CHAPTER 3 Recent PPBES Transformation in the Defense Department
By L.R. Jones and Jerry L. McCaffery
In 2003 the DoD announced significant changes to the Planning, Programming, and Budgeting System (PPBS), renaming it the Planning, Programming, Budgeting, and Execution System (PPBES) (Secretary of Defense, 2003a). The PPBES is the policy development, resource planning, and budgeting process for national defense. It is characterized by complexity and plurality. The DoD prepares its plan and budget using this system, which allocated almost $380 billion in federal funds in FY 2004, about 48% of the discretionary federal budget, and is projected to allocate $2.1 trillion from 2004 through 2008. This is almost six times the next-largest discretionary total, which is for the Department of Health and Human Services. Despite these trends, little attention has been paid to the defense budget process. This paper reviews and evaluates changes made to what is now termed the PPBES and also assesses other initiatives to reform the resource decision-making system employed by the DoD.
While the basic structure of PPBS remains, it was changed in several important ways. First, the reform merged separate programming and budget review into a single review cycle, combining the steps rather than performing them sequentially. Second, it incorporated a biennial budget process matched to national electoral cycles, with major strategic changes slated for the second and fourth year of a presidential term and minimal updating change made in the first and third years. Third, it fixed the process’s timing so that planning and budgeting were clearly derivative processes driven by the Quadrennial Defense Review and the National Military Strategy. Fourth, it changed the cycle for Office of the Secretary of Defense (OSD) provision of the top-level planning information to the military departments and services from annual to biennial.
The result of these changes was to create a two-year decision cycle, with a complete review in year one followed by limited incremental review in year two. This was meant to decrease turbulence and reduce unnecessary remaking of decisions. These changes made each on-year cycle quicker by compressing the programming and budgeting cycles, but preserved the decisions made in the on-year cycle through the off-year by limiting reconsideration of decisions to only the most necessary updates. Decisions would be made more quickly but would last longer.
In this report, we review the history of PPBS and indicate how the reforms of 2002–2003 have changed it. We also speculate about the limits to change within the DoD and at the political interface between DoD and Congress, which as Mayer and Khademian (1996) suggest (about defense procurement) will not give up its legislative oversight role without a fight and which makes certain demands on DoD about budget format and presentation (e.g., a line-item format) for the appropriation process, irrespective of what internal processes DoD may use. PPBS has served DoD well for almost four decades by balancing rational decision processes, neutral expertise from the military and others, and clearly political agendas over an extended planning and budgeting horizon. The relevant question here is the extent to which the new PPBES meets the defense needs of a world changed by the events of September 11, 2001, and thereafter.
PPBS HISTORY AND DEVELOPMENT
PPBS was implemented in DoD originally by Defense Secretary Robert McNamara and by Charles Hitch, Robert Anthony, and others during the administrations of Presidents Kennedy and Johnson in the 1960s (Thompson and Jones, 1994). Before 1962, DoD did not have a top-down coordinated approach for planning and budgeting (Puritano, 1981; Korb, 1977; Korb, 1979; Joint DOD/GAO Working Group on PPBS, 1983). Until this time, the Secretary of Defense (SECDEF) had played a limited role in budget review, as each military service developed and defended its own budget. McNamara had used PPBS when he was the president of the Ford Motor Corporation. He and Charles Hitch, his comptroller, had confidence that the system would be valuable for long-range resource planning and allocation in DoD. McNamara wanted PPBS to become the primary resource decision and allocation mechanism used by DoD. McNamara implemented the system after President John F. Kennedy tasked him to establish tighter control by the secretary of defense, a civilian, over the military departments and services. As a former member of Congress, Kennedy was highly distrustful of the military service planning and budgeting. He ordered McNamara to take control of DoD planning and budgeting away from the military and put it in the hands of civilian leadership. Consequently, the initial motivation for establishing PPBS had as much to do with control and politics as it did with rational resource planning and budgeting. By June 30, 1964, PPBS was operational within the Department of Defense (Thompson and Jones, 1994; Feltes, 1976; Korb, 1977; Korb, 1979).
Hitch implemented PPBS and systems analysis throughout DoD, but most of the program analysis was done by his “whiz kids” in the OSD under the comptroller and the Office of Program Analysis and Evaluation. The military departments were not anxious to implement PPBS but had to do so eventually to play in the new planning and budgeting game run and orchestrated by Hitch and his staff. After a few years, the military departments were fully engaged in learning how to compete in the new PPBS process. However, as noted, PPBS was not just budget reform—it was a new approach to analysis and competition among alternative programs, weapons systems, and, ultimately, multiyear programmatic objectives. Additional reforms beyond PPBS were to be proposed by DoD under the Johnson administration.
Charles Hitch was followed as DoD comptroller by Robert N. Anthony, a professor of management control on loan from Harvard University’s School of Business, who proposed an ambitious set of changes to DoD budgeting and accounting in 1966 in what was termed Project Prime. Among other things, Project Prime would have divided all parts of DoD into mission, revenue, expense, and service centers, consistent with management control theory according to Anthony, and required accrual accounting with reimbursable fee-for-service internal transactional payments (using negotiated or shadow prices) throughout DoD (Thompson and Jones, 1994: 66-68). What Anthony envisioned was a reimbursable accounting process similar to what was implemented in much of DoD by Comptroller Sean O’Keefe and Deputy Comptroller Donald Shycoff as part of the Defense Management Report initiatives of 1989–1992 under the first Bush administration and Defense Secretary Dick Cheney (Jones and Bixler, 1992). Project Prime also included accrual accounting and budgeting for DoD. Accrual accounting is required now under the Chief Financial Officers Act of 1990, which DoD has been unable to implement successfully. Clearly, Anthony was ahead of his time in his vision of how DoD accounting and budgeting should be organized (Thompson and Jones, 1994: 67-68).
Congress did not support Anthony’s proposed changes. Key members of the appropriations committees refused to allow the change to accrual accounting and rejected Project Prime, probably because they thought it would reduce their leverage to micromanage DoD through the budget. Opposition was so strong that it was suggested Anthony be asked to resign. Anthony was not asked to do so but chose to return to Harvard, and the experiment was ended (Jones, 2001b). Not until 2003 did DoD return to Congress with such a sweeping reform proposal—the Defense Transformation Act (Wolfowitz, 2003). Although not much of what DoD requested in 2003 was passed (approval of a new DoD personnel system was the only significant change Congress passed), the request marked a milestone on the long road of DoD reform.
The post-WWII sequence of budget reforms that led to PPBS in the 1960s started with performance budgeting in the 1950s. In essence, performance budgeting (Burkhead, 1959: chapters 6-7 and pages 133-81) attempts to connect inputs to outputs. As implemented by the President’s Bureau of the Budget (BOB) under the Eisenhower administration, performance budgeting (PB) in the 1950s was characterized by indicators of cost per unit of work accomplished, focusing on workload measures rather than on outputs or outcomes. The history of performance budgeting includes the Taft Commission of 1912, which recommended it be implemented, and its implementation in the Department of Agriculture in 1934 and the Tennessee Valley Authority in the later 1930s, as well as its having been strongly recommended by the Hoover Commission in 1949 (McCaffery and Jones, 2001: 69).
In 1949 Congress required that the DoD budget estimates be presented in performance categories. Performance budgeting was an executive branch managerial budget tool. During the 1950s, under the leadership of Bureau of the Budget Director Maurice Stans and others, executive budgeting was transformed somewhat radically through the institution of performance measures. Many of the measures had already been in use for decades as proxies that facilitated and simplified negotiations between the executive branch and Congress. However, in this first wave of performance budgeting (the second wave would hit in the 1990s), great effort was exerted to develop measures of performance and relate these to appropriations and spending. In fact, many of the measures developed in this era did not measure performance. Instead, because doing so was easier (and perhaps the only approach possible), workload and input cost data were used in place of real measures of performance. Still, budgeting in this era moved far from the simple line-item formats of the past. Formulas and ratios between proposed spending and actions were integrated into the executive budget, along with explanations of what the measures demonstrated and how they related to justifications for additional resources (McCaffery and Jones, 2001: 69).
The emphasis of budget reform shifted in the early 1960s to what was termed “program budgeting.” Program budgeting (Mosher, 1954; Novick, 1969) is a variation of or evolution from performance budgeting in which information is collected by program categories, without much of the detail of the performance-budget construction. These categories of spending are tied to specific objectives. Activities are grouped by department, agency, and then by mission objective and sometimes by function, and projected for a five-year period. The Department of Agriculture experimented with program budgeting in the early 1960s as reported by Wildavsky and Hammond (1965) and later adopted throughout the entire federal government through an executive order issued by President Lyndon Johnson in 1966.
The PPBS (Lee and Johnson, 1983: chapter 5; Hinricks and Taylor, 1969; Merewitz and Sosnick, 1972; Schick, 1966; Schick, 1973; McCaffery and Jones, 2001: 70) was intended to be a thorough analysis and planning system that incorporated multiple sets of plans and programs. Under Secretary of Defense Robert McNamara and DoD Comptroller Charles Hitch, PPBS drew on methods from various disciplines, including economics, systems analysis, strategic planning, cybernetics, and public administration to array and analyze alternative means and goals by program and then derive benefit/cost ratios intended to indicate which means and ends to choose. Budgeting under this system was to become a simple matter of costing out the goal chosen.
In theory, the program budgets that resulted from PPBS were supposed to provide the executive branch and Congress information on what the federal government was spending for particular categories, e.g., health, education, and public safety, across all departments and agencies. Program budgets may best be understood as matrices with program categories on one axis and departments on the other. Thus, in the fully articulated program budget, Congress could determine how much was spent on health or education in total in all departments and agencies, and this would promote deliberation over whether this was enough, too much, or too little.
President Lyndon Johnson thought that PPBS was so successful in DoD that in 1966 he issued an executive order to have it implemented throughout the federal government. Regrettably, although executive branch departments prepared their program budgets and related spending to objectives, Congress largely ignored what it was presented, preferring to stick with the traditional appropriations framework for analysis and enactment of the budget (Schick, 1973). Various reasons are advanced for why this was so. Perhaps program budgets presented too much information to be used and understood by Congress. Alternatively, and as likely, perhaps Congress perceived that program budgeting would reduce the power of members of appropriations committees because the budget in this format would be determined too much by formula, thus decreasing Congress’s political spending discretion (Jones and Bixler, 1992). Although the government-wide experiment with PPBS was suspended by President Richard Nixon in 1969, this was done more for political than efficiency reasons. However, PPBS was perceived in much of the executive branch and Congress as paper-heavy and consuming too much staff time for preparation and analysis (Schick, 1973). Still, DoD continued to use the system, in part because DoD purchases substantial long-lived capital assets, and PPBS requires long-range planning as its first component.
Thus, despite criticism that PPBS was a failure in the federal government, the process remained in use by DoD and has been modified incrementally so as to operate effectively despite some evident flaws (Wildavsky 1988: 186-202; Puritano, 1981; McCaffery and Jones, 2001). While the manner in which PPBS operates has varied under different presidents and secretaries of defense, the basic characteristics of the system have remained in place for more than 40 years. During this period, three significant reform initiatives have influenced the PPBS: the Laird reforms, the Goldwater-Nichols Act, and the Rumsfeld transformation in 2001–2003.
LAIRD REFORMS
In 1969 Melvin Laird was appointed secretary of defense by president-elect Richard Nixon to succeed McNamara. Laird brought a different management orientation to the DoD, one more in keeping with its historical predilections, emphasizing decentralization and military service primacy. If McNamara increased scientific decision-making in the Pentagon, he also installed a centralized management approach. Systems analysis, top-down planning, and benefit/cost analysis supported this centralized focus. One of the key bureaucratic players was the Office of Policy Analysis, which made use of the tools cited above to help McNamara centralize decisions in the OSD (Thompson and Jones, 1994: 68-73). Laird’s methods ran counter to this approach, emphasizing participatory management and decentralization of power. Beginning in 1969, Laird shifted decision-making power away from the DoD staff agencies to the military department secretaries because there were “… many decisions that should be made by the Services Secretaries and they should have the responsibility for running their own programs. I have no business being involved in how many 20mm guns should go on a destroyer. That is the Secretary of the Navy’s business. I must let the Services take a greater role” (Feltes, 1976). Laird also pursued a process of participatory management in which he hoped to gain the cooperation of the military leadership in reducing the defense budget and the size of the forces.
During Laird’s four-year tenure, U.S. troop strength in Viet Nam fell from 549,500 persons in 1969 to 69,000 in May of 1972 (Laird, 2003). Laird was preoccupied with disengaging from Viet Nam, but not to the exclusion of other issues, such as burden-sharing costs with other nations, maintaining technological superiority (e.g., B-1 bomber, Trident submarine), improving procurement, enhancing operational readiness, and creating strategic sufficiency and limitations on the nuclear build-up (Feltes, 1976; Armed Forces Management, 1969). He ended the selective service draft in January of 1973 and was persistent in his efforts to secure the release of American POWs.
Laird spent much time preparing for testimony and testifying in Congress, and he improved DoD relations with Congress. On the management side, Laird gave the military department secretaries and the JCS a more influential role in developing budgets and force levels, but he also returned to the use of service program and budget ceilings (fixed shares) and required services to program within these ceilings. This concept of ceilings, or “top-line,” endured for most of the next 40 years and still influences DoD budget requests today, as services are expected to balance their program and budget against the total obligational authority they are given at various stages in the planning and budget process.
Laird sought to provide a better balance between military and civilian judgment in the defense decision-making process by providing better and earlier strategic and fiscal guidance to the services and the JCS. Feltes suggests that the result of Laird’s emphasis on decentralized management was that responsibility for military planning was shifted back to the military services, and the role of OSD Systems Analysis was de-emphasized. While no abrupt shifts were made, the Laird era was marked by a steady and persistent shift away from McNamara’s emphasis on centralization of DoD decision-making under the secretary of defense (Feltes, 1976; Armed Forces Management, 1969).
THE GOLDWATER-NICHOLS ACT OF 1986
It may be argued that the creation of the DoD in 1947–1949 never really took hold in that, by and large, the military departments continued to go their separate ways within the envelope of the DoD until the reforms of the 1960s and, to some extent, until implementation of the Goldwater-Nichols Act of 1986 (Thompson and Jones, 1994: 78-79, 246). In the 1950s, presidents Truman and Eisenhower both fought arguably losing battles to strengthen the role of chairman of the JCS and the JCS itself (Thompson and Jones, 1994: 51-3).
By 1981 the sitting JCS chairman, General David Jones, was writing that the system was broken and was asking Congress to fix it (Jones, 1982). The fact that General Jones as CJCS was voicing such criticisms was in itself very significant (Chiarelli, 1993: 71). In 1981 Jones (1982) suggested that because of the decentralized and fragmented resource allocation process driven by parochial service loyalties, there was always more program than budget to buy it; that the focus was always on service programs; that changes were always marginal, when perhaps better analysis would have led to more sweeping changes; and that it was impossible to focus on critical cross-service needs—and the result was that an amalgamation of service needs prevailed at the JCS level.
General Jones argued that staff to the chairman of the JCS was so small that the chairman could focus only on a few issues. The result was that the defense budget was driven by the desires of the services (usually for more programs and money), rather than by a well-integrated JCS plan. In addition, he argued that all of this undercut the authority of not only the JCS but the entire unified command structure established in the Defense Reorganization Act of 1958 (Thompson and Jones, 1994: 51-53). General Jones noted this was particularly evident in acquisition, where weapons systems met performance goals 70% of the time, but schedules 15% of the time and cost goals 10% of the time. Jones explained:
“The lack of discipline in the budget system prevents making the very tough choices of what to do and what not to do. Instead, strong constituencies in the Pentagon, Congress, and industry support individual programs, while the need for overall defense effectiveness and efficiency is not adequately addressed.” (Jones, 1996: 27).
In 1986 Congress passed a sweeping reform plan, commonly referred to as the Goldwater-Nichols Act (for its congressional sponsors), over the ardent objections of many in the Pentagon, including Secretary of Defense Caspar Weinberger (Locher, 1996: 10; Locher, 2002), who thought it would break apart the DoD management system. The legislation is too complex to detail here, but among other things it strengthened the hand of the chairman of the JCS as chief military advisor and spokesman to the secretary of defense and to the president, provided the CJCS with a larger staff, and identified important phases in the PPBS process where the JCS would help in setting requirements and reviewing the plans of other players. It established the national command authority to run from the president to the secretary of defense to the unified commanders in chief (CINCs). This increased the unified CINCs’ formal authority, so that rather than using whatever forces the military services would allow them to use in their geographical area, the unified CINCs had warfighting and command responsibilities, and the military service roles were to provide them with the wherewithal to do so (Thompson and Jones, 1994: 51-53, 79, 223-224). This distinction clearly put the military services in the role of training people and providing personnel and equipment for the warfighting missions of the geographically based unified command CINCs. Goldwater-Nichols also created the position of vice-chairman of the JCS. Generally, the officers who have served in this spot have been strong innovators and, through various committee structures, have had a substantial impact on the resource planning process within DoD.
Goldwater-Nichols also emphasized the requirement for joint command officer duty assignment. Before Goldwater-Nichols, JCS and joint command assignments were viewed as almost career-ending assignments; thus, many of the best officers tried to avoid them. CJCS Jones observed that people serving joint tours did less well in the promotion process than those who had not served such tours (Jones, 1996: 28). While implementing it has been an evolutionary process, Goldwater-Nichols has changed this perspective—such assignments now may be career enhancing. The act also required all officers to pass certain levels of joint proficiency, and upwardly mobile officers now believe a joint tour is a must.
Most important, Goldwater-Nichols changed the caliber of advice given to the president and secretary of defense by the JCS. Former CJCS army general Shalikashvili praised this part of the Act: “… We have broken free from the ‘lowest common denominator’ recommendation that so often plagued us in the past” (Roberts, 1996: 1). Shalikashvili indicated there was still room for smoothing the role of the JCS in the planning and budgeting cycles, in the national military planning process, and in management of officers into joint billets. Nonetheless, it is clear that Goldwater-Nichols is a success, as Secretary of Defense Perry noted in 1995: “It dramatically changed the way that America’s forces operate by streamlining the command process and empowering the Chairman and the unified commanders. These changes paid off in … Desert Storm, in Haiti, and today in Bosnia” (Locher, 1996: 15).
On the resource allocation side, Goldwater-Nichols provides two classes of organizations: those who do the warfighting, under the unified command CINCs, and those who support them, the military departments and services, and their own CINCs. The military department secretaries hold most of the DoD budget authority, while the service CINCs play key roles in programming, with less leverage in budgeting. Most of the combatant commands, the unified CINCs, do not have their own budgets (except for their staffs). Rather, they use the personnel and weaponry provided them by the military departments and services. However, the military CINCs must pass their budget requests through the unified command CINCs before they move upward in the budget chain of command to the Pentagon. Before the mid-1990s this review by the unified command staffs used to be pro forma, but it has become a real review in many unified commands, e.g., CINCPAC review of CINCPACFLT budget proposals. The Special Operations Forces (SOF) command, headquartered at McDill Air Force Base in Florida, has its own sizable (and increasing) budget, but SOF budgets still are small compared to the military department budgets.
The unified CINCs also have an opportunity to identify requirements in the PPBES process, and the chairman of the JCS is responsible for advising the secretary of defense to certify the merit of these requirements as well as how well the budgets of the military departments satisfy the unified CINC needs. The JCS chairman also can submit alternative recommendations to SECDEF to meet unified CINC needs in the budget. In this matter, SECDEF is the final arbiter of what the military departments get in their budgets. The unified and service CINCs both have opportunities to give input to the chairman of the JCS in the PPBES planning process for development of the National Military Strategy and in the final draft of the defense guidance that leads to the POM process. In the POM process, the service CINCs make inputs by providing integrated priority lists (IPLs) that indicate their top warfighting needs (important information for the JCS and unified CINCs). Military service CINCs may indicate program deficiencies that exist and make recommendations to fix deficiencies to both the JCS and the military service chiefs. The IPLs are a part of the programming and budgeting process and are duly considered in several venues in OSD and the military departments.
An unresolved tension is evident here as the unified and service CINCs both have been criticized as sometimes tending to focus on short-term operational needs, warfighting issues, and the O&M accounts that support readiness. Simultaneously, the military departments have to keep an eye not only on the short-term and immediate items and issues, but also on weapons procurement and recapitalization issues, such as modernizing the aircraft or fleet inventory. Some players in the PPBES process believe this is a healthy tension. Others worry that immediate issues, and some long-term needs, may be slighted. DoD is currently in the middle of another significant change as Secretary Rumsfeld pursues his goal of transforming both military and business affairs while actively employing some part of the operating force in combat situations. This reform is somewhat of a return to a more centralized pattern of operations (Rumsfeld, 2001; 2003).
PPBES PROCESS OVERVIEW
Recent change in the venerable PPBS was separate from Secretary Rumsfeld’s efforts to transform defense and was not part of the Defense Transformation Initiatives presented and debated before Congress in 2003. None of the reforms implemented in 2001 through 2004 and continuing through 2006 and beyond in the PPB process needed congressional approval; they were internal matters concerning how DoD would organize its planning and budgeting process. While critics had pointed out flaws in PPBS for some time, the genesis of this set of reforms clearly appears to rest with Secretary Rumsfeld, who felt the process was too slow and cumbersome and did not deliver the “right stuff” on a timely basis. Thus, the change was intended to accelerate the cycle and to avoid unnecessary remaking of decisions. Implicit in this process is the idea that SECDEF and his staff will be brought into the decision loop more quickly, though this is not articulated in documents that describe the process changes.
Each year the secretary of defense has issued the Defense Planning Guidance (DPG) to guide the programming and budgeting processes. This comes after extensive analysis of world conditions, the threat situation, and U.S. options and strategies. The DPG describes SECDEF’s guidelines for creating force structure to meet the threat, including his appreciation of fiscal constraints. The DPG now will become a biennial guidance. The Office of the Secretary of Defense will no longer provide the military services and defense agencies this annual classified planning document designed to help them develop their budget and program requests for the upcoming fiscal year. The move away from developing the top-level DPG each year is part of the OSD move toward two-year budget cycles. If necessary, OSD may prepare “off-year” guidance documents reflecting minor strategy changes, according to Management Initiative Decision No. 913, issued May 22, 2003, by Deputy Secretary of Defense Paul Wolfowitz (Secretary of Defense, 2003a). The intent of reform is that the threat does not often change quickly. For example, the Cold War threat scenario lasted from about 1948 to 1990 and was followed by a “base force and reconstitute” scenario until 2001. However, there are some exceptions, e.g., after September 11, 2001.
Threat assessment has long-term salience, and complete annual reviews to it followed by a full budget cycle were seen as costly and inefficient. The essence of the reform places the biennial issuance of the DPG document in a two-year cycle within the four years that a presidential administration has to develop its national defense objectives and strategy. A series of documents has in the past guided this process, including the annual DPG, the Future Years Defense Program, the issuance of each new president’s national security strategy, and development of the Quadrennial Defense Review (QDR) for use by DoD and for reporting to Congress. The QDR consists of a comprehensive analysis of military readiness, capabilities, and force structure that helps to provide a reporting framework to permit a newly elected administration to develop its spending plan and budget. Since the early 1990s, the QDR has become the primary external and one of the major internal statements of policy by the secretary of defense. To explain these changes and how they are playing out, we first examine the PPBES process and then discuss how the Department of the Navy operated its budget process in the new PPBES process in 2002 and 2003.
The purpose of PPBES is to provide a systematic and structured approach for allocating resources in support of the national security strategy of the United States. The ultimate goal of the entire PPBES process is to provide the military commanders-in-chief with the best mix of forces, equipment, and support attainable within resource constraints. Before delving into the full complexity of PPBES, it is useful to review the system. Once we understand how PPBES operates in general, we then review changes initiated in 2001 to 2003 and beyond to significantly modify the former PPBS into what is now, PPBES—the result of significant reforms authorized by Secretary Rumsfeld under the administration of President George W. Bush. Then, when we understand the changes made during this period, we examine how the process operates in detail.
FOUR PHASES OF PPBES
PPBES has four distinct phases, with each phase overlapping the others phases (Jones and Bixler, 1992: 19-31).
The planning phase begins at the executive branch level with the president’s National Security Strategy (NSS), developed by the National Security Council. The NSS takes its input from several federal agencies (including the Department of State, the Central Intelligence Agency, and others in the intelligence community) to ascertain the threats to the United States in order to form the nation’s overall strategic plan to meet those threats, thereby outlining the national defense strategy. Subsequently, the Joint Chiefs of Staff (JCS) produce a fiscally unconstrained document called the National Military Strategy Document (NMSD). The NMSD contains their advice regarding strategic planning to meet the direction given in the National Security Strategy while addressing the military capabilities required supporting that objective.
As a follow-on to the NMSD, the Chairman of the Joint Chiefs (CJCS) advises the secretary of defense, in the Chairman’s Program Recommendation (CPR), regarding joint capabilities to be realized across DoD military components. The CPR provides the personal recommendations of the chairman of the JCS for promoting joint readiness, doctrine, and training, and better satisfying joint warfighting requirements to influence formulation of the DPG. The CPR is seen as a key joint staff input from the CJCS and his staff into the PPBES process. It is meant to help steer the Defense Planning Guidance (DPG).
All of the above inputs are provided to the SECDEF for drafting and ultimately issuing the DPG and the Future Year Defense Plan, a six-year projection of department-wide force structure requirements. The DPG provides the military services official guidance regarding force structure and fiscal guidelines for use in preparing their program objectives memorandum (POM) during the programming phase of PPBES. For purposes of reporting to Congress on defense planning, the DoD also prepares and transmits a comprehensive report referred to as the Quadrennial Defense Review (QDR). In the past decade, the QDR has enhanced the FYDP and DPG for purposes of planning for the OSD and DoD.
The purpose of the programming phase is for each military component to produce a POM to address how they will allocate resources over a six-year period. The development of the POM requires the services to consider numerous issues, including their CINCs’ fiscally unconstrained Integrated Priority Lists (IPLs) stipulating programs that must be addressed during its development. The POM also must support the guidance given in the DPG and operate under fiscal constraints issued within it, e.g., total obligational authority by military department by year. POMs are developed in even-numbered years and subsequently reviewed in odd-numbered years.
Woven within the POM are the sponsor program proposals (SPPs) developed by resource sponsors (e.g., the major commands, systems commands, and defense agencies) to address military service objectives and preferences of the CINCs. The SPPs must be developed within the constraints of military component Total Obligation Authority (TOA), defined as the total amount of funds available for spending in a given year, including new obligation authority and unspent funds from previous years.
The JCS reviews military department and service POMs to ensure compliance with the NMSD and DPG, assessing force levels, balance, and capabilities. Following the review, the CJCS issues the Chairman’s Program Assessment (CPA) to influence the secretary of defense decisions delineated in the program decision memoranda (PDM) marking the end of the programming phase. The CPA is another key steering device that the chairman uses to give his personal assessment of the adequacy and risks of service and defense agency POMs. He also proposes alternative program recommendations and budget proposals for SECDEF consideration before the SECDEF issues PDMs. The PDM issued by OSD approves or adjusts programs in each POM. The POM that has been amended by the PDM provides an approved baseline for military departments to submit their budget inputs. While the programming phase of PPBES operated as a separate cycle from the 1960s through the early 2000s, in August 2001 Secretary Rumsfeld merged the POM and budget review cycles.
In acquisition matters, the chairman of the JCS is supported by the Joint Resources Oversight Committee (JROC), a committee led by the vice chairman of the JCS and composed of the service vice-chiefs, who review all joint acquisition programs and programs where a joint interest in interoperability is evident. The chairman then makes recommendations about acquisition priorities. This is another change rising out of Goldwater-Nichols and out of the Grenada operation, where Army and Marine troops on the ground could not communicate with other units because the radios used were not interoperable. The JROC approves the mission need and conducts an analysis to see how well the suggested acquisition program meets these needs. The process of staffing a proposal up to the JROC decision level involves assessment and analysis by various committees ending at the flag level—and analytic effort by JCS staff—and can take four to five months. A successful program that is vetted and found to meet joint requirements then has a priority attached to it at the JROC level and is passed into the POM and later the budget for funding.
Part of the 2003 reform was intended to accelerate and improve the acquisition process. In April 2002 Defense Planning Guidance study #20 (Secretary of Defense, 2002b) concluded that the resource requirements process frequently produced stovepipe systems that were not necessarily based on required capabilities and incorporated decisions from a single service perspective. The study found that the acquisition process did not necessarily develop requirements in the context of how the joint force would fight. Rather, requirements tended to be more service focused. Moreover, duplication of efforts was apparent in the less visible and smaller acquisition programs. The study observed that the current culture aimed for the 100% (perfect) solution, and this resulted in lengthy times to field weapons. In addition, the process was still found to lack prioritization of joint warfighting demands.
Ongoing reform here resulted in reshaping of the JROC process so that decisions would be better set up for JROC to make its decision by two new oversight committees reporting to it, headed by flag officers and focused on functional areas. This is an ongoing part of the 2003 reform and is indicative of Secretary Rumsfeld’s interest in joint operations, joint warfighting, and a quicker acquisitions process.
The Future Year Defense Plan (FYDP) is the database of record in which POM and budget actions are tracked and recorded. The FYDP is updated after every major process action, e.g., submittal of the president’s budget, of the conclusion of program change proposals, and of budget change proposals or passage of an appropriations bill. Some see the planning and programming phases as planning and more planning, but they are very different. The planning phase involves ascertaining the threat and deciding how it may be diminished or avoided. For example, a national strategy that said the United States would intervene only in conflict situations in which there was an overriding national interest would have substantially decreased operating tempo (the analog of workload) for DoD in the 1990s. Agreements between nations in treaties and compacts may either increase or decrease workload for defense. In the programming phase the major options have been decided, the SECDEF introduces fiscal constraints, and choices are subsequently made about the shape of force structure given the decisions made in the planning process. Thus the planning and programming processes are planning processes, but they are very different. The planning phase involves other departments and agencies, assesses the threat on a global basis, and describes a U.S. response to the threat without fiscal constraint. The programming process is internal to DoD and focuses on providing the force structure (aircraft carriers, tanks, personnel) to meet the threat within a generally constrained fiscal environment. The budgeting phase buys and supports the force structure for a particular fiscal year.
The budgeting phase begins with the approved programs in each military service POM. Each military component costs the items that support its POM for the budget year and submits its part of the budget as its budget estimate submission (BES). The BES in even-numbered POM years is a two-year submission and is based on the first two years of the POM as adjusted by the PDM. The BESs are amended by the services during the POM update occurring in odd-numbered years and cover only one year. Every BES is reviewed by military secretariats under the authority of the military department secretaries because budgeting is a civilian function in DoD, as mandated by Congress in the 1970s. The budgets of the military department secretaries are then reviewed by the DoD comptroller, other OSD officials, the JCS, and ultimately by the deputy secretary of defense and the secretary of defense.
SECDEF cooperates in this review with the president’s Office of Management and Budget (OMB). This review attempts to ensure compliance with the DPG, the PDM, and the president’s national security strategy. SECDEF staff makes changes and provides rationale for these changes in the form of program budget decisions (PBD). Before becoming part of the President’s budget, which must be submitted to Congress no later than the first Monday in February, PBDs are issued to allow the military department secretaries and budget staff to respond with appeals of cuts (reclamas) to SECDEF/OSD comptroller staff. Once major budget issues have been resolved, the final defense budget is sent to OMB to become part of the president’s budget. This step constitutes the end of the budget proposal and review phase of PPBS. However, as noted subsequently, budget execution is a critical part of PPBS that is typically ignored in analysis of this system.
Budget execution consists of first gaining permission to spend appropriations approved by Congress through a separate budget submission process referred to as the allotment process. In allotment review, DoD must show how it intends to spend what has been appropriated, by quarter, month, or fiscal year for multiple-year appropriations. This is always somewhat different than what was proposed in the president’s budget since appropriations must now be attributed to programs and allocated into the months they will be obligated (usually by quarters). After allotment approval is received from OMB and the Treasury, DoD begins the process of separating and distributing shares of its budget to the military departments and services and other DoD commands and agencies. After they have received their spending allotment authority, these resource claimants begin to incur obligations to spend and then liquidate their obligations through outlay of money. During this process, comptrollers and budget officials at all levels of DoD monitor and control execution of programs and funding. At the midpoint of the spending year, the military departments and services typically conduct a mid-year review to facilitate shifting money to areas of highest need. At the end of the fiscal year in September, all DoD accounts must be reconciled with appropriations, and spending must be accounted for before closing the accounts from further obligation and outlay (for annual accounts). Financial and management audits by military department audit agencies, the DoD Inspectors General, the Government Accountability Office (GAO), and other entities follow the conclusion of execution and reporting. Figure 3.1 illustrates how these processes are combined with the PBBES process.
Figure 3-1. Funding the National Security Strategy.
PPBES BY YEAR
Year One: Review and Refinement
Management Initiative Decision 913 sets out a two-year budget and planning cycle within the framework of the four years in a presidential administration. Year one requires “review and refinement” of the previous president’s strategy and plans, including only limited changes in programs and budgets, an early national security strategy, and an “off-year DPG.” As stated in MID-913, “The off-year DPG will be issued at the discretion of the Secretary of Defense … The off-year DPG will not introduce major changes to the defense program, except as specifically directed by the Secretary or Deputy Secretary of Defense … However, a small and discrete number of programming changes will be required to reflect real world changes and as part of the continuing need to align the defense program with the defense strategy” (Secretary of Defense, 2003a: 5). A major objective of the off-year guidance is to provide the planning and analysis necessary to identify major program issues for the next DPG. One of the benefits of the new four-year cycle is that it fits the PPBS process into the electoral cycle. Incoming administrations usually struggle to get their people on board in the first year, and significant defense policy changes usually do not come until later. The new cycle recognizes this reality. Significant events do happen in the first year. The national security strategy is issued at about mid-year, and the quadrennial defense review begins shortly thereafter in June and is issued in February of year two. These are significant guidances for defense strategy and resource allocation. Also in year one, the new administration may take steps to insert its defense policy priorities in the budget submitted to Congress and to make changes caused by fact-of-life events in acquisition programs. Congress also may make changes in this year that have consequences for the following years. An overview of the new PPBES decision cycle is provided in Figures 3-2 and 3-3.
Figure 3-2. Summary of 2003 PPBES Cycle. Source: Secretary of Defense, Management Initiative Decision 913, 2003: 3.
Figure 3-3. PBES Cycle. Source: Secretary of Defense, Management Initiative Decision 913, 2003: 4.
Year Two: Full PPBES Cycle—Formalizing the Agenda
Year two in the new four-year framework is more intense in that the military departments and services and OSD will conduct full program, planning, budget, and execution reviews to formalize the President’s defense posture and strategy, including the resource portion of the strategy. In addition to a quadrennial defense review issued early in the year, the second year will include a full, “on-year” DPG that is issued in May and is designed to implement the QDR results. Previously, the QDR had been issued on September 30 in the first year of a presidential administration. However, in the FY 2003 Defense Authorization Act, Congress changed the QDR reporting requirement to the second year to provide new DoD leadership more time for analysis and preparation. Senior defense officials had argued to Congress that the requirement to submit a QDR in the first year was too much to ask of a new administration barely through the rigorous congressional process of confirmation of presidential appointees to lead the DoD and military departments. Year two will see, then, a full POM and a full budget build. This will result in a full FYDP build. A full execution review is also scheduled for years two and four. Giving this the same emphasis as the POM/budget build is new.
Year Three: Execution of Guidance
The new planning and budget process specifies that year three be used for execution of the president’s defense plan and budget agenda as provided in the QDR and the previous year’s DPG. Year three corresponds with FY 2005 in the budget cycle and could include an “off-year” DPG if the secretary of defense so desires. This off-year guidance could task new studies or incorporate fact-of-life changes in acquisition programs, including increased costs or schedule delays as well as congressionally mandated changes. In May 2003 DoD comptroller Dov Zakheim indicated that no 2005 DPG was to be prepared under the Bush administration and Secretary Rumsfeld. Year three is a year of refinement of objectives and metrics, with only the most necessary program or budget change proposals considered.
Careful examination of DoD execution of dollars and plans is a critical part of the new planning and budgeting process. Traditionally, budget execution has been left primarily to the military departments. However, the revised process provides OSD with greater opportunity to examine and critique the budget execution decisions of the military departments and services. Zakheim reported in February of 2003 a widespread agreement in the DoD not to return to a comprehensive annual budget and program review; rather, the intent was to use the off year to measure the “burn rate” (rate of spending) in an execution review. To this end, the comptroller said the review would include asking questions such as how money is being spent, if it should be moved to other areas and accounts, and what results have been achieved.
An important budget change initiated by the Bush administration announced in February 2003, and subsequently by the DoD comptroller, is implementation of “performance-based budgeting,” to focus more on the costs of achieving desired military and programmatic outcomes, rather than concentrating budget review on the details of program administration and production. The driving military concept behind performance-based budgeting (PBB) is the concept of “effects-based capabilities” for warfighting. The effects-based approach focuses on desired end results from a military action rather than the military action itself. Under this concept, military commanders specify the results, such as capture of territory, in addition to the amounts and types of forces needed to achieve the outcome.
Year Four: Full PPBES Cycle—Ensuring the Legacy
Year four in the budget and planning cycle is characterized in MID-913 as the point where the achievements of a four-year presidential administration are assessed. This year will include preparation of a full DPG to refine the alignment between presidential strategy and the DoD program and budget. As usual, the DPG will initiate and guide the cycle of military department and service POM and budget preparation, review, and submission (for FY 2006). The next full PPBES cycle will encompass fiscal years 2006 to 2011. See Figure 3-3 for a model of the PBBE system.
CONCLUSIONS
With respect to long-range DoD planning in the first phase of PPBES, we may observe that during the 1990s it was clear that the shift from a Cold War mentality to a new framework was proceeding slowly. The gist of what was necessary in the post–Cold War world did not seem clear. Despite all the discussion of asymmetric threat and successive preparation and reviews of the QDR, much of the defense budget seemed focused toward a Cold War scenario. Meanwhile the defense establishment was contracted by about 35% and U.S. forces were employed in a large number of small-scale conflicts in the 1990s after the first Gulf War. Still, the precise nature of the threat was not clear. Was our own activism a problem? Were there really inimical forces out there, and if so, what was needed? These were all difficult questions. The terrorist attack of September 11, 2001, ended this period of doubt and confused reflection.
It is routinely acknowledged that the planning component has been the weakest part of PPBS for decades. Part of this is due to the contingent nature of threat assessment, while other impediments include the sheer volume of information and absence of data coordination. For DoD to plan to counter-threat effectively, it seems to us that a capabilities-based planning process within PPBES, rather than a theater-based approach, is one way to tear loose from the old bipolar geographic analyses that focused on the USSR, potential enemies in Asia, or elsewhere. Instead, it is critical to ask what capabilities the United States needs to meet threats wherever they occur, especially given that the terrorist threat has a personal or group basis less geographically bound.
The deployment of U.S. forces since September 11, 2001, illustrates new concepts in joint operation, using special forces, and applying joint forces in unique ways, supported by traditional forces using traditional doctrines. Nonetheless, it is a new mix. Much of the transformation in military affairs that has been ongoing since the mid-1990s is driven by new threats that seem to emerge almost daily. All this points to the ties between changes in military warfighting and PPBES planning.
It is obvious that important changes have been made in the DoD planning and budgeting process. The simultaneous execution of the POM and budget review and its consolidation into one database is one such change. In the old system, a good POM could still be lost on the way to the final budget. In addition, sometimes the budget process ended up doing a lot of reprogramming and re-making of decisions that would have been better done in a POM exercise. For example, the 2003 POM process started with a pricing review of the shipbuilding budget. This is a budget drill and in the old PPBS would have been performed in the budget process long after the POM had been completed. Observers comment that when such drills (re-pricing the shipbuilding account for inflation, etc.) result in a big bill that has to be paid, it is good to have that bill considered and paid at the beginning of the process in the POM, where large-dollar changes can be made more easily. They also felt that doing the POM and budget simultaneously should result in fewer surprises and less re-programming of changes to the POM in the budget process than previously. They felt that the process should be quicker but less linear—a layered process rather than a sequential one. The routing of all products of the POM and the budget into one database was seen to be a significant move to help resolve some of this added complexity.
Second, the outcome focus of the process is an important change. Secretary Rumsfeld has emphasized outcomes, and the Navy approach illustrates this concern in two ways. The procurement accounts are focused around the outcomes each weapon system bought will provide, and the performance models for steaming hours and flying hours also are outcomes-focused. As has been stated above, this covers almost 90% of the Navy budget. Nevertheless, Congress still appropriates by line item, and DoD must be able to translate capabilities into budget items and make winning arguments for those translations. The fact is that budget lines (line-items) make it easy for Congress to buy things, and what has not changed is where the power of the purse is located. As one senior budget official remarked, “… There are a lot of changes, but what has not been changed is the Constitution. Changes will end when they bump into things that are Constitutional. The appropriation process is still a congressional process and changes in the Pentagon process have to be responsive to the needs of Congress. The menu of changes the Pentagon can pursue is not unlimited.”
Third, the new process put SECDEF into the process at the early stages. Decisions in the new PPBES are intended to reach the secretary before the decision has become a foregone conclusion, while options are still open, and while important and large-scale changes still can be made. When SECDEF inputs come at the end of the stream of decisions, some decisions that could be taken get pre-empted simply because they might cause too much breakage in other programs or because everyone has already become committed to the likely outcomes of the decision. Secretary Rumsfeld had a clear interest in transformation, but not all communities within the defense establishment were equally committed or committed at all to Rumsfeld’s vision (Rumsfeld, 2001; 2003). As previously noted, inserting SECDEF in the decision process early stands up as long as history proves that the decisions SECDEF makes are right. While this is true whether SECDEF input is early or late, inserting SECDEF early in the PPBES process puts a larger burden of proof on SECDEF. Veteran observers see these changes as an evolving process, cautioning officers bound for the Pentagon in a couple of years not to bother memorizing the new process until they get there, since it has changed significantly since 2001 and will continue to change.
Last, the new emphasis on execution seems an important change, but it is too early to speculate on how this will turn out. It seems clear that no one wants to be viewed as decreasing military effectiveness in the name of saving dollars. Through 2006 a continuing theme of administration critics was that the United States was trying to do Iraq “on the cheap,” with not enough troops or the right kind of troops. If the new emphasis on execution becomes a code word for efficiency and this is parsed into “doing things on the cheap,” then the emphasis on execution will not have important or long-lasting effects.
The final dilemma for DoD for the rest of the decade and beyond is that the cost of the global war on terrorism and Operation Iraqi Freedom will take up all of the “flexible” slack in the DoD budget, leaving little money left for weapons modernization, improving the quality of life for service members, and even routine operations and maintenance. The opportunity costs of the global war on terrorism and Operation Iraqi Freedom are such that process reform of the type analyzed here will assume a back-burner position among defense transformation priorities.