, do preliminary studies, and carry investigations to the....



Figure 3-2. NASA organization chart, 1 November 1961.

Figure 3-2. NASA organization chart, 1 November 1961.


[38] ....point at which the research could be usefully applied-in this case, by NASA itself.

In sum, the 1961 reorganization was a response to problems, certain of which stemmed from the circumstances of NASA's establishment, others from the prominence of the manned space program, while still others existed by virtue of NASA's status as a large organization involved in research and development. The changes discussed so far bound the headquarters offices and the centers more tightly together and at least temporarily checked the tendency of both to become quasi autonomous. The changes confirmed the role of the Associate Administrator as the official responsible for authorizing projects and approving budgets and provided him with a staff to review and evaluate all NASA programs-those in progress and those for which approval was sought. Nevertheless, the reorganization left certain problems in abeyance and created new ones. The role of the Office of Programs reduced that of the Office of Plans and Program Evaluation, with its Ten-Year Plan, to something of a fifth wheel. The substance of planning, top management came to believe, lay less in stating new objectives than in getting the maximum return on the total dollars invested in programs already approved. The Ten-Year Plan, as it stood, provided no real guidance. Moreover, the budgetary process, for NASA as for any other Federal agency, required that funds be sought on a year-to-year basis, with later years always being treated as less important. By 1962 relatively few new projects were being started; the funds requested were for work in progress. The cancelation of the Ten-Year Plan early in 1963 resulted from the logic of the situation: Planning had to be integrated into the budgetary process, not superimposed on it. The budgetary process became, in Seamans' words, "the mechanism by which new projects, or major reorientation of current projects, may be proposed."16

One difficulty occasioned by the reorganization was that neither the centers nor headquarters was really prepared for it. Holmes and the other program directors complained that they could not "task" the centers effectively enough for the work in hand. Many of the directives required to spell out the details of the changes were a long time aborning; the crucial instructions on the responsibilities of the program directors and the functions of the Office of Administration (responsible for financial management, personnel, security, and the like) were first drafted in July 1961, issued in "informational" form in June 1962, and promulgated authoritatively only in June 1963. But the most serious problem was that neither center directors nor headquarters program directors were quite certain of their functions and responsibilities. The management instruction on "Planning and Implementation of NASA Projects," issued by Glennan in January 1961, was already out of date.17 By September 1962 the situation had deteriorated to the point at which one official wrote to Seamans to complain of "gross inadequacies in major management systems . . . [and] confusion over the respective authorities of the Associate Administrator and Program Director.... The situation ranges from a lack of systems to duplicative systems and in some instances conflicts or disregard of established systems." He was especially disturbed because....



Figure 3-3.-T. Keith Glennan, NASA Administrator from August 1958 to January 1961.

Figure 3-4.-James E. Webb, NASA Administrator from February 1961 to October 1968.



Figure 3-3.-T. Keith Glennan, NASA Administrator from August 1958 to January 1961.

Figure 3-4.-James E. Webb, NASA Administrator from February 1961 to October 1968.



Figure 3-5.-Thomas O. Paine, NASA Administrator from March 1969 to October 1970.

Figure 3-6.-Hugh L. Dryden, NASA Deputy Administrator from October 1958 to December 1965.



Figure 3-5.-Thomas O. Paine, NASA Administrator from March 1969 to October 1970.

Figure 3-6.-Hugh L. Dryden, NASA Deputy Administrator from October 1958 to December 1965.


Figure 3-7.-(Left to right) Robert C. Seamans, Jr., Smith DeFrance, and James E. Webb.

Figure 3-7.-(Left to right) Robert C. Seamans, Jr., Smith DeFrance, and James E. Webb.


Figure 3-8.-Wernher von Braun (left) and Raymond Bisplinghoff.

Figure 3-8.-Wernher von Braun (left) and Raymond Bisplinghoff.


Figure 3-9.-(Left to right) Homer E. Newell? James E. Webb, and Abe Silverstein.

Figure 3-9.-(Left to right) Homer E. Newell? James E. Webb, and Abe Silverstein.


...Center officials feel that the Agency may be working toward a "bureau" type of organization and are uncertain about how they should act and react to their different and multiple "bosses," and the extent to which they may be held responsible by Program Directors vis-a-vis top management.... Waste is inevitable ... if Program Directors can delegate or withdraw program authorizations to Centers without serious consideration of the related impact such actions may have on manpower, facility, and related general resource requirements.18

At bottom, the problem was a management structure that was "inadequate and unrealistic."

At the semiannual management conference held that October at Langley, several center directors, especially those under the Office of Space Sciences, complained that the various organizational changes had made their work much more difficult. While the centers under Holmes had been given much greater operating authority, Goddard and JPL found their positions worse than before. The internal communications problem cut across all other organizational levels: It affected relations among headquarters and the centers, the program and functional offices, and the centers themselves. As Harry J. Goett, Director of Goddard, pointed out, five centers were engaged in designing and fabricating spacecraft with almost no exchange of information between them.19 Headquarters was too eager to involve itself in relations with contractors, too slow to approve projects submitted by the Centers, and too reluctant to encourage intercenter coordination. Furthermore, the [42] center directors had almost no direct contact with Seamans, although the ostensible reason for having the centers report to him was precisely to foster such contact. Changes in organization inevitably affected the ways in which programs were carried out-whether the approval of contracted advanced studies, the management of approved projects, or the center negotiations with contractors.

In summary, the 1961 changes failed for three reasons. They tended to create a "free for all" between the program offices at headquarters and the centers. The headquarters program director had authority only over specific, discrete projects.20 "Often a center was working on projects in several program areas. Any one field center could be involved in projects under the supervision of all five Headquarters program offices." 21 This tended to obliterate each center's orientation toward its specific mission. Moreover, the role assigned to Seamans was extremely demanding; it left him little time to visit the centers, confer with project managers, or grasp what was going on in the field. The problem was how to relieve him of part of this crushing workload and free him to take a broader view of his responsibilities. Reducing the scope of his office might be too drastic; delegating his authority, although less disruptive, would be easier to recommend than accomplish.22 Finally, until 1963 there was no set pattern for top management to meet with the program directors and functional staff to discuss operating problems as they arose. To be sure, there were semiannual conferences organized around some central theme, staff meetings chaired by Webb, and program reviews chaired, from time to time, by Seamans. But no real format had evolved by which Webb, Dryden, and Seamans could grasp and evaluate program developments before they advanced so far that change directed from above would be difficult to carry out.

Yet it would be too facile to interpret the 1961 changes as part of an experiment that could not be made to work. The arrangement under which the centers reported to Seamans rather than to the program directors was probably never intended as a lasting solution. Seamans would later remark-although he may have been wise after the fact-that "we realized that it would not be desirable to maintain this structure as a lasting arrangement."23 It had served its purpose reasonably well in reminding the centers that NASA had a single mission to which all local interests must remain subordinate. The 1963 reorganization was in no sense a return to the pre-1961 state of affairs, although the centers once more reported to the heads of the program offices. That NASA was once more "decentralized" signifies almost nothing, unless one adds that there were now central functional offices for Defense Affairs, Public Affairs, and Industry Affairs, to which NASA's Procurement and Supply Division (once part of the Office of Administration) was moved; for the first time a Technology Utilization Division reporting to Webb was set up to work out means by which the technical byproducts of space research and development-"spin-off," "fall-out," "second-order consequences," or "technology transfer"-could be disseminated most effectively to private industry and other users. In itself this was a major undertaking, involving identification of useful technology, evaluation of its potential, support of [43] research on technology transfer (often through grants to universities and research institutes), and matching of data collected with potential users.24

The 1963 changes marked NASA's arrival at organizational maturity (figure 3-10). The significant change, under which the centers once again reported back to the program offices, had been anticipated in October 1962 when Holmes was named Deputy Associate Administrator for Manned Space Flight Centers. What was new was the role envisaged for the program directors, now the associate administrators for their several offices. As Webb put it several years later, the purpose of the reorganization was to emphasize that the director "was a guy running his show as an Associate Administrator, and that he ought to think of himself as nearly as possible doing the total job. He had to present his program to Congress-he wasn't just an internal manager.... It was the idea that for his area he had almost as broad responsibility as the Administrator, subject to supervision and . . . to evolving his own relationships with functional staff and line staff." 25

As it turned out, this never occurred-at least not to Webb's satisfaction- and it is worth considering why. T he enhanced power of the program directors was, for Webb, only one side of the equation. The other side was the concept of "functional management," which, although not formulated in a comprehensive policy statement until 1966, had been official NASA policy since the 1961 reorganization.26 Basically, Webb tended to distinguish between program and functional offices, between those who prepared and carried out substantive programs and those who provided centralized, agencywide services in a specific professional discipline. The functional staff offices, most of which reported to Seamans, had two responsibilities: to serve as a central staff for the three top officials and "to emphasize the agency-wide importance of a particular area of specialization," that is, to check the tendency of the program offices to create their own supporting groups with their own parochial interests.27 But this never worked out. Five years later, Webb was complaining that the heads of the program offices had not reached up to use the central staff; that each of the three program associate administrators-as of 1963 George Mueller in OMSF, Homer Newell in the Office of Space Science and Applications (OSSA), Raymond Bisplinghoff in OART-tended to work in isolation from the other two; and that while the three were to have been ''associated with the triumvirate at the top to serve as an overall group devoted to getting the total NASA job done," this (he immediately added) "never really happened." 28 The program offices tended to build up their own functional staffs at the same time that the central staff offices were discovering the limits within which they could impose uniform policy, especially at the center level.

Three other features of the 1963 reorganization should be noted. The first entailed a shift in emphasis, a return to the NACA concept (in a very different context) of giving the field installation responsibility for technical decision making Top management stressed, then and afterward, that project management was the responsibility of the centers. For all flight projects except Apollo, there was to...



Figure 3-10. NASA organization chart, 1 November 1963.

Figure 3-10. NASA organization chart, 1 November 1963.


[45] ....be one lead center, regardless of how many installations actually participated. In the case of Apollo, the major elements were given to lead centers: for example, the spacecraft to Houston, the launch vehicle to Marshall, and the tracking system to Goddard. The tools for getting the job done had to be grouped in a related fashion; thus Applications, which used the same launch vehicles and centers as Space Sciences, merged with the latter in 1963. This meant that a particular center had (or was assumed to have) the capacity to manage large development contracts, the skills to integrate the subsystems of a project parceled out among two or three different centers, and the ability to draw on the resources of other centers instead of needlessly duplicating them. In 1963 even more than in 1961, there was an agency "program" instead of "programs," hence the need for the cross-servicing of one center by another center. (Only in 1963 was NASA able to establish a unified Launch Operations Center at the Eastern Test Range, where previously each center had its representatives-a Goddard team, a Marshall team, a JPL team. The Launch Operations Center, headed by Kurt Debus, was renamed the Kennedy Space Center in December 1963. 29

This system was not self-regulating. It depended on a mixture of formal delegations of authority and a network of informal relations too subtle to be put on paper; neither was fully understood throughout the agency. Under these circumstances the general manager really had to be prepared to manage. He had to know what was going on before problems got out of control. He had to make himself visible to the heads of the program offices, and he had to be able to say, "'Well, we need more data on this,' [or] 'this is something we can't decide here, it must go to the Administrator."'30

The second feature of the 1963 reorganization was to create a mechanism that enabled the general manager to do this. The NASA Management Committee, which met for the first time in October 1963, was chaired by Seamans and comprised nine key headquarters staff officials reporting to him. Its terms of reference extended to the discussion and resolution of management problems-relations with the centers, procurement policy, the format for project authorization, and so on.31

At the same time, Seamans instituted an intensive "monthly status review" at which he would sit down in turn with each of the program associate administrators for a searching examination of each project for which he was responsible. Such reviews would cover all substantive and administrative aspects: planned versus actual manpower allocations at the centers and at contractor plants, planned versus actual expenditures, key milestones in program and procurement schedules, and advanced studies prior to their completion. Each review was to be preceded by staff discussions, and the preparation of a formal agenda by the Office of Programming would be followed by presentations to Seamans. The drift of Seamans' thinking was to make short-range planning more realistic by seeing that the program offices provided him with the information on which key decisions had to be made. Consider for example the official monthly launch schedule, which was the Associate Administrator's responsibility to update. The changes made in 1963 [46] were in the direction of greater realism and precision. All launches scheduled within thirty days of a Management Committee meeting were to be scheduled by day, those within the ensuing year were to be specified by month, and launches planned for the following year were to be specified by quarter. Seamans wanted to eliminate unnecessary reporting, while insisting that the program offices assist him in making launch dates as realistic as possible.32

On balance, the system of status reviews worked well. The principal recurring meetings-the monthly status reviews chaired by Seamans, the internal reviews of the program offices themselves, the annual program reviews attended by Webb-did aid in bringing problems to the surface. In the annual reviews, for example, specific program areas were reviewed in depth from the standpoint of their overall objectives, scientific and technical content, organizational structure, and interrelationships with other Government agencies. The meetings were held on Saturday and lasted all day; the following Monday the presentation was repeated for NASA staff and senior officials of other departments. But shortcomings remained. The monthly status reviews covered only substantive programs, omitting the work of functional management, and the presentations for Webb outlined each program in a fragmentary way, since the programs could only be covered a segment at a time. Still, the 1963 reforms were a definite improvement over their predecessors. The Management Committee, the monthly reviews, and the annual reviews of each program gave Webb and Seamans access to information and views that had scarcely been tapped earlier. The meetings enabled top officials to use overlapping sources of information, hear all points of view, eliminate some middlemen in channeling information upward, and define the purpose of each meeting so that those attending would know precisely why they were there.33

Finally, long-range planning continued to be a vexation and a burden to top management. By early 1963 Webb had decided not to update the Ten-Year Plan; in October the Office of Plans and Program Evaluation, which had played an equivocal role-not being quite in on decision making nor entirely out of it-was abolished, to be replaced by a Policy Planning Board reporting to Webb and a Planning Review Panel attached to Seamans' office.34 Neither had effective authority, presumably because top management preferred not to give them any. It was one thing to bring officials together to discuss NASA programs in a general way; it was another, much more serious step to empower them to speak authoritatively for NASA. Planning could be no more effective than its organizational location and status allowed. Webb was reluctant to propose a long-range plan, not solely from a conviction that NASA's long-range mission could only be spelled out in the political arena, but also because he did not want to declare preferences that could set one part of the agency against another. The so-called plans of 1965- 1968 were more in the nature of shopping lists (most of them prepared by OMSF or OSSA staffs) than actual outlines of what NASA intended to do; the Policy Planning Board was abolished in 1965 because, as Webb disingenuously explained, nobody was using it.




The years from November 1963 to November 1965 were not marked by changes as sweeping as those already discussed. Top management, the heads of the line and functional offices, and the center directors had worked out a system they could live with. Two successive reorganizations had shaken up the agency, spotlighted the values top management deemed important, and reduced the inertial forces of custom-"this is the way things have always been done around here"- that might form in an organization that had inherited so much from its predecessor. But it was not part of Webb's philosophy to keep NASA in a state of, as it were, permanent revolution. Smaller (although not small) changes would serve just as well. As he explained to one official, top management consciously decided "to make relatively small changes on a fairly frequent basis. Thus we . . . have used our decisions on incremental improvements to teach the organization to expect change rather than contemplate static periods after a series of major changes."35 In his view the 1963 changes, if carried to their logical conclusion, would free himself and Dryden to concentrate on the major issues, particularly where they impinged on other agencies, Congress, and the White House. But it was absolutely essential that NASA officials understand their roles in what Webb called "the total milieu." As a matter of policy he believed in transferring headquarters officials to the centers and vice versa, assigning executives jobs outside their experience and beyond their area of proven competence, and using NASA officials to present their programs before congressional subcommittees. For all the shifts and turns of the NASA organization during Webb's tenure, there is a definite consistency of intent toward reducing the layering of authority so that fewer officials and fewer documents would be needed to authorize projects, and toward bringing as many senior officials as possible into decision making. The goal being posited, the means of attaining it had to be worked out on the basis of periodic reviews, staff meetings, and presentations to senior officials-a system at once more rigorous but less formal than a Ten-Year Plan.

Two examples of "relatively small" changes at headquarters are the revision of the system of management instructions and the formal establishment of the executive secretariat, both begun in 1964 and extending over several years, both closely tied to the major changes of the previous October. The so-called issuance system had been a longstanding grievance.36 The Management Manual instituted in 1959 could not adequately document the changes taking place in NASA. Often, there were no instructions (or occasionally, conflicting instructions) covering a subject area. Instructions tended to lose relevance within two or three years of publication but not to the point of becoming totally obsolete. Added to this, the coding scheme was inadequate, there was no indexing or cross-referencing, and definitive instructions took too long to prepare. These problems, recognized by 1962, had become acute following the 1963 reorganization that, in delegating maximum authority to the program offices and the centers, called attention to the gap between the directives and the purposes they were intended to serve.

[48] One of the first steps in revising the Manual was taken early in 1963, when the Division of Management Analysis (Office of Administration) conducted a study based on an earlier report prepared by the General Services Administration at NASA's request. These two studies, the former of which was discussed by the Management Committee in April 1964, served as the basis for subsequent revisions, but converting from the old Management Manual to the new directives system-entailing as it did the rewriting and converting of current directives, canceling others, and updating the remainder-took another two years to complete.37 The job of revision included several sorts of change: eliminating what was extraneous, issuing a new coding system to distinguish ad hoc notices from policy directives and the management instructions that supplemented them, and the preparation of a checklist to make cross-referencing possible. A particularly important step related to these reforms was the publication of "NASA Basic Administrative Processes" in February 1964.38 This manual, complete with flow charts, official definitions, and enumeration of basic policies and principles, established the guidelines to be followed by agency officials and, as such, was the substratum from which the issuance system was supposed to develop.

The history of the management documentation problem is interesting, less for its own sake than for what it reveals of the workings of the NASA management process. For brevity's sake, the important points may be summarized as follows. First, the changes of 1964 through 1966 followed logically from the decentralizing tendencies of 1963. In line with management's policy of delegating and dispersing authority, the new system authorized all officials reporting to top management (and, of course, top management itself) to prepare and to sign directives outlining their powers and responsibilities. It gave the functional offices the authority to determine the content, prepare the format, and coordinate drafts with other offices. Second, while this relieved general management of the responsibility of preparing all the detailed instructions, the functional offices were given the onerous burden of preparing directives that might affect nine or ten other offices, as well as the centers.

Finally, the new system failed in its purpose almost as much as its predecessor. NASA did not stand still, and no set of instructions or directives could possibly serve as a contemporary record of operating procedures. At the Office of Administration, Young tried in vain to satisfy Webb's demand for precise, up-to-date descriptions of every responsible person's job.39 "'Basic Administrative Processes" was already out of date by the end of 1966; it could not be replaced until it had been studied by one task group, worked over by the Office of Organization and Management (see below), discussed by the NASA Management Council in January 1968, and subsequently coordinated between Organization and Management and the newly revived Office of Management Development. In the end, agency officials found it impossible to summarize the processes between the covers of a small handbook. The compromise effected was to replace the handbook one chapter at a time-a process that was never brought to a formal conclusion.

[49] The creation of an executive secretariat was as much the result of the 1963 reorganization as the enhanced authority of the program offices was. Although discussed for some years prior to its establishment, the concept of a central secretariat for handling communications made little headway prior to 1963.40 It was not mentioned in the December 1958 McKinsey report on establishing headquarters functions, nor in the Kimpton report of October 1960. The 1961 staff papers prepared by Deputy Associate Administrator Jack Young did mention and elaborate on a "Central Secretariat," but no such staff office was included in the changes that followed. The real catalyst for change-another staff study supervised by Young-came toward the end of 1963, and the management instruction establishing the secretariat was only issued on 1 February 1964.41 Why it should have taken so long to recognize the need for a secretariat function in NASA is unclear. Prior to 1964 the responsibility for handling communications between top management and the rest of the agency had been delegated to an "Executive Officer," usually an officer detailed to NASA from the Army Corps of Engineers.**** 42

The creation of a secretariat did not immediately resolve the several administrative problems that had led to its creation. It was by no means clear whether the executive officer would be inside or outside the secretariat (ultimately inside); whether the secretariat would serve the Administrator alone, or the Deputy and Associate Administrators as well (with qualifications, the latter); whether its function was completed when the Secretary informed the Administrator of what he needed to know, or whether the secretariat had an additional responsibility to see that decisions made by the Administrator were carried out promptly (the office became responsible for both). The functions of a secretariat really depended on the Administrator's view of his own position vis-a-vis the rest of NASA. To a degree, Webb's concept of a secretariat was not radically different from the one held by the heads of other large Federal agencies, particularly the State Department and DOD. Webb needed and wanted a staff to handle all materials coming to his office, sparing him matters that did not require his personal attention, while placing before him everything that he had to know to meet his responsibilities. Webb's knowledge of secretariat functions was strongly influenced by his experience as Under Secretary of State; the State Department secretariat was manned by officials who frequently reported directly to the Secretary or Deputy Secretary of State.43 Webb envisaged a NASA secretariat that would assist in the making of-more precisely, in the implementation of-agency policy, but it still remained to be seen how it would act in following up decisions and in working with and through the headquarters offices and field installations.44

At the end of 1965 the secretariat's charter was conspicuously enlarged as part of the much broader changes brought about by Dryden's death on 2 December. In failing health since 1961, Dryden had nevertheless managed to carry on with few interruptions for the next three years. By late 1964, however, he had [50] become incapacitated to the point that his resignation or death was anticipated. Before Dryden's death, it had been informally decided that Seamans would succeed him as Deputy Administrator. What this would mean did not become clear for several months. During 1964-1965 Webb had started moving toward a more precise definition of his and Seamans' respective spheres. With Dryden gone, Webb decided to pull together into one unit functions and offices that had hitherto remained separate. In a memorandum of 29 December, which Seamans discussed with all top headquarters officials the next day, Webb announced the creation of the Office of the Administrator as a "single, uncompartmented entity to afford maximum, direct . . . contact between Dr. Seamans and myself and our associates."45 However, Seamans cautioned, this concept was "experimental in nature" and was designed to "assist in eliminating functional delays." 46 But the creation of the Office of the Administrator was intended to do more than this, and it would be well to summarize here the kind of organization Webb had in mind.

First, Seamans remained general manager. That is, he retained his function of overseeing the agency daily. He would continue to chair the monthly status reviews, update the NASA flight schedule, sign project authorization documents, and so on. But his role as general manager did not exhaust his responsibilities. Webb wanted him to move into several of Dryden's functions, for example, to involve himself in international scientific programs. He especially wanted Seamans to work closely with the Offices of Legislative Affairs and Public Affairs in preparing NASA's congressional statements and in meeting with those Congressmen who had most to say in determining NASA authorizations and appropriations. Seamans had been doing much of this before; he was now expected to carry it to the fullest extent. All staff and program offices (with certain exceptions shown in figure 3-11) would report directly to Seamans.

Second, Webb intended to make the secretariat visible throughout the agency. As the staff arm of the Office of the Administrator, it became responsible not only for reviewing incoming correspondence but for establishing and maintaining "a need-to-know reporting system of items of significance to general management . . . a Critical Reports and Correspondence Review System to keep general management currently informed of significant matters covered in written reports and communications by the heads of Headquarters offices" and "a Codified Policy Reference that will provide for recording policy directives . . . and for making them available to appropriate officials."47 By what he said and did, it is clear that Webb intended the secretariat to have a de facto role in policy making. It became the Secretary's responsibility to prepare the agenda for management reviews, to attend major staff meetings (for example, those attended by Mueller, Seamans, and Webb) and to prepare a memorandum for the record afterward, and, through his assistants, to obtain copies of the internal papers of all the program offices (including OTDA, which was elevated to program office status in December 1965). But the secretariat could assist in policy making in more direct ways. The Executive Secretary could, for instance, remind each of the program offices to coordinate an impending action with the functional offices; he could say in so many....



Figure 3-11. NASA organization chart, 2 January 1966.

Figure 3-11. NASA organization chart, 2 January 1966.


[52] ...words, "Have you seen the General Counsel? Have you talked with Procurement?" And he could coordinate the policies of the program offices with each other. It sometimes happened that two offices, ignorant of what the other was doing, would prepare separate policy statements on the same matter. The different approaches of OSSA and OMSF on the quarantine of outbound spacecraft converged only at the level of the Executive Secretary, who intervened to bring the matter to Seamans' attention and to gain a final resolution.48 Obviously, a great deal depended on the program offices' willingness to cooperate, for example, to provide the secretariat with copies of internal correspondence, to allow assistant executive secretaries to sit in on their staff meetings, and to tolerate another level between themselves and general management.

These changes were meant to reduce still further Webb's burden of routine business. Seamans was now completely responsible for the internal management of NASA and for many of its external relations as well. Moreover, even those groups lodged in the Office of the Administrator did not always report directly to Webb. The Executive Secretary, for example, reported to Willis H. Shapley, who had joined NASA in September 1965 as Associate Deputy Administrator. Shapley came to NASA from the military division of the Bureau of the Budget, where he had been involved in space policy making since NACA days. He was one of the drafters of the Space Act and had worked closely with Seamans to resolve issues between NASA and DOD. Arriving just before Dryden died, Shapley had first been assigned responsibility for interagency relations. But following the December reorganization, it was Seamans' notion to work out a three- or four-way division of labor among himself, Webb, Shapley, and Earl D. Hilburn, the Deputy Associate Administrator. In the broadest terms, Webb and Seamans would be the top two officials running NASA, and Shapley and Hilburn were to run the agency on a day-to-day basis. According to Seamans, "Mr. Webb and I would have available to us two senior people, and you wouldn't have a very definite rule as to what Shapley did or what Hilburn did, but a general area defined, and then I would work very closely with both of them." 49 In that sense, Seamans' plan never got off the ground. Webb would not buy it, preferring to have one person responsible for a given area. He saw to it that Shapley was placed in immediate charge of the secretariat, with a kind of overall responsibility for the "policy planning" function, to which the entire agency would be expected to contribute. In sum, Webb was delegating his authority for daily operations in the hope that he could totally involve himself in the key problems facing NASA.

This account rather understates the complexity of what Webb had in mind. At one end, he would delegate authority to Seamans and Shapley (Hilburn having left in the summer of 1966) to run NASA on a daily basis; at the other end, he was delineating a sphere of policy planning that would be his own bailiwick. This division of authority could never be laid down with any finality because no one knew precisely where "management" gave way to "policy planning" or "policy analysis." Either might be used to mean the determination of whether previously established policies were being carried out, the documentation of specific policy [53] decisions' or the study of how "programs are put together year by year, not only for the 5 year future, but 10 years and beyond." 50 Lack of definition was only one of Webb's problems. Another was that the key officials were not-nor did Webb intend them to be-just one thing or the other. In his view, Seamans was not just the general manager any more than Shapley was just the chief of staff or Mueller was just the Associate Administrator for Manned Space Flight. They had to be able to grasp the "total milieu," to understand what the agency would be years later, when the compartmentalization of manned and unmanned programs had broken down, when NASA was developing a technical capability that would enable its officials to "know enough to know what ought to be done in the world," and when Webb or his successor would be involved as "exponent[s] of advanced technology and advanced scientific thrust and the combining of the two for everything from agriculture to economics and military."51 To the engineer or program manager working full-time on Apollo Applications or Voyager, Webb's demand that he see the total picture was a very tall order indeed.

On balance, the changes of 1965-1966 accomplished a great deal without disrupting the existing roles of the program line management. Webb had introduced the concept of a central staff to serve the Office of the Administrator, whose chief of staff was Willis Shapley, had created an executive secretariat to handle internal communications flowing up to his office and policy directives moving downward, and had delegated to Seamans most of the authority for running NASA. None of these changes had anything final about them; rather, Webb was attempting to develop an agency that could run itself on a daily basis, while he, Seamans, Shapley, and the four program directors worked to extend NASA's influence into economic and political spheres hitherto untouched by the space program. The question was, would the agency accept this interpretation of NASA's mission? For that matter, would Congress understand it? And finally, how far was it within NASA's competence to determine its mission, since it had to go outside for the funds to sponsor it?




James Webb:

What I want you to realize is that you are not on a nice ship steaming across the ocean at a high speed with the flags flying. You are sort of on a raft that is partly at the mercy of the currents and . . . you are going to keep your feet wet and have a real hard time getting all of it done, but you must consider the total environment within which you operate.

Edgar Cortright:

I don't think that is quite a fair analysis.

James Webb:

It isn't. I agree with you. It is exaggerated for eflect.52


The year 1967 was probably the most eventful for NASA since 1961. Although the Apollo fire and its aftermath was the dominating factor for months, that year also marked the completion of Gemini, the first launching of the Saturn V, the cancelation of the Voyager program for the exploration of Mars, further reductions in the NASA budget, a Civil Service Commission report that was highly [54] critical of NASA personnel management, and a General Accounting Office investigation of NASA support service contracts. Even without the Apollo fire and the ensuing publicity, NASA officials would have had reason to worry about the future. Questions asked at congressional hearings had a disturbing way of recurring. What was NASA doing about post-Apollo planning? Why did NASA spend comparatively little on aeronautics and applications on the one hand and so much on manned spaceflight on the other? How could NASA guarantee that it could handle its contractors instead of becoming their captive? By the end of 1966-one month before the fire-Webb and Seamans were sufficiently worried about the NASA management structure to ask Harold B. Finger to head a task force to study the full spectrum of functional management with a view to restructuring it. He was invited to recommend how changes might best be implemented, how they would affect the program offices and the centers reporting to them, and whether they should be phased over time.53 By February, Finger was preparing the charter for an "Office of Management," and on 15 March Webb named him the Associate Administrator for Organization and Management-ushering in the last major reorganization of the 1960s.

Webb chose Finger because he had succeeded as a research and development manager; he was the very type of line operator Webb wanted to bring from the field to headquarters to understand the agency as a prelude to running it. As director of the NASA-Atomic Energy Commission Nuclear Propulsion Office since its establishment in 1960, Finger had done remarkably well in handling not only the usual range of project management problems but the added burden of working with another agency, with the Joint Committee on Atomic Energy, as well as with Senator Clinton Anderson (D-New Mex.), a member of the Joint Committee and chairman of the Aeronautical and Space Sciences Committee, which authorized the NASA budget. Finger had the experience and the ability to win the respect of those program offices with which he would have to work to restructure the agency in the aftermath of the Apollo fire.

The changes of 15 March can only be summarized here. As figure 3-12 indicates, the following functional offices would now report directly to Finger: the Office of Administration headed by William E. Lilly; the Office of Industry Affairs under Bernhardt L. Dorman; and the Offices of Technology Utilization and University Affairs headed, respectively, by Richard Lesher and Francis B. Smith. The other major change, by which Wyatt (who reported to Seamans) became Assistant Administrator for Program Plans and Analysis, was perhaps just as significant; in the process, the budgeting and programming functions previously under him were transferred to the Office of Administration (which reported to Finger) as part of an agencywide system "for resources management, including programming, budgeting, personnel management, and financial reporting."54 Within the year several more offices were brought within Finger's control: the Office for Special Contracts Negotiation and Review, established in May to monitor certain important contracts, particularly the North American Aviation contracts for the S-II stage of the Saturn launch vehicle and for the Apollo [55] command and service modules; the Audit Division; the Inspections Division; and the Headquarters Administration Office. By the spring of 1968, the role of management and supervision had passed from Seamans, who left NASA at the beginning of January, to Finger (figure 3-12).

In certain crucial respects, however, Finger's assignment differed from Seamans' Finger was brought in because self-policing had not worked and because, in Webb's view, there had been too much emphasis on programs and not enough on administration In the past, he argued, nobody minded how the job was done so long as it got done. The Apollo fire and the attacks by Congressmen hitherto favorable to NASA changed all that. The agency had been caught unprepared. Webb wanted the job done, but done properly and through the approved chains of command. The management instruction that established Finger's office stated that "all officials with leadership and executive responsibility are expected to accomplish agency programs . . . within the prescribed systems and procedures." Furthermore, it would be Finger's responsibility to develop criteria "for selection upward of R&D personnel who also demonstrate administrative competence . . . and the transfer to non-administrative specialties of those who cannot grow in both program and administrative areas."55

In private Webb was even blunter. Finger's office would be to the rest of NASA what the Bureau of the Budget was to the Federal structure.


We gave this Office of Organization and Management police authorities over the system. We say, "You've got to prescribe the system, you've got to monitor the system, you've got to audit performance under it, and these fellows can't get the money to go forward without you." . . . I am giving them real teeth. I am saying to Harry [Finger], "If these fellows don't satisfy you with respect to the components to the system, cut off the water. Don't give them any money." He's got the authority to allocate the money, he's got authority to issue in his own right a modification of a project approval document and say, "Boys, you used to think you're going to do it this way, and even when you don't like what I'm proposing, here it is-signed. Go do it this way." 56


Along with the Office of Program Plans and Analysis and the secretariat, the Office of Organization and Management would provide Webb with multiple layers of feedback; under Finger's leadership it was expected to penetrate almost every facet of administration. Now the heads of the program offices had to work through Finger in preparing and seeking approval for their project approval documents (PADs); that is, Finger was expected to certify to the Administrator that a draft PAD was sound-for example, that it not only had an updated launch schedule attached but also a reference to the research and ongoing projects that should feed the project. But Finger and his assistants were expected to do more than this. During 1967 and 1968 they planned and carried out several important studies of NASA management processes: an April 1967 report on the "functional review process," that is, the self-evaluations and internal reviews of the functional Offices themselves; a task force review of those "action and decision processes" leading to mission assignments and approved PADs; and the preparation of detailed guidelines for phased project planning. The power to review, inspect,....



Figure 3-12. NASA organization chart, 15 March 1967.

Figure 3-12. NASA organization chart, 15 March 1967.


[57] ....establish management systems, deal with the General Accounting Office, and allocate resources for research and development made Finger the most powerful staff official at headquarters. Reporting directly to Webb, a member of the NASA Management Council (see below), and the conduit through which all PADs had to pass before arriving on Webb's desk, Finger was both line and staff.

The creation of the Office of Organization and Management was only part of a larger strategy, a wider frame of reference. It assumes importance only in relation to the other functions and roles that comprised "NASA management." Where the three top officials of 1961-1965 had been able to substitute freely for each other, the key officials at headquarters after 1967 were charged with broad functional responsibilities delegated by the Administrator. Finger was, so to speak, both inspector-general and administrative housekeeper. Shapley, whose office was "subject to the overall administrative system established . . . by Mr. Finger," was once more in charge of the offices clustered around external affairs, especially relations with Congress and DoD.57 Newell, who became Associate Administrator in September 1967, was charged with organizing and coordinating NASA's long-range planning.58 At the same time, the heads of the program offices were expected to become more like group vice presidents; the more they had to deal with one or two officials like Finger and Shapley, the more they were to shoulder responsibilities for getting their work done. And that work was principally coordination: coordinating their long-range planning with Newell, their project planning with Finger, their congressional presentations with Shapley, and their review of management processes with each other, "as if each were the Administrator having to make decisions concerning and affecting all areas of the Agency's program and operations."59

The nature of the administrative changes in NASA from the beginning of 1967 to Webb's resignation on 7 October 1968 may be summarized as follows. First, Webb had a penchant for pouring new wine into old bottles-he liked to retain the name while changing the format. Thus the Office of Management Development, established in 1964 to enable one of his consultants to do some troubleshooting, was revived in February 1968 for the totally different purpose of reviewing all NASA management documentation.60 Similarly, the Office of Programming became the Office of Program Plans and Analysis and shifted from the preparation and validation of the NASA budget to long-range studies (see figure 3-13 for the NASA organization as of 1968). Such changes tended to emphasize continuity within the NASA administrative system. Indeed, most of the changes at headquarters after 1965 represented a shift in the functions of existing units rather than the addition of new ones.

Second, the changes of 1967-1968 were supposed to lead to group participation in decision making. One of the ways of effecting this was to have the top officials at headquarters meet together periodically for discussion and task assignment. The concept was nothing new: Seamans' monthly status reviews and Webb's attendance at an annual review of each program have already been mentioned. There were also various in-house reviews by the program offices (and....



Figure 3-13. NASA organization chart, 1 May 1968.

Figure 3-13. NASA organization chart, 1 May 1968.


[59] ....often attended by DOD representatives) and, from 1964, annual reviews of advanced mission studies in which each program office discussed its study plans for the upcoming fiscal year. But in line with the administrative fragmentation of 1967-1968, new panels, committees, and councils were set up to focus on NASA-wide problems. Webb established a Management Advisory Panel composed of NASA officials and part-time consultants to serve as a link with the public administration community.61 As Associate Administrator, Newell chaired the NASA Management Council, which, established in January 1968 and attended by representatives of all the program and functional offices, was supposed to evolve into a top management team sharing responsibility, that is, to assume the functions of a general manager like Seamans.

But the most interesting changes were those involving the program reviews. Each of the reviews suffered from serious flaws. The Administrator's reviews presented each program a segment at a time, and the monthly status reviews mainly covered substantive programs, although Seamans did attempt to include functional managers within the system. Neither considered programs as entities. Worse, none of the meetings made it possible to take a hard look at the total NASA program; none distinguished between program office reviews, which would naturally concentrate on the technical aspects of each project, and top-level reviews, which would examine a broader range of management problems. Such shortcomings in the review system had been aired at several meetings, notably one chaired by Seamans in November 1967. On 11 March 1968 Newell announced that the "'project status reviews" would become "general management reviews" to be attended by all key headquarters officials. The important change was that, instead of dealing with one program at a time, the reviews would "provide a regular forum for the presentation of issues, problems, and policy questions" involving more than one element of "the General Management team." This was not to be an arena for decision making as such. Rather, policy would be discussed "in such a manner that when formal approval channels are used on that particular item, there is a clear understanding by all organizations involved of the nature and substance of that particular issue." 62 The emphasis had shifted from the predictable to the unexpected, from what concerned one office to what affected all. The NASA Management Council would stress the responsibility that officials had both to make policy and to justify what they had done.

Third, the changes only intensified the fragmentation of authority at headquarters. No one below the Administrator's level could even pretend to see the agency as a whole. In particular, the responsibilities for planning and general management were now parceled out to separate officials. The three-man team, each member of which was prepared to substitute for the others, had been succeeded by interlocking councils and panels and by units like Organization and Management that were not so much "offices" as conglomerates of smaller units. Take, for example, the way in which policy planning was handled from 1967 on. Earlier, "policy planning" and "policy analysis" were shown to be thoroughly ambiguous terms. If they were understood to include long-range planning, then [60] no office and no official could really claim full responsibility for that activity. This was as true of the Planning Coordination Steering Groups over which Newell presided as it was of the analytical staffs within the program offices, the Office of Policy reporting to Newell, or Wyatt's Office of Program Plans and Analysis, which pulled together the so-called program memoranda and special analytical studies required by the Bureau of the Budget as part of the annual budget submission. Long-range planning admittedly represents a rather extreme case, and the difficulties in preparing a long-range plan were not only the result of fragmentation. Both the separation of planning from operations and the incomprehension of NASA officials as to the kind of plan to be drafted stemmed from the same source: The key officials, each preoccupied in his own sphere, could not be the general managers that Webb demanded they become.

Fourth, functional management tended to run counter to broader interests in two ways, one relating to the contact between the functional offices and the centers; the other, to the use of the central functional staff by the program offices. In the former case, functional offices like Procurement or Administration tended to work directly through their counterparts at the centers without first consulting with the program offices. As to the latter case, there is evidence showing that the program offices did not use the functional offices to anything like the extent that Webb had anticipated. At a meeting of the Management Advisory Panel in the spring of 1968, he complained that the program directors "didn't reach upward to use the general staff and functional staff in the way that those of us at the top reached down to use them . . . each one began to have around him his own people . . . [and] tended to be autonomous, run things his own way, and you didn't find him reticent to come up when he had a problem with a Senator or Congressman to get help at the top." 63

Finally, the new organizational philosophy failed to "take," as new men and problems appeared. Seamans left the agency at the beginning of 1968; Webb resigned that October, to be succeeded by Thomas O. Paine who had been appointed Deputy Administrator a few months earlier; and Finger became Assistant Secretary of the Department of Housing and Urban Development in April 1969. A number of new faces appeared at the end of 1969: George Mueller, the Associate Administrator for Manned Space Flight since 1963, returned to private industry, to be succeeded by Dale D. Myers, general manager of North American Rockwell's space shuttle program; George M. Low, the Apollo spacecraft manager since April 1967, became NASA's Deputy Administrator.

By 1970-1971 the organization painstakingly created by NASA general management during the 1960s was changing into something else. The various offices-Industry Affairs, University Affairs, Administration-no longer reported to Organization and Management, which would be dissolved in 1975. Similarly, the various "policy" offices and panels kept going with ad hoc assignments and played no central role in agencywide planning. As a third example, phased project planning, by which each project was broken into discrete sequential steps, was drastically modified in 1972 to permit the program offices to determine procedure [61] on a case-by-case basis In sum, the concept of group management, which was the heart of Webb's philosophy, was never widely shared by the program offices, the functional offices, or the centers




This account of headquarters organization may seem to have answered every question but one: What did top management do ? To answer this without having first reviewed the organizational cycles within headquarters did not seem very useful. But having reviewed the organizational changes, what seemed simple and definable is now hedged with qualifications, and the clear patterns of the organization chart must be replaced by complex interlocking networks of formal and informal relationships Even to assert that the Administrator was responsible, for example, for representing NASA to the rest of the Federal community ignores the fact that the program associate administrators were as much NASA spokesmen in their areas as the Administrator was for the agency as a whole. One must also be aware of the extent to which any top official could act for the others; that Seamans, first as Associate Administrator and then as Deputy Administrator, assisted in making the policies that he carried out; or that Webb insisted on bringing in Dryden and Seamans on the most important management decisions, such as the review of all source evaluation board recommendations involving contracts of $5 million or more. Further, the way in which a function changes or evolves is as important as its official definition at the outset. Much depended on the ways in which officials came to understand their assignments: whether they chose to concentrate or delegate their powers and to increase or narrow their span of control; or whether a function empowered the official to initiate action himself or to review actions taken on some other level.

The role of top management was complicated immensely by the tension between headquarters and the field centers that was built into NASA almost from the start. Each side had its characteristic viewpoint. Headquarters, itself almost a kind of rival installation, had certain key functions that were well understood: to prepare and defend the agency budget; to allocate funds for "research and development''' "construction of facilities," and "administrative operations"; and to act as a central point of control. Beyond this, there were problems that senior management could not resolve in the short term. One was whether the centers should report to the agency's general manager, the Associate Administrator, or to the heads of the program offices. The former would appear to be the logical solution when the centers were involved in a variety of projects; the latter, when each center had carefully defined tasks distinct from those of other centers. Another problem was how centers reporting to one headquarters office could work with those reporting to another. A third was the problem of project assignment:

whether to give the entire management to one center; to split it between a number of centers and, if so, how to select one as "lead"; or whether to put the entire [62] project management team in headquarters. A fourth problem was how to convert the older research-oriented centers into managers of large development contracts. All these problems were compounded by the necessity, for top management, of taking an agencywide perspective and considering NASA in relation to variables that, in a technological sense, were extraneous to the successful pursuit of the agency mission. If a program was canceled or shifted from one center to another, project managers and center directors might be excused for seeing politics at work. Just as agency management had to defend the NASA budget as if every penny counted, so each center director would see his programs as vital to agency needs. The problem in part was one of communications. The greater the pressures of time, the faster the rate of significant change in the environment, the more interrelated the various programs, the more difficult and necessary adequate communications would be.64

So much for any simple, straightforward explanation of administrative behavior. The question of what top management did remains open. This is best answered by describing not discrete functions but broad areas of functional activities, none of which was completely distinct from the others, and all of which served to form the structure of the organization and to settle some of the more important policy issues. These activities are the establishment of uniform organizational procedures; resource allocation; review and control; and the maintenance of intra-agency checks and balances.


Establishment of Uniform Organizational Procedures

Top management, particularly the Administrator, had the prerogative of establishing the organization and procedures most useful to carry out the NASA mission. When Webb came on the scene, he found an organization young enough not to be hidebound by tradition and small enough to be changed with relative ease. This chapter has listed some of Webb's organizational changes: the creation of bureaus with agencywide functional responsibilities for procurement, budget preparation, management development, public affairs, and the like; the use of internal task forces rather than outside consultants for management studies; the establishment of the Office of the Administrator with a supporting executive secretariat; the inauguration of annual reviews of each program area; the location of long-range planning in the line organization; and the invention of stereotyped formats for management instructions, delegations of authority, and (though Glennan had done some of the preparatory work) project authorization. Because of the work of Webb and his associates, there was no need to set up an informal organization to bypass an inherited and cumbersome formal bureaucracy.

Webb's freedom to maneuver in no way minimized the problem faced by top management in selling its philosophy to the rest of the agency-a problem compounded by the changing nature of the NASA mission. In the early 1960s NASA expanded prodigiously to develop its capabilities for manned spaceflight and space science; toward the end of the decade NASA was becoming an aging research and [63] development organization where emphasis had shifted from "the development of capabilities to that of using capabilities already created to meet established needs " 65 The task of organization building thus had a dual aspect: The organization had to be able to get the job done, but it also had to be sufficiently detached from the immediate task to weather any sudden expansion or contraction in its resources. In any case, the top officials had to persuade the rest of NASA to accept their concept of how the agency should be managed. One way of doing this was to handpick key line and functional officials and to make it clear to those already on board that they must either "sign on" or look for jobs elsewhere. A second way to achieve consensus was through face-to-face encounters with other NASA officials. Unlike Glennan, Webb preferred small recurring meetings with heads of program and functional offices. These gave him the flexibility to explain policy, to get feedback from the staff, to discuss unresolved problems, and to decide whether to take immediate action or to table an item for subsequent resolution.

Webb went to great lengths to enunciate the concepts behind organizational changes. Among those mentioned thus far are the concepts of functional management, the role of program associate administrators as group vice presidents, and making major top management decisions jointly. In essence, these concepts could be reduced to the basic principles of management by exception and collegiality. The former principle means that the higher levels of management were called on for decisions "only when something extraordinary occurred in the process of executing approved projects." 66 According to the latter principle, Webb accepted full responsibility for his agency, while delegating his functions and authority piecemeal to officials further down the line. Paradoxically, the option of decentralized decision making was available only to managers strong enough to delegate responsibility.


Resource Allocation

NASA management had the responsibility for (1) allocating resources, that is, determining how much of the agency funds should go for one program rather than another; (2) reviewing proposals and providing the resources for implementing them; and (3) continually evaluating goals, programs, and the allocations themselves. What this meant was that headquarters officials-from the general manager down-worked toward a system in which each program had its place in the NASA mission and in which all programs were covered by documentation sufficient to track them from preliminary studies to formal termination. The history of NASA project authorization policies in the 1960s demonstrates that top management basically wanted two things: the ability to intervene at all key decision points in the life of a project (phased project planning) and the use of the project approval document to control the authorization and funding of every item in the operating budget. In the last analysis, the power of the purse, whether exercised by Seamans or delegated to Finger, was one of the most important internal constraints in determining the character and scope of NASA programs. [64] Also note that the power to allocate resources was also the power to say no, for example, to deny Holmes the additional $400 million for Apollo that would have had to come out of unmanned space science programs.


Review and Control

NASA management was responsible for control, defined by Sayles and Chandler as "the measurement, primarily after the fact, of the extent to which funds are being spent wisely; schedules are likely to be met; and whether all federal personnel, contracting and other policies are being adhered to by both program offices and centers."67 But to exercise control without knowing what agency units were doing was meaningless. Hence the stress placed by Webb on means of gathering, screening, and disseminating information: the establishment of an executive secretariat for internal communications; Webb's appointment of consultants as links to the Departments of Defense and State and to the intelligence community; the institutionalization of monthly status reviews and annual program reviews; and the creation of the Office of Organization and Management, one of whose functions was to police the agency and to see that each unit was measuring up to agencywide and Government-wide standards.


Maintaining Intra-agency Checks and Balances

Webb and Seamans sought to preserve a system of checks and balances within NASA. Thus program offices such as OSSA and OART forwarded mission concepts different from those of OMSF, while functional management served, supposedly, as a check on the freedom of program offices and centers to pursue their ends at the expense of the overall NASA program. Webb's policy of management by exception meant that tradeoffs between one program office and another or between headquarters and the centers had to be built into the system itself. Hence the freedom of program directors to transfer funds between projects, subject to their ability to justify the transfers at management reviews. Hence the overlapping responsibilities of the program offices, the "space science" component within OMSF, the "manned" element within OSSA, and their joint responsibility for coordinating their supporting research and technology with OART. Each office also had agencywide responsibilities over and above its specific task: OART for aeronautics and supporting research and technology, OSSA for university grants, OMSF for the development of large launch vehicles. Much overlapping and duplication was deliberate. Officials were able to report to two or more superiors if necessary; program associate administrators had "line" and "staff" functions; and certain functions, for example, those of an inspector general, were dispersed among four or five offices instead of being concentrated in one.68 Such overlapping forced each program or functional manager to undertake responsibilities in an agencywide context.


* The location in Federal Office Building No.6 (FOB-6) in Washington, D.C., to which NASA Headquarters moved in the summer of 1961.

** The Office Of Programs became the Office of Programming in 1963 and the Office Of Program Plans and Analysis in 1967.

***The program offices abolished were Advanced Research Programs, Space Flight Programs, Launch Vehicle Programs, and Life Science Programs. Abe Silverstein, previously the Director for Space Flight Programs, returned to Cleveland in November to head the Lewis Center.

**** By selecting officers detailed to NASA, Webb hoped to keep the position "above politics."