Skip Navigation

Budgeting for Defense: Maintaining Today's Forces September 2000

The Congressional Budget Office's estimate of a sustaining budget for the U.S. military answers the question, What budgetary resources would be required to sustain today's forces into the future? In general, CBO estimated the funding it would take to retain forces at their current size; to modernize their equipment at a pace consistent with the expected service lives of the equipment and with continuing the military's technological advantage over potential adversaries; and to maintain current funding for readiness and training. Specifically, CBO's estimate of a sustaining defense budget would:

• Provide enough funds to maintain current operating tempos (the rates of military training and operations) and current levels of maintenance and support for equipment;

 
• Keep increases in military and civilian pay for Department of Defense personnel comparable with pay increases in the private sector;

 
• Provide enough funding for modernization to purchase new equipment to eventually replace all current items;

 
• Provide funding for research and development that is consistent with historical levels; and

 
• Provide funds to repair and replace the existing stock of military facilities and family housing units.

The concept of a sustaining budget represents the funding that DoD would require in a "steady state," when everything was held constant and nothing changed over time. In other words, CBO's estimate begins with the size and structure of today's military and calculates the annual budget that would be necessary to sustain it into the future. Of course, in reality, many things can and do change over time--including the number and structure of military units, technology, the pace of military operations, the civilian labor market from which military and civilian personnel are drawn, and countless other factors that affect how much money it would take to provide a given level of military capability. So many things can change that it is difficult, if not impossible, to compare an estimate of the defense budget that would be required for a given period with estimates for budgets covering other periods. In any use of CBO's sustaining-budget estimate, that caveat should be kept in mind.
 

The Scope of the Estimate

CBO estimated that a sustaining budget for all national defense activities (budget function 050) would total $340 billion. (CBO's estimates in this study are in 2000 dollars of budget authority.) Included in that total are funds for the Department of Defense and for the defense-related activities of the Department of Energy and other agencies of the federal government. In 2000, the Congress appropriated $289 billion for national defense: $276 billion went to DoD, $12 billion to the Department of Energy, and $1 billion to the remaining agencies. CBO's estimate of a sustaining budget for today's national defense activities is thus $50 billion higher than the actual appropriations for 2000. (The 2000 appropriations that are discussed in this paper do not include supplemental appropriations, which total about $9 billion.)

Since CBO's estimate of a sustaining budget is a long-run, or steady-state, concept, it is not an estimate of the national defense budget for 2001 or for any other fiscal year. That kind of calculation would have to take many factors into account, including the aging of specific types of weapons and equipment; the priority to be given in the short term to purchases of certain equipment or to certain operations compared with other equipment and operations; and the status of new weapons and equipment (whether they were ready to enter production). Such factors were not considered in developing CBO's estimate of a sustaining budget.
 

CBO's Estimate of a Sustaining Budget for the Department of Defense

CBO estimates that sustaining funding for DoD would total $327 billion. The discussions that follow break down that total by budget title. Most of the funds that the Congress appropriates for DoD fall into six titles: military personnel; operation and maintenance; procurement; research, development, test, and evaluation; military construction; and family housing. ⁽¹⁾ CBO developed separate estimates of funding for those categories for each of the three military departments and a total estimate for the rest of DoD's organizational components (mainly the defense agencies). ⁽²⁾ However, with the exception of procurement, the discussions in this section present only the total estimates for each title (see Table 3). As part of those discussions, CBO also assesses both the uncertainties surrounding those estimates and their sensitivity--how much the estimates would be affected by changes in the assumptions used to develop them.
 

Military Personnel Appropriations

The military competes with the private sector for its personnel. To keep the quality and quantity of today's forces in a steady state, their compensation must remain competitive with compensation in the private sector, which generally rises each year at a rate above inflation. ⁽³⁾ So a sustaining budget for military personnel must increase each year.

In 2000, the Congress appropriated $74 billion for military personnel. To calculate a sustaining budget for that category, CBO had to choose an actual period over which to project the increase in pay and benefits. Such a choice is necessarily arbitrary; CBO chose 2001 through 2015 as a reasonable span over which to make its calculations. (The effects of using a longer or shorter period are discussed below.) To maintain military pay and benefits at today's level over that period, military personnel appropriations would need to average $82 billion annually, CBO estimates.

What the Funds Purchase. In total, the services today have about 1.4 million members on active duty and another 0.9 million in the reserves, which include the National Guard. Military personnel appropriations cover pay and many benefits for active and reserve military personnel in the four services, as well as accrual charges for future retirement pay. The appropriations also include the cost of allowances that military personnel receive for subsistence and housing, together with moving costs to relocate personnel from one permanent duty location to another. (Housing benefits that the services provide to their members include housing in government-owned units and allowances that personnel can use to secure housing in the private sector. See the later discussion on family housing.)

Estimating Methods. CBO's estimate for this budget title began with the amount appropriated for military personnel in 2000. That figure reflects the changes that the Congress made in 1999 to the military's retirement system and pay table. In estimating a sustaining-budget amount, CBO held constant the total number of current service members as well as their pay grade and length of service.

As part of its estimate, CBO increased DoD's costs for military personnel during each year of the 2001-2015 period by an estimated rate of real growth that is specified in law. The National Defense Authorization Act for Fiscal Year 2000 directs DoD to adjust upward the nominal value of the pay portion of military personnel appropriations by certain amounts each year. In 2001 through 2006, that increase is set to equal the percentage change in the employment cost index (ECI) plus 0.5 percentage points; thereafter, the pay portion would rise at the same rate as the ECI. The sustaining-budget estimate reflects those mandated increases, which CBO based on a projection of future growth in the ECI averaging 3.3 percent per year through 2015. To calculate annual real growth, CBO deflated the nominal value of military pay by its projection for the gross domestic product (GDP) price index. Under that approach, the real cost of military pay would increase by about 1.5 percent annually, and total funding for the military personnel title would rise by about 1.2 percent per year. At those rates, the appropriations for military personnel would total about $88 billion in 2015; over the 15-year period, appropriations would average about $82 billion annually.

Uncertainty and Sensitivity of the Estimate. To maintain the current number and quality of military personnel, CBO assumed that increases in real pay would need to equal or exceed those in the economy as a whole. But private-sector pay could increase more or less than CBO has assumed. If real wages in the private sector grew substantially more than CBO projected, future policymakers might need to increase compensation for the military even more than estimated to maintain an equally qualified force. Conversely, private-sector wages that grew more slowly than in CBO's projections might result in smaller increases in the future.

If, for example, military pay increased annually by 2.2 percent rather than by 1.5 percent (the figure CBO used in its calculation), DoD would need about $85 billion each year, on average, over the 2001-2015 period. But if pay grew by only 1.2 percent, the department would need about $80 billion.

Moreover, CBO's decision to use the 2001-2015 period affected its estimate. A longer or shorter period would change the calculation. For instance, choosing the 2001-2010 period would yield an estimate of only $79 billion; using 2001 to 2020 would produce an estimate of $84 billion.

Pay is only one aspect of the civilian labor market, however. Unemployment rates can affect the propensity to enlist and reenlist, as can young adults' attitudes toward military service. Those factors and others could potentially affect the compensation rates needed to keep a force of the current size and quality. More generally, if the labor market for young adults tightens further or even remains as challenging for military recruitment and retention as it is today, the costs of labor (military pay and benefits) could rise faster than the costs of other aspects of support for military forces. However, an analysis of those issues is beyond the scope of this study.

Operation and Maintenance Appropriations

Together with the funding for military personnel, the operation and maintenance appropriations provide most of DoD's annual operating budget. ⁽⁴⁾ The adequacy of O&M funds, therefore, is an important determinant of whether military forces are trained and ready to fight on short notice.

Part of the O&M appropriations covers pay and benefits for most of the civilians who work for the Defense Department. To estimate a sustaining budget for those costs, CBO used the same period (2001 to 2015) and techniques that it used for military personnel. CBO estimates that O&M funding would need to average about $107 billion annually to maintain a civilian workforce equivalent to today's and to cover the cost of the items and services that are also funded through these appropriations. In 2000, the Congress appropriated about $102 billion for the O&M title.

What the Funds Purchase. Pay and benefits for most of DoD's civilian employees constitute about a third of the costs in the O&M category. (DoD employs the equivalent of about 700,000 full-time civilian personnel.) The other two-thirds of the appropriations cover many of the costs of diverse items and activities that include fuel, spare parts for DoD equipment, other supplies, operations at military bases, training and education of individual service members, and medical care.

Estimating Methods. CBO developed its estimate of sustaining O&M funding in two parts: one for civilian pay and one for the rest of the appropriations. In estimating the pay of DoD's civilian workers in each military department and in the other DoD organizations, CBO took the pay they received in 2000 and increased it each year by the projected annual percentage rise in the ECI minus the estimated rate of inflation (in the form of the GDP price index). In estimating funding for the remaining portion of the O&M title, CBO simply used the amount appropriated for 2000.

Uncertainty and Sensitivity of the Estimate. Any estimate of sustaining funding for the O&M title is uncertain to a significant degree. One reason is that the appropriations pay for so many disparate goods and services. Another is that O&M funding is influenced by numerous factors, including:

• The number, type, and condition of the services' equipment;

 
• The number, size, type, and condition of DoD's facilities;

 
• The number, size, and type of operations and deployments undertaken by military forces;

 
• The cost of medical care for beneficiaries of DoD's health care system; and

 
• The size and pay-grade structure of DoD's civilian workforce.

CBO's assumption that a sustaining budget for two-thirds of the funds in the O&M category (those spent on purchases) would be the same as this year's budget is based on its approach of holding constant as many factors as possible. Among the most important are the structure of DoD's forces, the number of personnel, and the way the department organizes and manages its business operations. Considerable uncertainty remains, nevertheless, about the level of O&M funding that would be required to sustain the readiness and operations of today's military.

Upward Pressures on O&M Spending. At least over the past 30 years, O&M spending per capita--that is, spending adjusted for the size of the active-duty military--has increased annually by an average of 2 percent to 3 percent in real terms. Indeed, the O&M budget, on a per-person basis, was about 35 percent higher in the 1980s than in the 1970s, and it averaged about 25 percent more in the 1990s than in the 1980s.

Some of DoD's leaders have suggested that per capita spending for O&M could continue to grow in the future. ⁽⁵⁾ Analysts have suggested that factors such as additional costs for training personnel to operate ever more complex equipment and funding for more-sophisticated spare parts might contribute to growth in the future and push O&M spending higher than CBO's sustaining-budget estimate. If, for example, spending grew by 2 percent per year over the next 15 years, annual O&M budgets would average about $13 billion more than CBO's estimate.

An example of a budget category that could easily grow is medical costs. Funding for military medical care totaled about $17 billion in 2000; about $12 billion of that amount was part of the O&M title. If DoD's medical costs maintain their past relationship to national health expenditures--which are projected to increase in the future--DoD's average O&M bill could be more than $2 billion higher than CBO's sustaining-budget estimate.

Downward Pressures on O&M Spending. The underlying determinants of past increases in O&M spending may not persist into the future. For instance, the increase in the 1990s in O&M spending per capita may reflect the cuts in military forces following the Cold War. Those cuts, it could be argued, spread DoD's fixed costs--many of which are funded through the O&M appropriations--over a smaller number of military personnel, which raised O&M costs per person. If that argument is valid, the increase will not be repeated as long as DoD's forces and the number of military personnel are held constant, which CBO's estimate assumes.

Other factors--the effects of management initiatives, possible additional base closures, and other economies, to name a few--might wholly or partially offset any tendency for O&M costs to rise. If future Administrations made DoD's business practices more efficient, reduced maintenance costs, or closed more bases, O&M spending could be cut considerably. Indeed, by DoD's estimates, an additional round of base closures comparable with those in the 1990s could bring annual savings of as much as $3 billion, some of which would come from the O&M title.

Also likely to dampen costs is a shift in the age distribution and pay-grade structure of DoD's civilian workforce. The current workforce, on which CBO based its sustaining-budget estimate, is more senior than the future workforce is likely to be. Thirty-five percent of today's civilian DoD workers are age 51 or older and will be eligible for retirement in the next few years; 75 percent of workers are over 40, and many of them are also likely to retire in the next two decades. If employees at lower pay grades replaced retirees, the O&M funding needed would be less than the amount CBO assumed. In fact, if the average pay grade among DoD's civilian personnel dropped by just two steps (from today's average of grade 9, step 9, to grade 9, step 7), O&M funding for civilian salaries would fall by between $1 billion and $2 billion a year.

Finally, as for military personnel, CBO's decision to use 2001 through 2015 as the period for its estimate affects overall O&M funding in its sustaining budget. Basing the estimate on the 2001-2010 period would yield a total of $106 billion in sustaining funding for the O&M title ($1 billion less than the estimate presented here). In contrast, using the 2001-2020 period would yield an estimate of $108 billion.

Procurement Appropriations

Funding for procurement buys new weapons and other equipment that DoD needs to carry out its missions in peacetime and to prepare for war. The funds cover a wide array of items ranging from aircraft, ships, and missiles to automobiles and air conditioners.

The Congress appropriated $53 billion for defense procurement in 2000, but by CBO's estimate, annual sustaining funding for procurement would total about $90 billion. That figure falls within the range of past experience and is only about 15 percent below the average for the 1980s--a period when DoD was buying large quantities of many systems. (In 2000 dollars, funding for procurement averaged $64 billion in the 1970s, $104 billion in the 1980s, and $59 billion in the 1990s.)

Because major procurement programs overlap, funding for procurement might need to exceed $90 billion in some years, but those years would be offset by years in which the budget could fall below $90 billion. CBO's estimate is not an average; it represents the steady-state amount that would be sufficient over time to replenish the equipment associated with current forces.

CBO's estimate of a sustaining budget for procurement has two parts: an estimate for major weapon systems such as aircraft, ships, and tanks, for which DoD provides detailed procurement plans; and an estimate for all other equipment, for which DoD provides less detail. The latter category includes such items as communications equipment, trucks, and computers.

Procurement of Major Systems. CBO's estimate of the annual funding needed to procure major air, sea, and land systems to sustain DoD's current forces is $40 billion. Of that amount, roughly $5 billion would pay for Army systems, $20 billion for Navy (including Marine Corps) systems, and $15 billion for Air Force systems. CBO based its estimate on the number of systems that DoD requires for its current forces, the projected service lives of those systems, and the costs of replacing them.

Estimating Methods. In calculating sustaining funding for procurement of DoD's major equipment, CBO took into account only the numbers and types of systems to be purchased and not the current procurement budget or the age distribution of the systems in a particular "fleet," or inventory. If DoD purchased all of its systems in the quantities reflected in this estimate, its inventories would eventually be evenly distributed throughout a range of ages--that is, from new deliveries to systems ready for retirement. Retirements from inventories with such a distribution would be steady instead of varying from year to year, as they would if systems were purchased unevenly. Thus, the average fleet age of each system (the average age for all systems of that type) would come to equal half the equipment's service life and would neither increase nor decrease thereafter.

The following example illustrates how CBO arrived at steady-state dollar values for the quantities of major systems it assumed would be procured. The inventory of aircraft that CBO considered in calculating a sustaining-budget estimate for the Air Force totals about 5,100. The expected service lives of those planes stretch from 30 years for fighters to 80 years for bombers. The costs for replacing them range from $10 million to $400 million per plane. To calculate an annual sustaining budget for each type of aircraft, CBO determined how many DoD needed to buy each year (the annual requirement), dividing the inventory by the plane's expected service life and adding an adjustment for expected yearly losses during peacetime. Those annual purchases were then multiplied by CBO's estimate of the unit cost for replacing each aircraft.

Uncertainty and Sensitivity of the Estimate. CBO based its estimate of a sustaining budget for major procurement on a number of assumptions, all of which are uncertain to varying degrees. ⁽⁶⁾ Those assumptions cover projected inventories, estimates of the ages of equipment at the time of replacement, and costs for replacing systems.

For example, CBO's assumptions about annual purchases of systems, which underlie its estimate of sustaining funding, call for replacing systems on a one-for-one basis. But that might not happen in every case. DoD might find that the services could perform their missions with less equipment than they use today--perhaps because of improvements in the equipment's capabilities. The inventory DoD required for comparable capability would then be smaller and annual procurement costs lower than in CBO's estimate. In fact, DoD has reduced the amount of equipment used for certain missions. (For example, the Navy cut the number of fighter and attack aircraft in its carrier-based air wings from 60 to 48.)

Another crucial group of assumptions underlying CBO's estimate of major procurement are those about retirement ages. CBO's estimate assumed that equipment would be retired at ages consistent with DoD's current plans (see the column in Table 4 labeled "Based on Longer Service Lives"). But in the case of many systems, the services have never kept them for as long as they now plan to. If shorter service lives--based generally on historical experience--proved to be more accurate indicators of how long systems actually remained in service, then an estimate using those assumptions would better reflect sustaining quantities for major procurement (see the corresponding column in Table 4). Those quantities are much larger than CBO's estimate of the quantities associated with DoD's plans. Consequently, under those shorter-service-life assumptions, CBO's estimate of a sustaining budget would not be sufficient to maintain today's inventories.
 

In estimating the costs for replacing systems, CBO used the prices that DoD paid for similar units as a base. For some systems, CBO adjusted those prices to reflect expected improvements in technology and growth in costs. However, its assumptions about costs may not be accurate. DoD's analysts, for example, might argue that using historical prices--with or without CBO's adjustments--may overstate costs in the future. The Defense Department and private industry, they might contend, are committed to reducing the cost of procuring major systems--for instance, through so-called lean manufacturing techniques. ⁽⁷⁾ In addition, for some missions, the services are planning to buy systems that could be less expensive than a current-system replacement. For example, DoD is considering replacing some manned reconnaissance systems with lower-price unmanned equipment. CBO's estimate does not reflect such possible changes to the composition of DoD's forces.

In contrast to the problem of overstating costs, some of the prices for weapons and equipment that CBO used for its estimate could be too low. Replacements for many current systems are either early in the development stage or not in development at all. Historically, as systems have moved from early stages into full production, their price has grown. As mentioned earlier, some of CBO's estimates of prices for replacement systems include the likely growth of costs. But the estimates may not be high enough to cover the full price of improved capabilities or further modifications.

Other Procurement. The other part of CBO's estimate for procurement covers equipment and systems for which CBO lacked the data to develop individual cost estimates. ⁽⁸⁾ (Examples include some trucks, communications and civil engineering equipment, and ammunition, as well as programs for modifying existing systems.) CBO's overall estimate of the total costs for all of those systems and programs--almost $50 billion annually--exceeds the total costs of the programs that procure major systems. The Army's share of the estimated funding for other procurement totals more than $10 billion, the Navy's almost $15 billion, the Air Force's more than $20 billion, and the defense agencies' almost $5 billion.

Estimating Methods. CBO estimated a sustaining budget for purchases of equipment other than major systems by using historical data (for the 1974-1998 period) on total spending for procurement. In particular, CBO based its estimate on the relationship between total current spending on procurement and past spending on equipment similar to the replacement items, as well as on the relationship between spending on major systems and spending on other kinds of procurement.

Uncertainty and Sensitivity of the Estimate. CBO's estimate of the funds needed to sustain DoD's other procurement is uncertain for at least two reasons. First, it is based in part on CBO's estimate of the sustaining funding needed to procure major systems; as a result, it is affected by all of the uncertainties associated with that estimate--specifically, changes in the costs, service lives, and required inventory for individual systems. (For example, if CBO used the shorter service lives described in Table 4 in its calculations, its estimate of $90 billion for steady-state procurement would increase by $25 billion.) Second, the estimate for other procurement relies on statistical analysis that is inherently imprecise.

To try to lessen some of those uncertainties, CBO developed alternative methods for calculating a sustaining budget for this category. One approach broke down spending for other procurement into subcategories and developed an estimate for each one using detailed statistical relationships. Like CBO's original approach, that method generated an estimate of about $50 billion for other procurement funding.

Research, Development, Test, and Evaluation Appropriations

In 2000, the Congress appropriated $38 billion for the programs that make up the research, development, test, and evaluation (RDT&E) category of DoD's budget. At $40 billion, CBO's estimate of the RDT&E funding necessary to sustain today's forces is quite close to the appropriations for 2000.

What the Funds Purchase. Appropriations in this title pay for basic and applied research, fabrication of devices for demonstrating new technologies, development and testing of prototypes, and testing of full-scale models of weapon systems before they enter production. Development funds also pay for operational testing of systems, when they are first taken into the field and when they are modified during the course of operations.

Estimating Methods. The RDT&E budget title has four major components. The Army, Navy, and Air Force each have RDT&E appropriations (research and development for the Marine Corps is included in the Navy's appropriation); the fourth component takes in DoD-wide programs such as national and theater missile defense. For the three military departments, CBO based its estimate of sustaining funding on the relationship of each department's RDT&E funding to its total budget. First, CBO calculated the shares of their annual budgets that the Army, Navy, and Air Force devoted to RDT&E between 1974 and 1999, the period CBO used for this estimate. The average share of its budget that each service had allocated over the period was then used to compute CBO's estimate of the service's sustaining budget for RDT&E.

Two important considerations support the assumption that estimates of sustaining budgets for research and development should be tied to total sustaining-budget estimates. First, much of DoD's funding for RDT&E pays for the salaries of engineers and scientists employed under defense contracts. Those costs should rise with costs in the civilian labor market and thus should be tied to CBO's estimates of sustaining funding for O&M and military personnel. Second, because most of the funds for RDT&E go toward developing and testing new systems, an estimate of sustaining funding for RDT&E should be linked to the number and costs of DoD's weapon programs--and thus to the total sustaining budget for procurement.

The fourth component of DoD's RDT&E spending is not tied directly to the services' budgets, since it funds DoD-wide research and development activities. The largest of those are the national and theater missile defense programs. Those programs, which have substantially increased DoD-wide research, were established to develop systems to protect the United States and its forces from attacks by enemy missiles. To estimate a sustaining budget for this component, CBO used the amount appropriated for it in the 2000 budget.

Uncertainty and Sensitivity of the Estimate. CBO's estimate of sustaining funding for RDT&E is highly uncertain. A change of as little as 1 percentage point in CBO's estimate of the RDT&E share of DoD's total budget would lead to a shift of $3 billion in the sustaining budget for RDT&E. And the RDT&E share of DoD's budgets has varied by more than 1 percentage point in the past. Moreover, an estimating approach that is founded on budget shares links CBO's calculation of a sustaining budget for RDT&E to its estimates of sustaining funding for other appropriations. As a result, CBO's estimate for RDT&E is subject to all of the uncertainties that surround those other calculations.

As the foundation for procuring future weapons and equipment, RDT&E activities--and by extension, funding--should be a function of DoD's plans for those items (for example, when they should be bought or the capabilities they should deliver). Indeed, that connection is the basis for CBO's budget-shares approach to estimating a sustaining RDT&E budget. But determining the exact relationship between plans for future weapon systems and the RDT&E funding that would be needed to implement those plans is very difficult. The relationship can vary widely from system to system, depending, for example, on the specific technologies involved and the schedule for introducing a new system. In principle, CBO's approach might balance those effects across the mix of new systems being planned. But CBO's methods are sensitive to shifts in the shares of funding allocated to RDT&E. The historical variations in those shares, combined with the variations that have occurred in funding for DoD-wide research and development activities, point up the considerable uncertainty in CBO's calculation. For instance, CBO's estimate of about $10 billion for DoD-wide research and development programs could either over- or understate the funding needed to sustain those activities. Over the past 10 years, funding for DoD-wide RDT&E has varied from about $9 billion to about $11 billion (in 2000 dollars).

Military Construction Appropriations

In 1997, DoD operated about 1.7 billion square feet of facilities, ranging from office buildings to schools for the dependents of military personnel to facilities on air bases. ⁽⁹⁾ Construction and replacement of those facilities and improvements to them are funded under the military construction title of the defense budget, which also covers many of the costs associated with base closures. In 2000, the Congress appropriated about $5 billion for this category, and CBO's estimate of a sustaining budget for it is about $5 billion as well.

Estimating Methods. CBO based its estimate of a sustaining budget for military construction on what DoD spent on facilities in the 1980s. (During the 1990s, rounds of base closings and realignments--shifts of the activities at a variety of facilities--affected this category of funding and made that period less representative.) CBO's estimate of $5 billion is the average of two calculations that used the 1980s data. In the first calculation, CBO took average spending for military construction per active-duty service member in the 1980s and multiplied that amount by the number of such personnel today. In the second calculation, CBO took average spending in the 1980s per square foot of building space and multiplied it by DoD's current space. CBO's estimate of sustaining funding for military construction makes no provision for the effects of additional rounds of base closures and realignments.

Uncertainty and Sensitivity of the Estimate. As with assumptions made for other categories of appropriations, there is considerable uncertainty about whether relationships for construction spending in the 1980s are the right measure to use in developing CBO's sustaining-budget estimate. Again, as with other estimates, CBO's calculation is quite sensitive to changes in its assumptions. One way to demonstrate that sensitivity would be to compare the two calculations noted above without averaging them together. Using only the approach based on spending per square foot results in an estimate that is about 15 percent greater than CBO's averaged estimate. And using the approach based on spending per active-duty service member produces an estimate that is about 15 percent less than the average.

Family Housing Appropriations

Appropriations for family housing in 2000 totaled about $4 billion, and CBO's estimate of sustaining funding for that budget title is the same. The appropriations finance the costs of constructing, improving, operating, maintaining, and leasing military family housing units.

Estimating Methods. CBO's assumptions about military family housing began with the premise that DoD would continue to maintain its current inventory of housing units. CBO also assumed that operation and maintenance costs for the units would match those for 2000. The new construction that each military department would need each year to sustain its forces was estimated by taking the portion of the inventory that the department owned and dividing it by 50 years--a unit's assumed service life. CBO estimated that a new unit would cost slightly more than $150,000, on average, to construct and that each unit would be revitalized once during its life at an additional cost of roughly $80,000. (Those costs are actual averages from the 1998-1999 period.)

Uncertainty and Sensitivity of the Estimate. In constructing its estimate, CBO assumed that DoD would continue to rely on publicly funded construction to maintain its housing stock. Recently, however, DoD has begun to broaden the private sector's involvement in constructing and maintaining its family housing. If DoD increased that involvement in the future and consequently produced substantial savings, a sustaining budget for family housing might be lower than CBO has estimated. If DoD eliminated any current deficits in family housing, it might need more money than CBO has estimated.
 

CBO's Estimate of a Sustaining Budget for Defense Activities in Other Agencies

In addition to the Department of Defense, other federal agencies conduct programs or activities that contribute to national defense. In 2000, the Congress appropriated approximately $13 billion for those activities, allocating more than $12 billion to the Department of Energy (budget subfunction 053) and $1 billion to other agencies (budget subfunction 054). CBO's estimate of a sustaining budget for the activities covered under both of those categories also totals $13 billion.

What the Funds Purchase

The Department of Energy (DOE) uses its funds to maintain stockpiles of U.S. nuclear weapons and develop technologies to better detect, identify, and respond to the proliferation of nuclear and other weapons of mass destruction. DOE's funding also pays for environmental cleanup of sites that formerly contained facilities for producing nuclear weapons. Funds for other agencies support diverse activities that range from overseas deployments of the Coast Guard and its enforcement of U.N. sanctions to counterintelligence and surveillance activities by the Federal Bureau of Investigation.

Uncertainty and Sensitivity of the Estimate

CBO has not evaluated potential changes to funding for the above activities in any detail, and as a result, its estimate of sustaining funding is particularly uncertain. One factor that could drive up DOE's budget requirements is higher-than-expected costs for cleaning up some facilities. CBO, however, has not performed a detailed analysis of such potential budgetary pressures.
 

The Limitations of CBO's Estimate

As the Congress debates future defense budgets and evaluates proposals for future strategies and missions, CBO's analysis and estimates of a sustaining budget could prove to be useful. But several limitations of CBO's work should be noted.

First, the estimates presented here do not apply to the period for which DoD typically constructs detailed plans. For example, CBO's estimates of sustaining funding for procurement reflect the cost of a number of weapons and systems that either are not yet in production or are in production only at initial low rates. Thus, CBO did not estimate the level of procurement DoD would require in the near term. In fact, as noted earlier, CBO's estimates are not associated with any specific budget year.

Second, because CBO based its estimates on methods that used broad totals as their foundation, the estimates are also broad. As a result, CBO's numbers are not specific enough to be used in a formal cost estimate of the type that CBO prepares for legislative proposals.

Third, and most important, CBO has not evaluated the current national security strategy and the threats it is intended to counter, nor has it assessed whether the forces and the goals for modernization that are now in place adequately support that strategy.

In fact, today's national security strategy has been the subject of considerable debate. Some observers argue that its two-war focus overstates the conventional combat forces that DoD needs to sustain. Others contend that DoD requires larger forces than those in place today because the strategy also emphasizes smaller-scale contingencies and peacetime presence and those operations add to the forces needed for combat. Next year, a new Administration could markedly change the country's national security strategy. And that might significantly change the numbers and kinds of forces DoD would need to maintain and the budget that would be required to build and sustain them. But even if today's strategy continued to form the basis for the defense requirements of a new Administration, the forces that constitute the current U.S. military might not be able to support that strategy. The leaders of the military services and others have suggested that the military is being asked to do more than its current forces can sustain. This analysis does not address that issue but only the question of what funds are necessary to maintain those forces in the future.

CBO's analysis also does not take into account any alterations that DoD and the Congress might make in the military's support services or DoD's infrastructure. Even if plans for strategies, forces, and modernization remained substantially unchanged, such alterations could affect CBO's estimate of sustaining funding. If DoD reduced the number of bases it operated, for example, or improved its business practices to provide services more efficiently, those changes could decrease the sustaining budget for a given level of forces. Conversely, increasing support services or enhancing infrastructure could increase DoD's long-term operating costs. CBO's analysis does not reflect those possibilities.

1. Two other categories, revolving and management funds and trust funds, are intermediate financing mechanisms and are not discussed in detail here.

2. There are three military departments: the Army, the Navy, and the Air Force. The Marine Corps falls under the Department of the Navy. In addition, DoD includes a number of separate support agencies such as the Defense Intelligence Agency and the Defense Logistics Agency. Funding for defense agencies comes from multiple appropriation titles. CBO's forthcoming study, Budgeting for Naval Forces: Structuring Tomorrow's Navy at Today's Funding Level, addresses funding for the Department of the Navy in more detail.

3. One measure of the "quality" of military forces is the percentage of high school graduates among the services' recruits.

4. See Budget of the United States Government, Fiscal Year 2001: Appendix.

5. "Department of Defense Press Briefing on the Fiscal Year 2001 Budget Submission," February 7, 2000. Despite the potential for growth in per capita spending, the budgets DoD has planned through 2006 do not reflect such increases.

6. For an alternative estimate based on markedly different assumptions, see Daniel Gouré and Jeffrey M. Ranney, Averting the Defense Train Wreck in the New Millennium (Washington, D.C.: Center for Strategic and International Studies and Management Support Technology, Inc., November 1999). Gouré and Ranney's estimate is substantially higher than CBO's. For an estimate of DoD's budget needs that is lower than CBO's, see Steven Kosiak and Elizabeth Heeter, Cost of Defense Plan Could Exceed Available Funding by $26 Billion a Year Over Long Run, CSBA Highlight (Washington, D.C.: Center for Strategic and Budgetary Assessment, April 2, 1998).

7. Those techniques include delivering parts to the production line just before they are needed, thereby reducing storage costs; using computers more efficiently in the design and manufacturing process; and using more generic equipment to assemble systems, which reduces spending on specialized tooling.

8. CBO's category should not be confused with accounts labeled "Other Procurement" in the budgets of the Army, Navy, and Air Force. Although CBO's category includes funds from those accounts, it also includes money for items that are funded through many other accounts.

9. The figure of 1.7 billion square feet represents buildings at facilities for both active and reserve components of the services. The last year for which complete data were available from DoD was 1997.

---