• Option 5B -make the Deputy Under Secretary of Defense (Policy) responsible for coalition matters.

Again, the logic behind this proposal is to assign one official with the responsibility of raising coalition considerations in DoD decision-making processes.

• Option 5C-strengthen the position of the Deputy Under Secretary of Defense (International Programs and Technology).

One of the major failures of our coalition efforts has been poor defense industrial cooperation with our allies. The Deputy Under Secretary (International Programs and Technology), located in the Office of the Under Secretary of Defense (Research and Engineering), is a key force for multinational armament cooperation. Strengthening his role in relevant decisions may result in enhanced cooperation.

• Option 5D -create mission-oriented assistant or under secretaries who would be assigned responsibilities for coalition matters in their mission areas.

As the mission-oriented assistant or under secretaries, proposed in Options 1A, 1B, and 1C, would have both policy and resource elements, they may have more success in coordinating the various aspects of our coalition policies and programs.

6. PROBLEM AREA #6-INADEQUATE REVIEW OF CONTINGENCY PLANS

Two options have been developed to overcome this perceived problem area.

• Option 6A -create an OSD office, staffed by a combination of civilian and military officers, to review contingency plans. This office would report to the Under Secretary of Defense (Policy) in the current organization or to the mission-oriented assistant secretaries proposed in Option 1A. In organizational arrangements (Options 1B and 1C) with mission-oriented under secretaries and an Assistant Secretary (Strategic Planning), it could report either to appropriate under secretaries or to the strategic planning office, or to both. Given the need for tight security for these contingency plans, it would appear appropriate to consolidate this work in one office-most logically, the Assistant Secretary (Strategic Planning).

• Option 6B-create a joint OSD/OJCS office to review contingency plans.

This office would be manned by both civilian and military officials and would report to both the Secretary of Defense and the JCS Chairman or their designees.

F. EVALUATION OF ALTERNATIVE SOLUTIONS

This section evaluates the specific options for reforming OSD that were set forth in Section E. No effort will be made here to compare these options with each other or to identify the most promising options for legislative action. Rather, this section seeks to set forth in the most objective way possible the pros and cons of

each alternative solution. The options will be identified by the same number and letter combination used in the preceding section. 1. OPTIONS FOR DEALING WITH THE PROBLEM OF LIMITED MISSIONS INTEGRATION

• Option 1A -create an Under Secretary of Defense for Policy and Program Integration whose office would include assistant secretaries for three major mission categories: nuclear deterrence, NATO defense, and regional defense and force projection.

This option would essentially entail the creation of three powerful positions in OSD, whose occupants would be able to cut across functional areas and Service priorities in order to ensure that fundamental DoD missions receive the highest priority. Under the present arrangement, the responsibility for these major missions is divided among so many offices and officials that their priority has become obscured and a certain focus has been lost. As Samuel Huntington has argued in his paper, "Defense Organization and Military Strategy":

The most striking deficiency in U.S. defense organization today is the absence of any single official or office in the Pentagon with overall responsibility for any one of these strategic missions and only for that mission. Individual officials and organizations are responsible for parts of each of these missions; other officials, such as the Chairman of the JCS and the Undersecretary for Policy, have a general responsibility for all these missions. The Secretary of Defense knows where to turn when he wants the individual officials responsible for the Air Force or the Marine Corps, for research and development or intelligence, for manpower or the budget. But where does he find an official with overall and exclusive responsibility for strategic deterrence? There is none. Nor is there any single official responsible for NATO defense or for force projection in the Third World. These are precisely the major strategic purposes of American defense policy, and they are virtually the only important interests in defense that are not represented in the defense organization. (page 33)

There is, at present, no senior OSD official below the Secretary and Deputy Secretary who watches out for these mission priorities, and the military officers who do so the unified commanders-do not have a strong voice or advocate in Washington. The three mission-oriented assistant secretaries could become important spokesmen within OSD for the interests of the unified commands. Their very existence would tend to draw attention to how various procurement, research and development, and operations and maintenance decisions and trade-offs affect the overall capability to fulfill key military missions.

Creating mission-oriented offices also has benefits in terms of other OSD and PPBS problem areas. It could strengthen strategic planning by diminishing OSD's focus on resources (Option 1A of Chapter 7) and by strengthening the mission orientation of organizations that contribute to the strategic planning process (Option 11

of Chapter 7). In addition, mission-oriented offices could reorient OSD's attention away from functional micro-management (Option 4C) and strengthen efforts to achieve coalition-oriented planning and programming (Option 5D). In sum, mission-oriented offices would help to overcome the serious deficiencies of a functional structure in a large and complex organization.

On the negative side, the creation of these three assistant secretaries and the transfer of numerous offices and subunits to their jurisdiction would cause considerable confusion during the transition period. While it is true that more attention needs to be paid to major missions, it is less clear that the creation of three civilian assistant secretaries is the best way to achieve this. Alternative approaches involving the JCS system might be more effective and less disruptive.

Moreover, in some cases at least, the transfer of various units and subunits to the purview of the proposed mission-oriented assistant secretaries might result in less efficient or useful analysis and work. For example, if the program analysis and evaluation (PA&E) function were divided among three mission-oriented offices, there would be more attention devoted to cost-benefit tradeoffs within mission categories, but less attention devoted to tradeoffs that cut across mission categories and that embrace the entire defense budget (although this need could be fulfilled by the smaller PA&E office to be assigned to the Assistant Secretary (Comptroller)). Why break up functional offices that may require a certain critical mass in size in order to accomplish their function?

These arguments on the disadvantages of a mission organization versus a functional organization represent the traditional business dilemma of a product line versus a functional organization.

• Option 1B -create under secretaries in OSD for three major mission categories: nuclear deterrence, NATO defense, and regional defense and force projection

This option might be more disruptive than Option 1A, primarily because it would create three powerful mission-oriented under secretaries. On the other hand, a single under secretary with missionoriented assistant secretaries under him would have considerably less ability to cause mission-oriented integration to actually happen than three under secretaries who focus on well-defined areas of responsibility. Moreover, given the fact that officials heading these mission-oriented offices would be responsible for the central strategic purposes of DoD, it would seem reasonable that they should be among the most senior officials in OSD and not lower in the hierarchy than functional-oriented officials. In addition, decisions on policy and resource allocation priorities among these three mission areas are among the most fundamental and important ones to be made in DoD. It can be argued that the Secretary or Deputy Secretary of Defense should be making these decisions and not the proposed Under Secretary for Policy and Program Integration. In many respects, the influence and decision-making responsibilities of the Under Secretary for Policy and Program Integration, proposed in Option 1A, could exceed those of the Secretary and Deputy Secretary of Defense.

52-314 0-85-5

It is clear from organizational trends in OSD that Secretaries of Defense are searching for improved mechanisms to help integrate the overall defense effort. Part 4 (Functional Organization of OSD) of Section B of this chapter indicates that: "most of the additions [to OSD functional areas since 1953] have been to strengthen the Secretary's policy, program review, and oversight responsibilities." These capabilities are primarily oriented toward seeking improved integration of the policies and programs of the Military Depart

ments.

While these relatively new integration capabilities in OSD have not taken an explicit mission orientation, there has been a recent precedent for establishing mission-oriented offices. During the early years of the Carter Administration, Ambassador Robert W. Komer served as the Advisor to the Secretary of Defense for NATO Affairs. While he did not have a formal organizational structure to support his work (as is proposed for the assistant or under secretary for NATO defense), he was able, primarily due to his hierarchical position, to cut across functional and Service lines to give the NATO mission high priority. In this regard, Ambassador Komer made substantial contributions, including development of NATO's Long-Term Defense Program and planning the deployment of Intermediate-Range Nuclear Forces in Europe. Ambassador Komer essentially served as the proposed under secretary for NATO defense.

• Option 1C-create a matrix organization with mission-oriented under secretaries and functional-oriented under and assistant secretaries.

If one were convinced of the need for continued functional coordination in OSD as well as the need for mission-oriented offices, a mission-function matrix organization could be employed. The advantage would be effective coordination on both a mission and functional basis.

The major disadvantage would be the complexity of a matrix organization. The complexity problem would be compounded by the fact that OSD would just be emerging from a traditional functional organization to one that included mission-oriented offices. Adding a matrix at the same time that mission offices were created may be too much organizational change in OSD at one time. It might be better to follow a two-step process: create mission-oriented offices first and add the mission-function matrix later.

On the other hand, it might be preferable to make all of these changes at one time. It is clear that creating mission offices would be the more disruptive change. The matrix would be a rather modest step by comparison and might serve to ease the transitional process by providing continued functional coordination. Furthermore, the matrix proposed for OSD is a simple one. In any case, a mission-function matrix organization in OSD probably ought not to be a matter for legislation, but at most a recommendation to the Secretary of Defense.

The business literature, especially Davis and Lawrence in their book, Matrix, indicate that organizations turn to matrix organizations when three conditions apply:

There is considerable pressure for balanced decision-making that focuses on two or more organizational dimensions -in OSD's case, on both missions and functions;

There is considerable pressure for high rates of information exchange because of uncertainty, complexity, and interdependence in the issues confronting the organization; and

There are internal demands to achieve greater economies of scale and to meet high quality standards with scarce financial and human resources.

All three of these conditions apparently apply to OSD.

A matrix organization has numerous advantages. The matrix's most basic advantage over more familiar structures is that it facilitates a rapid management response to changing requirements. Multiple expertise from the various matrix dimensions is brought to bear on a problem to solve it in a manner that benefits the entire organization. The matrix forces simultaneous consideration of all relevant factors mission, function, and geographic and enhances prospects for agreement on the best course of action. Resources can be allocated more rationally and with greater effect, primarily because the matrix helps middle managers (assistant secretaries and their deputies) to make trade-off decisions from a general management (Secretary of Defense) perspective, an orientation which is not now possible in OSD.

A matrix organization also increases the potential for more effective control and coordination. The matrix permits better control over mission and functional issues because it avoids an exclusive focus on one dimension. More than any other structural format, the multiple reporting relationships and flexibility of a matrix encourage communication and coordination.

The disadvantages of a matrix are associated with making it work. Peter F. Drucker has argued that the matrix "will never be a preferred form of organization; it is fiendishly difficult". Key among the disadvantages is the potential for power struggles between the matrix dimensions. Because the matrix formalizes the conflict that already exists between mission and functional points of view, power struggles could result because the authority and responsibility of the two dimensions would overlap. This would be less of a problem in OSD because power would not be balanced between the mission and functional dimensions; the mission dimension would be dominant.

In their book, In Search of Excellence -Lessons From America's Best-Run Companies, Peters and Waterman are critical of matrix organizations in large corporations:

Along with bigness comes complexity, unfortunately. And most big companies respond to complexity in kind, by designing complex systems and structures....Our favorite candidate for the wrong kind of structure, of course, is the matrix organization structure. (page 306)

However, this criticism is focused on those organizations that have created large, complex, and often four-dimensional matrices. For those companies who have kept their matrices simple, Peters and Waterman are more positive: